Factors Associated With Medication Adherence Among Tuberculosis Patients in Timor Leste

FACTORS ASSOCIATED WITH MEDICATION
ADHERENCE AMONG TUBERCULOSIS

PATIENTS IN TIMOR-LESTE
Juliao dos Reis

BHlthSc, GradDipEnvH, MPH
A thesis submitted in fulfilment of the requirements for the degree of

Doctor of Philosophy
School of Public Health and Social Work

Institute of Health and Biomedical Innovation
Queensland University of Technology
2016
ii
factors associated with medication adherence among tuberculosis patients in timor-leste
Abstract
Tuberculosis (TB) is an infectious disease that damages lungs and other

parts of the human body, causing serious illness and death. Timor-Leste has
one of the highest rates of TB in Southeast Asia. The estimated prevalence
of TB was 758 per 100,000 population, and more than 800 people died as a
result of TB in Timor-Leste in 2012. The country has undergone major
conflicts as a result of colonial influences, most notably the conflict in 1999,
which resulted in the loss of many lives, major damage to health
infrastructure, and a shortage of public utilities. The most recent civil unrest in
2006 also resulted in major displacement of the locals, forcing a large
number of people into poorly set-up, temporary camps where they
experienced overcrowding. According to the United Nations Human
Development Index, 2014, Timor-Leste continues to be one of the poorest
nations in the world. Due to the impoverished living conditions, including poor
housing, high levels of malnutrition, low individual earnings and unstable
employment, it is difficult for the country to fight against TB. Despite these
challenges, the local government continues to work collaboratively with
international and local non-government agencies, and private and churchbased clinics in the expansion of the Directly Observed Treatment Shortcourse (DOTS), the World Health Organizations (WHO) internationally
recognized strategy to fight against TB. However, to date, the effectiveness
of the program is yet to be determined and health professionals have little
reliable data on the levels of treatment adherence and the barriers to or
enabling factors for adherence. One study has qualitatively determined
factors affecting implementation of the DOTS strategy, while another RCT
study has investigated nutrition as a factor in promoting adherence in Timor
Leste. Yet, no study has quantitatively assessed TB treatment adherence. In
order to fill this gap, this study used a mixed-methods approach to determine
levels of TB treatment non-adherence and their relationships to various
iii
factors, including but not limited to service delivery, social & demographic
and individual and disease related factors.
The study involved two main parts and the first part involved qualitative,
semi-structured, face-to-face interviews of ten health professionals and ten
TB patients from the private and public sectors. The health professionals
interviewed included a national TB program manager, a clinic doctor, a TB
program adviser, district TB coordinators, TB program assistants and
international and local non-government TB program implementers. The
patients were conveniently selected from a private clinic and a public health
centre through the TB registers as they came to visit their respective health
service for consultations and medication refills. The selection was based on
the subjective assessment of non-adherence of individual patients (i.e.
missed appointment), as observed by the clinic staff throughout the treatment
period. The second part of the study involved a cross-sectional survey of 347
TB patients in six community health centres and three specialist clinics with a
response rate of 97% (n = 347/359).
The cultural appropriateness of the qualitative and quantitative instruments

was carefully considered in the design and implementation process. The
survey instrument was developed through the use of a systematic procedure
and the repetitive test of the tool was conducted to ensure its suitability.
Furthermore, to improve the validity and reliability, the tools were reviewed by
experts with national and international disciplinary background and
experiences.
The qualitative study revealed themes that were common to health

professionals and patients in relation to treatment adherence. Health
professionals cited many factors associated with medication non-adherence,
including lack of access, distance between patients residences and health
services, poor nutritional status, and lack of treatment knowledge. Patients
reported that a lack of financial support, limited access to transport, aging,
and low energy levels affected treatment adherence. Health professionals
iv
and patients suggested that non-adherence to TB treatment was relatively

common among people with TB in Timor-Leste.
The quantitative survey used two measures of non-adherence, with a cut-off

of <80% regimen intake classified as suboptimal adherence and 80% as
optimal adherence. A Visual Analogue Scale (VAS) estimated the prevalence
of the past month of non-adherence to be 39.6% and the Medication
Adherence Rating Scale (MARS), which measures patients adherence
behaviours in the past month, found non-adherence rates as low as 25.3%.
The observed agreement between the two measures was 69.4% with a
kappa coefficient of 0.323 (SE = 0.052, p < 0.0001).
Analysis of the interviews and administered surveys with the 347 patients
found that medication non-adherence was significantly associated with low
wages or lack of family financial support, alcohol use, untreated depression,
stigma, use of traditional healers for health care, lack of TB knowledge, lack
of caregivers support in taking medication, and chance health locus of
control. No association was observed between TB medication adherence and
gender, age, education and marital status, disclosure of treatment status,
distance between clinic and patients residences, cigarette smoking, patienthealth professional communication and social connectedness with family,
friends, neighbours and health volunteers.
To date, this is one of the largest studies in Southeast Asia to examine TB

medication non-adherence and its associated factors. The evidence
suggested practical measures that program planners and service providers
can implement to improve the outcome of TB treatment. The results
suggested that clinics and outreach services should integrate professional
counselling for patients with depressive symptoms and heavy alcohol use.
Health education should target common but potentially harmful beliefs that
chance or luck determines health outcomes.
Further, the results also showed that there is a strong need to collaborate
with traditional healers to work cooperatively in assisting patients with TB
treatment. In addition, incentives should be provided to those who struggle

financially. Despite the lack of significant correlations between social
supports and adherence in the findings, the data showed that patients who
received sufficient supports were more likely to have better mental health and
this may lead to better treatment adherence than those who received no
support. Adherence may be maximised by improving patients psychological
well-being. Patients mental health and overall medication adherence may be
improved by assisting patients to maintain social connectedness with friends,
family, health professionals and people in the community. The findings of this
study also highlighted the benefits of using a mixed-method approach in
examining complex barriers and facilitating factors for medication adherence.
vi
Keywords
Adherence, adherence knowledge, medication adherence, social stigma, traditional

healer, Timor-Leste, tuberculosis
vii
viii
Table of Contents
Abstract ...................................................................................................................................................... iii

Keywords ..................................................................................................................................................vii
Table of Contents ................................................................................................................................... ix
List of Figures ........................................................................................................................................xiii
List of Tables ..........................................................................................................................................xiv
List of Abbreviations ............................................................................................................................xvi
Statement of Original Authorship ................................................................................................ xviii
Acknowledgements.............................................................................................................................. xix
Chapter 1: Introduction.......................................................................................... 1
1.1
Background ....................................................................................................................................1
1.2
Aim and Research Questions .................................................................................................5
1.3
Significance of the Study ..........................................................................................................5
1.4
Structure of the Thesis ..............................................................................................................6
Chapter 2: Literature Review ............................................................................... 9

2.1 Tuberculosis Treatment Adherence ...................................................................................... 11
2.1.1 Drug Side-Effects and Multi-Drug Resistant Tuberculosis ........................... 16
2.1.2 Recommended TB treatment regimens ............................................................... 17
2.1.3 Definition of TB treatment completion................................................................... 17
2.1.4 Treatment Interruptions .............................................................................................. 18
2.2 Determinants on Sub-optimal Adherence ........................................................................... 19
2.2.1 Health Service Delivery Factors .............................................................................. 22
2.2.2 Socio-demographic & Economic Factors ............................................................ 24
2.2.3 Individual Factors Related to Adherence ............................................................ 35
2.3 Summary of Research into TB Medication Adherence ................................................. 41
2.4 Tuberculosis Program and Research on Treatment in Timor-Leste ........................ 42
2.5 Framework ....................................................................................................................................... 47
ix
Chapter 3: Methodology ...................................................................................... 53

3.1 Study Design .................................................................................................................................. 53
3.2 Study Sites ...................................................................................................................................... 57
3.3. Qualitative Study .......................................................................................................................... 58
3.3.1 Sample Size .................................................................................................................... 59
3.3.2 Data Collection ............................................................................................................... 60
3.4 Survey Development ................................................................................................................... 60
3.4.1 Dependent Variables ................................................................................................... 62
3.4.2 Independent Variables ................................................................................................ 64
3.4.3 Translation ....................................................................................................................... 68
3.4.4 Pilot Test ........................................................................................................................... 70
3.4.5 Integration of Findings on Composition of the Study Questionniare ........ 72
3.5 Cross-sectional Survey .............................................................................................................. 72
3.5.1 Study Population ........................................................................................................... 72
3.5.2 Method of Sampling ..................................................................................................... 73
3.5.3 Statistical Power Analyisis ......................................................................................... 74
3.5.4 Procedure ......................................................................................................................... 75
3.6 Data Analysis and Management............................................................................................. 82
3.6.1 Exploratory Qualitative Data Analysis ................................................................... 82
3.6.2 Quantitative Content Analysis .................................................................................. 82
3.7 Ethical Considerations ................................................................................................................ 84
Chapter 4: Qualitative Study Results ............................................................... 87

4.1 Demographic Characteristics ................................................................................................... 87
4.1.1 Health Professionals .................................................................................................... 87
4.1.2 Tuberculosis Patients .................................................................................................. 87
4.2 Framework of Qualitative Data Analysis ............................................................................. 88
4.3 How TB Patients Adhered to Treatment.............................................................................. 89
4.4 Factors Influencing TB Treatment Adherence .................................................................. 90
4.4.1 Severity of Illness .......................................................................................................... 90
4.4.2 Drug Side-effects........................................................................................................... 91
4.4.3 Psychological Distress ................................................................................................ 92
4.4.4 Traditional Healers........................................................................................................ 93
4.4.5 Education & Knowledge ............................................................................................. 93
x
4.4.6 Cigarette Smoking and Alcohol Consumption ................................................... 95

4.4.7 Cost and Treatment Accessiblity ............................................................................ 96
4.4.8 Patient-Carer Communication.................................................................................. 97
4.4.9 Family, Social Support & Treatment Confidentiality ........................................ 98
4.5 Discussion ....................................................................................................................................... 98
Chapter 5: Main Survey Results ...................................................................... 103

5.1 Characteristics of study participants ...................................................................................103
5.1.1 Demographic Variables ............................................................................................104
5.1.2 Family and Social Characteristics ........................................................................105
5.1.3 Alcohol, Tobacco and Illicit drug-use ..................................................................106
5.1.4 Health Professional and Patient Communications ........................................108
5.1.5 Characteristics of Participants' Psychological Distress ...............................109
5.1.6 Descriptive Statistics for Health Locus of Control ..........................................110
5.1.7 Descriptive Statistics for Traditional Healer ......................................................111
5.1.8 Tuberculosis knowledge ...........................................................................................111
5.1.9 Tuberculosis Stigma ..................................................................................................112
5.1.10 Social and Family Supports..................................................................................114
5.2 Subjective Non-adherent Factors Listed by Patients ...................................................116
5.3 Level of Adherence (two measures) ...................................................................................117
5.3.1 Visual Analogue Scale ..............................................................................................117
5.3.2 Medication Adherence Rating Scale ...................................................................118
5.3.3 Comparison between the Two Measurements................................................119
5.4 Univariate Logisctic Regression Analysis .........................................................................120
5.5 Univariate Analysis of Participants Perception of Stigma .........................................123
5.6 Multivariate Analysis of Factors Associated with Medication Adherence ............125
Chapter 6: Discussion ....................................................................................... 129

6.1 Factors Associated with TB Medication Adherence .....................................................130
6.1.1 Health Service Delivery Factors ............................................................................132
6.1.2 Social and Economic Factors.................................................................................134
6.1.3 Individual and Disease-Related Factors ............................................................136
6.2 Limitations of the Study ............................................................................................................141
6.3 Significance of the Study .........................................................................................................143
6.4 Public Health Implications .......................................................................................................144
xi
References ............................................................................................................ 149

Appendices ........................................................................................................... 171
xii
List of Figures
Figure 1. Literature mapping of TB treatment adherence................................................. 21

Figure 2 Survey Conceptual Framework ................................................................................. 50
Figure 3 Study Flow Diagram ....................................................................................................... 56
Figure 4. Translation Process ...................................................................................................... 69
Figure 5. VAS Distribution............................................................................................................ 118
Figure 6. MARS Distribution ....................................................................................................... 119
Figure 7 Map of Timor Leste ....................................................................................................... 185
Figure 8 Management of TB Treatment Interruption......................................................... 213
xiii
List of Tables
Table 1 Evidence-Based Rating System ................................................................................. 10

Table 2 WHO TB Treatment Category ..................................................................................... 13
Table 3 WHO TB treatment Outcome and Definition.......................................................... 14
Table 4 Summary of Literature Review on TB Treatment Adherence ........................ 45
Table 5 Matrix for Sampling in Qualitative Research.......................................................... 59
Table 6 Various Methods of Measuring Medication Adherence..................................... 61
Table 7 Population size, density and percentage of TB in the selected sites ........... 73
Table 8 Data Collection Sites, TB Cases and Participant Numbers for the
Study Districts ........................................................................................................... 75
Table 9 Percentages and Frequencies for Demographic characteristics ................. 104
Table 10 Percentages and Frequencies for Social and Family
Characteristics ........................................................................................................ 106
Table 11 Descriptive Statistics for Risk Behaviour and Clinical
Characteristics ........................................................................................................ 107
Table 12 Chi-Square P-Values for Risk Behavioural Characteristics by
Gender ....................................................................................................................... 107
Table 13 Means and Standard Deviations for Health Professional-Patient
Communication Item, Total and Overall Satisfaction Scores .............. 108
Table 14 Means and Standard Deviations for Psychological Distress Item
and Total Scores.................................................................................................... 109
Table 15 Means and Standard Deviations for Chance Health Locus Control
Item and Total Scores ......................................................................................... 110
Table 16 Frequencies, Means and Standard Deviations for Traditional
Healer Item and Total Scores ........................................................................... 111
Table 17 Frequencies and Percentages for Knowledge of TB Patients ................... 112
Table 18 Patients Perceptions of stigma .............................................................................. 113
Table 19 Family and Social Support, Mean and Standard Deviation ........................ 114
Table 20 Percentage and Frequency of Non-adherent Factors Listed by
Patients ...................................................................................................................... 116
xiv
Table 21 Agreement between the VAS and MARS on TB medication

adherence................................................................................................................. 119
Table 22 Univariate Analysis of Factors Associated with VAS Adherence.............. 121
Table 23 Univariate Analysis of Factors Associated with VAS Adherence
cont. ............................................................................................................................ 122
Table 24 Univariate Analysis of VAS with Perceive Stigma........................................... 123
Table 25 Multivariate Analysis of Factors Associated with VAS Adherence
Cont. ........................................................................................................................... 127
Table 26 Multivariate Analysis of Factors Associated with VAS Adherence
Cont. ........................................................................................................................... 128
Table 27 Summary of Worldwide Qualitative Studies on TB treatment ................... 172
Table 28 Summary of Quantitative Studies on TB............................................................. 176
Table 29 In-depth Univariate Analysis .................................................................................... 215
Table 30 Health Professional Interview Transcripts .......................................................... 233
xv
List of Abbreviations
AIDS
Acquired Immune Deficiency Syndrome
BCG
Bacille CalmetteGurin
CHC
Community Health Center
DHS
District Health Service
DOT
Directly Observed Treatment
DOTS
Directly Observed Treatment Short-course
DTA
District Tuberculosis Assistance
DTC
District Tuberculosis Coordinator
EMB
Ethambutol
FDC
Fixed Dose Combination
HIV
Human Immunodeficiency Virus
INGO
International Nongovernmental organization
INH
Isoniazid
KAP
Knowledge Attitude and Practice
LNGO
Local Non-government Organization
MARS
Medication Adherence Rating Scale
MDR-TB
Multidrug-resistant tuberculosis
MoH
Ministry of Health
NGO
Nongovernmental Organization
NSP
New Smear Positive
NTP
National Tuberculosis Program

xvi
PAF
Population Attributable Fraction
PLHIV
People Living with HIV
PZA
Pyrazinamide
QUT
RCT
Randomized Control Trial
RIF
Rifampin
SAT
Self-administered Therapy
SD
Standard Deviation
SM
Streptomycin
TB
Tuberculosis
VAS
Visual Analogue Scale
WHO
World Health Organization
xvii
Statement of Original Authorship
The work contained in this thesis has not been previously submitted to meet
requirements for an award at this or any other higher education institution. To
the best of my knowledge and belief, the thesis contains no material
previously published or written by another person except where due
reference is made.
QUT Verified Signature
Juliao dos Reis

Date: March 23rd 2016
xviii
Acknowledgements
I would like to express my sincere appreciation to everyone who helped me

throughout my study to complete this piece of work successfully at the
Queensland University of Technology (QUT). All this would not have been
possible without your support, and for that I thank you.
First and foremost, I would like to start with my supervisors. Special thanks
and appreciation to my principal supervisor Associate Professor Xiang-Yu
Hou (Janet) for everything she has done. I have been taught by her since I
was an undergraduate, and I am very thankful for her support throughout my
years of studying at QUT. When I started my PhD, I was ambitious and
hoped to solve all the problems in the Timor-Lestes health system and I was
not entirely sure about the specific problem that I should investigate.
However, after a number of discussions with her, she was able to guide me
to choose the right topic, which suited my interest. I wish to thank her for her
dedication and support day in and day out, looking at my work and refining
my ideas, providing critical input and editing my thesis.
From the bottom of my heart, I would also like to extend my deepest gratitude
to my Associate Supervisor, Professor Michael Dunne. I have been inspired
by his work and the way he motivates his students, and I am thankful for his
supervision. I thank him for his tireless support, whether through constructive
feedback, critical input, thesis editing and writing of the application to seek
funding support for data collection and so much more.
Thank you to Associate Professor Mark Brough for the valuable feedback on
my thesis. Thank you for your time and support during my final seminar. I
would also like to thank Dr Joseph Debattista who has taken part as the
external reviewer for my confirmation and final seminar. He helped me with
editing my thesis and helped to link me with an external organization for
funding opportunities. My sincere gratitude goes to Associate Professor
Ignacio Correa-Velez who was one of the panel members for my confirmation
xix
seminar. Ignacio has also been very helpful throughout my PhD journey as
he was able to provide feedback on my document and guide me with
statistical analysis of my research.
I would also like to acknowledge the QUT Faculty of Health for facilitating the
journey to complete my undergraduate, postgraduate, master and PhD
degrees. Special thanks to the Health Research Services and Language
Support Team for their exceptional work; Emma Kirkland, Jessica Harriden,
Kerry Fesuk, and Dr Martin Reese. Thank you to Professor Mary-Lou
Fleming for her support throughout my years of studying at QUT. PhD
colleagues; Khalid Aboalshamat, Nidup Dorji & Paraniala Silas Celebi Lui
who gave statistical advice, ideas and identified typographical errors in my
work. I also want to acknowledge professional editor, Dr Bill Wrigley, who
provided copyediting and proofreading services, according to the guidelines
laid out in the university-endorsed guidelines and the Australian Standard for
editing research theses.
Special thanks to the Endeavour Awards Scholarship of the Australian
Government. I am honoured as the first East Timorese to be selected for a
merit-based competitive scholarship. I thank the staff of Global Fund Division
in the Ministry of Health of Timor-Leste, national tuberculosis program staff
from sub-district to national levels, questionnaire administrators and the staff
of Bairopite, Motael, ISMAIK and Klibur Domin Tibar, Catholic Relief Services
for their support throughout my study. I would like to thank also Bonifacio
Barreto, Oldegar Massingga and Joao dos Reis for the questionnaire
translation. My sincere gratitude to the Terrace Timor Network (TTN) who
supported me with data collection overseas.
I would like to express my sincere gratitude to Wes Meurant, Stuart Delaney

and everyone at St Peters Lutheran College (SPLC) boarding for a stable
working environment.
My deepest appreciation to my family and friends for the support they have
shown throughout my study journey. Thanks to Geraldine Horan and Paul
xx
Murray, whom I have met during our undergraduate studies in 2005 at QUT.
Id like to thank the Silvas family; Chan, Adriana, Johann, Sheran, Jivan and
Jordan for all their support. My appreciation to Doug Martin and Marion
Crawford for taking me in as their adopted son and their endless support.
Sally Bannah, an important person in my whole journey from Timor to
Australia and I thank her for all she has done. I would also like to thank my
friend Mark Andrews for his support during my PhD study. It is difficult to put
into words but I hope you all understand how appreciative I really am for all
the support you provided.
My parents could not understand why I left a fulltime job to continue studying,
but they are proud of what I have accomplished. My parents have taught me
many important lessons about life, giving without questioning and loving me
unconditionally. My father used to say If you work like a king, you will live life
like a slave, but if you work like a slave you will enjoy life more like a king.
By which he meant if you work hard you will succeed, but if you live life like
you are entitled to privileges and put in no effort, you will end up struggling in
life. This is one of many ways he encouraged me to work hard regardless
how small the job.
To my amazing fianc Nela, thank you with all my heart for your tireless
support and encouragement. Even though our families had doubts about our
long distance relationship, we have shown determination, persistence and
trust in each other over the long period of my study. I am deeply thankful for
having you in my life. Thank you to my brother and my sister, Uncle Joao,
Aunty Veronica, Aunty Rosa and all my cousins and many other family
members for all their support.
Finally, this thesis is dedicated to Apa Boot Duarte da Costa Barros who died
during my study. He was a freedom fighter, a father, a friend and a teacher. I
wish he was able to see what I have achieved, but I am thankful for his
influence in my life by teaching me how to be humble and be respectful of
others.
xxi
Chapter 1: Introduction
1.1 Background
Tuberculosis (TB) is an infectious disease that leads to death, morbidity and
disability. It is ranked the second most fatal communicable disease
worldwide after HIV/AIDS (Armijos, Weigel, Qincha, & Ulloa, 2008; Ayisi et
al., 2011; Clark, Karagoz, Apikoglu-Rabus, & Izzettin, 2007; Naidoo &
Mwaba, 2010). There are approximately 9 million new cases of TB per year
(Stop TB Partnership, 2015), and TB itself contributes to 2.8% of the global
disability burden (Armijos et al., 2008). Over 95% of TB deaths occur in low
and middle-income countries and approximately 10 million children
worldwide have been orphaned due to death of a parent from TB (Stop TB
Partnership, 2015; WHO, 2014). In 2013 alone, 1.5 million people died from
the disease, and an estimated 550,000 children became ill, and of these,
80,000 HIV-negative children died of TB (WHO, 2014b). Southeast Asia and
the Western Pacific Regions collectively accounted for 56% of the cases in
2013 (WHO, 2014b).
In an effort to eliminate the TB disease worldwide, many multilateral

agencies provide much-needed funding every year. However, it is estimated
that there is a funding gap of $1.6 billion per year from the Global Fund of
the $4.8 billion total resources needed each year from 2014-2016. This
funding is needed to combat TB in 118 countries eligible for financial support
from the Global Fund (WHO, 2013b).
Historically, TB had been known as a prime example of a social disease,
the control of which requires social, economic and environmental
interventions (Lnnroth, Jaramillo, Williams, Dye, & Raviglione, 2009). After
World War II, a medically oriented TB control model emerged (Berridge,
2005), and much hope was placed on mass vaccination with Bacille
CalmetteGurin BCG (Brimnes, 2007). Effective chemotherapy became
available for TB treatment at the end of 1940s. Availability of the
chemotherapy resulted in switching the TB control model to a mostly

curative focus. It was predicted that effective TB treatment would result in
the rapid decline of TB incidence (Hinman, Judd, Kolnik, & Daitch, 1976).
Yet, recent observations have indicated that the effectiveness of the current
TB control strategy has been lower than expected (Behera, 2008; Lienhardt
et al., 2012; WHO, 2010). The ongoing quest for a better TB vaccine and
enhanced chemotherapy for preventive treatment should improve TB
prevention (Lnnroth et al., 2009). Furthermore, actions to address the
social determinants of TB as well as behavioral and psychological risk
factors associated the disease are required. Many factors directly and
indirectly influence the risk of exposure to TB, the risk of acquiring TB
infection and the likelihood of progression to active TB (Lnnroth et al.,
2009). Effective control programs require high levels of adherence to TB
medication, yet this goal is difficult to achieve.
Researchers have identified numerous factors associated with medication

non-adherence, including the psychological challenges faced as a
consequence of having TB (Ailinger, Martyn, Lasus, & Lima Garcia, 2010;
Armijos et al., 2008; Ayisi et al., 2011; Naidoo & Mwaba, 2010), the quality
of health care received, and the use of traditional healing systems, such as
herbal or animal derivatives (Naidoo & Mwaba, 2010). Psychosocial factors,
including feelings of helplessness, depression, and a lack of appropriate
social support were found to act as barriers to adherence to the Directly
Observed Treatment Short-course (DOTS) program (Naidoo & Mwaba,
2010).
Factors associated with TB medication adherence can be grouped in in

three ways. First, the health service-related factors which includes the
quality and availability of services, complexity of regimens and management
of medication side-effects, and patient-provider relationships. Second, social
connectedness (trust, confidentiality, disclosure, type and source of support
and relationships). Third, socio-demographic characteristics, personal
beliefs, mental health, and risk behaviours such as drinking alcohol and
tobacco smoking (Peltzer, 2014). These factors can directly or indirectly
2
influence treatment adherence. Therefore, understanding the characteristics

of each factor is necessary for the design of effective strategies that can
optimize adherence.
Most studies on TB medication adherence have been conducted in the

United States, Africa and Asian countries, and there is considerable
variation in the estimates of adherence found among these studies (Ailinger,
Black, Nguyen, & Lasus, 2007; Eticha & EKassa, 2014; Kaona, Tuba,
Siziya, & Sikaona, 2004; Nackers et al., 2012; Weiguo et al., 2009). The
most common approach used by these studies to gather information was
self-reported questionnaire in which participants described their adherence
over a specified time interval (Ahmad, Mahendradhata, Utarini, & de Vlas,
2011; Ailinger et al., 2007; Eticha & EKassa, 2014; Kaona et al., 2004; Kipp
et al., 2011a; O'Donnell, Wolf, Werner, Horsburgh, & Padayatchi, 2014;
Weiguo et al., 2009). Although, the recommended threshold for TB
adherence has been determined, those literature reviews identified (see
Table 4) had used a cut-off point that varies. However, the use of different
thresholds has led to potential under- or over-estimation of the actual
adherence.
The current study was conducted in Timor Leste, a small nation located in
Southeast Asia, with an estimated total population of 1.28 million in January,
2015 and a population density of 86.3 per square kilometer as of April,
2015. According to World Bank data, in 2015 almost 53% of people lived on
less than US$1.25 per day. In developing nations such as Timor-Leste
where poverty is widespread, people are facing high incidences of infectious
diseases, such as TB, malaria, measles, and many others. The country has
high rates of illiteracy, low health literacy and low community TB awareness,
and relatively poor use of health facilities. It also has mountainous terrain,
long distances between residences and health centres, poor road conditions
and undeveloped health systems, especially in rural areas (Ministry of
Health Timor Leste, 2010).
Timor-Leste remains one of the least developed nations in Asia, ranking 128
out of 179 countries on the Human Development Index (HDI) in 2014
(United Nation Development Programme, 2014). In 2005, the government
established the National TB Program (NTP), which is now implementing
grants from the Global Fund, to strengthen government TB services,
including national-level institutions, six referral hospitals, and sub-district
community health centres (CHCs), of which there are 65 across the country,
183 health posts and 162 mobile clinics. The CHCs, health posts and mobile
clinics provide primary care while the referral hospitals provide mainly
secondary and tertiary care. The total health expenditure was US$96 per
capita (WHO, 2013a). In addition, the MoH in Timor-Leste is supported by
bilateral aid, multilateral agencies, such as the United Nations and the
Global Fund through financial, and technical assistance for various health
priority programs. The WHO's DOTS approach is the cornerstone of the
NTP strategy and is theoretically available at all CHCs. However, adherence
to the DOTS strategy in practice is variable across health services in the
country.
Despite the growing evidence on TB treatment in resource-rich and

resource-poor settings, information about adherence to TB treatment in
Timor-Leste is lacking. Through funding support from many different
international donors, Timor-Leste has expanded its TB program across
health services in an attempt to improve treatment services. However,
despite this expansion, program staff still have inadequate data on the
levels of TB adherence and its associated factors to help them optimise
treatment efficiencies. Thus far, in Timor Leste, there has been only one
randomised control trial (RCT) that investigated the impact of nutritional
support on the overall outcome of TB treatment (Martins, Morris, & Kelly,
2009), one qualitative study on factors associated with successful
implementation of DOTS in Timor-Leste (Martins, Grace, & Kelly, 2008) and
a knowledge, attitude and practice survey of households with TB patients
(Ministry of Health Timor Leste, 2011). To date, no study has quantitatively
investigated factors associated with TB treatment. Hence, the current study
will use a mixed-methods approach to investigate barriers and enabling
4
factors, involving health service delivery factors, education, financial,

personal, social, psychological, and disease related-factors associated with
medication adherence among TB patients in Timor Leste.
1.2 Aim and Research Questions

Aim
The overall aim of this study was to identify factors affecting medication adherence
among TB patients in Timor-Leste, including health service delivery, education,
financial, personal, social, psychological, and disease-related factors.
Research Questions
1. To what extent do tuberculosis patients in Timor-Leste adhere to their

treatment regimen?
2. Which factors in service delivery, social and financial and/or individual
disease-related aspects influence TB medication adherence in Timor
Leste?
3. What are the practical implications for future program interventions to
improve tuberculosis medication adherence?
1.3 Significance of the Study

To date, health professionals and TB program staff in Timor-Leste have limited
evidence on TB treatment adherence and barriers or enabling factors to
treatment optimization. The current study contributes to the field of TB care
and treatment in Timor-Leste in a number of ways. First, the results of the
study are likely to inform the design of a comprehensive approach to effective
interventions, which is necessary to practically measure a level of medication
adherence and explore barriers and enabling factors associated with treatment
adherence (WHO, 2008a).
This study combines qualitative exploratory methods and a quantitative survey

to learn medication adherence practices from TB patients and health
professionals from public and private sectors. The study aims to provide indepth evidence on the cultural, contextual and structural constraints and
enabling factors on TB care programs. As Timor-Leste is striving towards the
elimination of TB, the study is likely to provide a comprehensive guide to policy

makers and key stakeholders to develop appropriate and effective
interventions, in which adherence to treatment regimens can be improved.
1.4Structure of the Thesis

Chapter 2 reviews the international literature related to TB medication
adherence. It begins with an overview of the TB program and health system in
Timor-Leste and reviews the guidelines for TB treatment, treatment
classification, definition of treatment completion, and management of
treatment interruptions. This is followed by a review of the determinants of
adherence, including aspects of health services, socio-economic and
demographic factors and individual disease-related factors. The next section
provides a summary of quantitative studies on adherence. Chapter 2
concludes with an introduction to the main research aims and conceptual
framework of the study.
Chapter 3 describes the design, methods and measurement instruments

employed in this study. It begins by describing the study design, the research
process, study sites and population, sample size and data collection methods.
This is followed by a description of the survey instrument development and
testing, data analysis management, and ethical considerations.
Chapter 4 describes the results from the qualitative study. The chapter starts by
describing the treatment adherence of TB patients and factors influencing their
adherent behaviours. The main purpose of conducting the qualitative study is to
gain a better understanding of TB treatment from patients and health
professionals points of view. At the same time, the qualitative study also
enabled the researcher to identify potential factors that may not have been
foreseen in the literature to be included in the final instrument. Finally, Chapter
4 provides the insights of health professionals and patients on treatment
adherence, barriers and enabling factors on TB treatment.
Chapter 5 reports the results from the quantitative survey of 347 TB patients.
The chapter begins by describing the characteristics of the study sample. This
is followed by the estimates of non-adherence levels for TB treatment among
the sample, as measured by the Visual Analogue Scale (VAS) and Medication
Adherence Rating Scale (MARS). Finally, the chapter presents the results of the
univariate and multivariate statistical analyses of various factors associated with
the VAS adherence.
Chapter 6 triangulates findings from the cross-sectional survey, and considers

these findings in relation to previous research. The chapter then discusses the
strengths and limitations of this study. Finally, the chapter includes discussion of
the future research implication and recommendations for public health practice.
Chapter 2: Literature Review

INTRODUCTION
To undertake a literature review of studies of TB medication adherence, online

search engines and electronic databases such as EBSCOhost, ProQuest,
Science Direct, Web of Science, informit and Google Scholar and publications
by both government and non-government organizations were utilized. Keywords
included tuberculosis AND medication adherence, tuberculosis AND patient
compliance, tuberculosis AND social/economic factors and tuberculosis AND
treatment AND adherence OR compliance.
As defined by many authors and international organizations, the term

adherence (also compliance or concordance) describes the degree to which
patients take medications correctly as prescribed by their health care providers
(Bell, Airaksinen, Lyles, Chen, & Aslani, 2007; Dick, Jaramillo, Maher, &
Volmink, 2003; Elliott RA, 2009; Haynes, 1979; National Institute for Health and
Clinical Excellence, 2008; Osterberg & Blaschke, 2005; Tilson, 2004).
The word, adherence, is preferred by many health care providers, as
compliance suggests that patients passively follow doctors orders rather than
the treatment plan being based on a therapeutic alliance established between
the patient and the physician. The term, concordance, is not synonymous with
either compliance or adherence (Bell et al., 2007). Concordance does not
refer to the patients behaviour of taking medications, but the nature of patienthealth care providers interaction. The word, "compliance", is based on the
notion that the consultations between the prescriber and the patient are equally
negotiated, and they both agree on the regimen that the patient will take
(Aronson, 2007). Major barriers to treatment adherence have been identified by
researchers as health system-related, contextual and patient-related factors
(Ailinger et al., 2010; Armijos et al., 2008; Ayisi et al., 2011; Clark et al., 2007;
Gough & Kaufman, 2011; Naidoo & Mwaba, 2010; Qian, Smith, Tuohong,
Shenglan, & Garner, 2011). Worldwide qualitative studies on these sub-themes

are summarised in Appendix A.
It is strongly recommended that patient-centred care be the initial management

strategy. This strategy should include an adherence plan that emphasizes
DOTS, in which patients are under direct observation during the ingestion of
anti-tuberculosis medication. Programs utilizing DOTS as a central element in a
comprehensive, patient-centred approach maximize the likelihood of treatment
completion (Bock et al., 2002; Horsburgh, Feldman, & Ridzon, 2000; Peloquin,
2003; Zierski, 1976).
In order to make informed treatment decisions, an evidence-based rating

system was developed by McGowan, Chesney, Crossley, and LaForce (1992).
The rating recommendation assigns letters indicating the strength of the
recommendations, and a Roman numerals indicating the quality of the evidence
(see Table 1) (Ma, Lienhardt, McIlleron, Nunn, & Wang; McGowan et al., 1992).
Health care professionals can differentiate the recommendations based on the
results of clinical trials and those clinical practices and scientific rationales when
clinical data is not available (Horsburgh et al., 2000; Peloquin, 2003).
Table 1 Evidence-Based Rating System

Infectious Diseases Society of America/United States Public Health Service
rating system for the strength of treatment recommendations based on quality
evidence.
Strength of the recommendation
A
Preferred, should generally be offered
Alternative; acceptable to offer
Offer when preferred or alternative regimens cannot be given
Should generally not be offered
Should never be offered
Quality of evidence supporting the recommendation

I.
10
At least one properly randomized trial with clinical end point

II. Clinical trials that either are not randomized or were conducted in
other populations
III Expert Opinion
Note. Adapted from (McGowan et al., 1992). Guidelines for the use of
systemic glucocorticosteroids in the management of selected infections. Working
Group on Steroid Use, Antimicrobial Agents Committee, Infectious Diseases Society of
America
2.1 Tuberculosis Treatment Adherence

The overall goal for TB treatments is to cure the individual patient and to
minimize the transmission of Mycobacterium tuberculosis to other persons.
Hence, successful treatment of TB has benefits both for the individual patent
and the community in which the patient resides (Dick et al., 2003; Peloquin,
2003; WHO, 2008b).
Adherence to TB treatment is similar to other forms of infectious and chronic

diseases. Adherence appears to be better during the initial, acute phase of the
illness when patients are more symptomatic. Those patients who continue to
adhere to the TB treatment program are hopeful that they will recover fully
because they can see the effectiveness of the drugs. The history of treatment
success of family members who have been affected by TB also motivates
patients to comply with their treatments. Naidoo, Dick, and Cooper (2009) found
that many patients report that TB medication had cured members of their family
and friends who were infected with the disease.
Complexity of the treatment and occurrence of side-effects are reported to be

associated with patients stopping or interrupting their treatments (Hand &
Bradley, 1996; Haynes, 1979; Liam, Lim, Wong, & Tang, 1999; Sockrider &
Wolle, 1996). Moreover, patients who experience medication side effects and
notice no improvement in their health are more likely to choose other treatment
options, such as the traditional healer. This is worse in cases where there is no
health service available in their communities, and the traditional healer is more
11
accessible to them (Armijos et al., 2008; Coreil, Lazardo, & Heurtelou, 2004;
Jaramillo, 1998; Liefooghe, Baliddawa, Kipruto, Vermeire, & De Munynck,
1997).
A literature review on adherence to TB care in a Canadian Aboriginal population

found that poor adherence to TB treatment was the most common cause of
initial treatment failure and disease relapse (Orr, 2011). Consequently, such
failure contributed to higher levels of transmission of the disease and higher
chances of developing multi-drug resistance (Baylan, 2011; Qian et al., 2011;
Tahir et al., 2006). TB adherence is a general concern worldwide (Orr, 2011).
To better facilitate potential interventions, reduce barriers and improvement to
adherence, further understanding is needed in the aspects of health service,
personal and social determinants of adherence behaviour (Garner, Smith,
Munro, & Volmink, 2007).
Globally, 79% of people with TB do not have access to DOTS, the WHOs
internationally recognized approach to TB control (WHO, 2015a). It is estimated
that up to 50 million people may be infected with drug-resistant TB. Thus far,
there are no affordable cures for developing and under-developed nations
(WHO, 2015a). A new analysis focusing on the trend of multi-drug resistance
TB (MDR-TB) from 2008-2013 found that, at the global level, the proportion of
new cases for MDR-TB has remained unchanged, at around 3.5%. The
seriousness of such epidemics in some countries jeopardises the progress
against the disease (WHO, 2015b).
In ensuring effective treatments, the DOTS is the most cost-effective strategy

and is recommended globally (WHO, 2003b). The DOTS strategy requires
every individual patient to be categorized according to their conditions and TB
treatment history, so that appropriate drugs can be administered to individual
patients at the time of diagnosis. Patients who have severe forms of
tuberculosis are allocated into category one, and those who have interrupted
treatments are allocated to a different category to those who are newly
diagnosed (refer to Table 2 and
12
3 for descriptions and definitions of TB
treatment outcomes) (Ministry of Health Timor Leste, 2008; WHO, 2003b,

2008a, 2008b).
Table 2 WHO TB Treatment Category
Category Description
New cases of smear-positive tuberculosis and other newly diagnosed

seriously ill patients with severe forms of tuberculosis i.e. disseminated
tuberculosis, tuberculosis meningitis, tuberculosis spondyolitis with
neurological complications, tuberculosis pericarditis, peritonitis, bilateral or
extensive pleurisy, smear negative pulmonary tuberculosis with extensive
parenchymal involvement, intestinal tuberculosis, genitor-urinary
tuberculosis etc.
II
Relapse or failure patients, those who interrupted treatments or patients

who previously treated for more than one month not under a DOTS
treatment program
III
New cases of new-smear negative tuberculosis and extra-pulmonary

tuberculosis.
Note. Adapted from WHO. (2008a). Implementing the Stop TB Strategy: A handbook
for national tuberculosis control programmes. Retrieved from
http://whqlibdoc.who.int/publications/2008/9789241546676_eng.pdf
If patients take medication regularly, the one-year cure rate can reach up to
95% or higher (WHO, 2003b). As a part of the DOTS standardized strategy,
WHO (2003b) recommended that the case detection and treatment success
rates are to be at least 70% of case detection of new smear positive and 85% of
those detected cases expected to be successfully treated. Research has shown
that 20-50% of TB patients do not complete their treatment successfully or
interrupt the regimens by themselves (Hsieh et al., 2008).
13
Table 3 WHO TB treatment Outcome and Definition

Outcome
Cure
Description
Patients are considered as cured if the initial smear/culture was
positive before the onset of treatment and had a negative sputum
smear results, on at least two occasions, one of which was at the end
of the treatment.
Treatment
Sputum smear-positive patient who has completed treatment with
completed
negative smears at the end of the intensive phase, but the smear was
done at the end of treatment. Or a sputum smear-negative TB patient
who has received a full course of treatment and has not become
smear positive during or at the end of treatment. Or extra-pulmonary
tuberculosis patient who has received a full course of antituberculosis chemotherapy but failed to have the required number of
smear or/and cultures.
Death
Patients who died during treatment for TB, regardless of the cause of
death.
Treatment
Patient who failed to achieve bacteriological conversion within 5
failure
months after the start of treatment or after previous conversion

becomes sputum smear or culture positive again. Also patient who
was initially smear negative before starting treatment and became
smear positive after completing the initial treatment.
Transfer out
Patient who has been transferred before the completion of his/her

treatment to another recording and reporting unit and for whom the
treatment outcome was unknown.
Note. Adapted from WHO. (2008a). Implementing the Stop TB Strategy: A handbook
for national tuberculosis control programmes. Retrieved from
If patients take medication regularly, the one-year cure rate can reach up to
95% or higher (WHO, 2003b). As a part of the DOTS standardized strategy,
WHO (2003b) recommended that the case detection and treatment success
rates are to be at least 70% of case detection of new smear positive and 85% of
those detected cases expected to be successfully treated. Research has shown
14
that 20-50% of TB patients interrupt or do not complete treatment regimens by

themselves (Hsieh et al., 2008).
A qualitative study in South Africa by Naidoo and Mwaba (2010) examined the
willingness of TB patients to adhere to treatment regimens found that many
factors were associated with adherence and non-adherence to the DOTS
strategy (Naidoo & Mwaba, 2010). Influencing social, economic and diseaserelated factors such as poverty, having one or more co-morbid health
conditions, alcohol consumption and tobacco smoking were significant
predictors of treatment non-adherence (Naidoo et al., 2013)
Major progress in global TB control followed the widespread implementation of

the DOTS strategy in countries with a high burden of TB. Yet, global statistics
have indicated that DOTS alone has not been sufficient to achieve TB control
and elimination (Behera, 2009). To successfully implement the strategy of
DOTS, it is vital to understand the components of the strategy and its
implementation approaches (Ministry of Health Timor Leste, 2011). Hence,
addressing the TB control problem requires sustained efforts and good
resources, both human and physical infrastructures as well as strong
commitments from every government, and this is clearly highlighted in the Stop
TB strategy (WHO, 2006). Unfortunately, as a young nation, Timor-Leste faces
multiple obstacles towards its development. There are priorities which need to
be addressed by its leaders and TB program implementation may not be a
sufficiently high priority. This is evident in the Timor-Leste Stop TB Strategy
Plan that the allocation and utilization of domestic resources at district levels for
TB control specifically, is low (Ministry of Health Timor Leste, 2010).
Furthermore, data showed that almost 80% of financing for TB control in 2014
was funded by the Global Fund (WHO, 2015c).
The WHO (2003b) suggested that a successful DOTS program would depend
on political commitment with increased and sustained financing, case detection
through quality-assured bacteriology, standardized treatment with supervision
and patient support, and an effective drug supply and management system. To
ensure maximum results, a monitoring and evaluation system must be in place
15
for impact measurement (Ministry of Health Timor Leste, 2008; WHO, 2003b,
2011). Realistic achievement of these recommendations will reflect the good
performance of a system. Consequently, it enhances effectiveness of DOTS
and ultimately, it will contribute towards the progress of eliminating TB globally.
2.1.1 Drug Side-Effects and Multi-Drug Resistant Tuberculosis

Research indicates that side-effects of TB drugs are responsible for the
termination of therapy in up to 23% of cases during the intensive phase
(Schaberg, Rebhan, & Lode, 1996). Furthermore, defaulting is found to be
significantly associated with medication side-effects (Awofeso, 2008). The likely
chance of encountering side-effects is higher for patients being treated with
second-line drugs for multi-drug resistance TB (MDR-TB), during which as
many as 86% of patients may develop drug side-effects (Anderson et al., 2013;
Leimane et al., 2005; Torun et al., 2005). In Jiangsu Province, China, Weiguo et
al. (2009) found that 37.8% of patients listed the main reasons for being nonadherent to treatment were drug side-effects. To reduce the adverse effects of
medication adherence in TB treatment, it is crucial that health staff are
adequately trained for the management and recognition of side-effects.
Anti-tuberculosis (TB) drug resistance is currently considered a major public

health issue worldwide, which threatens the progress made in TB care (WHO,
2012c). MDR-TB is defined as tuberculosis that is resistant to the two most
powerful first-line anti-TB drugs, named isoniazid (INH) & rifampicin (RMP).
MDR-TB develops during the treatment of fully sensitive TB when the course of
antibiotics is interrupted and there is a lack of drugs in the body to fully kill 100%
of bacteria (WHO, 2008a, 2009). The strongest concern in relation to MDR-TB
is that the complete treatment will take as long as 24 months (6 months of the
intensive phase and 18 months of the continuation phase). Patients are likely to
develop MDR-TB when they inconsistently adhere to anti-TB prescribed drugs,
making it much harder to achieve a higher cure rate (Naidoo et al., 2009). In
addition to the length of treatment, the cost of treatment is 100 times higher
compared to the cost of curing a TB patient infected with drug-sensitive TB
(Fourie, 2001; Naidoo et al., 2009). MDR-TB is fatal, as it spreads from person
16
to person as readily as drug-sensitive TB and in the same manner. In other

words, whoever develops the active disease with a drug-resistant TB strain can
transmit this form of TB to other individuals (WHO, 2008a, 2009). Given the
severe impact of drug resistance in TB program interventions, prevention and
control should be emphasized through TB program case detection, quality
assurance and systematic treatment observation, especially with patients who
are at high risk (Shao et al., 2011).
2.1.2 Recommended TB treatment regimens

There are five basic regimens recommended for adults who have been
diagnosed with or presumed to be susceptible to TB. These regimens include
isoniazid (INH), rifampin (RIF), ethambutol (EMB), pyrazinamide (PZA) and
streptomycin (SM).The recommendation that a 4-drug regimen be used initially
for all patients is based on the current proportion of new TB cases caused by
organisms that are resistant to INH (2008). In the presence of INH resistance
there were fewer treatment failures and relapses if a regimen contained the four
drugs that were used in the initial phase (Peloquin, 2003). Each regimen has an
initial phase of 2 months followed by a continuation phase of 6 months
(Peloquin, 2003). The initial phase, it is indicated by a respective number (1, 2,
3 or 4) and a continuation phase is denoted by the respective number and a
letter (a, b or c) (Peloquin, 2003).
2.1.3 Definition of TB treatment completion

TB treatment completion is defined by the number of doses taken by a patient
regardless of treatment duration. Thus, the determination of treatment
completion is based on the total doses taken, not the duration of therapy
(Peloquin, 2003). For instance, the 6-month daily regimen of 7 days a week
treatment should consist of at least 182 doses of INH and RIF, and 56 doses of
PZA (Bumburidi et al., 2006; Peloquin, 2003; Zierski, 1976). In some cases, due
to toxicity or non-adherence to the treatment regimen, specified doses cannot
be administered within the target period. Consequently, the remaining doses
need to be administered within a recommended maximum time of 9 months. If
17
treatment is not completed within the maximum period given, the patient should
be assessed to determine the appropriate treatment regimen. This will either be
that the patient continues treatment for a longer duration or restarts treatment
from the beginning with more restrictive measures (Bumburidi et al., 2006;
Gross et al., 1994; Horsburgh et al., 2000; Peloquin, 2003).
2.1.4 Treatment Interruptions

Numerous reasons for the interruption of medication adherence in TB
treatments have been revealed in research, including patients' lack of
knowledge and poor understanding of the treatment. Some patients may halt
their antibiotics when feeling better after a few months. Patients may also forget
or skip medication from time to time (Armijos et al., 2008; Ayisi et al., 2011;
Corless et al., 2009; Craig et al., 2007; Mulenga et al., 2010; Naidoo et al.,
2009; Qian et al., 2011).
Interruption of TB treatment is common and healthcare professionals

responsible for supervision must decide upon the available treatment option
(Connolly, Davies, & Wilkinson, 1999; Peloquin, 2003). If such a case is
encountered, the patient would either continue the regimen as originally
intended or restart a complete course of treatment. It is critical that the
treatment option chosen is based on the duration of the interruption and the
bacteriological status of the patient before and after the interruption (Peloquin,
2003).
An approach modified from the New York City Bureau of Tuberculosis Control
Clinical Policies and Protocols presented an example of management treatment
interruption (refer to Appendix J) (Gross et al., 1994). If the treatment
interruption occurs in the intensive phase and the lapse is 14 days or more, it is
recommended that the treatment be restarted from the beginning. Additionally,
during the continuation phase, if the patient has received less than 80% of the
total doses and the lapse is 3 months or more in duration, treatment should be
restarted (Gross et al., 1994; Peloquin, 2003; Zierski, 1976).
18
There are numerous relevant factors that can be taken into consideration when
defining treatment interruptions, and these factors include the health service
delivery system, and socio-cultural, economic and individual-related factors. In a
circumstance where the patients prescribed regimen is interrupted, a decision
by the responsible health professional, based on his or her knowledge, would
shape the outcome of an individual patients treatment. The overall aim of this
research study is to measure non-adherence of a patient with a TB prescribed
regimen. By using the VAS and a number of non-compliant questions, the level
of non-adherence can be measured.
2.2 Determinants on Sub-optimal Adherence

Numerous factors that influence adherence to TB treatment have been
identified. These have been grouped in various ways, such as an ecological
framework that involved individuals, through to organization levels or other
frameworks that include more detailed categories. The WHO has grouped these
categories as structural and economic factors, patient-related factors, regimen
complexity, supportive relationships between the health providers and patients,
and the patterns of health care delivery (Dick et al., 2003). The structure of
categories involved in this studys framework was drawn from the ecological
framework and the WHOs way of grouping. However, this thesis focuses on
literature presented in appendices A & B, that specifically addresses factors
associated with TB adherence that are divided into three categories
(summarised in Figure 1). The first level involves factors related to the health
service delivery, including accessibility and availability of services, delivery
settings (i.e. availability of equipment and accessibility of health facilities), the
patient-provider relationship and regimen complexity. The second level involves
the
social
and
economic
factors,
which
include
socio-demographic
characteristics of individuals, such as gender, age, education, occupation,

marital status and socio-economic status. The third category comprises the
individual and disease-related factors, such as illness characteristics and
severity, treatment duration (intensive and continuation phase), treatment
progress over time, risk behaviours (tobacco smoking, illicit drug and alcohol
19
use), individual knowledge, beliefs and attitudes, and mental health problems.
The demographic characteristics and individual disease-related factors directly
affect the level of medication adherence, while service factors and social
supports directly or indirectly influence patients adherence behaviours through
individual-related factors.
20
Health Service Delivery
Social & Economic
Accessibility
Individual & Disease-Related
Social and economic

characteristics
Cost, time and distance
Treatment
Illness characteristics and severity,
treatment duration, and treatment
progress over time
SES, gender, age, education,

occupation & marital status
Availability
Information, presence of
Regimen
health carer & quality of

Complexity, dosage level & side effect
service, supply continuity,

Social motivation
working hours
Risk behaviour
Family support, financial support &

health professional support
Other supports: friends, relatives
& other social supports
Service delivery settings
Alcohol, drug use & smoking
Equipment and facilities

Relationship
Knowledge, belief & attitude

towards treatment
Relationship between health

care providers and patients
Dose adherence
Pill pick up
Treatment complexity, dosage

level & side effect
management
Appointment
Regimens
Level of
ADHERENCE
Follow instruction
Figure 1. Literature mapping of TB treatment adherence.
Literature Review
21
Mental health problems (depression,

anxiety, stress & isolation)
Locus of control (internal, chance and
other power)
2.2.1 Health Service Delivery Factors

There are a number of factors associated with health service delivery. This study aimed
specifically at factors associated with service delivery such as cost associated with
treatment, distance to and from an available service and patient-health professional
relationship.
Delivery Factors
Structural factors in the health system affect adherence to TB treatments
(Shargie & Lindtjrn, 2007). Structural factors identified as barriers, which have
an impact on individual levels, include economic, social, policy, organizational
and other environmental aspects (Sumartojo, 2000). Health systems that do not
function in a manner designed to meet patients needs create a barrier to TB
treatment adherence. A study in Zambia by Mulenga et al. (2010) found that
healthcare access is correlated with treatment adherence. In North Africa, a
case control study by Tachfouti, Slama, Berraho, and Nejjari (2012) showed
that the distance to the nearest health services and the low quality of health
professional and patient communication were associated with poor adherence
(Tachfouti et al., 2012). Another study investigated the level of adherence and
factors associated with the completion of TB treatment between an urban and a
rural district in Timor Leste, where there was a different level of accessibility to
health services. In-depth interviews were carried out with 28 TB patients and
their nurses. Focus group discussions were carried out with patients, health
workers and community members. The study found that non-adherence to TB
treatment was due to local cultural beliefs, lack of access to services, and socioeconomic challenges experienced by individuals. These factors may cause
have people to seek alternative treatments, such as traditional healers. Patients
from rural districts provided additional reasons for defaulting, such as not being
able to walk to the clinic, taking traditional medicine, and being busy with work
(Martins et al., 2008).
The following is a list of system barriers presented by a study on TB treatment

adherence (Orr, 2011)
22
Literature Review
Health workers or systems which fail to provide services according to the

recommended guidelines or standards cannot expect adherent behaviors
from their patients.
Long wait times for care would likely discourage patients to adhere to
treatment.
Health care facilities or mechanisms which actively or passively promote

feelings of physical insecurity or rejection.
Care that is apparently available, yet in reality not accessible due to an

operational culture that does not serve patient needs, i.e. limited service
availability and hours of operation, transportation barriers etc.
Lack of trust, respect and dignity in the relationship between patients and
health workers.
Staff having a lack of knowledge and skills in the diagnosis and

management of the side-effects.
Poor TB program management system resulting in drop-outs and losses

from treatment and follow-up appointments
Care that focuses only on TB, yet fails to acknowledge and assist patients to
overcome barriers, assist patients to counseling, concerns, homelessness,
food, and other individual associated factors faced by TB patients.
Patient- Care Provider Relationships

Health care professionals play a vital role in TB treatment adherence. A lack of
communication and support from health care workers discourages patients from
adhering to treatment, while positive experiences and competent support
creates meaningful connections and relationships with patients (Liam et al.,
1999; Martins et al., 2008; Munro, Lewin, Swart, & Volmink, 2007; Naidoo et al.,
2009). As a consequence, patients are likely to follow instructions and adhere to
their treatment regimen. For example, in the study by Naidoo et al. (2009),
patients were asked about their perception of the quality of the health care they
received. Their responses clustered around one theme, namely, the attitude of
the health care team at the clinic. Good communication between the health
Literature Review
23
practitioners and patients was one of the most powerful predictors of adherence
(Naidoo et al., 2009). Ayisi et al. (2011) and his team conducted a qualitative
study in three district hospitals in Kenya that provided basic curative and
preventive services. The study demonstrated that the most common reason for
defaulting on treatment was the lack of communication with providers and the
lack of their involvement in the treatment process. Evidence from other settings
also showed the importance of patient-health provider relationship, for instance,
in China, research indicated that regular home visits by health workers
appeared to reduce the risk of non-adherence (Weiguo et al., 2009).
The relationship between the health care provider and the individual patient is
indicative of a good therapeutic alliance, a process of collaboration between
provider and patient, in which the practitioner effectively communicates with the
patients about their problems and formulates a management plan that is
conducive to patient compliance. A therapeutic alliance is underpinned with
values of empathy, trust and positive support. This is evident in a mental health
systematic review, which found that the clinician-patient alliance and
communication were associated with favourable outcomes of adherence
(Thompson & McCabe, 2012).
2.2.2 Socio-demographic & Economic Factors

There are many socioeconomic factors that directly influence TB treatment
adherence. According to the WHO evidence for action, factors, such as a lack
of social support network, culture, beliefs and treatment, ethnicity, gender and
age, and the high cost of transport directly or indirectly affect adherence (Dick et
al., 2003).
Research indicates that TB rates are often higher in urban areas compared to
rural locations. One reasonable explanation is that TB transmission increases
due to population density. Crowded living conditions and poor nutritional status
increase the risk of infection, especially in urban areas (Aparicio, Capurro, &
Castillo-Chavez, 2002; Holtgrave & Crosby, 2004; Lienhardt et al., 2012;
Rieder, 1999; Shimao, 2005). Other social factors, such as living alone while
24
Literature Review
struggling to cover health insurance also pose major constraints to treatment

adherence. This is supported by the findings in a survey conducted in Jiangsu
Province, China that found that those who were divorced or widowed and
patients who lacked health insurance were more likely to be non-adherent
(Weiguo et al., 2009).
One other factor that direct or indreclty influences treatment adherence is

patients living conditions. Those who live in remote areas with poor conditions
and lack of access to services are more likely to be non-adherent to treatment.
This is evident in research which found poor resource settings pose barriers to
TB treatment adherence (Munro, Lewin, Swart, et al., 2007; Munro, Lewin,
Smith, et al., 2007). The challenges faced by TB patients undergoing treatment
are beyond their control when they are associated with poverty and inadequate
health services. It is well established that poor living conditions contribute to the
spread of TB, often referred to as a disease of poverty, and it is certainly a
barrier to treatment adherence (Naidoo et al., 2009). As a result of living in a
poor resource setting, participants in the study by Naidoo (2009) found that the
majority of participants had to endure the experience of being poor and living
with the consequences of poverty, including crowded, dangerous living
conditions, unemployment, limited formal education.
A cross-sectional study conducted by Mulenga (2010) in Zambia indicated that

seeking alternative treatment is common for those living in poor resource
settings. As much as 64% of the sample responded that their way of coping with
TB symptoms prior to visiting health care providers was self-treatment. In
addition, the author emphasized that nearly all respondents (98%) presented at
a health centre only when they felt really ill (Mulenga et al., 2010). Some
patients were likely to seek alternative treatments depending on availability,
affordability and accessibility of the various potential sources, such as
traditional, spiritual and Western medicine. The uncertainty of treatment options
and low availability of the sources of care can lead to considerable delays in
treatment (Ayisi et al., 2011).
Literature Review
25
Many studies have estimated the burden of TB in specific, vulnerable population

groups, for instance, prisoners and the homeless, and found that there is a
strong association between social deprivation and TB risk (Bobrik, Danishevski,
Eroshina, & McKee, 2005; Buskin, Gale, Weiss, & Nolan, 1994; Holtgrave &
Crosby, 2004; Rieder, 1999). Yet, the underlying pathways linking poverty and
low socio-economic status to higher TB risk and low treatment adherence are
not fully understood, suggesting that more research in diverse populations
would be beneficial.
The Impact of Financial Constraints in Resource Limited

Settings
Studies conducted in diverse global populations have demonstrated the
importance of low household income as a major contributor to the diagnostic
delay of tuberculosis, poor treatment adherence, and treatment failure (Armijos
et al., 2008; Cambanis et al., 2005; Farmer, Robin, Ramilus, & Kim, 1991;
Jackson, Sleigh, Wang, & Liu, 2006; Needham, Godfrey-Faussett, & Foster,
1998). Many patients experience severe financial constraints to cover transport
cost or purchasing of additional drugs, and the challenges often include
competing financial demands that participants face (Naidoo et al., 2009). This
finding is consistent with a study in Ecuador by Armijos (2008) that found that
38.9% (7/18) of TB sufferers lacked enough money to buy medications.
Even when TB drugs are provided for free, costs still incur for sufferers and this
evidence is shown in numerous studies across different settings. A qualitative
descriptive study by Zvavamwe & Ehlers (2008) in the Omaheke, region of
Namibia, Southern Africa, involved interviews with 40 community-based TB
nurses. Although TB drugs were freely available to patients for treatment, the
geographical inaccessibility of TB services, lack of transport and the time
required to travel to a health service would cost much more than purchasing the
drugs (Zvavamwe & Ehlers, 2008).
Another study emphasized the effect of the distance and cost of transport to
treatment facilities on the overall treatment (Armijos et al., 2008), while a study
26
Literature Review
by Kebede et al. (2012) found that 23.7% of participants in the study missed
doses due to a lack of money for transport. This finding is supported by a study
in Myanmar that concluded that those who spent less amount of time to travel to
and from the clinic were more likely to be adherent (Myo Su, Win Myint Oo, &
Khay Mar, 2015 ). Evidence also suggests that many patients consider the cost
of treatment and the implications for their employment, often choosing to remain
at work rather than risk losing their jobs to undertake treatment (Munro, Lewin,
Swart, et al., 2007).
This complex interaction between the necessity of seeking care and balancing
work commitments is also seen in other settings. For example, a qualitative
study was conducted in Ho Chi Minh City, the largest city in Vietnam, with a
population of approximately eight million (2014 estimate). The study used nonprobability sampling to identify 26 persons with diagnosed TB while under
treatment. Some of the questions explored perceptions and experiences of the
various health care providers competence, attitudes, perceptions and
experiences of the quality of services, waiting times and costs. The results of
the study indicated that associated treatment costs and the aspect of time
became a major constraint for treatment completion. Almost all respondents
reflected on the importance of avoiding time-consuming components of health
care contacts, and thus avoiding loss of time from work (Lnnroth, Tran,
Thuong, Quy, & Diwan, 2001).
Balancing adherence to the medication regimen with other factors was also
illustrated in a systematic review of research into medication and treatment
adherence by Salla et al. (2007) that screened 7,814 papers by title, abstract
and full texts derived from multiple international studies (from 1966 2005). Of
these, 44 papers were included in the synthesis. These studies highlighted
major themes covering structural factors, contextual and individual factors, such
as the financial burden of treatment, knowledge, attitudes, beliefs, and
community and household supports for adherence to treatment. The systematic
review found that even when patients have the willingness to maintain
adherence, contextual issues such as poverty may prevent them from doing so.
Literature Review
27
With the global pervasiveness of poverty, adherence to long-term therapy for

chronic disease remains an issue worldwide.
In resource-limited settings where income is generated solely based on daily

effort, the demand affects patients to risk their chances to adhere to their
treatment regimens. In a study in Zambia by Mulenga et al. (2010), participants
were randomly selected by accessing registered smear-positive TB patients,
new and retreatment cases across 26 treatment centers, from January, 2006 to
July, 2007. The authors expected adherence to treatment of 50% 10% with a
non-response rate of 10%. The results showed that there was a high level of
knowledge among participants. However, delays in seeking treatment and
hoping that the symptoms would go away or believing it was not serious even
when they suspected that they had TB, increased the severity of the disease.
Only 17% of the participants were in formal employment and, for most
respondents, income was dependent on their daily efforts. Therefore, they could
not afford the time to seek and present themselves to an available health center
for diagnosis (Mulenga et al., 2010). A study by Weiguo et al. (2009) was
conducted in Jiangsu Province, China that showed the effect of financial burden
on treatment adherence. The study found that those who experienced a heavy
financial burden were less likely to complete treatment successfully, despite the
Chinese government's policy of free TB service (Weiguo et al., 2009). In Kenya,
a cross-sectional survey was conducted among new TB patients receiving 6
months of standard TB treatment. Level of adherence was assessed with urine
testing for isoniazid (INH), pill count, an interviewer-administered questionnaire
and the VAS. The authors reported reasons for non-adherence such as running
out of pills, were likely to be explained by missed clinical appointments or loss
of pills, and being away from home, mostly due to economic activities (Nackers
et al., 2012).
Financial constraints were associated with patients in Timor-Leste was not

necessarily about the cost of drugs. Generally, in Timor Leste, people
understand that Government health services are free of charge. A survey
conducted in 2011, investigated the ability of community members to pay to
reach health facilities (Ministry of Health Timor Leste, 2011). The results
28
Literature Review
showed that the average cost for TB treatment was USD 80.55 ranging from
USD 0 up to USD 3,300. The estimated cost varied as some patients may have
used more than one health facility. Other costs associated with the treatment
included diagnosis, additional drugs recommended by health professionals, and
the cost of seeing traditional healers. Additional detail on the expenditure
showed that the most expensive component was transportation (Ministry of
Health Timor Leste, 2011). Also, the Timor-Leste National Stop TB Strategy
Plan, 2011-2015, supports poor patients and those living in remote areas
through compensation for indirect costs of TB treatment and food supplements
where required (Ministry of Health Timor Leste, 2010). Yet, the effectiveness of
such support in Timor-Leste is yet to be measured. The treatment process over
6-8 months may require patients and/or family members to pay for
transportation
expenses,
additional
treatment
recommended
by health
providers, self-remedies, and other associated costs (Martins et al., 2008;

Ministry of Health Timor Leste, 2011). It is clear that treatment non-adherence
in resource-limited settings is associated with a range of financial constraints,
such as low household income, lack of transport, inaccessibility and poor quality
of services, the cost of time and the risk of losing jobs. Each of these factors
directly influences patients to be non-adherent to treatment (miss doses)
because they miss appointments, away from home, and likely to seek
alternative and easier treatment options, such as the traditional healer.
Financial hardship also affects patients to delay treatment. It was evident that
most often, patients delay seeking treatment by having self-treatment as they
consider affordability cost associated with the treatment.
Impact of nutrition on TB treatment

Another important socio-economic factor that influences TB treatment
adherence is nutrition. The first RCT study to assess the impact of food
incentives on TB treatment in Timor-Leste was carried out in 2009 (Martins et
al., 2009). The study employed newly diagnosed TB patients from three primary
clinics in Dili, the capital city of Timor Leste. The majority of patients with TB
were very poor and malnourished. The results showed that there was significant
improvement in weight gain at the end of treatment. However, the study
Literature Review
29
concluded that food incentives had no significant benefit on the outcome of

treatment adherence (Martins et al., 2009). One reason which was not
highlighted in the study was that the participants had undertaken TB treatment
in the capital city, where food is plentiful compared to the remote areas of the
country. Participants feelings of impotence within the context of poor living
conditions with limited access to nutritional support were evident by the fact that
they had the information that anti-TB drug treatments were more effective if an
individual is well-nourished, but they sometimes could not afford to buy basic
food items, let alone the nourishing food recommended by the health care
workers.An estimation of the cost associated with food incentives had been
given and it was highlighted that the integration of food incentives on TB
treatment programs can be very costly. However, studies have found the
importance of nutritional intake by patients undertaking TB treatment (Baldwin
et al., 2004; Gupta, Gupta, Atreja, Verma, & Vishvkarma, 2009; Ramakrishnan,
Rajendran, Jacob, Fox, & Radhakrishna, 1961). A study conducted in India
showed similar results, where the overall treatment response was similar in both
groups. However, it was also noted that those receiving nutritional support
tended to show more rapid clearance of bacteria and radiographic changes in
addition to greater weight gain (Ramakrishnan et al., 1961). Nutritional
supplementation may represent a novel approach for a fast recovery for TB
patients in underdeveloped areas of the world and it may be an effective
measure to control TB.Those studies have reported inconsistent findings;
hence, further research to validate the influence of nutritional support and other
barriers or enabling factors to TB treatment is encouraged. (Baldwin et al.,
2004; Gupta et al., 2009; Ramakrishnan et al., 1961).
Family and Social Support

One of the enabling factors to treatment adherence is family and social support.
Family and social support have been found to affect adherence behaviours both
directly (such as through positive encouragement and reinforcement) and
indirectly (through stress relief, reduced anxiety and depression). Do (2011)
investigated medication adherence among 615 people living with HIV (PLHIV)
30
Literature Review
across five clinics in Northern Vietnam using a multi-methods approach.

Although medical regimens for care of people with HIV and TB are different, the
influences of family and social support on individuals undergoing treatment may
be similar. The study showed that the disclosure of information to others
regarding treatment status was an important facilitator for medication
adherence, while not being able to share treatment status was reported to be a
major constraint. In Dos (2011) study, a significant number of respondents
believed that the sharing of information with relatives and friends had the
potential to improve adherence. Conversely, some of the participants perceived
that disclosure of information might cause risks, such as loss of business, loss
of job, emotional injury and stigmatization from others. A qualitative study about
the social impact of tuberculosis in Thailand similarly showed that social support
aided patients in their adherence to treatment. However, stigma caused some
patients with TB to conceal their status and prevented them from advocating for
their healthcare. Similar findings regarding stigma and social support have also
been documented from studies in other settings (Barnhoorn & Adriaanse, 1992;
Eastwood & Hill, 2004; Jaramillo, 1998; Johansson, Long, Diwan, & Winkvist,
2000; Johansson & Winkvist, 2002).
Lack of social support is known to reduce individuals well-being and is

positively associated with depression (Revenson & Schiaffino, 1990). It is also
known to have adverse effects on an individuals ability to cope with stressful
events, such as suffering from an illness (DiMatteo & Martin, 2002). Balanova et
al. (2006) found that there was relatively low default rates with the implantation
of DOTS program in Russia, and this may be due to the extremely financed
social support. Such success demonstrated that the sustainability of DOTS
programs would depend on the integration of social support.
TB control at both the national and international levels includes tools to identify
patients with adverse social risk factors who might benefit from social support
packages (Craig et al., 2007). A prospective cohort study by Craig et al. (2007)
was conducted in London, employing a sample of 250 newly diagnosed
tuberculosis patients between 2003 and 2005. The study aimed to examine the
Literature Review
31
impact of social factors on tuberculosis program management. The data was

collected from medical records and risk assessment tools. Over a third of the
patients faced difficulties adhering to medications and 30.3% had nobody to
remind them about taking medications. A number of social risk factors were
identified in the literature, including unstable housing, which were associated
with poor adherence (Craig et al., 2007; Fourie, 2001). Lack of social support
was also associated with extended periods of hospitalization (Craig et al.,
2007). A cluster RCT in Senegal, a resource-poor setting in African country,
was carried out by Thiam (2007) between June, 2003 and January, 2005. The
health centers were randomized for the intervention (n = 778) and control (n =
744) groups (Thiam et al., 2007). The outcome of the RCT showed that 5.5%
defaulted from the intervention group while 16.8% did so in the control group.
This suggested that the risk of defaulting was significantly higher in the control
group. Social and cultural influences beyond the patients control were found to
have caused suboptimal adherence. Low socio-economic status was found to
be related to poorer TB medication adherence, especially for those individuals
who were chronically ill (Thiam et al., 2007) or who experienced a lack of family
support (Awofeso, 2008; Mishra, Hansen, Sabroe, & Kafle, 2005). Limited or a
narrowly defined social network, poor infrastructure, and general ill-health in
their family and the community (Naidoo et al., 2009).
Social Stigma
Social stigma has been reported as an important factor that directly contributes
to treatment non-adherence. Discrimination related to TB in most societies has
been identified as one of the most important socio-cultural aspects that
influences the illness experiences of those affected by the disease (Baral, Karki,
& Newell, 2007; Brassard, Anderson, Menzies, Schwartzman, & Macdonald,
2008; Dodor, Neal, & Kelly, 2008; Gibson, Cave, Doering, Ortiz, & Harms, 2005;
Harper, 1987; Johansson, Long, Diwan, & Winkvist, 1999; Kipp et al., 2011a;
Sengupta et al., 2006; Somma et al., 2008; Yang, Wonpat-Borja, Opler, &
Corcoran, 2010). This is due to stigma being a socially constructed
phenomenon that can shape the attitudes and behaviours of others towards
32
Literature Review
those affected by the disease (Harper, 1987; Karim, Chowdhury, Islam, &
Weiss, 2007; Somma et al., 2008).
A number of qualitative studies using focus group and face-to-face interviews

have
identified
various factors associated
with
TB
stigma,
including
socioeconomic status, level of education, gender and religion are significantly

associated with TB stigma (Dodor & Kelly, 2010; Johansson et al., 2000; Kipp et
al., 2011a; Rubel & Garro, 1992; Sengupta et al., 2006; Somma et al., 2008;
Woith & Larson, 2008; Zhang, Liu, Bromley, & Tang, 2007). Several studies
have found that incorrect knowledge of what causes TB, the mode of
transmission and treatment also associated with levels of discrimination (Baral
et al., 2007; Dodor & Kelly, 2010; Eastwood & Hill, 2004; Johansson et al.,
2000; Karim et al., 2007; Long, Johansson, Diwan, & Winkvist, 1999; Rubel &
Garro, 1992; Sengupta et al., 2006; Somma et al., 2008). Evidence from
numerous studies also shows that levels of contagiousness of TB transmission
increases the likelihood of individual patients being stigmatized (Baral et al.,
2007; Dodor et al., 2008; Eastwood & Hill, 2004; Mavhu et al., 2010). TB stigma
has also been found to be associated with marginalized groups, such as the
poor, ethnic minorities, prisoners and those who have co-infections with
HIV/AIDS (Baral et al., 2007; Dodor & Kelly, 2010; Eastwood & Hill, 2004; Karim
et al., 2007; Long et al., 1999; Mavhu et al., 2010; Ngamvithayapong, Winkvist,
& Diwan, 2000; Sengupta et al., 2006; Somma et al., 2008).
A cross-sectional study was conducted in southern Thailand involving 480

patients and 320 community members. (Kipp et al., 2011a). The study found
that community members believed eating or drinking with a TB-infected person
was a cause of TB, while patients believed working hard had caused TB.
Stigma was found to be associated with low levels of education, which can lead
to incorrect beliefs that TB increases the chance of getting HIV/AIDS (Kipp et
al., 2011a). The study also found that stigma was higher among community
members than patients.
Patients perceived being stigmatized as the major challenge of being infected

with TB. It was evident that many patients withdrew from the members of their
Literature Review
33
close social network because they felt embarrassed. It is also evident that there
is an association of stigmatization with psychological stress through negative
reactions of family and community members, as they feel ashamed and blamed
for having TB (Armijos et al., 2008). The experience of facing prejudice,
discrimination, and disadvantage is termed enacted stigma by Kelly et al.
(1980), as cited in Porter et al. (1999). Patients who have TB may believe that
family members betray them because they feel ashamed of them for having TB.
As a consequence, patients feel devalued in their family home (Naidoo et al.,
2009). The loss of control over their lives can lead patients to develop a sense
of helplessness and other mental health problems, which are believed to be
associated with poor treatment outcomes (Craig et al., 2007; Fourie, 2001;
Naidoo & Mwaba, 2010).
A systematic literature review of TB stigma by Courthwright et al. (2010) found

that the most common cause of TB stigma related to the risk of transmission of
the disease by TB-infected individuals to susceptible members of the
community. Furtermore, TB stigmatization is common due to its association with
HIV, poverty and low social class (Courtwright & Turner, 2010). Knowing
someone who has died of TB was also associated with higher community TB
stigma compared with not knowing anyone with TB (Kipp et al., 2011a). Further
results of the study showed that a low level of education, a belief that TB
increases the chance of getting AIDS, and the stigma of AIDS were associated
with higher TB stigma TB (Kipp et al., 2011a). To date, few quantitative studies
on TB stigma and the potential factors associated with TB stigma have been
published. These studies investigated education, beliefs, gender, family and
social support from numerous cultures and drew similar conclusions regarding
stigma towards TB patients. Generally, community or family members who are
less educated with strong beliefs that TB is a disease of the poor or it is a sign
of punishment for bad behaviours are more likely to discriminate patients with
TB (Baral et al., 2007; Brassard et al., 2008; Dodor et al., 2008; Eastwood &
Hill, 2004; Harper, 1987; Johansson et al., 1999; Karim et al., 2007; Kipp et al.,
2011a; Long et al., 1999; Sengupta et al., 2006; Somma et al., 2008; Yang et
al., 2010).
34
Literature Review
A knowledge, attitude and practice (KAP) survey in Timor-Leste (Ministry of

Health Timor Leste, 2011).found that, when participants were asked about their
readiness to look after TB patients among their household members, 91%
believed they were ready to look after. In relation to confidentiality, 18.6% of
males, and 16.4% of females thought it was important that information on a
household member suffering TB be kept secret. Although it appears that family
and social supports are strong for most affected family members in this specific
population, there were a significant number of people who said they were likely
to refuse patients to live among them and were willing to isolate them. This may
be due to those who have limited knowledge would not understand that once
chronic patients undergo treatment and take regular medication for two weeks,
they will no longer be contagious. Additionally, it was reported that lack of
confidentiality increase levels of discrimination, especially if the information was
shared with people who have limited knowledge (Ministry of Health Timor Leste,
2011).
2.2.3 Individual Factors Related to Adherence

There are a number of factors involved in the individual disease-related factors,
including level of education and patients knowledge about TB, health locus of
control which; ones belief that his or her health outcome is determined by
chance or luck, risk behaviours such as alcohol consumption and/or cigarette
smoking and mental health problems.
Knowledge of tuberculosis patients

Numerous studies have found that the higher the education level of an
individual, the more successful treatment adherence seems to be (Armijos et
al., 2008; Ayisi et al., 2011; Liam et al., 1999). A case control study conducted
in Morocco, North Africa, employed 290 TB patients (85 defaulters and 205
controls). A defaulter was defined as a TB patient who had interrupted
treatment for 2 months or longer, while a non-defaulter was defined as a patient
who followed the same treatment protocols and completed the whole treatment
Literature Review
35
regimen. The study found non-adherence was significantly associated with low
education and poor treatment knowledge (Tachfouti et al., 2012).
One study in Ecuador involved a sample of 212 adults undergoing diagnostic

TB testing. The results suggested that misconceptions or a lack of key
knowledge could adversely affect early diagnosis, treatment, adherence to
treatment and a further spread of the disease. It was also highlighted that
education was the most important predictor of knowledge, beliefs, perception
and attitudes (Armijos et al., 2008). The study by Liam et al. (1999) supported
these finding, showing that TB patients with tertiary education had significantly
higher knowledge scores compare to the rest of the participants with nontertiary education. Further, education level was an important determinant of the
patients level of knowledge of tuberculosis (Liam et al., 1999) and treatment
adherence (Kebede & Wabe, 2012). Patients with low education levels admitted
that either they or their family members had stopped taking their TB medication
for one or more weeks without their physicians consent (Armijos et al., 2008),
while 6 of 18 patients said they felt cured after a few months of treatment.
A survey in China concluded that illiterate patients were more likely to be nonadherent (Weiguo et al., 2009). A study by Mulenga et al. (2010) also showed
that 55% of participants stated that the most common reasons for stopping
treatment were that they felt better and this was due to lack of knowledge about
the treatment duration. Adhering to treatment was significantly associated with
patients' knowledge. Periods of non-adherence to the TB treatment program
were associated with participants stopping the treatment because they felt
better (Naidoo et al., 2009).
Similarly, a case control study of 384 TB patients conducted in Ndola, Zambia,

found that 39.1% of females and 29.8% of males reported having stopped
taking medication within the first two months of commencing treatment. Overall,
29.8% of the sample had stopped taking TB drugs at some point during the
treatment. The most common reason given for stopping treatment by adherent
and non-adherent patients was feeling better (45.1% and 38.6% respectively).
Additionally, adherent patients mentioned a lack of knowledge of the benefits of
36
Literature Review
completing treatment (25.7%), TB drugs being too strong (20.1%) and a lack of
food at home (11. 4%). (Kaona et al., 2004)
Another study was conducted in a rural South African community by GonzalezAngulo et al. (2013). Fifty adult TB suspects were interviewed prior to diagnosis
and 50 newly diagnosed TB patients were interviewed at pre- and posttreatment. The results showed that almost 60% of the participants reported they
knew of TB, while 25% responded correctly that TB was caused by a microbe.
Other reported causes of TB included alcohol consumption (5%), poor personal
hygiene (5%), tobacco smoking (12%), low ambient temperature (15%) and
hereditary disease (6%) (Gonzalez-Angulo et al., 2013).
Research shows that patients would decide to interrupt TB treatment if they had
a lack of knowledge of the duration of the course and they were unsure about
the importance of completing treatment (Ayisi et al., 2011; Khan, Walley,
Newell, & Imdad, 2000; Sagbakken, Frich, & Bjune, 2008a). Non-adherence to
treatment often due to a limited understanding and lack of treatment knowledge
of the disease and its treatment process (Hand & Bradley, 1996; Haynes, 1979;
Liam et al., 1999; Sockrider & Wolle, 1996). Availability of treatment options
would enable choices for individual patients and the decisions made to be
based on their level of knowledge about the disease (Harper, Ahmadu, Ogden,
McAdam, & Lienhardt, 2003; Lienhardt et al., 2012; Munro, Lewin, Swart, et al.,
2007)
A number of qualitative studies have observed care-seeking and attitudes

towards treatment compliance (Arlt, Lindner, Rsler, & von Renteln-Kruse,
2008; Ayisi et al., 2011; Blackwell, 1979; Liam et al., 1999; Mesfin, Newell,
Walley, Gessessew, & Madeley, 2009; Needham, Bowman, Foster, & GodfreyFaussett, 2004; Zwi et al., 2009). These studies concluded that a lack of
awareness of the duration of treatment and a lack of family support potentially
contrubted to the reduction of adherence to treatment (Arlt et al., 2008; Ayisi et
al., 2011, p. 9; Liam et al., 1999; Zwi et al., 2009).
Literature Review
37
Recently, a survey was conducted by the Ministry of Health Timor Leste (2011)
to determine the knowledge attitudes and practices (KAP) regarding TB among
communities in Timor Leste. Additionally, the survey explored the socioeconomic burden of TB on patients and their families in the community. The
study covered five (40%) of the districts in Timor Leste, and was geographically
representative. Multi-stage systematic random sampling was applied for the
sample selection, and data were collected from patients and their household
members. A number of variables, such as housing characteristics, access,
distance and affordability of TB treatment, TB knowledge and health-seeking
behaviour were included. For social and family support, participants were asked
whether TB patients should be isolated to prevent the further spread of the
disease, 48.6% (49% males & 48.2%) females) believed that TB patients should
be isolated.The survey found that there was a low level of community
awareness of TB and only 10% of the participants knew the mode of TB
transmission. When assessing the participants knowledge of tuberculosis, such
as mode of transmission, symptoms, diagnosis and treatment, the survey found
that 65.3% reported to have only ever heard about TB. In comparing the
knowledge of TB between rural and urban areas, the highest proportion of
respondents who had heard about TB was 83.8% from urban areas. When the
respondents were asked about the mode of tuberculosis transmission, only
6.6% were correct, 71.5% answered incorrectly, and 21.9% did not know about
tuberculosis transmission. (Ministry of Health Timor Leste, 2011).
The finding of the Timor-Leste Ministry of Heath Survey is consistent with the
ethnographic study by Martins et al. (2008) that concluded that participants with
a low level of knowledge about TB and its treatment may cause them to seek
alternative treatment such as traditional healers. Other barriers to treatment
completion included the patients reasons for defaulting, such as feeling better,
or no symptomatic improvement. Patients were willing to complete their
treatment when they had strong knowledge about the treatment, a high
willingness to be cured, or evidence of symptomatic improvement and
satisfaction as a result of taking the drug. Furthermore, the study suggested that
due to high illiteracy rates and low levels of knowledge about the disease, it was
38
Literature Review
a common belief that TB was a punishment for negative behaviour (Martins et

al., 2008).
Health Locus of Control

Locus of control refers to the extent to which individuals believe they can
control events affecting them, regardless of their actions being desirable or
undesirable (Rotter, 1966). The causal attributions that individuals assign to their
illness condition will affect their way of coping with it. If a treatment regimen
recommended by a health practitioner is consistent with an individuals
understanding of what caused his or her illness, then the individual is more
likely to adhere (Do, 2011). An individual who ascribes external factors that are
out of his or her control as causing the illness is less likely to be self-motivated
to improve his/her health status, as compared to an individual who believes they
can control what happens to them (Do, 2011). Patients perceptions of the
implication of tuberculosis symptoms, their beliefs and attitudes are potential
determinants of compliance (Haynes, 1979).
Risk Behaviour
Risk behaviour behaviour in this study describes specific behaviours such as
alcohol consumption, cigarette smoking and/or illicit drug use that direct or
indirectly impact or causing significant harms to individuals health. Clinical
experience suggests that drug use and alcohol problems are associated with
treatment non-adherence and poor treatment outcomes (Craig et al., 2007;
Fourie, 2001). A prospective cohort study in London by Craig (2007) with a
sample of 202 participants suggested that drug and alcohol use were
significantly associated with patients having difficulties taking their TB
medication. Additionally, drug or alcohol use increased the duration of
hospitalizations (Craig et al., 2007). Missed appointments were also associated
with alcohol and drug use (Craig et al., 2007). Similarly, in Hong Kong, China, a
study of TB defaulters found that tobacco smoking was one of the many factors
that contributed towards treatment default, or poor adherence (Chang, Leung, &
Tam, 2004). In Uganda a study found that alcohol consumption and tobacco
Literature Review
39
smoking was directly correlated with TB medication non-adherence (Amuha,

Kutyabami, Kitutu, Odoi-Adome, & Kalyango, 2009). The findings from studies
in various settings are also consistent with a cross-sectional survey in South
Africa which found that high risk of alcohol mis-use and tobacco smoking are
significant predictors of treatment non-adherence (Naidoo et al., 2013).
Mental Health Problems

Depression is another potentially significant impediment to effective treatment.
Within the treatment context, negative thoughts, anxiety, hopelessness and
depression are often associated with low adherence to health care instructions
(Do, 2011). Feelings of helplessness and depression have been found to be
significantly associated with an individuals ability to cope with stressful events,
especially when suffering from an illness (DiMatteo & Martin, 2002). Naidoo and
Mwaba (2010) considered mental health to be critically important for an
individual infected with TB because individuals are likely to cease trying to
achieve a positive outcome when they are unable to control events in their lives.
The psychological wellbeing of an individual is important to the overall

management of the disease (Peltzer et al., 2012). In his study, Seligman (1975)
classified helplessness into three different categories, namely, motivational,
when an individual stops trying to change an outcome, cognitive, when
individuals have learnt no new responses that could help to prevent aversive
outcomes in the future, and emotional, when a feeling of depression sets in
(Seligman, 1975).
Conversely, high self-efficacy has also been found to be associated with better
disease management, such as adhering to prescribed medicines, managing
stress, and following a recommended nutrition program (Dodge & Clark, 1999).
One might also surmise that these individuals do not believe that adhering to
the recommended treatment of the local health authorities will cure their illness
condition and possibly view the treatment methods with scepticism and mistrust.
Optimistic bias refers to the belief that other people but not oneself will develop
40
Literature Review
a disease, have an accident, or experience other negative events (Brannon &

Feist, 2000, p. 106).
Other social determinants associated with psychological stress include

continued anxiety, low self-esteem, social isolation and control over ones home
and working life (Wilkinson & Marmot, 2003). A descriptive qualitative study by
Armijos (2008) summarized feelings of isolation and loneliness 44%, and
sadness 42%. Overall, 78% of the participants indicated that they had suffered
significant psychological and emotional trauma after learning that they had to be
tested for possible TB and had to receive treatment.
Therefore, fighting to eliminate TB is not a simple task. It is an approach of all

involved stakeholders, including international donors that provide sustainable
funding, committed local government, the cooperation of health workers and
communities, and the awareness of patients to undertake their treatment
seriously. Failure in any of these areas is likely to affect the achievements of the
program.
2.3 Summary of Research into TB Medication Adherence

The previous section discussed the findings from various studies on TB
treatment and factors associated with TB sub-optimal adherence. From all the
studies that have investigated factors associated with TB treatment, the majority
are qualitative and have covered a whole range of contextual and individual
factors associated with TB treatment (see Appendix A).
The present review identified a number of published articles on TB medication

adherence studies using quantitative research methods. These studies were
published from 2004 up to 2014, and most were conducted in regions of Africa
(refer to Appendix B for a detailed summary of these quantitative studies).
There are various measurement methods used to determine treatment
adherence in these studies; structured questionnaires, pill counts, patients
medical records and self-reported methods. However, levels of adherence to TB
Literature Review
41
treatment were most commonly assessed by asking patients about the number
of doses missed over a specified recall period.
A study conducted in Kenya used the VAS and other measures to determine
adherence. The study found that VAS non-adherence was 7.5% with 92.5% of
the participants reported to have taken 100% of their medications. Such high
adherence rates may have been influenced by the method used, as selfreported adherence measures are prone to social desirability bias, in which
patients tend to provide answers that they believe would fit interviewers
expectations, which is high adherence to treatment in this case (Nackers et al.,
2012). The majority of the studies had non-adherence rates between 20-30%
(see Table 4). However, one of the studies conducted in Ethiopia on TB/HIV coinfected patients used a 90% threshold and found that with recall periods of 3
and 5 days, the levels of non-adherence were 53.3% and 55.8%, respectively
(Eticha & EKassa, 2014).
2.4 Tuberculosis Program and Research on Treatment in TimorLeste

TB is recognised as one of the major public health problems in Timor-Leste
(Ministry of Health Timor Leste, 2010), although large-scale research into TB
prevalence and incidence is yet to be undertaken. It was estimated from smear
positives data that the incidence rate of TB is between 115 and 175 new smear
positive cases per 100,000 per year. As emphasized in the National TB
Strategy Plan, 2010 and the Health Sector Strategy Plan, 2007, the high TB
burden is linked to social and economic indicators in the country (Ministry of
Health Timor Leste, 2007).
Timor Lestes National Stop TB Strategy Plan (2011-2015) aims to provide
quality Community DOTS (C-DOTS) services through the CHCs. C-DOTS is a
method used to strengthen DOTS by increasing case detection and improving
patient outcomes within the community. It consists of training community
members to support DOTS therapy in various ways, including, but not limited to,
42
Literature Review
identifying and referring suspect cases, assisting with collection of sputum

samples, providing and observing the intake of daily drugs with patients in their
homes, tracking down defaulters, and referring patients with side-effects to
health personnel at the CHC. Successful C-DOTS programming is reliant on
consistent supervision, clinical support, drug supplies from the CHC and high
adherence of TB patients for treatment over 6 to 8 months. Timor-Leste can
only begin to work toward the internationally recognized DOTS targets of 70%
case detection and 85% cure rate if there is a well-functioning C-DOTS
program. Fighting the elimination of TB will only be achieved if there are
effective strategies and sufficient resources to improve TB patients adherence
to treatment.
The WHO handbook for implementing the Stop TB Strategy identified behavioral
and environmental risk factors, such as smoking, malnutrition, indoor air
pollution and alcohol abuse (WHO, 2008a). There are many other barriers in
implementing the Stop TB strategy in developing countries such as Timor
Leste. As a result of colonization, major health infrastructure was destroyed,
causing a shortage of equipment, supplies and public utilities with scarce
financial and technical support (Martins, Heldal, et al., 2006; Martins, Kelly,
Grace, & Zwi, 2006). In addition, a huge volume of marginalized and
disadvantaged groups in the country experience high levels of malnutrition and
a low socio-economic status. At the same time, those in remote areas live in
poor housing with a high level of indoor air pollution (Ministry of Health Timor
Leste, 2011), while experiencing a lack of support, with limited access to health
services (Martins et al., 2008).
Thus far, there has been little research conducted in Timor-Leste to show how
TB patients comply with their treatments or to investigate the factors influencing
treatment adherence. One ethnographic study
(Martins et al., 2008) has
qualitatively addressed issues associated with TB treatment, such as local

cultural practices, knowledge and socio-economic factors associated with
treatment adherence. The findings of this study have been used as the
cornerstone in modifying the DOTS expansion strategy in Timor-Leste to
overcome treatment-associated factors. A survey was also conducted by
Literature Review
43
Ministry of Health Timor Leste (2011) to determine the level of knowledge,

attitudes and practices among community members and TB patients, and an
RCT study investigated the impact of nutritional support on TB treatment
(Martins et al., 2009). To date, there is no quantitative, peer-reviewed journal
article reporting the level of TB treatment adherence and factors impeding
optimal adherence in Timor-Leste.
44
Literature Review
Table 4 Summary of Literature Review on TB Treatment Adherence
Source
Country
Study design
Setting &
sample size
NonThreshold
adherence
%
Measurement and
recall period
Structured
Mixed
Weiguo et
al. (2009)
China
questionnaire,
methods,
13 counties
multi-stage
(districts).
sampling
670 TB patients
90%
12.2%
Recall period: The

whole duration of
strategy
treatment (6-8
months)
Subjective tools
Questionnaire:
VAS: 7.5%
miss >25% pills,
Nackers et
Kenya,
al. (2012)
Africa
Cross-
Health TB
sectional
clinic, 279
survey
patients,
(questionnaire &
VAS) & objective
last 4 days.
Urine test:
adherence
VAS: <80, last
2.4%
monitoring tools
month
(urine test & pill
INH test:
Pill count:
count)
negative
17.7%
VAS: last month
Pill count: <80%
2 days for IP and 8

days for CP.
Kaona et al.
(2004)
Zambia,
Africa
Crosssectional
study
Pre-structured
6 clinics, 400
patients
100%
30%
survey
questionnaire, 8
month period
Self-reported.
O'Donnell et
South
Prospective
Kwazulu-Natal,
al. (2014)
Africa
cohort study
South Africa
Taking all pills in

N/A
32.3%
the previous 7
days for whole 6
months.
Myo Su et
al. (2015 )
Cross
Myanmar
sectional
survey
Morisky Medication
A specialist
hospital
100%
25.3%
Adherence Scale
(MMAS), 4 items &
pill count
TB/HIV coEticha and

EKassa
Ethiopia
(2014)
Cross-
2 public
sectional
hospitals, 3
study
health centers
90%
infected
Structured
patients, 3
questionnaire, 3-5
days=53.3
days period
%&5
Literature Review
45
Source
Country
Study design
Setting &
sample size
NonThreshold
adherence
%
Measurement and
recall period
days=55.8
%.
Interviewer
Amuha et
al. (2009)
CrossUganda
sectional
study
1 Hospital, 140
participants
administered
N/A
25%
questionnaire, 5day period prior to

interview.
MSMN Scale,
Likert scale,
Corless et
South
al. (2009)
Africa
Descriptive
2 clinics, HIV
Exploratory
(n=149), TB
Design
(n=159)
symptoms
MSMNS: 75%
N/A
assessment scale,
QoL scale, Missed
appointment, SOC
Scale, MOS Scale.
Ailinger,
Moore,
Nguyen,
USA
and Lasus
Cross-
1 Urban public
sectional
health clinic, 53
study
patients
2nd month:
N/A
South
Cross
Wabe
West
sectional
(2012)
Ethiopia
study
Ailinger et
al. (2007)
USA
Descriptive
Study
1 district
hospital, no
N/A
N/A
153 participants
20.8%
16%, 8th
month:
66%
and medical
records. Number of
month client
complete therapy.
1st month:
1 urban public
health clinic,
month:
28%
(2006)
Kebede and
2%, 9th
Brief questionnaire
Self-administered
questionnaire
Medical records &

number of pills
taken
-
Clinical
attendance
3 clinics, 270
Martins et
Timor
Randomized
al. (2009)
Leste
Control Trial
patients
(intervention =
N/A
DOTS
Interview
N/A
and
137, control =
Pill count
133)
IP: weekly,
CP:
fortnightly
46
Literature Review
Source
Country
Study design
Setting &
sample size
NonThreshold
adherence
%
Measurement and
recall period
Improve
adherence
Ailinger et
al. (2010)
USA
Pre-
9 clinics, 131
Experimental
patients
design
medical records
N/A
Interventio
Self-report of
n vs control
number of pills
groups
taken, 9 months
(62%vs39
%)
Armijos et
al. (2008)
Ecuador
Cross
1 Public health
sectional
facility, 212
study
people
Chiang, Lin,
Retrospective
Lee, Lee,
observational
and Chen
China
(2012)
case control
study
Face-to -ace
N/A
N/A
structured
questionnaire
1 tertiary
medical center,
interview using
Electronic records
N/A
N/A
302 patients
and medical
charts, 6 months
1 HospitalQuestionnaire,
based TB clinic
Kipp et al.
(2011a)
Cross
Thailand
treatment < 1
(480 patients),
sectional
2 Hospitals
survey
(300 healthy
N/A
N/A
month
community
members)
Note. IP: Intensive Phase; CP: Continuation Phase; VAS: Visual Analogue Scale; INH: isoniazid; N/A: Not Applicable,
MSMNS: Morisky Self-reported Medication Non-adherence Scale, SOC: Sense of Coherence; MOS: Medical
Outcome Scale.
2.5 Framework
Several frameworks were reviewed to find the theoretical framework that is
most likely to capture the determinants of medication adherence. They are
presented below.
2.5.1 Behavioural Perspective

The behavioural perspective integrates the behavioural learning theory which
focuses on the environment and the teaching of skills to patients to manage
Literature Review
47
adherence (Dick et al., 2003). It is characterised by the use of the principles of

antecedents (which are either internal thoughts or external environmental cues)
and consequents, and patients influence on behaviour (Munro, Lewin, Swart, et
al., 2007). The consequents are rewarded if they are a positive behaviour or a
punished if they are a negative behaviour. Despite studies indicate the
importance to compliance of patients organization and comprehension, there is
a little chance in the attitude that leads to compliance with the treatment
instructions (Leventhal & Cameron, 1987)
2.5.2 Self-Regulation Perspective

Self-regulatory theory is the most popular theory in the field of medication
adherence. The theory is developed to conceptualize the adherence process in
a way that focuses on the patient (Leventhal & Cameron, 1987). The theory
assumes that people are motivated to avoid or treat illness and are active, selfregulating problem solvers. The patients need to be interested in improving
his/her health for treatment to be effective (Nouwen, Urquhart Law, Hussain,
McGovern, & Napier, 2009).
The behavioural perspective focuses on the environment and patients skill
development to cope with treatment and self-regulation theory assumes
patients are active, motivated and problem solvers for their own health.
Therefore, both models focus on inviduals behaviours towards their health.
However, the current study investigates 3 main aspects, namely health service
delivery, social and demographic and individual factors. These two models were
used as a guide for the individual disease related factors. However, the
conceptual framework for the study was derived from the literature.
2.5.3 Conceptual Framework

The literature review revealed many factors associated with adherence to
tuberculosis treatment. Figure 1 presents the mapping of literature related to
health service delivery, social and economic and disease related factors.
Arrows in Figure 1 explains the links and relationships among all variables in the
literature. For example lack of health service accessibility would force individual
48
Literature Review
patients to travel to and from an available health service. As a consequence,

patients would require financial support to meet such need.
In comparison to Figure 2, the arrows of health service, social and eocnomnic

and personal and disease related factors directly linked with adherence. This
explains by the fact that the current study investigated the relationship of each
aspect with treatment adherence rather than the interaction of each of the three
main domains.The structure of the literature mapping in Figure 1 is retained in
Figure 2 to ensure consistency of variables in the categories. However, only
variables that were relevant to be investigated in the current study were

included in the conseptual framework in Figure 1.
Those factors are divided into three main categories, as follows:

a. Health service and delivery factors
There are numerous aspects identified in the literature which are related
to health service delivery. However, this study will mainly focus on
investigating the aspects of accessibility and availability factors due to
limited time and resources. Accessibility aspects include cost, time and
distance, and availability includes information, presence of health care
workers and quality of services i.e. counseling.
b. Social and economic factors
Social and economic factors are two broad areas which also cover a lot
of factors. This includes demographic characteristics, including socioeconomic status, gender, age, education, occupation and marital status.
Social motivation, such as family support, financial support and other
social support i.e. support from friends and relatives, will also be
examined.
c. Individual and disease-related factors including:
treatment: illness characteristics and severity, treatment duration, treatment
progress over time. Regimen: Complexity of treatment, dosage level and sideeffects, knowledge, beliefs and attitudes towards treatment and mental health
problems such as depression, helplessness, stress and isolation (refer to Figure
2).
Literature Review
49
Health Service Delivery Factors
Social & Economic Factors
Accessibility
Socio-demographic
characteristics
Treatment

occupation & marital status
Illness characteristics and severity,

treatment duration, and treatment
progress over time
Cost, time and distance
Personal & Disease Related Factors
Regimens
Complexity, dosage level & side effect
Availability
Social motivation
Information, presence of health

carer & quality of service
Family support, financial support &

health professional support
Other supports: friends, relatives &
other social supports
Alcohol, drug use & smoking
Knowledge, belief & attitude towards

treatment
Level of dose
ADHERENCE
Mental health problems (depression,

anxiety, stress & isolation)
Locus of control (internal, chance and
other power)
Figure 2 Survey Conceptual Framework
50
Risk behaviour
Literature Review
SUMMARY
Treatment adherence has been recognised as a pivotal aspect of the fight

against TB, as poor adherence has a negative influence at both individual and
population levels. A major public health problem that threatens the progress
made in TB care and control worldwide is the development of multi-drug
resistance TB (MDR-TB). It has been documented in many different settings
around the world that drug resistance arises as a result of improper treatment
and failure to ensure that the patients complete the whole course of their
treatment regimen successfully. Research continues to show thresholds that
vary between 75% and 100% as optimal adherence, while the recommended
threshold for TB adherence is at 80%. There are a number of ways in which
adherence to TB treatment can be measured. A common measure is the selfreport of the estimated percentage of prescribed doses taken over a given
period of time. Yet, the self-report method sometimes leads to social desirability
bias, which may influence the overestimation of medication adherence.
Understanding the barriers and enabling factors to TB medication adherence is
of paramount importance, especially in designing effective strategies to optimize
adherence.TB is a key public health issue in Timor-Leste, and key stakeholders
for TB programs have limited data available on the level of treatment
adherence.Therefore,
this
research
is
designed
to
develop
better
understanding of the situation for TB treatment adherence among patients in

Timor Leste.
Literature Review
51
52
Literature Review
Chapter 3: Methodology
INTRODUCTION
Chapter 3 describes the methodology used in this study. This chapter begins
with the study design, the research process, description of study sites, study
participants and the data collection procedure. Then, it presents the instrument
development in details, which includes the variables and measurements of the
correlates, translation of the tool and field-testing. Finally, this chapter describes
the data analysis, data management and ethical considerations.
3.1 Study Design

This study employs qualitative and quantitative methods sequentially. The
overall research design consists of four main phases. Phase one involving
literature review, qualitative study and survey drafting. Phase 2 including
instrument revisions, piloting and main survey. Phase 3 involving administration
of survey, and phase 4 include survey analysis, finding synthesis and
implication for future research (see Figure 3 for the study flow diagram).
3.1.1 Phase 1: Development of the Study Survey

Development of the questionnaire involved two sub-phases:
Phase 1a: Derivation of independent variables and their relationships from

the literature review
In phase 1a of the study, research questions were defined and methodologies
were determined. The study then began with a literature review to identify
important factors related to tuberculosis medication adherence. Reviewed
studies related to this specific health issue enabled the researcher to develop
tools for both a qualitative semi-structured questionnaire and a tentative draft of
cross-sectional survey.
Methodology
53
This process involved identification of relevant studies on TB treatment

worldwide.
Independent
characteristics,
risky
variables
lifestyle
such
behaviours
as
patients
(cigarette
demographic
smoking,
alcohol
consumption and illicit drug use), belief, psychological wellbeing, TB knowledge,

social stigma, social and family supports were retrieved. Relationships among
selected independent variables with TB treatment were also investigated.
Phase 1b Derivation of additional independent variables developed from
qualitative interview analysis

In phase 1b, the study continued with the exploratory qualitative research to
capture an in-depth understanding of various aspects related to tuberculosis
medication adherence. The qualitative study involved face-to-face interviews
with health professionals from communities to the national level and currently
registered patients undergoing treatment at the time of the interview. The
rationale of conducting the exploratory qualitative research was to gather ideas,
approaches, and clues regarding any specific health issues that may not have
been foreseen from the literature (Creswell & Plano Clark, 2011; King, 2009).
The qualitative results were utilized to refine a draft of the quantitative survey
prior to pilot testing. Details of the qualitative findings are presented in Chapter
4.
3.1.2 Phase 2: Construction, Revision & Pilot Study of Survey

Phase 2 included a number of components;
1. The questionnaire was reviewed by four experts with 10 plus years of
experience nationally and internationally from multidisciplinary backgrounds
and experiences in international public health, the management of TB and
HIV/AIDS research projects and research in TB and HIV/AIDS treatments,
health research methods, cross-cultural research, health inequality, the
social determinants of health, clinical epidemiology, mental health, and
culturally and logistically diverse communities.
2. The questionnaire was further examined by two qualified translators with
health backgrounds and the principal researcher to ensure the content was
54
Methodology
valid and culturally appropriate (see Figure 4 for the translation process).
The instrument was piloted and detailed field-testing process is outlined in
(Section 3.4.5).
3.1.3 Phase 3: Administration of Survey

Phase three, administration of questionnaire involved recruitment and training of
survey administrators, and data collection which included administration of
questionnaires to 347 TB patients who currently under treatment at the time of
data collection.
Methodology
55
Literature Review
Phase 1
variables and relationships

Qualitative Study
Data Collection
(10 patients & 10 health
professionals)
Study Survey
Qualitative Results
Questionnaire drafting
Development of study survey
Potentially important
Derivation of potential
Integrating results of
Experts Feedback
qualitative study
Tools were reviewed by 4 experts

with 10 plus years of national and
Survey Piloting
international experience
(Repetitive testing of tools

with 50 TB registered
patients)
Recruitment & training of survey

administrators
Phase 3
Final Survey
Survey administration
Survey Administration
(347 TB patients)
Survey results
Phase 4
Construction, revision & piloting
Revision
Implementation
Phase 2
independent variables
Finding Synthesis and
Results
Discussion
Implication for Practice and

Future Research
Figure 3 Study Flow Diagram

56
Methodology
3.2 Study Sites

Qualitative Interviews Sites
The qualitative data was collected in Dili and Liquica Districts for the health
professional group, and Bairopite clinic and Comoro CHC for patient group.
3.2 Study Sites

3.2.1 Qualitative Study
The qualitative study was conducted in Dili and Liquica districts for the health
professional group, and Bairopite Clinic and Comoro CHC for the patient group.
3.1.2 Cross-Sectional Survey Sites

The survey data was collected in nine health services (six community health
centers, and three private clinics) in three districts. The community health
centers included Liquica CHC (urban), Maubara CHC (rural), Bazartete CHC
(rural), Maliana CHC (urban), Centro CHC (urban), Comoro CHC (urban),
Motael clinic (urban), Bairopite clinic (urban), and ISMAIK Tibar (rural).
Bairopite Clinic is a private clinic located in Dili, founded by Dr Dan Murphy and
operated by himself and Timorese health professionals with the support of the
Ministry of Health Timor Leste. The clinic plays a vital role in providing all
consultations to meet patients needs on a daily bases throughout the year. The
clinic provides diagnostic services, manages approximately one third of all TB
cases in Timor Leste, and provides treatment to more than 1,000 people every
year. It works collaboratively with the National TB program in providing DOTS,
education and follow-up care.
There are 16 beds in two TB wards dedicated to patients who are too sick for
out-patient treatment in Bairopite Clinic. These beds are also used to facilitate
patients from remote areas of the country that lack medical supervision during
the intensive phase of treatment. Bairopite Clinic has also initiated a program to
provide preventive treatment to young children exposed to TB. The program is
Methodology
57
recommended by the WHO and the Timor-Leste National TB Program to be

implemented, but it has been proven to be difficult to implement. Bairopite Clinic
is the first centre in Timor-Leste to establish such a program on a large scale.
Both St. Antonio Motael and ISMAIK Tibar Clinics are church operated while
supported by the MoH Timor Leste. All patients undertaking treatment at
ISMAIK Tibar are diagnosed at Bairopite Clinic. After diagnosis, those whose
families live far away from a nearby clinic, or unable to walk with no support
undertake intensive treatment in ISMAIK Clinic prior to returning home. Patients
who are able to gain their strengths after a few months of treatment are able to
return home and continue the continuation phase of treatment. The clinic has a
full time nurse to provide DOTS to patients. The patients are also frequently
visited by health professionals from Bairopite Clinic.
Comoro and Centro CHCs are located in Dili, and Luqica, Bazartete, Maubara
CHCs are located in Liquica district, Maliana CHC is located in Maliana district,
and all CHCs are operated by the Ministry of Health. Generally, each of these
CHCs has a pharmacist, nurse, doctor, laboratory technician, and a district TB
assistant. Due to high demands and limited health resources, health
professionals who work in each clinic often play multiple roles in assisting
others to provide general consultations to the public. The district TB assistant
and the laboratory technician are regularly supervised by the district
tuberculosis coordinator (DTC). The DTC also assists them with the crosscheck of slides and sputum testing. CHCs provide DOTS to TB patients
undergoing intensive phase treatment and patients under a continuation phase
of treatment while living at home.
3.3. Qualitative Study

As discussed in the literature review, medication adherence is a complex issue.
To gain a better understanding of the subject, it is crucial to take into
consideration of contextual, individual and psychological aspects that directly or
indirectly influence adherence. Hence, qualitative method is appropriate to
58
Methodology
capture perspective of study participants (Malterud, 2001). Such method will

also allow researcher to uncover beliefs, values and personal motivations that
underlies health behaviours (Berkwits & Inui, 1998; Curry, Nembhard, &
Bradley, 2009; Inui, 1996). The exploratory qualitative research was used in this
study to validate contents of the quantitative survey, to explore issues related to
TB adherence, and provide insights into how participants perceive adherence
and reasons associated with non-adherence.
3.3.1 Sample Size

The qualitative study employed 10 health professionals from a national level
(i.e. National TB Program Manager) to a sub-district level (i.e. District TB
Assistant).
Participants
from
I/LNGOs
that
contribute
towards
the
implementation of TB programs were also selected. Ten patients who undertook

treatment at the time of interview were recruited from a private clinic and a
public service (CHC) (seeTable 5).
Table 5 Matrix for Sampling in Qualitative Research

Qualitative
technique
Public Sector
Private Sector
Participants
Health providers
Total
- 1 NTP Director
- 1 clinic doctor
- 1 NTP Adviser
- 2 clinic TB staff
semi-
- 1 District TB
- 2 NGO staff
structured
Coordinator
director, TB Adviser
10
or NGO staff
interviews
TB patients
Methodology
2 non-adherent females
2 non-adherent males
2 non-adherent males
2 adherent females
1 adherent male
1 adherent male
10
59
3.3.2 Data Collection

The qualitative data collection was conducted between September and
November, 2012. In order to recruit patients, the researcher worked
collaboratively with the District TB Assistants, District TB Coordinator or clinic
staff to select patients through the TB registry list. Five male and five female
patients were conveniently selected based on the subjective assessment of
patients' missing appointments, as observed by the health staff. The interviews
were undertaken in locations that were most convenient to the patients. Almost
all participants were selected at the time they came to the clinic or CHC to
receive their medications. Those who were willing to participate were asked for
verbal consent prior to the interview. The semi-structured interviews were based
on
the
themes of perceived
benefits of
medication
adherence
and
consequences of non-adherence, possible facilitating factors to adherence,

knowledge on adherence, associated cost with treatment, views and
experiences and why they adhere or not adhere to treatment regimens (see
Appendices C & D).
The semi-structured interviews were conducted with health staff in Dili and
Liquica Districts to discuss their views on the quality of health care, the impact
of demographic characteristics on adherence, lifestyle, beliefs and social
support that may directly or indirectly influence patients' adherence to treatment.
The qualitative study provided important findings to assist in modification of the
research questions, conceptual framework and revision of the final survey
instrument to ensure their relevance to the context of Timor-Leste (See
Appendix L for the interview transcripts).
3.4 Survey Development

The questionnaire was adapted from a number of studies worldwide (see
Appendix H). There are a number of tools available for the measurement of
medication adherence and each has its strengths and weaknesses (see Table
6) Although self-reported questionnaire and face-to-face interview methods may
be susceptible to social desirability bias and recall bias, they are simple,
60
Methodology
inexpensive and demonstrated as the most useful method for clinical settings.
The VAS and MARS were selected for this study due to their simplicity to be
administered to patients, and VAS has been widely used to measure medication
adherence in other settings that have similar social and demographic
characteristics wit Timor Leste.
A pilot study to test the tool was undertaken to ensure its appropriateness in
Timor-Leste context. The survey was then revised based on the findings from
the pilot study.
Table 6 Various Methods of Measuring Medication Adherence
Method
Advantages
Disadvantages
Most accurate
Patients can hide pills in the mouth and then discard them;
Direct methods
Directly observed
therapy
Measurement of the
impractical for routine use
Objective
level of medicine or
Variations in the metabolism and white coat adherence

can give false impression of adherence; expensive
metabolite in blood
Measurement of the
Objective; in clinical
Requires expensive quantitative assays and collection of
biologic marker in
trials, can also be
bodily fluids
blood
used to measure
placebo
Indirect methods
Patient questionnaire,
Simple; inexpensive;
Susceptible to error with increase in time between visits;
patient self-reports
the most useful
results are easily distorted by the patient
method for the

clinical setting
Pill counts
Objective,
Data easily altered by the patient (e.g., pill dumping)
quantifiable, and
easy to perform
Rates of prescriptions
Objective; easy to
A prescription refill is not equivalent to ingestion of
refills
obtain data
medication; requires an accessible pharmacy system
Methodology
61
Method
Advantages
Disadvantages
Assessment of the
Simple; easy to
Factors other than medication adherence can affect clinical
patients clinical
obtain data
response
Electronic medication
Precise; results are
Expensive; requires return visits and downloading data
monitors
easily quantified;
from medication vials
response
tracks patterns of
taking medication
Measurement of
Often easy to
Marker may be absent for other reasons (e.g., increased
physiological markers
perform
metabolism, poor absorption, lack of response)
Patient diaries; when
Help to correct for
Easily altered the patient susceptible distortion
the patient is a child,
poor recall; simple;
questionnaire for
objective
(e.g., heart rate in

patients taking betablockers)
caregiver or teacher
Note. Adapted from Osterberg, L., & Blaschke, T. (2005b). Drug Theraphy:
Adherence to Medication the New England Journal of Medicine, 353(5), 487497. Retrieved from
http://www.nclnet.org/sosrx/membersonly/sept21/adherenceNEJM08042005.pdf
3.4.1 Dependent Variables

The study used two measures of the dependent variable of medication
adherence - a VAS and the Medication Adherence Rating Scale (MARS) (see
Appendix H). The visual analogue scale (VAS) was adapted from Nackers et al.
(2012) and estimates adherence along a continuous line between 0% and
100%, indicating the percentage of TB doses taken by respondents in the past
month. Patients were asked to determine about how much of their medications
they have managed to take during the last month. For instance, zero means no
medication has been taken; 50% means a half and 100% means they have
taken every single dose of the prescribed regimens. The VAS had also been
62
Methodology
widely used to measure TB and ART adherence (Do, Dunne, Kato, Pham, &
Nguyen, 2013; Nackers et al., 2012; 1982).
The second dependent variable was the MARS which measures adherence in
the past month. It was adapted with minor modification to cross-validate the
VAS adherence estimate (Thompson, Kulkarni, & Sergejew, 2000b). The MARS
items were drawn from the combined items of the Drug Attitude Inventory (DAI)
and Morisky Medication Adherence Scale (MMAS) to enable greater validity
and clinical utility (Thompson et al., 2000b). The MARS asks patients to
respond to a set of dichotomous statements in the questionnaire by circling 'yes'
or 'no' to describe their behaviours or attitudes towards medication during the
past month (Thompson et al., 2000b).
The VAS is the most commonly used tool to estimate adherence for both TB
and HIV/AIDS studies in various settings (Do et al., 2013; Nackers et al., 2012).
The MARS scale has been used to measure behaviour and attitudes of patients
towards medication adherence (Lee et al., 2013; Morisky, Green, & Levine,
1986; Oliveira-Filho, Barreto-Filho, Neves, & Lyra Junior, 2012; Thompson,
Kulkarni, & Sergejew, 2000a).
There were five questions selected from the MARS scale. These questions ask
whether a patient ever forgets to take medication, is careless about taking
medication, stops taking medication when feeling better or worse, or only takes
medication when feeling sick. If patients correctly answer the statement 4 of 5
(80%) our 5 of 5 (100%), they are deemed as adherent. Both measurements
used a cut-off point of <80% of drug intake classified as suboptimal adherence
and 80% as adherence (Nackers et al., 2012; WHO, 1982).
The questionnaire includes a set of subjective reasons for missing doses in a

checklist of 21 items. These items have been adapted from an ART adherence
Methodology
63
study by Do et al (2013). In this study, reasons for missed medications used

'yes' or 'no' for patients agreement or disagreement on each statement.
3.4.2 Independent Variables

The first section of the questionnaire assessed participants demographic
information such as sex, age, education, marital status and income. Second
section consisted of questions which collected information related to patients
lifestyle, especially cigarette smoking, alcohol consumption and illicit drug use.
The third section included questions that measured psychological wellbeing of
patients, their beliefs. This followed by measuring patients level of TB treatment
knowledge. The fourth section contained a list of non-adherence reasons which
patients could nominate to explain the reasons for missing their medications.
The final section of the questionnaire contained a set of questions that
measured the helpfulness of their communciations with health professionals as
well as the benefits of receiving social supports.
The demographic variables such as gender, participants are given two options
such as (1) male, (2) female to be chosen from. For their ages, they were asked
for their dates of birth, while an additional question was also asked to determine
their current age in order to confirm their actual age (see Appendix H for the
survey questionnaire English version). Measurements for other indepent
variables were as follows.
Alcohol use: participants were asked about their alcohol consumption status. If
they answered in the affirmative, they were then asked how often in the last
month they had at least one drink of alcohol, using a response category of (1)
daily or nearly every day (>4 time/week), (2) sometimes per week (1-4
times/week), (3) sometimes per month (1-3 time/month), (4) drink when
available. The quantity of the drink every time they consumed alcohol and the
frequency of drinking more than 6 units were also asked (Do et al., 2013). A
picture that showed a glass of wine, a pint of regular beer or a bottle of wine for
64
Methodology
estimated alcohol units was presented. The picture was used to guide
participants to roughly estimate amount of alcohol consumed as per each of the
measurement. For instance, if a participant drink a full bottle of wine, the
estimated measurement is roughly around 9 units.
Cigarette smoking: the participants' smoking status was determined by asking

the following whether they were: (1) current smoker, (2) smoked in the past, (3)
had stopped smoking at the time of treatment, or (4) never smoked. Participants
who responded as smokers were asked a number of questions which
highlighted how soon after they awoke they smoked their first cigarette,
measured in minutes of (1) within 5 minutes, (2) 6-30 minutes, (3) 31-60
minutes or (4) after 60 minutes. The number of cigarettes smoked in a day was
also asked: (1) 10 or less; (2) 11-20; (3) 21-30; or (4) 31 or more. The last
question about cigarette smoking asked participants to recall how many
cigarettes they had consumed in the past 24 hours.
Illicit drug use: Participants were asked if they had ever used any kind of illegal
drugs such as illicit drug. Three options were given to assess their drug use
status, including (1) current user, (2) used, but in the past, and (3) never. Those
who were current users or who had ever used illicit drugs were asked their
frequency of usage (Do et al., 2013).
Psychological distress was measured by the Kessler Psychological Distress

Scale (K10) (Slade, Grove, & Burgess, 2011). This is a 10-item, self-report
rating scale which assesses psychological distress of participants in the last 30
days. The scale uses a 5-point Likert scale, with a response ranging from 1
(none of the time) to 5 (all of the time). The tool is a desirable method of
assessment as it is a genuine attempt on the part of the clinician to collect
information on participants current condition. Furth, the scale has been widely
used and reported to have strong psychometric properties, indicating that it
accurately measures patients anxiety and depressive symptoms (Slade et al.,
2011).
Methodology
65
Multidimensional Health Locus of Control (MHLC): The MHLC scale (Malcarne,

Fernandez, & Flores, 2005; Wallston, Wallston, & DeVellis, 1978) was used to
assess the respondents health-related locus of control. A shortened, modified
9-item version of the MLHC by Malcarne et al. (2005) was used. The scale has
been widely used in many settings to measure patients beliefs that their health
outcomes are determined by chance or luck. A study investigated treatment
adherence for HIV/AIDS in Vietnam reported a strong psychometric property,
indiciating that the tool is effective in measuring patients beliefs about their
health outcomes (Do, 2011). The four subscales of locus of control on this 9item scale are internal health, powerful others, and external health. The items
are scored on a 4-point response scale ranging from 1 (strongly disagree) to 4
(strongly agree).
Tuberculosis knowledge: The Tuberculosis Knowledge Scale (TKS) was

adapted from a survey conducted in Sudan (Mohamed, Pharm, Yousif, Ottoa, &
Bayoumi, 2007). This 7-iten scale measures patients understanding about
tuberculosis treatment, which includes knowledge on possible causes of
tuberculosis, screening of family members and the duration for completing TB
treatment.
Stigma: The Tuberculosis Stigma Scale (TSS) (Van Rie et al., 2008) is a12-item
scale that measures the patients' understanding of the communitys perspective
towards tuberculosis patients and their own perspectives towards tuberculosis.
This scale had been widely used to measure stigma for both tuberculosis and
HIV/AIDS diseases (Kipp et al., 2011a; Sengupta et al., 2006; Van Rie et al.,
2008). The first seven questions cover the community perspective towards
tuberculosis. This includes patients understanding of whether the community
would feel that they might not want to eat or drink with a person who has
tuberculosis, feel uncomfortable and keep their distance. The section which
highlights the patients perspective towards tuberculosis assesses the feelings
experienced by individual patients undergoing treatment, such as if the patient
66
Methodology
feels hurt when others react knowing he or she has tuberculosis, feeling alone,
afraid to tell others outside of family, causing a burden for their family or afraid
to share their treatment status with family members(Van Rie et al., 2008). In a
given statement, a patient answered yes or no according to their perspectives
whether he/she agreed or disagreed with the statement.
Provider-patient Communication: This was measured using the Patient

Reaction Assessment (PRA) scale by Galassi et al. (1992). This scale has been
shown to differentiate effective and non-effective relationships between patients
and caregivers (Malcarne et al., 2005). It contains five items, using a response
scale ranging from 1 (strongly disagree) to 4 (strongly agree).
The final section of the questionnaire was about social and family support.
These items have been used in the ART adherence study by Do et al. (2013),
which aimed to measure social and family support.There are 30-items included
in the scale which measures disclosure of treatment to family, friends,
neighbours, peer educators and social organizations and satisfaction of patient
on the support received during treatment.
Patients are asked to rate the
support received from each group during their treatments. For example to what
extent do the following people help you to remember to take TB medication?
Six of the same statements are given to each group, and each with a response
scale ranging from 0 (not at all), 1 (a little), 2 (somewhat) to 3 (a lot). The higher
or lower the total score in each group indicates more or less support is receive
by a patient.
The K10 was used to measure anxiety and depression symptoms experienced
by the participants in the past month. Those who scored under 20 are likely to
be well, a score of 20-24 are likely to have a mild mental disorder, a score of
25-29 are likely to have moderate mental disorder, and a score of 30 or above
are likely to have a severe mental disorder (Slade et al., 2011). The internal
consistency (Cronbachs for K10 was 0.806. The health locus of control was
measured with a subscale of the multidimensional Health Locus of Control
Methodology
67
Scale, with an internal consistency (Cronbachs = .91). The Tuberculosis

related stigma scale measures both community and patient perspective towards
TB (Van Rie et al., 2008), internal consistency of (Cronbachs = .84). A patient
assessment tool to measure effectiveness of communication between patient
with care provider, (Cronbachs = .90) (Do et al., 2013), knowledge
(Cronbachs = .60), traditional healer (Cronbachs = .62) and social support
(family, friend, neighbour, peer educator and social organization) (Cronbachs
=.76) (Do et al., 2013; Munro, Lewin, Swart, et al., 2007).
3.4.3 Translation
The translation process was carried out by adapting methods from the WHO
(2012a) as well as a number of guidelines for translational validity, including the
Census Bureau Guideline for the Translation of Data Collection Instruments and
Supporting Material (Pan & Puente, 2005), the Principles of Good Practice for
the Translation, and the Cultural Adaptation Process for Patient-Reported
Outcomes (Wild et al., 2005). The survey was translated from the original
language English to Tetum (Timor-Leste national language), and then it was
back translated to English. The translation to Tetum from English was carried
out by a qualified translator with health background. The translator (English to
Tetum) worked for the Timor-Leste Ministry of Health as monitoring and
evaluation specialist for the National TB Program. He also has years of
experience
in
translating
documents
for
various
organizations.
The
questionnaire was then back-translated from Tetum to English by a different

translator. He also has years of translation experience and worked for an
international non-government organization as a health manager for TB.
Although the guidelines for translation validity were closely followed, it was
important to highlight that due to the incompleteness of the Tetum language
(Timor-Leste National Language) (Golden, 2004), some translation required
additional explanation to ensure that the actual meaning of the English version
was not lost (see Appendix I). Refer to Figure 4 for the translation process.
68
Methodology
Forward
Original source of language (English)
translation
Translated to Tetum by a professional translator
with health background
Reviewed by a panel
A health Professional
An accredited translator with
health background &
The principal investigator
The final version of the Tetum questionnaire

Compare original
document with
Back translation
outcome
Translated into English by a professional
translator
Reviewed by a panel
A health Professional
An accredited translator with
health background &
The principal investigator
The final version of the back-translated English

questionnaire
Figure 4. Translation Process
Methodology
69
3.4.4 Pilot Test

The aim of the field-test of the survey was to observe the effects the interviewer
had on the question responses and to establish the feasibility of the question
being administered by the interviewer. It was conducted to ascertain errors the
respondents made on the questionnaire, the clarity of each item for the
respondents and the respondents ability to answer the questions without errors
or confusion, and whether the respondents understood the meaning of all of the
items in the questionnaire.
The patients were informed about the research activities three days before they
came to their respective clinic. Participants were conveniently selected, as they
appeared to receive their medication at the clinic, using the inclusion and
exclusion criteria. Verbal consent was given at the time prior to filling out the
pilot survey. Those who were unable to participate due to other circumstances
or chose to not participate were excluded.
Testing the questionnaire involved recruitment of 50 currently enrolled

tuberculosis patients in several clinics across Dili and Liquica Districts in Timor
Leste. The plan was that 50 patients were to be recruited to fill out the pilot
questionnaire at one time. However, the pilot testing was conducted, it was
necessary to make changes. Therefore, the questionnaire was piloted with 10
participants per session and the same process was repeated four times. It was
noted that many participants were not able to read and write; therefore, the
questionnaire was read out for them. There were some participants who
claimed to read and write; however, as they started to respond the questions, it
was noted that they could not fill the questionnaire independently. Hence,
participants who experienced difficulties filling the pilot questionnaire were also
assisted and items were read out to them by the survey administrator.
The responses by the participants were observed and noted according to the
following factors: the appropriateness of each response choice in every
question, for example if the respondent placed an "X" or tick the box available
as directed; whether the response choice ranges were used. For instance,
70
Methodology
(strongly disagree, somewhat disagree, neither agree nor disagree, somewhat

agree, strongly agree); whether the respondents followed the directions
correctly; checked the appropriateness of some questions within sections of the
questionnaire and observed if more or less than half the participants responded
to the specific items.
Length of time it took to complete the questionnaire was also recorded. During
the first stage of the testing process, it took an average of 75 to 90 minutes for
an individual patient to complete the questionnaire. However, after several
revisions, the final test of the questionnaire took an average of 40-50 minutes.
This was due to the removal of a number of irrelevant questions within each
section of the questionnaire. Participants were asked if they had used a device
as a facilitator to remind them of the medication time. For example using a
clock or an alarm to remind medication time. This item was removed as none of
the participants admitted to having used any form of these devices as a
reminder.
In all sections of the questionnaire testing, the omission, inclusion and

duplication of questions were taken into consideration. In the first part of the
questionnaire testing, the main focus was on the structure of the questions,
appropriateness of multiple response options and use of scientific and medical
words. Revisions were undertaken before the second phase of questionnaire
testing. After the first section some explanatory information was added to
several parts and some questions were simplified. For instance, instead of
directly asking 'how often did you miss taking your medication in the last month',
it was necessary to ask whether at any time they had missed their medication
prior to asking about missed medication in the previous month.
In the second and third sections of the field-testing, some of the multiple options
given in each questions were added and irrelevant options were removed. An
example of this was that the questions asking about the frequency of using a
type of transport, local mini bus, bus and angguna (mini truck) were combined
into one option. Other options, such as by car as a driver were irrelevant and
were, therefore, removed.
Methodology
71
3.4.5 Integration of Findings on Composition of the Study

Questionniare
A number of key factors need to be considered as a result of conducting the
pilot study. Worldwide literature reviews identified a list of factors associated
with medication adherence, and the draft questionnaire was developed based
upon these studies. However, after piloting the questionnaire, it was clear that
not all factors identified in the literature were relevant for investigation in the
present study's specific population. For instance, life style illicit drug use was
excluded from the questionnaire as the health professional group in the pilot
interviews stated that there had been no illicit drug users known. The previous
qualitative studies conducted in Timor-Leste had not assessed drug use
(Martins et al., 2008; Ministry of Health Timor Leste, 2011) and it was assumed
that it may be unaffordable given the poor economic background of the
population. Although the drug scale was initially removed after the pilot study,
through collaborative discussions with a group of panel members, it was
advised that the scale be included in the final version of the questionnaire.
The official stats suggest that approximately 65% are smokers. However,
participants in the health professional group advised that through years of
experience in the field, it appeared that almost all male patients were smokers.
Therefore, the tobacco consumption variable was included in the study. Most
commonly, smoking is also embedded as part of the culture, where presence of
guests would be complimented with cigarette-smoking. This was evident in the
latest WHO Global Tobacco Epidemic report which showed the latest trend of
tobacco use for youth and adult groups. The survey reported 65.7% males and
2.9% females as current cigarette smokers (WHO, 2015).
3.5 Cross-sectional Survey

3.5.1 Study Population
The Democratic Republic of Timor-Leste (DRTL) consists of 13 districts,
covering an area of 14,874 square kilometres (refer to appendix E). The data
72
Methodology
collection was conducted in the three districts of Dili (capital city of Timor),
Liquica and Maliana, including six Community Health Centers and three private
clinics.
3.5.2 Method of Sampling

The districts were chosen based on the highest prevalence of TB cases in
urban, semi-urban and rural areas. The classification was determined using the
NTP data from the Ministry of Health of Timor Leste. Furthermore, in attempt to
achieve a representative sample, the sites selected were ensured to cover a
variety of areas with different geographic characteristics, such as mountain
terrain, seaside, urban, rural and remote areas of the country and population
density (see Table 7).
Table 7 Population size, density and percentage of TB in the selected sites
Districts
Population
Population
Annual Report for
(census 2010)
Density
TB Case Finding
2011
Urban
Rural
Total
Maliana
15,800
73,987
89,787
61/km2
410 (9.28%)
Dili
193,563
40,768
234,331
4,000/km2
1970 (44.60%)
Liquica
5,152
58,177
63,329
100/km2
164 (3.71%)
Note. Adapted from
Ministery of Health Timor Leste (2011) Knowledge,
Attitude and Practice (KAP) towards Tuberculosis and Socio-economic

Impact of the Disease in Timor-Leste.
The annual number of tuberculosis cases recorded in the NTP database in

2011 for Dili, Liquica and Maliana districts were 1970 (44.65), 410 (9.3%) and
164 (3.7%) respectively (Ministery of Health Timor Leste, 2011). The study
aimed to randomly select samples from across a number of patients registered
in each clinic until the required sample size for each district was achieved.
Methodology
73
However, due to the mountain terrain and distance required to travel to sites
and time limits, some CHCs in remote areas of Maliana district were not
reached. Therefore, the convenient sampling method was used to interview
patients who came to the hospital for treatment. Patients were conveniently
selected through hospital records and those who did not meet pre-defined
criteria were excluded. Selection was based on the pre-defined inclusion and
exclusion criteria. Patients were included if they were undergoing TB treatment
at the time of interview, able to speak Tetum fluently and agree to complete the
questionnaire. Participants under 18 years of age, out of reach geographically
with no contact provided in the TB registry and unwell or refuse to participate
were excluded.
3.5.3 Statistical Power Analyisis

The study estimated the level of non-adherence and its associated factors
during tuberculosis treatment among patients in Timor Leste. The estimated
power analysis was calculated using a proportion formula of the sample size for
estimating proportions (Daniel, 2009; Lachenbruch, 1991):
N z 2 p (1 p )
d 2 ( N 1) z 2 p (1 p )
The following parameters were used to calculate the sample size. Prior to the
study there were no official estimates of adherence of TB treatment in Timor
Leste. However, the National Tuberculosis Program Data for 2011 showed that
the treatment outcomes of new smear positive (NSP) patients registered in
2010 were 80% cured, 8% completed treatment, 4% failed, 4% defaulted & 4%
transferred out. From this figure an estimation of 88% adherence (p) was given.
The absolute precision (d) was 5% and the confidence level was 95%. This
calculation gave a sample size of 164. In order to increase the statistical power
and to incorporate a design effect due to the clustering within the districts
selected, the sample was multiplied by two, which increased the sample size to
328. The sample size was further increased by 5% to allow for any incomplete
interviews. A minimum sample size of 345 was required. The actual data
collected is presented in Table 8.
74
Methodology
Table 8 Data Collection Sites, TB Cases and Participant Numbers for the Study
Districts
District
Place of interview
Annual TB Case
(Home/Clinic/CHC)
Finding 2011
Participants
(n)
(n)
Dili District
1970
CHC Centro
CHC
14
CHC Comoro
CHC
35
Bairopite Clinic
Clinic
106
Motael Clinic
Clinic
Home visit
Home
111
Liquica District
164
CHC Bazartete
Home
11
CHC Liquica
CHC
10
CHC Maubara
CHC
13
ISMAIK Clinic
Clinic
Maliana
CHC Maliana
19
410
Hospital
Total
20
2544
347
3.5.4 Procedure
3.5.4.1 Recruitment of Survey Administrator
The National Tuberculosis Program and the office of the MoH/Global Fund were
visited. Key officials were given a brief explanation of the objective of the study,
and data collection method was also briefly explained. These visits included the
Global Fund Program Manager and National Tuberculosis Officer. Positive
support for the research data collection was shown and a facility such as a
training venue was made available to conduct the training of the survey
administrators. The ethics committee office under the Ministry of Health was
contacted to seek approval for research variations and after few days the
approval was obtained.
Methodology
75
The office of the Global Fund recommended the survey administrators, as those
candidates had been involved in research activities previously conducted by the
Ministry of Health. Recommended candidates were contacted and a brief
introduction of the study was given. Those who were available and interested in
the research activity were scheduled for a have a face-to-face meeting with the
principal researcher.
Positions were offered for two months, receiving a salary of USD 100 per
month. The survey administrators were expected to work from Monday through
to Saturday with equivalent to eight hours per day conducting the surveys
across nine health services in three districts. Accommodation and transport
costs were provided to individual survey administrators as part of the agreement
package.
The main roles of the research administrators were to assist the principal
researchers with conducting a survey. This included assisting Tuberculosis
patients who may not understand any questions or content of the survey
questionnaire; assisting by reading out, interpreting and explaining content from
the questionnaire and filling out the questionnaire for patients who were
illiterate.
Throughout the data collection process, the survey administrators were

expected to frequently travel and stay overnight in districts for up to a week or
more, depending on the task required. They were also advised to travel safely,
as there was no health care cover for any unpredicted accidents.
3.5.4.2 Training of survey administrators

Training of the survey administrators was conducted once they agreed to sign
the agreement and abide by it. Out of five people that were interested, three
were selected. Soon after the training and prior to the departure to the first
district in the research, a survey administrator decided to leave as he had
another job offered. As a consequence of only able to recruit two survey
administrators, the principal investigator had to involve directly in the survey
administration process to prevent delays of data collection in the selected sites.
76
Methodology
However, the changes made to the work plan resulted in deferral of data entry,
which was supposed to be completed prior to principal researchers return to
QUT.
We were fortunate to have the offer of a conference room by the Global Fund
Division in the Ministry of Health. Such support reduced any associated cost.
The training period took less than the two weeks originally planned. The content
of the training began with a brief presentation of the study, including
background, literature review and current TB status in Timor Leste. A brief
overview on the content of the questionnaire and elaboration of the differences
between qualitative and quantitative data was given. The team reviewed each
question of the questionnaire. Then, the survey administrators were asked to
explain the contents of each question in different ways to show their
understanding. The idea was to ensure that they were able to assist patients
who could not understand a question and provide support for patients to
complete every survey successfully. Each participant was given a task to think
about for each of the questions and to provide suggestions based on their
experiences of how certain questions could be best asked. This was to avoid
any cultural sensitivity for any question that may contain sensitive information,
i.e. drug usage and marital status. Additionally, the principal investigator was
aware from experience that data falsification by interviewers had occurred in
past work in Timor Leste. Extra care was taken to ensure data quality control.
First, the training given to all survey administrators highlighted their honesty in
the data collection. They were advised that falsifying data was a practice of
dishonesty and it would affect their future careers. They were also asked to
collect accurate data from patients which would help to determine the problems
associated
with
the
issue
investigated,
and
their
contributions
were
acknowledge. Second, the survey administrators were assisted prior to

collecting data from patients independently. For instance, the principal
investigator demonstrated by administering a survey with a patient, where the
survey administrators were able to observe how they should approach and
greet patients.
Methodology
77
3.5.4.3 National Tuberculosis Program Support

The Principal researcher worked closely with the NTP Timor Leste. Their
authorization, support and collaboration allowed further contact to the selected
sites. The NTP Director was approached to request support with access to
contact details of NTP staff in the districts. Fortunately there was no need for a
letter to request approval and contacts were retrieved. The districts TB
coordinator for each selected site was contacted to inform him/her about the
research and that there would be a visit to each CHC.
3.5.4.4 Administration Protocol

An overview of the study was given to the District TB Assistants to be
distributed prior to the survey administration. Participants who received the
information were asked by the clinic staff to come to the clinic. The
administration was conducted after agreement for the most suitable time for the
participants.
Those who agreed to participate were asked if they understood the information
delivered. Participants who were illiterate and had no family members to explain
the information were assisted to ensure that they understood what was
expected of them in completing the survey. During the survey administration
process, effort was made to ensure that participants felt comfortable, and were
not interfered by the health providers in their respective clinic. The
administration protocol was closely followed by providing a brief session,
confirming their willingness to participate and obtaining consent and ensuring
confidentiality of information.
For every participant, the interviewer was expected to introduce him or herself,
explain the purpose of the study, and emphasise that the researchers did not
represent any government body or other organization that may have been
directly involved with the participants treatment. The participants were
encouraged to provide accurate and honest responses, and the importance of
their participation in the survey to help identify ways in which TB treatment may
be improved was acknowledged.
78
Methodology
Participants with no formal education through to high school level were

interviewed face-to-face, while those with university qualifications were asked to
complete the questionnaire independently. A brief guide on how to answer the
questions was given at the beginning of the administration. The team and the
principal researcher were also available at all times to assist participants in any
difficulties they faced in completing the questionnaire.
3.5.4.5 Procedure in each Study District

The survey was conducted from the beginning of June until the end of
September 2013 in the following districts.
Liquica District
The Liquica District TB Coordinator was contacted and an information summary
about the research was dropped to District Health Service (DHS). However, the
visit to the site was delayed one week later due to an annual workshop
conducted in the district. A quick meeting with the DHS vice director was
arranged during the first visit to the DHS office. As all sub-district CHC directors
were present at the workshop, a brief explanation of research activities and the
research overall plan was shared with each sub-district director. They were also
informed that at some point the research team would visit their respective CHC
to interview patients under treatment. Further contact details of who should be
contacted were also provided by those CHC directors themselves. Contacts
were made, but the activities were not carried out as planned due to other
event, which required the district health staffs participation.
In the following week, district TB assistants (DTCs) for Bazartete, Liquica

districts were contacted. Klibur Domin local NGO was visited to discuss the
availability of the DTA who work directly to support the local community health
center. A short meeting was held with the LNGOs director to provide
background information of the study and the data collection process.
The DTA at Klibur Domin NGO agreed to have home visits for each patient
undergoing the continuation phase of treatment to provide prior information
Methodology
79
regarding the study. Those who were illiterate were given a short explanation
about the study process and they were asked to stay at home for the following
days. During each home visit, patients were asked again to clarify their
understanding about our presence.
Eleven patients in Bazartete sub-district were interviewed face-to-face during inhome visits. On the same week the team was able to visit Maubara sub-district
and collected data from 13 patients as well as 10 patients from Liquica subdistrict in the following week. It was fortunate as the Liquica District TB and
CHC staff were supportive of the research activities. The in-home interview was
supported by DTAs to provide translation for illiterate patients who only speak
the local language.
Maliana District
In the following week, the team travelled to Maliana district. On the first day a
meeting was arranged with the District Tuberculosis Coordinator (DTC) who the
principal investigator knew personally from previous work. On the same day,
after meeting with the DTC, the research team was also able to meet with the
District Health Service (DHS) Director. Due to the busy schedule, the research
team managed to convince the director to verbally approve the research
activity. The DHS directors requested for an evidence of approval for data
collection for his reference and copy of the approval letter from the National
Tuberculosis Program was provided.
On the following day, the research team travelled to the hospital to wait for
patients who came to the clinic for medication refill. Summary information
regarding the study was not given prior to arrival due to distance required to
travel to this site. However, the documents were prepared and presented to
each eligible patient. Explanation was also given to all patients regarding the
study. Those who were unsure about their participation were given the option to
take home the information and return to the clinic in two to three days, if they
decided to participate.
80
Methodology
The majority of the patients chose to participate after explanation was given.
This was due to the fact that most of them had travelled from the countryside.
Only a few decided to come back within the following days for interviews as
they lived close to the clinic. After three days, there were not many patients that
visited the CHC. Therefore, we consulted with the DTC and DTA to accompany
us for home visits.
Phone credits were provided to CHC staff to make appointments with patients
under the continuation phase of treatment. Those patients contacted were
asked to stay at home on the next day for home visit interviews. A brief
explanation of the study was given, including what was expected from the
patients if they chose to participate. Unfortunately the team was not able to
reach patients who lived a long distance away from the CHC due to the limited
time and resources.
Dili District
Due to a change of plan for interview in Liquica District, which was delayed until
one week later, the research team had to interview patients in some other sites.
Therefore, staff of Bairopite Clinic, which is located in Dili was contacted, so that
the team could start interviewing patients that were readily available and easily
accessible as they visit the clinic for DOTS. Very luckily a verbal approval was
acquired directly from the clinic director to conduct the interview. This was also
due to the fact that the director was aware of the research study during the
qualitative data collection. Summary information regarding the study was also
given to the clinic TB staff to provide to patients who came to receive their
medications.
After returning from field trips to Maliana and Liquica Districts, the team
continued to interview patients under treatment at Bairopite Clinic, Motel Clinic,
CHC Centro and CHC Comoro. Those patients under the continuation phase of
treatment were contacted to discuss the possibility of coming to the clinic the
same day, the following day or any other day that suited their daily schedule.
The interview was conducted from Monday to through to Saturday until the
required target sample was met.
Methodology
81
The research team continued to work collaboratively with TB staff at the

selected sites throughout the recruitment and interview process, especially with
patients in rural areas of every district. Prior to questionnaire refill or interview,
the local clinic staff who were directly responsible for the patients treatment
approached the patient at first to provide a brief overview on the purpose of the
data collection. Then, a research interviewer was introduced to the patient for
further interview or refill of the questionnaire.
3.6 Data Analysis and Management

3.6.1 Exploratory Qualitative Data Analysis
All interviews were digitally recorded and transcribed to English. A framework
for the analysis was developed according to the objectives and research issues.
The content analysis of each transcript was conducted thoroughly to identify
themes emerged, keywords mentioned repeatedly, and identified themes that
related to each other.
3.6.2 Quantitative Content Analysis

To prevent duplication or typographical errors in the process of data entry,
EpiData version 3.1 was used. The program allows users to create simple data
entry and documentation. It offers features that detect errors such as double
entry verifications and codebook overview of data. A database was created with
the same format and structure as the questionnaire using EpiData. To minimize
errors during data entry, options for answering each question were created and
specifically coded, where acceptable values for the question only can be
entered. For instance, a question with a scale of one to four, a coding can be
assigned where the program only accepts numbers one, two, three or four.
However, other number such as zero, five or a combination of two numbers will
not be accepted, and the program would prevent skipping to the next question
until the correct answer is given to the specific question. The data was then
converted to Statistical Package for Social Sciences (SPSS) statistical software
82
Methodology
for Windows Version 20.0 (Inc., Chicago Illinois) for analysis (Allen & Bennett,
2012).
Steps to Data Analysis
Data screening and preparation
- Developing a coding manual
- Checking to ensure there were no duplications
- Conducting frequency distributions for all variables to check for invalid
response and degree of missing data
Assumptions were checked through the following methods:
-
Distributions of all continuous variables were examined.
A correlation test using multi-collinearity was performed to check

variables that measured similar concepts.
Homegeneity of variance was considered when comparing the mean

(SD) of two or more continuous variables to ensure similar spread.
Bivariate data analysis

A bivariate analysis was conducted to determine the association between the
outcome variables and independent factors. Categorical independent variables
were analysed using chi-square and univariate logistic regression and t-test for
continuous independent variables.
Multivariate data analysis
The VAS scale was dichotomized to adherent and non-adherent. Prior to the
multivariate logistic regression analysis, tests for linearity of trends with
dependent variables and multicollinearity among independent variables was
performed. Correlation tests were conducted to determine the inter-relationships
of independent variables.
In multivariate regression analysis all independent variables were entered in the
model while simultaneously removing each insignificant variable when
observing the impact it buffers on other variables. The same method was
Methodology
83
carried out by entering all variables associated with non-adherence at p 0.1 in

univariate analysis. The logistic regression analysis was appropriate to
determine predictors of adherence and assess the independent and multiple
effects of those factors on adherence. The study used the probability level of
95% for all statistical tests at the significant level of p0.05 and all prevalence
estimates were calculated with 95% confidence intervals.
3.7 Ethical Considerations

All research activities involved in this project were carried out according to the
ethical approval from both the University Human Research Ethics Committee,
QUT (approval number 1200000458) and the Ministry of Health of Timor-Leste
National Institute of Health (approval number CHRD-2011-0028) (see
Appendices F & G).
Given that the surveys were administered with TB patients who were chronically
ill, all health and safety precautions were taken into consideration. This included
the training of survey administrators before the data collection, masks were
worn when administering the questionnaires in isolated areas, using a space
with good ventilation and airflow system, otherwise an open area was preferred.
Because the study investigated TB medication adherence and its associated

factors, some of which were sensitive to assess, especially about illegal drug
use and mental health, several methods were applied during the data collection
to minimize the discomfort or emotional distress of the participants. If
participants showed emotional distress, they were referred to their respective
clinics for support and follow-up. Throughout the data collection, it was
emphasised that the information provided was anonymous and confidential. It
was ensured that respondents felt as comfortable as possible and they
understood that each individual participant had control in completing the survey.
There were no complaints reported by health professionals of these clinics in
the study throughout the whole data collection period.
84
Methodology
Participation in this study was voluntary, but a small package containing

noodles, lollies, soap and washing powder (approx. AUD 2) was provided to
patients after the interview as a symbol of appreciation of their participation in
the qualitative data collection. For the main survey, participants were given an
incentive worth 3 US dollars per person (approx. AUD 4) to reimburse their
transport cost to and from their respective clinics.
Methodology
85
86
Methodology
Chapter 4: Qualitative Study Results

INTRODUCTION
Chapter 4 presents the findings from the qualitative study. The chapter begins by
describing the socio-demographic characteristics of participants, and the facilitators
and barriers to their treatment behaviours. In addition, the qualitative study was
conducted to gain a deep understanding of TB treatment adherence from the point of
view of health professionals and TB patients, and to identify important factors to be
included in the final survey.
4.1 Demographic Characteristics

4.1.1 Health Professionals
Of the 10 health professional participants for the interviews, nine were males.
This is common in Timor Leste, especially in the TB program employment,
where there was only one female out of the 13 District Tuberculosis
Coordinators (DTC) in the country at the time of the interviews. This could be
due to the nature of the position, where part of the responsibilities include travel
to the remote areas of the country, which sometimes can be very challenging,
due to the long distance and poor road conditions. The age range of the
interviewees was 25-70 years and the level of education across all participants,
including government staff, clinic and I/LNGO was high school, college or
university degree.
4.1.2 Tuberculosis Patients

The following is the demographic characteristics of the TB patients who
completed the qualitative interviews. The patients ages ranged from 18 to 60
years old at the time of the interviews. There were equal numbers of males and
females, as the participants were purposely selected. All participants were
undertaking treatment in Bairopite Clinic and Comoro CHC in Dili district. Over
half of the participants had not completed high school (n = 6), more than half
Qualitative Study Results
87
were married (n = 6) and a majority (n=7) of the participants were unemployed

(having a stable job) other than farming. Participants who were unemployed
reported receiving financial support from their children for transport costs to
travel to and from the clinic. Others had received social welfare support from the
local government if they were aged 55 years or older.
4.2 Framework of Qualitative Data Analysis

Several important themes were identified in the qualitative data analysis
process.
Treatment-related factors
-
Long distance to the nearest health service and cost associated with
travelling
Poor patient-health care relationships and limited counselling services
Lack of confidentiality and trust
Unreliable service
Poor management of side-effects
Lack of information on treatment
Complexity of treatment regimen.
Personal factors
-
Gender (women tended to put priority on family over themselves)
Age
Depression
Risk behavior (alcohol and tobacco use)
Loss of appetite to take medication
Personal belief and perception of treatment regimen in comparision to

traditional healer
Treatment knowledge (stopped treatment when feeling better, unaware of

consequences for not completing treatment, considered alternative
treatment, such as visiting traditional healers).
88
Consequents of social stigma

-
Avoided taking medication
Prevented patients from disclosing treatment status
Determined level of adherence
Family and Social Support

-
Having children and being a parent
Husband/wife
Friends and relatives
Health professionals and community volunteers.
4.3 How TB Patients Adhered to Treatment

Of the 10 patients who participated in the qualitative study, four were in the
intensive phase of treatment (admitted for treatment within 2 months) and six
were in the continuation phase (had been under treatment for over 2 months but
less than 8 months).
When the patients were asked if at any time they had missed medications, four
patients answered that they had not missed medications because they wanted
to get better. They also further stated that it was important to take care of
themselves and follow the doctors instruction because it was important for their
health.A 53-year-old patient stated I have been sick, [that is why] I come to see
the doctor, and I have been asked to take all my medications I want to get
better, so I follow their instructions. Another added The doctors know what is
best for us, and if they say we have to take all medications, we have to follow
what is asked. It is for our wellbeing.
When asked how important it was for a patient to follow instructions and adhere
to their treatment regimens, their responses clustered around the themes of
following doctors instructions and wanting to get better.
89
A female patient stated:

It is very important for us to take all our medications. I dont know why,
but if I want to get better I will have to take all my medicines and follow
what the doctors ask.
Conversely, six patients stated that they had missed their medication at various
times and gave different reasons for this. The most common reasons were that
they needed to work on the farm, lived far away from the clinics, sometimes had
a conflicting obligation such as a family-related activity, or that they simply
forgot to take their medication. A distinctive factor which affected patients
missed medication is quoted below.
A 41-year-old male patient stated:

Yes, I have missed medication because I was busy with daily activities
working on the farm. Sometimes I have no appetite to take them.
The health professionals also agreed that patients would adhere to the
treatment when they felt there was an improvement in their health. However,
they also recognized that patients who had family priorities may unintentionally
miss their medications, even though they were aware of the consequences.
4.4 Factors Influencing TB Treatment Adherence

4.4.1 Severity of Illness
The severity of the illness had an influence on how soon a patient sought
treatment and how seriously he/she followed the prescribed treatment
regimens. When asked, how do symptoms of TB affect the level of adherence
to treatment?, the health professionals responded as follows:
A CHC staff member stated that those who were too sick would fear death and,
therefore, followed treatment seriously. However, those that were diagnosed
with TB while the sickness was not very severe might not follow the treatment
regimen if the health workers provided limited counselling and information about
90
TB and its treatment process. Furthermore, the CHC staff member stated
Theres a tendency to complete treatment if severity of the disease is high. This
severity is measured through the sputum test on a scale of 1 as moderate to 3
as severe.
A district TB coordinator also added that according to his experience, he

observed that some patients would take medications for a few months but after
the symptoms were gone and they felt better, they would stop. This was worse
for those who lived some distance from available health services while busy
with daily activities. One clinic staff member commented on the difficulties
individual patients faced when they had to travel a long distance to access
health services. Such a burden reduced access to medications for those who
were chronically ill. Consequently, patients might spread the disease further to
their family, relatives and neighbours. Several health professionals also
mentioned that patients who started treatment and experienced drug sideeffects were likely to stop treatment, especially if their sickness was not severe
at the time of diagnosis.
4.4.2 Drug Side-effects

In this study, the impact of medication side-effects on adherence was evident from
the experiences of the health professionals.
A DTC stated:
patients would discontinue treatment if health workers do not describe
clearly what would happen in terms of side-effects when taking the
medication.
While, CHC staff also agreed that if health professionals did not provide sufficient
information, patients were be less likely to adhere to treatment.
Through personal experience, an NGO staff member suggested that:

side-effects are a common challenge for TB patients. Therefore, it is crucial
that health professionals spend at least 15 minutes per session explaining
about the treatment process prior to treatment enrollment.
91
A clinic staff member supported the argument that if patients were given sufficient
information regarding the side-effects of the drugs, they would understand, and
when encountering side-effects, they would still have the willingness to adhere to
the treatment regimens. Additionally, another health professional suggested from
personal experience that identifying the side-effects early and treating them was
vital. At the same time it was also important to strengthen patients' knowledge to
help them to adhere to the treatment regimen. It was reported that, on a number of
occasions, patients who experienced medication side-effects and were unable to
get support tended to drop out of treatment.
4.4.3 Psychological Distress

The psychological wellbeing of individual patients was also an important aspect
that directly determined treatment success. When participants were asked,
"How does mental health affect patients' treatment adherence?", a number of
insights emerged.
One of the health professionals had come across numerous cases throughout
his work. He stated:
"Patients that are mentally affected, even family members would refuse
to look after them. This is due to the fact that some patients do not want
to take their medication at all"
Similar circumstances were reported by two other health professionals:

"Those who have mental health problems would be less likely to adhere
to the treatment regimen. This is because patients do not follow
treatment instructions given"
At the same time, a district TB coordinator stated:

"People with a mental health disorder would not follow what is expected
in the treatment. When the sickness becomes chronic, they go away to
hide or runaway".
92
4.4.4 Traditional Healers

Involvement of traditional healers in TB treatment was a locally derived element
directly impacted on TB treatment. A number of health professionals shared
their experiences of non-adherent patients who had considered the traditional
healers as a form of treatment facility.
A district TB coordinator stated:

when patients experience treatment side-effects, they are more likely to
seek the support of the traditional healers, and it is worse when patients
live far away from available health services.
Another health professional added:

Peoples belief about the traditional healers to have healing power is a
big thing in the country, and some patients take traditional medicine and
falsely guided by the traditional healers that it is ok for them to consume
alcohol and smoke cigarette.
The same health professional continued to share a case where there were 50
patients admitted to a traditional healer for treatment and 37 of these patients
were found to have had positive TB when diagnosed. This suggests that the
traditional healers had a significant influence on patients with TB in the remote
areas of the country.
4.4.5 Education & Knowledge

The health professionals seemed to have some knowledge and understanding
of various factors associated with patients treatments. When asked, Do you
think patients socio-demographic characteristics, such as age, gender, job
status, education, income etc. affect treatment completion?, their responses
were as follows:
93
For education and knowledge, a health professional stated:

"Patients education and their knowledge of TB treatment are vital to
treatment completion. Patients would stop treatment when they feel
better, if they lack knowledge about TB"
A DTC staff member also agreed with this statement and added that those who
lacked knowledge, who had moderate TB symptoms felt frustrated if they were
diagnosed with TB.
At the same time, a district TB coordinator (DTC) stated:

"patients may have good knowledge of treatment. However, if they are
concerned about loss of income as a result of treatment while receiving
no family support, coupled with other factors such as residing distance
from available health services, they will be highly likely to not adhere to
treatment."
A general comment was also given by a health professional:

"those who are considered to be living below the poverty line would have
a lack of food, suffering from malnutrition and living in overcrowding and
unhygienic conditions, and such obstacles would lead them to ignore
treatment"
The statements made by each health professional appeared to be consistent,

indicating that patients with low treatment knowledge, along with poor financial
support and who lived a long distance from available health services were less
likely to fully follow their treatment regimens. Furthermore, TB treatment
knowledge was repeatedly mentioned when discussing other factors, such as
illness severity, drug side-effects, psychological distress and involvement of
traditional healers in the treatment process. This was due to patients stopping
treatment if they felt better after some months of treatment. Moreover, if patients
experienced medication side-effects they were likely to feel stress and drop out
or consider other treatment alternatives such as seeing a traditional healer.
Therefore, TB treatment knowledge appeared to both directly and indirectly
influence other factors that contributed to the overall TB treatment outcome.
94
The level of knowledge and understanding of the treatment process also

depended on the availability and accessibility of information by individual
patients. More importantly, sufficient information on TB treatment provided by
the health professionals to patients at the time of diagnosis was critical to avoid
treatment non-adherence.
4.4.6 Cigarette Smoking and Alcohol Consumption

When the health professionals were asked whether TB patients' lifestyles, such
as alcohol consumption or tobacco smoking affected treatment adherence, a
few statements were given as follows:
A CHC staff member stated:

"Alcohol addicted patients may agree not to consume alcohol within the
first two months. However, once they feel better they would be less likely
to adhere to treatment and might start to drink again. This is common for
those patients who live in homes where many family members are
tobacco smokers and alcohol drinkers."
An NGO staff member agreed with this statement, and said:

"Patients would normally promise that they would not drink and smoke
again at the time of diagnosis and treatment. However, in most cases
when the family members of patients were visited, they constantly
reported that patients had started to smoke and drink again while under
treatment."
Another health professional confirmed that patients' family members have

reported that the patients had started smoking cigarettes and drinking alcohol
again, therefore disobeying the agreement they had with health professionals
prior to treatment.
95
4.4.7 Cost and Treatment Accessiblity

When patients were asked whether age and distance from health services had
an impact on treatment, most highlighted numerous non-adherent factors. The
most common reasons given were, old age, distance from health services, time
and travelling costs.
A 48-year-old female patient stated:

Distance is a problem for us. If we walk, the weather is too hot and the
clinic is too far. If we travel by transport, it costs money and it becomes a
problem because we do not have money to travel all the time.
A 51-year-old stated:
Distance away from the clinic is a problem as I have to walk to the clinic
every day. I want to catch a taxi, but it costs a lot of money. I walk every
day, but with my old age it is very hard.
While under treatment, patients are expected to regularly visit their respective
clinics for checks of their weight, medication side-effects, and general
improvement of their health. A health professional also explained that distance
to health services would involve transport costs, which many could not afford.
Such demands discourage patients from travelling to health centers to refill their
medication and, therefore, failed to comply with the treatment requirements.
In addition, it was difficult for patients with other family priorities that may require
them to travel, especially to the countryside. This was evident in the case of a
male patient who agreed to be interviewed but insisted on finishing quickly as
he had to travel to the countryside for a family-related activity. It is a common
practice in most parts of Timor that throughout the year, people have family
reunions and cultural celebrations where everyone is required to be present.
Under these circumstances, health carers and patients negotiate to find the
most suitable method, that is, provide sufficient drugs for the duration of time
that patients are in the countryside. However, sometimes patients were
96
provided drugs that were sufficient for only a week and may not return on time
for their medication refill, and so they ran out of pills to take.
4.4.8 Patient-Carer Communication

Communication is a fundamental aspect between health professionals and patients
in determining a successful treatment. Poor communication may result in patients
having limited information about the treatment process. Consequently, patients may
seek alternative treatments or drop out if they encounter a treatment-related factor.
When patients were asked whether having a good relationship with the health
professionals will help their treatment progress, their responses are as follows:
A 32-year-old female patient stated:

Yes, it is important to have a good relationship with our doctors,
because if they dont look after us, we may stop treatment.
A 47-year-old male patient added:

Yes, we need to have a good relationship because it is important for us.
The health workers attend to us and they help us with our treatment.
Although it was difficult to probe the patients to describe their relationships with
their health carers further, it was clear that all participants agreed about the
importance of having good relationships with their respective health carers.
While discussing the patient-health professionals relationship, patients were
asked about information they received from the health care providers at the time
of diagnosis. Three of ten patients were able to recall that they were told about
the duration of the treatment. Three participants recalled that they were asked
by the health professionals to take medication until they were cured. One
responded that the information given by health professionals should be kept
secret, while several gave no answer.
97
4.4.9 Family, Social Support & Treatment Confidentiality

In terms of who would remind patients to take their medicines, most stated
family that family members, including children or husnbad/wife reminded them
to take medicine, while one said he had nobody to remind him about medication
time. In relation to how comfortable they were to share information about their
treatments, one patient said it was important to share information about his
treatment with his family because they were supportive of the treatments. Two
people said that they did not tell the family members at all. One respondent said
that he felt very uncomfortable sharing his treatment status because he was
afraid of being stigmatized. For the support each patient received during the
treatment only one patient said he received no support, and nine patients stated
that they had support from family members and friends for associated travel
costs and food. Nine of the ten patients also confirmed that they received
support from family members with their treatment. However, they also admitted
that information regarding their TB diagnosis was not shared.
4.5 Discussion
Analysis of the qualitative study found that participants from the higher level of
government health services highlighted factors that were commonly found in the
scientific literature reviews, such as the impact of poverty, overcrowded
environments and poor access to health services. In comparison, those who
worked for I/LNGOs were more likely to share their understanding of the
weaknesses in the system, for example, a shortage of human resources and
lack of training for health professionals, resulting in poor delivery of services.TB
patients understood the importance of treatment adherence and were aware of
the consequences of poor non-adherence, though it was difficult to probe them
to provide further details about the treatment process due to their limited
experience with participating in a survey. However, both patients and health
professionals provided important insights on TB treatment in this specific
population.
98
The current national tuberculosis program (NTP) strategy in Timor-Leste

encourages patients to undertake treatment while living at home, as in-patient
clinics are too costly (Ministry of Health Timor Leste, 2010). However, patients
who are chronically ill with contagious TB, and are considered to have a low
socioeconomic status while living in remote areas distant from accessible health
services are eligible for short-term support to live in TB specialist clinics.
Patients with these circumstances are also given the option to stay with their
family. They are encouraged to stay at home as long as there is a family
member to care for them, agree to adhere to their treatment and regularly visit
the clinics as scheduled.
The qualitative study found that both health professionals and patients had
similar perceptions of TB medication adherence. Both agreed that poor access,
the long distance from and to health services, and treatment-associated costs
were major issues affecting treatment adherence. These findings are supported
by Martins et al. (2008) in a qualitative study in Timor-Leste and a systematic
review of qualitative research on TB by Munro, Lewin, Smith, et al. (2007). In
their study, they found that health professionals also believed lower adherence
to treatment was due to poor nutritional support (Munro, Lewin, Smith, et al.,
2007), and limited understanding of the treatment process (Ayisi et al., 2011). If
they lacked knowledge at the time of admission, once they encountered
treatment side-effects they were more likely to consider alternative treatments
such as seeing a traditional healer (Martins et al., 2008).
In the current study, two health professionals suggested that aged people
normally lacked knowledge about TB, yet they seemed to strictly follow
treatment as they wanted to get better. Furthermore, few added that aged
patients who lacked knowledge did not care about discrimination. However,
younger patients with good education and better understanding of TB feared
stigmatization. However, they cited instances where well-educated patients
tended to not adhere to treatment because they were afraid of being noticed
and feared being stigmatized. Furthermore, patients families sometimes found
it hard to believe that their family members had TB.
99
Health professionals further suggested that if a patient was given thorough

information regarding the treatment, they would be more likely to adhere and
complete their treatments successfully. The information given by health
professionals prior to admission of the patients to the program should highlight
the importance of the side-effects of the drugs, length of treatment,
consequences of not adhering to treatment, counselling, and all available
supports.
Most health professionals stated that employed patients were less likely to
adhere to their treatment. This was due to being more concerned about losing
their jobs than completing their treatment. Most health professionals agreed that
patients who were educated and had substantial knowledge of TB diseases had
higher adherence to their treatment, especially if they had been given a
thorough explanation of the treatment process. A review of the qualitative
literature found that both health professionals and patients may interpret the
themes of illness and wellness differently (Munro, Lewin, Smith, et al., 2007)..
The review also reported that patients knowledge, attitudes, beliefs and
interpretation about the disease were likely to be determinants of the overall
treatment outcome (Munro, Lewin, Smith, et al., 2007).
Patients highlighted that the most common factors associated with nonadherence were lack of finance to support travel from and to health services
(Munro, Lewin, Smith, et al., 2007). Patients also stated that personal
characteristics, such as being old limited their ability to travel independently to a
clinic. A systematic review of the qualitative research by Munro, Lewin, Smith,
et al. (2007) discussed the organisation of treatment and care for TB patients,
where their access to health facilities depended on the availability of transport,
distance required to travel and their physical condition (Munro, Lewin, Smith, et
al., 2007).The findings of the current research were consistent with Martins et
al. (2008) results. Participants in the current study stated that having a good
relationship with health professionals and stable family and social support was
vital to the treatment process, and health professionals consistently stated that
having good treatment knowledge was important for treatment success (Munro,
Lewin, Smith, et al., 2007).
100
When health professionals were asked about the impact of unhealthy lifestyle
behaviours such as tobacco smoking and alcohol abuse on TB treatment, some
suggested that cigarette smoking and alcohol consumption directly affected
treatment adherence. Some stated that patients were lost to follow up due to
alcohol problems. Additionally, at the time of admission, some smokers agreed
that they would give up smoking during treatment, yet several reports from their
family members indicated that they continued to smoke. Some of these cases
completed treatment over a longer duration than expected and were more likely
to default. Peltzer (2014) recommends preventative programs because conjoint
alcohol and tobacco use is associated with low adherence.
Mental health was raised as an important aspect of TB treatment. Seven out of

the ten health professionals stated that treatment was more complicated with
patients who had mental health disorder. Several discussed incidents where
family members did not want to take care of their relative with TB who
repeatedly refused to take their medications. Often, patients experienced
severe mental health due to experiencing no changes in their health while
encountering treatment side-effects. It has been widely reported that some
patients stopped treatment because of adverse side-effects, and did not know
what to do because they were not made aware of such issues (Munro, Lewin,
Smith, et al., 2007).
The health professionals also highlighted that the patients knowledge of TB
treatment was an important aspect in determining treatment success. Martins et
al. (2008) also concluded that patients knowledge of the disease was one of
the predictors of overall treatment completion. However, the qualitative finding
of the current study found that less educated and unemployed patients are
more likely to adhere to treatment compared to those who are educated and
currently employed. This finding contradicts with the prevous qualitative studies
which have reported that better education was directly influence treatment
outcome (Armijos et al., 2008; Ayisi et al., 2011; Baral et al., 2007; Eastwood &
Hill, 2004). This may be explained by the fact that patients who are less
educated more likely to be unemployed. Thus, they have more time to
101
concentrate on their treatments. In comparision, those who are educated and

employed are afraid of being stigmatised in their work place settings as result of
having TB (Armijos et al., 2008; Finnie et al., 2011; Karim et al., 2007; Martins
et al., 2008; Sengupta et al., 2006). A number of barriers identified in the current
study have also been reported by Martins et al. (2008), indicating the
consistencies of the current findings with the previous study on TB in Timor
Leste.
102
Chapter 5: Main Survey Results

INTRODUCTION
Chapter 5 presents the results from the main survey data analysis. The chapter begins
by describing the characteristics of the study sample, including demographic variables,
family and social features, clinical and risk behaviour characteristics (smoking, alcohol
and drug use), health professional-patient communication, psychological distress,
health locus of control, traditional healer, patient knowledge and stigma. The second
section of the chapter provides the estimates of non-adherence to TB treatment among
the sample, measured using the VAS and MARS and the degree of agreement of
adherence and non-adherence between the two measurements. The second section
also reports the subjective non-adherent factors listed by the patients.
The last section of Chapter 5 reports the results of the statistical tests examining a
range of factors associated with the VAS 1-month dose non-adherence. To determine
the associations of the independent variables with VAS non-adherence, categorical
independent variables were analysed using chi-square tests and univariate logistic
regression, while for the continuous independent variables, chi-square and univariate
logistic regression analysis were used. All the variables associated with non-adherence
in the univariate analysis at p 0.1 were entered in the multivariate logistic regression
for the final analysis. Logistic regression was used to determine the correlations of the
outcome variables and to estimate the independent and multiple effects of the selected
independent variables on the dependent variable. To examine the strength and
direction of the observed association, adjusted odds ratios were examined. All
hypotheses were tested at the 95% probability level. A draft paper for the final analysis
of this study for publication is attached in Appendix M.
5.1 Characteristics of study participants

This section presents demographic variables, characteristics of family and
social support, risk behaviours of participants, health professional patient
communication, psychological wellbeing, health locus of control, and traditional
Main Survey Results
103
healers involvement in treatmet, knowledge, social stigma and family and social
support.
5.1.1 Demographic Variables

The characteristics of the participants are summarized in Table 9. There was a
10 percent difference between males and females (55% and 45% respectively),
and the age ranged from 18 to 73 years (M = 38.05, SD = 16.3). Just over a
quarter (28.0%) had no formal education and 26.6% had only reached primary
education. Over one third of the population (38.7%) were unmarried and 61.3%
were currently married. Approximately a quarter of the sample (25.4%) had no
job while for the majority (44.5%), the means of subsistence was from personal
income (i.e. government subsidy for senior citizens) and 38.9% received family
support.
Table 9 Percentages and Frequencies for Demographic characteristics

Demographic characteristic
n (%)
Gender
Male
191 (55)
Female
154 (45)
Age (Min: 18; Max: 73; M = 38.05 16.3)

18-29
138 (39.8)
30-49
89 (25.7)
50+
120 (34.6)
Marital status
Unmarried
134 (38.7)
Married
212 (61.3)
Education (schooling attended)

No formal education
97 (28)
Primary school
92 (26.6)
Secondary/high school
101 (29.2)
College/University
56 (16.2)
Current Job
104
No job
102 (25.4)
Farmer
57 (14.2)
Main Survey Results
Demographic characteristic
n (%)
Labour work
70 (17.5)
Civil servant
24 (6.0)
Student
85 (21.2)
Housewife
43 (10.7)
Other
20 (5.0)
Main source of subsistence

Income
176 (44.5)
Support from husband/wife
62 (15.7)
Support from family
154 (38.9)
Other
4 (1.1)
Wages/family financial support

< 60USD/month
193 (55.6)
> 60USD/month
154 (44.4)
5.1.2 Family and Social Characteristics

Descriptive statistics for the family and social characteristics of the sample are
displayed in Table 10. Almost 90% of the population lived and shared meals
with four or more people: 46.1% with their husband/wife, 42.9% with children
and 34.7% with extended family members. Living alone, with friends or coworkers represented 3.2%, 1.5% and 1.2% of the sample respectively. In terms
of care-giving received, the majority (79.5%) said they took care of themselves,
28.5% from care providers and approximately one in five people received caregiving support from children (19.9%) or a husband/wife (19.6%).
Main Survey Results
105
Table 10 Percentages and Frequencies for Social and Family Characteristics
Family and social characteristics
n (%)
Number of people living in the same house and sharing meals

0
4 (1.2)
1-3
36 (10.4)
4 or more
303 (88.4)
Person(s) living in the same house and sharing meals

Live alone
11 (3.2)
Parents
77 (22.4)
Children
147 (42.9)
Husband/wife
158 (46.1)
Other family members
119 (34.7)
Friends
5 (1.5)
Co-worker(s)
4 (1.2)
Person most often care-giving

Husband/wife
68 (19.6)
Parents
50 (14.4)
Children
69 (19.9)
53 (15.3)
Friend(s)
27 (7.8)
Co-worker(s)
12 (3.5)
Community volunteers
11 (3.2)
Health care providers
99 (28.5)
Take care of myself
276 (79.5)
5.1.3 Alcohol, Tobacco and Illicit drug-use

Almost half the population (44.6%) stated that they had stopped drinking alcohol
at the start of treatment, while 12.7% continued to drink alcohol during
treatment (see Table 11). The percentage of people who had ever smoked was
48.3%. For illicit drug use, 1.2% admitted to having used illicit drugs in the past.
When stratified by gender, it was clearly evident that gender significantly
influenced these behaviours (see Table 12).
106
Main Survey Results
Table 11 Descriptive Statistics for Risk Behaviour and Clinical Characteristics

Variables
Never drink
147
42.6
Used to drink (quit drinkin start of tment)
154
44.6
Currently drinking
44
12.7
Never smoke
179
51.7
Used to smoke
155
44.8
Currently smoking
12
3.5
337
98.8
1.2
Public health services with CBTC
34
9.8
Private clinics
134
38.6
Public health services
69
19.9
Home visit patients
110
31.7
Less than 30 minutes
197
58.6
More than 30 minutes
139
40.4
Alcohol use (last month)
Cigarette smoking status
Illicit drug use

Never
Yes, in the past
Health Services
Travel time to nearest health service
Note. CBTC = Community Based TB Care Program.
Table 12 Chi-Square P-Values for Risk Behavioural Characteristics by Gender

Variables
Male (%)
Female (%)
Status of drinking (alcohol)
p-value
< 0.001
Never
36 (19.0)
111 (75.5)
Used to drink
121 (64.0)
33 (21.4)
Currently drinking
32 (17.0)
12 (27.3)
Cigarette Smoking status
< 0.001
Never smoke
39 (20.5)
140 (89.7)
Used to smoke
140 (73.7)
15 (9.6)
Main Survey Results
Chi-Square
107
Variables
Male (%)
Currently smoke
Chi-Square
Female (%)
11 (5.8)
p-value
1(0.6)
Status of illicit drug use
0.039
Never use
181 (97.8)
156 (100)
Ever use
4 (2.2)
5.1.4 Health Professional and Patient Communications

The descriptive statistics of patient-health care professional communication are
presented in
Table 13. The total mean score (M = 19.14, SD = 3.11), range 5 to 25. The
mean score of each item assessing the information received from their healthcarers shows that almost all participants were satisfied. This is also consistent
with level of satisfaction with the overall communication and support. It ranges 1
- 4, where 1 was very dissatisfied, 4 very satisfied, the scale showed a mean
score of (M = 3.27, SD = 0.54).
Table 13 Means and Standard Deviations for Health Professional-Patient

Communication Item, Total and Overall Satisfaction Scores
Item
M (SD)
Scale of communication between patients and HCPs

HCP told me what the possible side-effects of each of the TB drug are
3.64 (1.075)
HCP told me what the treatment would do
4.02 (.728)
HCP told me how the TB treatment would be
4.04 (.703)
HCP told me the changes to expect in my health when taking TB
4.07 (.741)
medication
Treatment procedure was clearly explained by the HCP
4.10 (.702)
Mean total
19.14 (3.11)
Overall level of satisfaction

Satisfaction with the overall communication support from HCP
3.27 (.541)
Note: Scale, range is 1-5, assessed degree of agreement with each sentence: (1) Strongly
Disagree, (2) Somewhat Disagree, (3) Neither, (4) Somewhat agree (5) Strongly agree.
Satisfaction scale, range is 1-4, assessed degree of satisfaction of support from health professional:
(1) very dissatisfied, (2) somewhat dissatisfied, (3) Somewhat satisfied, (4) Very satisfied.
108
Main Survey Results
5.1.5 Characteristics of Participants' Psychological Distress

Table 14 displays the scores of the degree of psychological distress of the
sample. The total score was calculated by aggregating the items scores
accordingly with the guideline from Kessler Psychological Distress Scale (K10).
A score under 20 was likely to indicate being well, 20-24 was likely to indicate a
mild mental disorder, 25-29 was likely to suggest a moderate mental disorder
and a score of 30 or over was likely to indicate a severe mental disorder. The
mean score for the degree of psychological distress in this sample was 26.83
(SD = 8.0), indicating that individuals in this population likely to suffer a
moderate mental disorder.
Table 14 Means and Standard Deviations for Psychological Distress Item and
Total Scores
Items
In the past 30 days, about how often did you feel tired out for no good
M(SD)
3.14(1.461)
reasons?
In the past 30 days, about how often did you feel nervous?
1.98(1.236)
In the past 30 days, about how often did you feel so nervous that
1.76(1.110)
nothing could calm you down?

In the past 30 days, about how often did you feel hopeless?
2.98(1.397)
In the past 30 days, about how often did you feel restless or fidgety?
2.60(1.351)
In the past 30 days, about how often did you feel so restless you could
2.44(1.310)
not sit still?

In the past 30 days, about how often do you feel depressed?
2.66(1.297)
In the past 30 days, about how often did you feel that everything is an
3.28(1.284)
effort?
In the past 30 days, about how often did you feel so sad that nothing
3.09(1.317)
could cheer you up?

In the past 30 days, about how often do you feel worthless?
2.85(1.427)
Mean total psychological distress
26.83(8.00)
Note: Scale, range is 1-5, assessed psychological distress in the past month. (1) None of the time,
(2) A little of the time, (3) some of the time, (4) Most of the time and (5) All of the time.
Main Survey Results
109
5.1.6 Descriptive Statistics for Health Locus of Control

The descriptive statistics for the Chance Health Locus of Control are presented
in Table 15. As indicated in the table, the mean of this scale (with possible
scores ranging from 9 to 36) for the sample is 29.98 (SD = 3.44). The Chance
Health Locus of Control consists of 9 items that covers internal (I am in control
of my health, the main thing that affects health is what I myself do and if I take
care of myself, I can avoid illness), powerful others (regular visit with my
doctor will avoid illness, I dont feel well, consult with a medical professional),
and chance (luck plays a big part in how soon I will recover from an illness, my
good health is largely a matter of good fortune, if it is meant to be, I will stay
healthy). The scores for this sample population in all 3 health locus of control
domains are almost similar (ranged 3.18-3.42). However, the items describe
that chance or luck determines ones health were with the highest mean scores.
This indicates that the average population in the sample having beliefs that their
health outcomes are strongly determined by chance or luck.
Table 15 Means and Standard Deviations for Chance Health Locus Control Item
and Total Scores
Chance Health Locus of Control items
M (SD)
I am in control of my health
3.28(.531)
The only thing that affects my health is what I myself do
3.30(.570)
If I take care of myself, I can avoid illness
3.38(.522)
Having regular contact with my doctor is the best way to avoid illness
3.34(.509)
When I don't feel well, I should consult a medical professional
3.18(.642)
When I recover from illness, it's because other people have been taking care of me
3.27(.614)
Luck plays a big part in how soon I will recover from an illness
3.36(.522)
My good health is largely a matter of good fortune
3.42(.523)
If it's what meant to be, I will stay healthy
3.42(.555)
Chance HLC
29.98(3.44)
Note: Scale, range is 1-5, assessed degree of agreement with each sentence: (1) Strongly
Disagree, (2) Disagree, (3) Neither (4) Agree, (5) Strongly Agree.
110
Main Survey Results
5.1.7 Descriptive Statistics for Traditional Healer

Table 16 displays the descriptive statistics for the use of traditional medicine.
The mean of this scale (with possible scores ranging from 4 to 20) for the
sample was 11.77 (SD = 2.45).High score indicates that patients have higher
use of traditional healers as a form of health service.
Table 16 Frequencies, Means and Standard Deviations for Traditional Healer
Item and Total Scores
Items
SD
Taking traditional medicine
124
3.48
0.841
Could not be cured at the clinic
121
2.79
0.906
There are no clinics close by
122
2.78
0.905
Cost of transport is too expensive
122
2.75
0.923
Total
121
11.77
2.45
Note: Scale range is 1-5, assessed degree of agreement with each sentence. 1
Strongly disagree, 2 Agree, 3 Neither, 4 Agree, 5 Strongly agree
5.1.8 Tuberculosis knowledge

A descriptive study of tuberculosis knowledge is displayed in Table 17 below.
When participants were given more than one option to choose for the
understanding of TB transmission, 64.4% cited the correct answer, as being
through droplet infections, while 54.2% wrongly recalled it as sharing utensils,
and 13.3% believed that it was through the punishment of bad behaviour. Thirty
nine percent indicated that they will not eat with family members as a result of
having TB. Almost half (47.7%) responded that it was not important to screen
family members that directly in contact with TB patients.
Over 90% of the participants understood that TB was curable with proper
treatment and 87% of the population understood that TB treatment required 6-8
months to be completely cured. However, as high as 86.4% stated that they
could stop treatment after feeling better, even if the treatment duration had not
been reach.
Main Survey Results
111
Table 17 Frequencies and Percentages for Knowledge of TB Patients

Items
Infectious droplets
224
64.4
Sharing utensils
188
54.2
Punishment for bad behaviors
46
13.3
Tuberculosis is caused by
Does a person with TB eat together with members of the family?

Yes
208
61.0
No
133
39.0
Do family members with direct contact need to screen for TB?

Yes
163
52.3
No
179
47.7
Is tuberculosis is curable with proper treatment?

Yes
322
93.6
No
22
6.4
What is the length of treatment time required to cure TB?

0-2 months
15
4.3
6-8 months
301
87.0
Rest of life
1.2
Varies according to individual
26
7.5
Can you discontinue treatment after feeling better, even treatment duration
has not been reached?
Yes
298
86.4
No
47
13.6
5.1.9 Tuberculosis Stigma

Table 18 displays descriptive statistics for patients perceptions of stigma. A set
of questions were asked to assess the patients views on community members
reactions towards those with TB. The majority believed the community felt
uncomfortable about being near those with TB (84.1%), may not want to eat or
drink with friends with TB (83. 5%), relatives (68.4%), kept their distance from
others with TB (78.3%), were afraid of those with TB (75.3%), did not want to
talk to others with TB (65.8%) and would not try to touch others with TB
(50.4%).
112
Main Survey Results
The patients' perspectives towards tuberculosis, showed that 81.7% believed

that people who have TB felt hurt with how others reacted, knowing their TB
status, 62% believed those with TB felt lonely, 53.3% agreed that TB patients
were afraid to tell those outside their family members, 51.5% agreed that TB
patients felt guilty being a burden to their family, and 45.9% believed that those
with TB were afraid to tell family members that they had TB.
Table 18 Patients Perceptions of stigma
Yes
No
n (%)
n (%)
Items
Community perspective towards tuberculosis

Do you think people may not want to eat or drink with friends who
289 (83.5)
57 (16.5)
290 (84.1)
55 (15.9)
Do you think people keep distance from others with TB?
270 (78.3)
75 (21.7)
Do you think people do not want to talk to others with TB?
227 (65.8)
118 (34.2)
Are people afraid of those with TB?
259 (75.3)
85 (24.7)
Do you think people would not try to touch others with TB?
174 (50.4)
171 (49.6)
Do you think people may not want to eat or drink with relatives
236 (68.4)
109 (31.6)
282 (81.7)
63 (18.3)
Do people have TB feel alone?
214 (62.0)
131 (38.0)
Do you think people who have TB are afraid to tell those outside
183 (53.5)
159 (46.5)
177 (51.5)
167 (48.5)
have TB?
Do you think people feel uncomfortable about being near those
with TB?
who have TB?

Patient Perspectives towards tuberculosis
Do you think people who have TB feel hurt with how others react
knowing they have TB?
their family members?

Do you think people who have TB feel guilty because their family
Main Survey Results
113
Yes
No
n (%)
n (%)
Items
has the burden of caring for them?

Do you think people who have TB are afraid to tell their family that 158 (45.9)
186 (54.1)
they have TB?
5.1.10 Social and Family Supports

Descriptive statistics for family social and supports are presented in Table 19.
The mean for disclosure of treatment status was 12.68 (SD = 2.82), with scores
ranging from 5 to 20. For each group, the mean of disclosure level to family was
the highest (M = 3.51, SD = 0.75) and the disclosure level to social organization
was the lowest (M = 1.29, SD = 0.68). High score in disclosing treatment to a
group indicates that patients are more open to share their treatment status.
Similarly, the sources of support include family, friends, neighbour, peer
educator and social organization. Among these sources of support, the mean
total for family support was the highest (M = 12.65, SD =3) and the lowest was
organization support (M = 5.3, SD = 2.02). This means that family provided
more support to patients compare to the other groups, especially the
organization support as it is being the lowest.The level of satisfaction for the
support received also reflected a similar pattern, where family support had the
highest satisfaction (M = 3.3, SD = 0.66) and social organization was the lowest
(M = 1.6, SD = 0.86).
Table 19 Family and Social Support, Mean and Standard Deviation
M(SD)
Disclosure of TB treatment
114
Family
3.51(.750)
Friends
2.58(.992)
Neighbors
2.51(.976)
Peer educators
2.79(.945)
Social organization
1.29(.677)
Main Survey Results

Total
M(SD)
12.68(2.82)
Family support
Help to take TB medication
3.29(.898)
Emotional support
3.17(.880)
Tangible support
3.04(1.026)
Information support
3.16(.854)
Total
12.65(3.00)
Friends Support
2.33(1.037)
Emotional support
2.34(.926)
Tangible support
1.98(.988)
Information support
2.39(.940)
Total
9.03(3.14)
Neighbors Support
2.20(1.012)
Emotional support
2.21(.866)
Tangible support
1.87(.936)
Information support
2.28(.917)
Total
8.53(2.92)
Peer educators support

2.70(.999)
Emotional support
2.51(.981)
Tangible support
1.96(1.062)
Information support
2.47(.936)
Total
9.60(3.212)
Social organization support

1.27(.659)
Emotional support
1.24(.580)
Tangible support
1.14(.453)
Information support
1.26(.623)
Total
5.30(2.02)
Satisfaction level
Family
3.33(.665)
Friends
2.83(.841)
Main Survey Results
115
M(SD)
Neighbors
2.75(.875)
Peer educators
2.97(.851)
Social organization
1.65(.858)
Total
13.56(2.758)
Note: Scale, range is 1-4, assessed level of discloser, support and satisfaction with
family, friends, peer educators and social organizations: (1) not at all, (2), a little, (3)
somewhat, (4) a lot.
5.2 Subjective Non-adherent Factors Listed by Patients

To further explain factors associated with the level of medication adherence, a
number of possible reasons as to why patients do not take their medications
were analysed. Table 20 shows an extensive number of reasons patients
missed their medications. The three most common reasons indicated by this
specific group were that they simply forgot (72.4%), were busy with other things
(64.7%) or were away from home (50.3%). The least common reason for not
taking their medication was that they illicit used drugs (1%). The top 3 reasons
for missed medication were simply forgot or relapse in memory or out of their
normal routines.
Table 20 Percentage and Frequency of Non-adherent Factors Listed by
Patients
Responses
Items
116
N (%)
Simply forgot
207 (72.4)
Felt good
83 (29.0)
were away from home
144 (50.3)
Had too many pills to take
27 (9.4)
Wanted to avoid side effect
33 (11.5)
Clinic is too far
96 (33.6)
Had a change in daily routine
93 (32.5)
Felt like the drug was toxic/harmful
49 (17.1)
Felt asleep/slept through dose time
64 (22.4)
Main Survey Results
Responses
Items
N (%)
Felt sick
31 (10.8)
Felt depressed/overwhelmed
24 (8.4)
Had problems taking pills at specific times
58 (20.3)
Health clinic did not give pills to me
18 (6.3)
Were busy with other things
185 (64.7)
Use drugs
3 (1.0)
Drink alcohol
14 (4.9)
Took them, then get nausea, but did not take the
replacement
Don't have money for eating and health care seeking
Ran out of pills but could not go to the clinic to get
them
54 (18.9)
81 (28.3)
86 (30.1)
Did not want others to notice you taking medication
36 (12.6)
Don't believe in the usefulness of the drug
14 (4.9)
5.3 Level of Adherence (two measures)

The VAS and the MARS scales were dichotomized with 80% cut-off point.
Patients whose scores were 80% or under are deemed to be suboptimal
adherent and 80% or higher are optimally adherent to their regimens. The
results show that the VAS suboptimal adherence was almost 40% and
suboptimal adherence was just over a quarter for the MARS.
5.3.1 Visual Analogue Scale

When considering adherence as a continuous variable on the VAS, the mean
adherence score was 76.24 (SD = 19.29), indicates that the average sample of
adherence was just below the recommended 80% cut-off point. The median
score was 80 and the scores ranged from 10 to 100. Figure 5 displays the
distribution of the VAS adherence scores for the study sample. Figures 5 & 6
shows that the data for VAS and MARS are left skewed. A dichotomous
adherence variable representing optimal and suboptimal adherence was used
Main Survey Results
117
to examine its association with other factors in the logistic regression analysis.
In total, 39.6% of patients whose scores fell below 80% were categorized in the
suboptimal adherence group and 60.4% who had scored equal or greater than
80% were categorized as optimally adherent.
5.3.2 Medication Adherence Rating Scale

The MARS consist of five questions, in which participants are asked to respond
to a statement of each question by circling yes or no to best describe their
behaviour towards medication during the past month. The answer was coded
zero for no and one for yes. When considering the MARS adherence as a
continuous scale, the mean of the adherence score was 3.94 (SD = 1.004), with
scores ranging from 0 to 5.The total values were added and participant who
scored 4 were deemed as optimally adherent and a score of < 4 was
considered to be suboptimally adherent (see Figure 6)
Figure 5. VAS Distribution
118
Main Survey Results
Figure 6. MARS Distribution
5.3.3 Comparison between the Two Measurements

The prevalence of 1-month suboptimal adherence measured by the VAS was
39.6% (134/338), while the past month suboptimal adherence for the MARS
was 25.3% (85/336). The observed agreement between both measurements
was 69.4%, with a Kappa coefficient of 0.323 (SE = 0.052, p < 0.0001). This
suggested that the majority of individuals were classified as either adherent or
as non-adherent by the two measures, while 30.6% had discordant
classifications (see Table 21).
Table 21 Agreement between the VAS and MARS on TB medication adherence

MARS Score
Sub-optimal
VAS Score
Main Survey Results
adherence
Suboptimal
Optimal
adherence
adherence
n (%)
n (%)
58(17.6%)
75(22.7%)
Total
n (%)
133(40.3%)
Optimal adherence
26(7.9%)
171(51.8%)
197(59.7%)
Total
84(25.5%)
246(74.5%)
330(100%)
119
Suboptimal
adherence
Optimal
adherenc
Total
58(17.6%)
26(7.9%)
171(51.8%)
84(25.5%)
246(74.5%)
75(22.7%)
133(40.3%)
197(59.7%)
330(100%)
Note. MARS = Medication Adherence Rating Scale; VAS = Visual

Analogue Scale.
5.4 Univariate Logisctic Regression Analysis

A summary of the univariate logistic regression analysis associated with
treatment adherence is presented in At the same time, those who never
smoked or used to smoke were four times more likely to adhere to treatment
regimens compared to those who smoked while under treatment (OR=0.235, p
< 0.05).
Table 22 and Table 23.The socio-demographic factors of gender, age, marital
status, and education were not significantly associated with treatment
adherence. However, for those who received family financial support or whose
wage was more than USD 60 per month, the likely chance of adhering to
treatment was multiplied 2.96 times compared to those who had a lack of
support or with a salary less than USD60 (OR = 2.96, p < 0.001).
In relation to service providers, the quality of services provided by clinics,
hospitals or community health centers contributed to the likely chance of a
patient adhering to a treatment regimen. Those who believe they received poor
overall services were less likely to adhere to treatment compared to those who
were sufficiently satisfied with the services (OR = 0.79, p = 0.037).
Behavioural risk factors such as alcohol consumption, that is, those who
currently drank alcohol while under treatment were found to be four times less
adherent compared to those who never drank alcohol (OR = 0.24, p < 0.001). At
the same time, those who never smoked or used to smoke were four times
more likely to adhere to treatment regimens compared to those who smoked
while under treatment (OR=0.235, p < 0.05).
120
Main Survey Results
Table 22 Univariate Analysis of Factors Associated with VAS Adherence
Factors
Optimal
Suboptimal
Crude OR
adherence n
adherence n (%)
(95%CI)
p-value
(%)
Socio-demographic factors
Gender
Male
110(59.8)
74(40.2)
Female
94(61.0)
60(39.0)
1.05(0.68-1.03)
18-29
82(60.3)
54(39.7)
30-49
53(60.2)
35(39.8)
0.99(0.58-1.73)
0.992
50+
69(60.5)
45(39.5)
1.01(0.61-1.68)
0.970
Single
63(53.8)
54(46.2)
Married
139(63.5)
80(36.5)
1.45(0.94-2.35)
No formal education
63(66.3)
32(33.7)
Primary school
49(55.7)
39(44.3)
0.64(0.35-1.16)
0.141
57(58.2)
41(41.8)
0.71(0.39-1.28)
0.244
College/University
34(60.7)
22(39.3)
0.78(0.40-1.56)
0.488
60USD/month
94(49.5)
96(50.5)
>60USD/month
110(74.3)
38(25.7)
2.96(1.86-4.71)
Never drink
92(64.3)
51(35.7)
Used to drink
97(64.7)
53(35.3)
1.02(0.63-1.64)
0.953
Currently drinking
13(30.2)
30(69.8)
0.24(0.12-0.50)
<0.001
Non-smoker
107(61.5)
67(38.5)
Used to smoke
93(61.2)
59(38.8)
0.99(0.63-1.54)
0.954
Currently smoking
3(27.3)
8(72.7)
0.24(0.06-0.92)
0.037
197(60.1)
131(39.9)
2(50.0)
2(50.0s)
0.67(0.09-4.78)
0.814
Age
Marital status
0.087
Education
Income/family support
<0.001
Behavioural factors
Tobacco use (last month)
Illicit drug use past month

Never
Yes, in the past
0.685
Note: current smoke and ever smoke were combined due to small cell sizes.
Main Survey Results
121
Table 23 Univariate Analysis of Factors Associated with VAS Adherence cont.
Factors
Optimal
Suboptimal
adherence
adherence
(n = 204)
(n = 134)
Crude OR
(95%CI)
Univariate analysis
p-value
Quality of service received
2.66
2.89
0.79(0.63-0.99)
0.037
HP-patient communication
22.91
23.41
0.96(0.91-1.02)
0.223
Disclosure of treatment
12.61
12.72
0.98(0.91-1.07)
0.726
People remind medication time
0.88
1.48
0.60(0.49-0.74)
<0.001
Social support
44.71
45.38
0.99(0.97-1.02)
0.603
Patients TB knowledge
5.94
5.50
1.34(0.10-1.64)
0.004
Psychological distress
25.70
28.52
0.96(0.93-0.98)
0.003
Patients perception of stigma
7.49
8.74
0.88(0.82-0.95)
0.001
Traditional healer
3.36
6.08
0.93(0.90-0.98)
<0.001
External Health Locus Control
28.13
30.26
0.83(0.78-0.89)
<0.001
A lack of caregivers and people to remind the TB patients of medication time

were significantly associated with non-adherence (OR = 0.60, p < 0.001). The
higher the score on the stigma scale, the more likely an individual patient having
stronger perception of being stigmatized. Those who perceived TB patients to
experience some form of stigma from the community were sub-optimally
adhered to their treatment regimen (OR = 0.88, p = 0.001). Similarly, patients
who visited a traditional healer to seek treatment options were less adherent to
treatment compared to those who did not visit a traditional healer (OR = 0.93, p
< 0.001).
For every unit decrease in the psychological distress scale, the odds of
suboptimal adherent were multiplied (OR = 0.96, p = 0.003). In other words, the
higher the psychological distress score the higher the likely chance of being
non-adherent. Also, for every unit increase in the patients scores on change
health locus of control, the odds of non-adherent were multiplied by (OR = 0.83,
p < 0.001).
122
Main Survey Results
For every unit increase on the knowledge scale score, the odds of being
adherent was multiplied by (OR =1.34, p = 0.004). In other words, patients with
a better TB treatment knowledge optimally adhered to treatment compared to
those that lacked treatment knowledge.
5.5 Univariate Analysis of Participants Perception of Stigma

When conducting Chi-square analysis of individual questions on the stigma scale, the
results showed significant associations of these individual questions with suboptimal
adherence. From the community perspective of tuberculosis those who believed or
may have experienced other people feeling uncomfortable while being near a person
with TB were significantly associated with non-adherence. Participants who believed or
may have experienced people who refused to eat or drink with a person with TB were
also significantly associated with non-adherence.
With the participants beliefs about TB patient perspectives the analysis found that
patients who were afraid to tell other people outside their family that they had TB were
significantly less likely to be adherent. Also, patients who felt guilty that their family
needs to care for them were significantly associated with medication non-adherence,
and those who were afraid to tell family members that they had TB were also
significantly associated with treatment non-adherence (see to Table 24).
Table 24 Univariate Analysis of VAS with Perceive Stigma
Items
Optimal
Suboptimal
adherence
adherence
n (%)
n (%)
Chi-square
Community perspective towards tuberculosis

Do you think people may not want to eat or drink with friends who have TB?
Yes
166 (58.9)
116 (41.1)
No
37 (67.3)
18 (32.7)
0.244
Do you think people feel uncomfortable about being near those with TB?
Main Survey Results
123
Optimal
Suboptimal
adherence
adherence
n (%)
n (%)
Yes
161 (56.9)
122 (43.1)
No
41 (77.4)
12 (22.6)
Items
Chi-square
0.05
Do you think people keep distance from others with TB?

Yes
155 (58.9)
108 (41.1)
No
47 (64.4)
26 (35.6)
0.400

Yes
132 (59.7)
89 (40.3)
No
70 (60.9)
45 (39.1)
Yes
151 (59.7)
102 (40.3)
No
50 (61.0)
32 (39.0)
0.839

0.836
Yes
107 (63.7)
61 (36.3)
No
95 (56.5)
73 (43.5)
0.181
Do you think people may not want to eat or drink with relatives who have TB?
Yes
130 (56.5)
100 (43.5)
No
72 (67.9)
34 (32.1)
0.047

Do you think people who have TB feel hurt because other peoples reaction knowing they have
TB?
Yes
164 (59.9)
124
110 (40.1)
0.835
Main Survey Results
Optimal
Suboptimal
adherence
adherence
n (%)
n (%)
38 (61.3)
24(38.7)
Yes
120 (57.7)
88 (42.3)
No
82 (64.1)
46 (35.9)
Items
No
Chi-square
Do people who have TB feel alone?

0.247
Do you think people who have TB are afraid to tell those outside their family members?
Yes
84 (47.5%)
93 (52.5)
No
116 (74.4)
40 (25.6)
<0.001
Do you think people who have TB feel guilty because their family has the burden of caring for
them?
Yes
80 (46.5)
92 (53.5)
No
121(74.2)
42 (25.8)
<0.001
Do you think people who have TB are afraid to tell their family that they have TB?
Yes
64 (41.8)
89 (58.2)
No
138 (75.8)
44 (24.2)
<0.001
5.6 Multivariate Analysis of Factors Associated with Medication

Adherence
Table 25 presents a number of factors that were found in a multivariate logistic
regression model to influence non-adherence, after taking into account other
risk factors for non-adherence.
After adjustment, factors that remained significantly associated with nonadherence were quality of health service providers, low income or lack of family
Main Survey Results
125
financial support, lack of a care giver or people to remind the participants to

take their medication, low treatment knowledge, use of a traditional healer,
perceive stigmatization and chance health locus of control.
Controlling for other variables in the model, for every unit increase in the income
scale, the odds of adherence was increased by 1.87. Poor health service
provision was significantly associated with non-adherence (OR = 0.72; p <
0.05). Similarly, patients who lacked caregivers or people to remind them of
their medication time were associated with non-adherence (OR = 0.70; p =
0.009).
For every unit increase on the knowledge scale, the odds of adherence were
multiplied by 1.35. In contrast, for stigma, traditional healer and health locus of
control, for every unit increase in each of these scales, the odds of being nonadherent to treatment were multiplied by 0.91, 0.94 & 0.86 respectively.
The model, as a whole, accounted for 32.3% of the variance in TB medication

non-adherence prevalence (R2 = 0.323), to which the service provider variable
contributed 1.8%, the income variable contributed 8.5%, the lack of caregivers
or people to remind medication time variable contributed 9.3%, the treatment
knowledge variable contributed 4.2%, the stigma variable contributed 4.8%, the
traditional healer contributed 5.9% and the health locus of control contributed
11.9% of the variance.
126
Main Survey Results
Table 25 Multivariate Analysis of Factors Associated with VAS Adherence Cont.
Optimal
Suboptimal
adherence
adherence
n (%)
n (%)
Male
110(59.8)
74(40.2)
Female
94(61.0)
60(39.0)
1.10(0.60-1.98)
18-29
82(60.3)
54(39.7)
30-49
53(60.2)
35(39.8)
1.10(0.48-2.21)
0.805
50+
69(60.5)
45(39.5)
1.09(0.39-3.04)
0.871
Single
63(53.8)
54(46.2)
Married
139(63.5)
80(36.5)
1.84(0.81-4.15)
No formal education
63(66.3)
32(33.7)
Primary school
49(55.7)
39(44.3)
1.32(0.55-3.15)
0.538
57(58.2)
41(41.8)
0.91(0.31-2.64)
0.854
College/University
34(60.7)
22(39.3)
1.63(0.48-5.47)
0.431
<60USD/month
94(49.5)
96(50.5)
>60USD/month*
110(74.3)
38(25.7)
1.87(1.01-3.47)
Never drink
92(64.3)
51(35.7)
Used to drink
97(64.7)
53(35.3)
1.01(0.51-2.33)
0.975
Currently drinking
13(30.2)
30(69.8)
0.41(0.15-1.12)
0.083
Non-smoker
107(61.8)
66(38.2)
Ever smoke
81(59.1)
56(39.4)
1.67(0.63-4.40)
197(60.1)
131(39.9)
2(50.0)
2(50.0s)
Factors
Adjusted OR
p-
(95%CI)
value
Gender
0.778
Age
Marital status
0.143
Education
0.048
Behavioural factor
Tobacco use (last month)
0.302
Illicit drug use

Never
Yes, in the past
Main Survey Results
127
Table 26 Multivariate Analysis of Factors Associated with VAS Adherence Cont.

Optimal
Suboptimal
adherence
adherence
(n=204)
(n=134)
Adjusted OR
(95%CI)
Quality of services received*
2.66
2.89
0.72(0.54-0.95)
0.021
22.91
23.41
1.10(0.96-1.17)
0.233
12.61
12.72
0.92(0.77-1.10)
0.923
People remind medication time*
0.88
1.48
0.70(0.54-0.92)
0.009
Social support
44.71
45.38
0.99(0.97-1.03)
0.929
Patients TB knowledge*
5.94
5.50
1.35(1.05-1.72)
0.018
25.70
28.52
0.99(0.95-1.04)
0.815
Perceive stigmatisation*
7.49
8.74
0.91(0.83-1.00)
0.049
Traditional healer*
3.36
6.08
0.94(0.90-0.98)
0.006
Chance Health Locus Control**
28.13
30.26
0.86(0.79-0.93)
<0.001
Quality
Multivariate analysis
p-value
Note. OR = odds ratio mutually adjusted for all other variables in the table; CI = confidence
interval for true estimate of adjusted odds ratio; p-value = statistical significance of the
adjusted OR. Model used enter method (-2 Log likelihood: 287.334; R2 :0.323)
CONCLUSION
Chapter 5 reports the characteristics of the study sample and an estimate of the
agreement between the VAS and MARS measurements of adherence. The VAS
measurement showed the proportion of patients missing a dose in the previous month
was 39.6%, and MARS estimated 25.3%. The chapter also analysed the VAS
adherence and subjective reasons listed by the participants with various potential
factors directly or indirectly associated with non-adherence. These independent factors
included the participants characteristics, treatment-related factors, and social, family
and individual-related factors. The multivariate logistic regression analysis shows better
quality of services by service providers are more likely to support patients to adhere to
treatment, and having a caregiver to remind of medication time is also found to be a
determinant factor to treatment adherence. However, those with minimum income, no
additional financial support, a lack treatment of knowledge, those who consider using
traditional healers as a form of health service, having strong perception of being
stigmatized or believe that health outcome is determined by chance or luck are less
likely to be adherent to treatment regimens.
128
Main Survey Results
Chapter 6: Discussion
The qualitative study found that the health professionals and patients had
similar perceptions of medication adherence. The health professionals often
thought that adhering to treatment regimens was a challenge due to a lack of
access to health services, patients residing too far from available health
services, poor nutritional intake and a lack of treatment knowledge. The patients
reported that a lack of financial support, limited access to transport, and old age
were the major constraints affecting their goal to achieve optimal treatment
adherence. The list of reasons nominated by patients most often was consistent
with the qualitative results. These barriers included forgetting to take medication
on time, being busy with daily activities or away from home, running out of pills,
and unexpected changes in daily routines.
Using the VAS measurement with 80% cut-off point, the analysis found an
overall 39.6% suboptimal adherence. When broadly comparing this result to
research in some developing countries, the level of non-adherence varies.
Adherence in Timor-Leste was much lower compared to most studies. For
example, a study conducted in Ndola, Zambia that measured adherence in
patients taking TB drugs daily for 8 months found
that 29.8% of patients
stopped taking TB drug at some point during the treatment (Kaona et al., 2004).
A cross-sectional survey assessed TB adherence with numerous tools, in which
the VAS for patients who missed medication in the previous 4 days was
reported to be as low as 7.5% (Nackers et al., 2012). Also, in Jiangsu Province,
China, a study measuring adherence using a structured questionnaire with a
90% cut-off point found a TB medication non-adherence rate of 12.2% (Weiguo
et al., 2009).
Other studies reported lower percentages of non-adherence compared to the

current study (Kaona et al., 2004; Myo Su et al., 2015 ; Nackers et al., 2012;
Weiguo et al., 2009). The differences may be explained by time frames, in
Discussion
129
which this study used one month rather than the four days used in other
studies. Moreover, the previous studies used cut-off points for adherence which
varied between 75% and 100%. However, according to the International Union
Agent Tuberculosis Committee on Prophylaxis (1982), the recommended cut-off
point for TB adherence is 80%. The differing cut-off points used to measure
adherence in other studies may lead to inconsistent findings, where the actual
levels of adherence may be under- or overestimated.
6.1 Factors Associated with TB Medication Adherence

A number of important themes were identified from the analysis of the insights
from both the health professionals and patients in the qualitative study. These
themes
included
service
factors
(service
delivery,
human
resources,
information), discrimination, low social and family support, disease-related

factors, medication side-effects, personal beliefs, age, risk behaviours,
knowledge, and illness severity.
Most of the findings in the qualitative study were consistent with the quantitative
findings, where the critical barriers to adherence were identified as those with
low family income of less than USD 60 per month, alcohol use, high levels of
psychological distress, perception of stigmatization, a preference for a
traditional healer over modern medicine, lack of treatment knowledge and an
external health locus of control. The facilitating adherent factors included the
support of care providers, and patients who had caregivers or somebody to
remind them of medication time.
Various supports of the patients involving peer educators, health professionals,

friends and family were not association with adherence in this study. However,
the analysis showed a significant interaction among these variables with
psychological distress. This may be explained by patients who received
substantially adequate support improved their overall psychological well-being
which, therefore, contributed towards their better treatment adherence.
130
Discussion
There are other factors that can be considered either as barriers to or facilitators
of adherence. For instance, those who lacked treatment knowledge were more
likely to be non-adherent with their treatment regimens. This finding is
consistent with many studies that found that the better the knowledge, the
higher the likelihood of treatment adherence (Armijos et al., 2008; Ayisi et al.,
2011; Demissie & Kebede, 1994; Gebremariam, Bjune, & Frich, 2010; Getahun
& Aragaw, 2001; Hoa, Chuc, & Thorson, 2009; Martins et al., 2008; Sagbakken,
Frich, & Bjune, 2008b; Shargie & Lindtjorn, 2007; Tekle, Mariam, & Ali, 2002).
Additionally, the present study found that disclosure of treatment status could
be either a facilitating factor to gain support or a barrier to treatment. This
depended on the people to whom the patients disclosed their treatment status.
If those people had a better knowledge and understanding of the treatment,
they were more likely to be supportive and vice versa. Consequently, positive
responses could encourage patients to complete treatment and negative
responses could cause patients to feel being discriminated, which was directly
linked with non-adherence. Research has reported that those who disclose their
treatment status were afraid of losing their jobs or feeling hurt, insulted, shamed
and isolated as a result of being stigmatized. Therefore, disclosure of treatment
status can be a barrier to patients under treatment (Do et al., 2013; Karim et al.,
2007; Somma et al., 2008). Despite this, research has suggested that
adherence can be improved through the sharing of treatment status, so that
individual patients receive social support throughout the treatment. Family and
social support benefits patients both directly (positive encouragement and
reinforcement) and indirectly (reduced depression, anxiety and stress relief) (Do
et al., 2013). Moreover, improving the patients and family members knowledge
and understanding of the treatment is likely to be a key factor in adherence
(Armijos et al., 2008; Ayisi et al., 2011; Gebremariam et al., 2010; Getahun &
Aragaw, 2001; Hoa et al., 2009; Martins et al., 2008; Tekle et al., 2002).
The qualitative findings showed the factors of distance from available health
services and a good relationship between the patient and health carers were
potentially associated with medication adherence. However, the survey analysis
Discussion
131
did not detect an association of these variables with medication adherence,

which is consistent with the literature on antiretroviral therapy (Mills et al.,
2006).
6.1.1 Health Service Delivery Factors

The qualitative interviews revealed a number of health service factors
associated with TB treatment. Almost all health professionals suggested that
one of the main reasons for the poor treatment services was the staff being
responsible for multiple roles. This was also associated with the acknowledged
lack of human resources in the health workforce in Timor Leste. (see Appendix
L).
Consequently, TB patients were likely to receive a short duration of
individualized service with insufficient information, and no proper counselling

services, which means they were less likely to adhere to treatment after
enrolment (Awofeso, 2008; Chaulk, Kazandjian, & Panel, 1998; Garner et al.,
2007; Munro, Lewin, Swart, et al., 2007; Munro, Lewin, Smith, et al., 2007;
Nguyen et al., 2003; Orr, 2011; Sumartojo, 1993).
Such poor services to meet patients needs also result in patients extending
their treatment duration, delaying seeking treatment, and/or being more likely to
consider alternative treatments such as seeing a traditional healer (Finnie et al.,
2011). Suboptimal adherent factors listed by patients in the quantitative findings
also suggested that 30% of the sample ran out of drugs and could not get them
and more than a quarter of the sample (33.6%) believed that the clinic was too
far. These findings are consistent with the results of the univariate and
multivariate analyses, where those who received poor services from certain
community health centers (CHC) or clinics were associated with suboptimal
adherence.
Additionally, the health professionals in the qualitative study stated that the
CHCs and clinics across the country may not have complete testing tools,
involving an X-ray, sputum culture, skin test and other necessary equipment to
assist in the early detection of TB and enrolment in treatment. As a
consequence, in some cases patients were likely to wait a long time for the final
132
Discussion
results before starting treatment. Hence, for those who required immediate
treatment, many were encouraged to travel to other clinics or CHCs to receive
TB testing. Such processes required time and effort from individual patients.
Consequently, they discouraged patients from using health facilities, especially
if they had to travel a long distance to an available health service. The interview
results from the patients also confirmed the health professionals views, in which
they found that distance from a health services was a major constraint in
treatment adherence. This is especially a challenge when patients have to travel
to available health services for drug refills, as patients are responsible for
associated travel costs that many cannot afford. The current findings are
consistent with a study in Ethiopia, which reported that a lack of access to
treatment facilities was a major factor in treatment non-compliance (Shargie &
Lindtjrn, 2007).
A study by Needham et al. (2004), which investigated barriers to TB medication

adherence in Zambia, found that the cost of transport to travel to and from
health services accumulated to a considerable16% of patients' monthly income.
This percentage doubled if patients required a caregiver to accompany them
(Needham et al., 2004). Furthermore, quantitative studies in developing
countries have also found that distance to health services is a barrier to TB
treatment adherence (Awofeso, 2008; Garner et al., 2007; Gebremariam et al.,
2010; Ministry of Health Timor Leste, 2011; Munro, Lewin, Swart, et al., 2007;
Munro, Lewin, Smith, et al., 2007; Nackers et al., 2012; Needham et al., 2004;
Nezenega, Gacho, & Tafere, 2013; Sagbakken et al., 2008a; Tachfouti et al.,
2012; Thiam et al., 2007; Zvavamwe & Ehlers, 2008). However, the findings of
the present study on distance travelled to and from available health services
were inconsistent with findings from the previous studies. This may be
explained by the majority of the participants admitting to treatment in Dili, the
capital city of Timor Leste, and temporarily residing with family and relatives,
which allowed them easy access to health services. Therefore, further studies
on TB treatment for people travelling from different places to detect associations
between distance and overall treatment outcomes are warranted.
Discussion
133
6.1.2 Social and Economic Factors

Social and economic factors contributed to overall TB treatment adherence. If
individual patients were socially connected and economically stable, it was
more likely that these factors influenced their treatment adherence in a positive
way. However, those who experienced social isolation and economic instability
would likely to find it difficult to meet the expectations of medication adherence
throughout the treatment process. This was likely to be due to patients who
underwent treatment being absent from their jobs or lacking the ability to earn,
and, therefore, being less likely to sustain an income. Such concerns may be
likely to affect their psychological well-being, knowing that it would be six to
eight months before the treatment could be completed. Furthermore, patients
may feel guilty that they may become a burden to their family members who
have to look after them throughout the duration of the treatment. This was
supported with evidence which showed that 51.5% of the patients indicated that
they experienced guilt because their family had the burden of caring for them
(Table 18).
Financial security for a patient undergoing treatment has been identified as one
of the most important factors in adherence. A number of studies have found that
low income can be a barrier to TB treatment adherence (Armijos et al., 2008;
Cambanis et al., 2005; Farmer et al., 1991; Jackson et al., 2006; Needham et
al., 1998). Although TB drugs are free for patients in some countries,
inaccessibility of services, which requires commitments to travel and transport
can cost more than purchasing the drugs (Zvavamwe & Ehlers, 2008).
Additionally, evidence has suggested that many patients estimate the
opportunity cost of treatment, such as time to travel and the implications for
their employment status (Munro, Lewin, Smith, et al., 2007). The results of a
KAP survey in Timor-Leste also supported the finding that financial constraints
are a major issue when it comes to TB treatment (Ministry of Health Timor
Leste, 2011).
The qualitative findings of the present study were consistent with the results of
other studies, while the quantitative analysis also yielded consistency, in which
134
Discussion
those who earned or received financial support of less than USD 60 per month
had higher levels of non-adherence. Adequate income is a vital supporting
element of treatment, as those who have sufficient support are likely to have
enough to spare for transport and other treatment costs, therefore making it
more likely that they successfully complete treatment.
A report from the WHO on the evidence for action to sustain long-term therapies
also identified finance as a contributing factor to adherence (Dick et al., 2003).
Furthermore, a synthesis of cross-sectional survey and qualitative research in
China found that cost was the most cited reason for treatment default. It has
been found to be worse in cases where providers utilize patients as a way of
generating revenues (Qian et al., 2011). A mixed-methods study by Weiguo et
al. (2009) also found that a heavy financial burden was directly associated with
medication non-adherence. Cost associated with treatment forced patients to
give up due to conflicts between work and treatment, treatment-related costs,
expenses exceeding available resources (Dick et al., 2003; Lnnroth et al.,
2001; Martins et al., 2008; Ministry of Health Timor Leste, 2011; Munro, Lewin,
Smith, et al., 2007; Needham et al., 2004; Noyes & Popay, 2007; Qian et al.,
2011; Sengupta et al., 2006; Weiguo et al., 2009), and unexpected expenditure,
that is, the cost of a taxi to a clinic varies depending on the negotiation with the
driver, which is common in developing and third world countries.
Along with financial problems, stigma related to TB has been identified as an

important element that influences those affected by TB (Baral et al., 2007;
Brassard et al., 2008; Dodor et al., 2008; Gibson et al., 2005; Kipp et al., 2011a;
Sengupta et al., 2006; Somma et al., 2008; Yang et al., 2010). Stigma is a
socially constructed phenomenon which reflects prevailing attitudes and
behaviours of others towards individuals affected by a disease or disability
(Harper, 1987; Karim et al., 2007; Somma et al., 2008). The experience of
stigma can cause people to socially isolate themselves from family, health
professionals and other people to avoid becoming a subject of gossip (Baral et
al., 2007). Consequently, patients who encountered stigmatization have severe
suffering, which, in turn, results in non-compliance (Dodor et al., 2008). Several
studies have described the impact of stigma on tuberculosis medication
Discussion
135
adherence (Baral et al., 2007; Dodor et al., 2008; Karim et al., 2007). The
univariate logistic regression analysis in the present study also found a negative
association between patients who did not want others to know that they were
taking medication and their level of treatment adherence. Research has shown
that patients may not want others to know of their regimen intake, as they often
encounter stigmatization, and negative reactions and/or responses related to
their treatment status (Baral et al., 2007; Brassard et al., 2008; Dodor et al.,
2008; Gibson et al., 2005; Harper, 1987; Johansson et al., 1999; Kipp et al.,
2011a; Sengupta et al., 2006; Somma et al., 2008; Yang et al., 2010). Further
analysis in both the univariate and multivariate logistic regression analyses
confirmed that experience of stigma was strongly associated with suboptimal
adherence. Studies of TB stigma and factors associated with adherence, such
as level of education, gender and religion (Dodor & Kelly, 2010; Johansson et
al., 2000; Kipp et al., 2011a; Rubel & Garro, 1992; Sengupta et al., 2006;
Somma et al., 2008; Woith & Larson, 2008; Zhang et al., 2007), incorrect
knowledge of its cause, mode of transmission and treatment (Baral et al., 2007;
Dodor & Kelly, 2010; Eastwood & Hill, 2004; Johansson et al., 2000; Karim et
al., 2007; Long et al., 1999; Rubel & Garro, 1992; Sengupta et al., 2006;
Somma et al., 2008), and contagiousness of TB transmission (Baral et al.,
2007; Dodor et al., 2008; Eastwood & Hill, 2004; Mavhu et al., 2010) have also
been conducted in other countries.
6.1.3 Individual and Disease-Related Factors

The qualitative interviews yielded evidence that health professionals believed
patients were more likely to complete the treatment if their sickness was severe
prior to admission. The most cited reasons were that when the level of sickness
was severe, people were afraid of death and, therefore, were willing to follow
the treatment instructions. Regardless of their ability to comply with the
expectations for the treatment, in these cases, family members were more likely
to share responsibility for the patient's care.
However, health professionals also believed that the patients were less likely to
seriously undertake treatment after two months of enrolling in the program when
136
Discussion
they felt better. In addition, those who had less serious symptoms at the time of
diagnosis did not take treatment seriously as when they started the treatment.
This was due to their belief that they did not have TB, despite the confirmation
of test results indicating they had positive TB. The qualitative findings also
showed that those who had mild TB symptoms tended to feel frustrated when
informed that they had TB. This finding also highlighted that most often those
with mild TB symptoms did stop taking their medication after two months of
treatment due to a lack of knowledge about the treatment process. It was also
likely that the health professionals had provided limited information relating to
the treatment process at the time of the treatment enrolment. Hence, the
patients may have had a limited understanding about the whole treatment
process and the consequences of not fully following treatment as expected. In
comparison, patients who had severe symptoms at the time of diagnosis
adhered to treatment and closely followed the instructions even if they felt
symptomatic improvement over time after taking medication. This also
depended on the level of knowledge and understanding patients had about the
treatment as well as the support they received during treatment. Patients who
experienced symptomatic improvement after enrolling in a treatment were more
likely to follow the treatment regimens as they believed that they were
recovering. Moreover, there were also a number of important factors that
directly influenced the treatment outcomes, such as knowledge and
understanding of the treatment process, in terms of side-effects, and
consequences of not completing the treatment successfully (i.e. death,
development of MDR-TB, further spread of the disease).
Studies have also reported that social connectedness is crucial in encouraging

patients to successfully follow the treatment regimens (Do et al., 2013; Weiguo
et al., 2009). Additionally, those who lacked support are expected to have a
poor treatment outcome. This is evident in the findings of the current study in
which patients who lacked caregivers and people to remind them of medication
times were associated with non-adherence. These findings are consistent with
the results of a study which investigated TB/HIV co-infected patients in Ethiopia
(Eticha & EKassa, 2014)
Discussion
137
Individual risk behaviours, such as alcohol use and cigarette smoking were
found to influence the overall outcome TB treatment. The health professionals
were aware of the negative impact of tobacco smoking and alcohol intake on TB
treatment adherence. Indeed, health professionals collaborated with family
members to work towards ensuring that the patients did not consume alcohol or
smoke tobacco while undergoing treatment. Therefore, this result suggests that
interventions to effectively manage such behaviour for patients in treatment are
vital.
Alcohol consumption has been reported from other settings to be directly

associated with non-adherence. According to an HIV/AIDS study in Vietnam
that investigated factors associated with optimal antiretroviral medication
adherence, alcohol use was found to be significantly associated with medication
non-adherence (Do et al., 2013). Furthermore, alcohol was also frequently
mentioned in the context of TB treatment non-adherence by a study conducted
in Russia (Jakubowiak, Bogorodskaya, Borisov, Danilova, & Kourbatova, 2007).
The findings of the present study were consistent with both these studies,
suggesting that alcohol consumption was significantly associated with
suboptimal medication adherence.
Although some research has shown an influence of illicit drug use on treatment
adherence, the current study found no association. This was likely to be due to
there being only four of the 347 patients in the sample who stated that they had
used drugs in the past. Culturally, in Timor-Leste, illicit drugs are known as
harmful substances, also perhaps many could not afford them if they were
illegally distributed. Previous studies in Timor-Leste have not assessed illicit
drug usage, could be due its perceived irrelevancy and low incidence (Martins
et al., 2008; Ministry of Health Timor Leste, 2011). The current study also found
that less than 1% of the sample have ever taken illicit drug in their past.
Given the direct impact of risk behaviours on patients health and overall
treatment outcome, people with TB who are depressed may also experience
impaired ability to follow treatment (Do, 2011). This may be because within the
treatment context, patients have negative thoughts, anxiety and feelings of
138
Discussion
hopelessness associated with their treatment. Therefore, such feelings may

impact on their ability to cope with the illness, thereby increasing the likelihood
of ceasing treatment because they feel they are unable to control events in their
lives (Abadia et al., 2010; DiMatteo & Martin, 2002; Do et al., 2013). The
present qualitative study indicates that people who had mental health disorders
were not likely to follow treatment. It was also reported that a few incidents had
been encountered where patients who were mentally ill had family and relatives
who were unwilling to take care of them. Some of these findings are consistent
with the qualitative component of a HIV/AIDS adherence study in Vietnam (Do,
2011) and a study in Ethiopia that investigated the pattern of depression in
patients with TB (Adem, Tesfaye, & Mohammed, 2014). In addition, the
univariate analysis of the current study also complimented the notion that
depression was strongly associated with suboptimal medication adherence.
Individual disease-related factors such as duration of treatment and severity of

sickness are more likely to affect the treatment recovery. In a case with TB, the
sickness severity experienced by individual patients varied throughout the
duration of treatment. This particularly depended on how chronic the sickness
was at the time of admission and the length of time a patient was on treatment.
The findings in the qualitative study showed that those who experienced illness
severely were afraid of death, which produced a higher tendency and
willingness to adhere to treatment. Due to the severity of the sickness at the
start of treatment, a patient may be unable to take care of him or herself. Under
these circumstances, it was most commonly observed that a patient would
receive extensive support from immediate family members, explained perhaps
by their fear of losing a loved one.
As explained in Chapter 2, locus of control is a psychological term related to
individuals' beliefs about who is responsible for their own actions, regardless of
it being desirable or undesirable (Do, 2011). Locus of control includes the belief
that chance or luck significantly influences ones destiny, including the
individuals health outcome. Health locus of control is the degree to which an
individuals health is believed to be controlled by internal or external factors.
The causal attributions that individuals assign to their illness are likely to affect
Discussion
139
their way of coping with it. If a treatment regimen recommended by a health

practitioner is consistent with an individuals understanding of what caused his
or her illness, then the individual is more likely to adhere. An individual who
ascribes external factors that are out of his or her control to the cause of the
illness is less likely to be self-motivated to improve his or her health status,
compared to an individual who believes they are in control of what happens to
them.
Do et al. (2013) suggested that adherence to the treatment regimen of

individuals depends upon their attitude. Energetic people are believed to adhere
to treatment better compared to those who have a 'do not care' attitude.
Therefore, those who have a positive attitude, which perhaps is influenced by
the support received, or who personally accept their sickness status are more
likely to adhere to treatment. In comparison, those who believe that their health
is controlled by others or is determined by a deity are likely to let things be, and
accept what will happen. Our study used the multidimensional health locus of
control scale, shortened version, which has been used by Do and her team (Do
et al., 2013). The analysis showed that chance health locus of control was
significantly associated with poor adherence. This result has significant
implications for future care and treatment of TB patients as it can suggest that
health carers can provide health education and effective counselling by
emphasizing the harm of the belief that chance or luck determines an
individuals health outcomes, as such beliefs are likely to negatively influence a
patient's adherence.
The use of traditional healers for treatment is common across many societies.
This may be explained by the strong traditional beliefs that many people still
have about traditional healers influence on their health. In Sub-Saharan African
countries, approximately 40% of people use traditional healers, according to
some studies (Salaniponi et al., 2000; Wilkinson, Gcabashe, & Lurie, 1999).
One study also found that women use traditional healers more often than men
(Eastwood & Hill, 2004). Traditional healers are thought to provide an
understandable explanation of ones health symptoms (Eastwood & Hill, 2004).
The study also found that the use of traditional healers may be related to a lack
140
Discussion
of knowledge, as traditional healers are used more by less educated people. It

is reported that patients who prefer a traditional healer as an alternative source
of treatment delay seeking early treatment and do not follow treatment
recommendations after diagnosis (Finnie et al., 2011). A survey conducted
across five districts (urban & rural) in Timor-Leste indicated that 14.5% of the
total population used traditional healers as some form of health facility and
34.6% considered traditional healers as an option for health care (Ministry of
Health Timor Leste, 2011). Due to the mountainous, high terrain and poor road
conditions in the remote areas of Timor Leste, visiting traditional healers have
been common. In addition, traditional healers may have been approachable and
accessible both socially and geographically. The univariate and multivariate
analysis confirmed patients were less likely to adhere to treatment if they used a
traiditional healer as a form of treatment facility. This may be due to patients
being distracted from the effort to seek and maintain treatment. Given that TB
treatment requires six to eight months for a patient to be fully cured, those who
have a second option for treatment may believe that if a traditional healer can
use magic to treat the sickness, they no longer have to undergo treatment for
such a long period of time. Given the presence of traditional healers, and that
their efforts to cure people with their magical abilities have been falsely believed
to be true for generations, an effort should be made to educate and work
collaboratively with them in providing health care services. This proposed
strategy could be achieved by identifying the existing traditional healers in the
communities where patients reside and provide basic helath education related
to TB which will enable them to conduct case detection and referral to health
services.
6.2 Limitations of the Study

The current study had some limitations that may have affected the validity and
generalizability of the results. The following biases are identified and the
methods used to attempt to minimize these problems are addressed.
Acquiescence bias, also known as the friendliness effect, is created when
respondents tend to agree with what an interviewer presents to them. This
Discussion
141
could have happened if respondents felt the tendency to agree and to give the
right answer as they wished not to offend the interviewer. Because a
considerable percentage of people in Timor-Leste are illiterate and it is most
likely that those who participated had limited experience of being interviewed or
completing a questionnaire, it was possible that acquiescence bias was present.
To reduce such bias, interviewers were expected to provide thorough
information about the research activity. They would stress the importance that it
would be much more helpful if they provided information that was accurate and
true; that is, if they had forgotten to take medication, they were asked to
accurately respond to the questionnaire with statements which best described
their behaviour. To the best of the interviewers ability, brief but thorough
information about the data collection was provided to ensure that the patients
provided accurate information about what they had encountered throughout the
duration of treatment.
Selection bias may have occurred, as it could have been that those who turned
up to replenish their medications at the clinic and agreed to participate in the
study were also those who were more likely to adhere to treatment. Additionally,
there were patients who met the inclusion criteria, but were excluded as they
could not be reached or lived a long distance from the health service. These
patients may have had important information that may have not have been
captured in the study. In an attempt to minimise these problems, data were
collected from patients across nine health services in three districts. For most of
the clinics, the principal investigator worked collaboratively with clinic staff to
randomly select patients through the TB registers.
According to the National Tuberculosis Program, Timor-Leste database, from

January 2013 to September 2013, 1160 TB patients at the age of 18 years or
older were registered. Of these registered patients, 29.9% (n = 347) participated
in the study. Due to limited time and funding, convenience sampling was used
in most of the clinics. The results may be generalizable to these health services,
but not necessarily reflect treatment practices in Timor Leste.
142
Discussion
6.3 Significance of the Study

A review of the literature suggested that the present study was the first mixedmethods study to investigate TB medication adherence in Timor-Leste. The
research provided evidence for the factors associated with medication nonadherence among TB patients.The study provided findings related to service
delivery, socioeconomic and cultural issues, individual and disease related
factors that could be addressed to improve the level of adherence to better
treatment and care in a resource-limited setting in the Southeast Asian region.
A strength of the study is the fairly large sample that enabled the identification
of TB medication non-adherence in different contexts. The team ensured that
the design and implementation of both the qualitative method and quantitative
survey were carefully considered to be appropriate to social and cultural values.
Thorough procedures for survey development and validation were followed
systematically. The instrument was first drafted based on worldwide literature
and revised to integrate findings from the qualitative study. The qualitative study
finding also helped to improve the validity of the quantitative measures. Input
from experts in the field and other relevant fields was incorporated in the
survey. Then, it was translated and back translated by professional language
translators with health backgrounds. The WHO guide for questionnaire
translation was used to ensure that the accuracy and meaning of the questions
was not reduced or lost. Finally, repetitive pilot study testing of the
questionnaire was conducted. Further comments, suggestions and observations
from the pilot study were incorporated to ensure the reliability and validity of the
questionnaire.
There were a number of issues that arose from this study that warrant further
research. The worldwide literature found a significant association between nonadherence and some factors not observed in the present study. It is
recommended that factors should be further researched, such as education
level, patient-health professional communication with family, friends, neighbours
and social supports, treatment disclosure, level of satisfaction with support
Discussion
143
received and distance to nearby service provider. In addition, illicit drug use has
been reported to be significantly correlated with non-adherence, yet, due to a
small number of people admitting to having used drugs in our sample, there
was no significant association found. However, due to unclear policy and free
trading in the country and the likelihood that the number of drug users could
increase, future research on the subject is encouraged. The present study
contributes to the literature on TB medication adherence by presenting new
patterns of factors that were not investigated in previous studies especially
psychological distress, support of caregivers or people to remind medication
time and health locus of control. It is also vital for future research to examine
levels of adherence throughout the whole period of treatment. This is likely to
reveal further knowledge and understanding on levels of adherence in each
month, while allowing the identification of any other behavioural and
psychological factors that may hinder adherence. Medication adherence does
not solely depend on a single factor. It is a complex, yet pivotal subject, and
therefore requires further studies into a range of factors that are associated with
individual patients.
6.4 Public Health Implications

Given a rate of almost 40% of patients having poor adherence, practical, yet
simple guidelines on strategies to manage adherence are important. These may
include guidelines on ongoing counselling for adherence to prescribed doses,
monitoring and follow-up, and methods to identify the influential factors
associated with adherence (Martins et al., 2009). Furthermore, incentives of
reimbursement of travel cost and the expenses of attending treatment centers,
and ongoing financial support for patients who struggle with financial security
are likely encourage and attract patients, thereby reducing the chance of
missing appointments and scheduled drug replenishment. Despite the results of
the randomized control trial study in Timor-Leste indicating that food incentives
had no significant benefits for TB treatment (Martins et al., 2009), studies from
other settings have demonstrated the importance of food incentives on the
overall outcome (Baldwin et al., 2004; Gupta et al., 2009; Ramakrishnan et al.,
144
Discussion
1961). Martins et al. (2009) may not have found a correlation due to
geographical location, because patients who participated in the study had been
undertaking treatment in Dili, where food is plentiful compared to remote areas
of the country. Evidence suggested that nutritional support showed more rapid
clearance of bacteria and radiographic changes in addition to greater weight
gain. Hence, nutritional supplementation assists patients in a fast recovery,
particularly in underdeveloped areas of the world where food is scarce, and it
may be an effective measure of TB control (Ramakrishnan et al., 1961).
The present study found that alcohol consumption impacted on adherence, and
of the 12.7% that drank while under treatment, it was more common with men
(72.7%) than women (27.3%) in Timor-Leste. Therefore, it is recommended that
the national TB care and treatment program should assist patients with alcohol
issues. Counselling and special treatment can be offered in an attempt to
reduce alcohol intake during treatment.
It is crucial to provide effective counselling and treatment of patients with

depressive symptoms, given the likely severity of its effect on adherence.
Depression is common among patients with chronic medical illness, such as
diabetes, HIV/AIDS and TB (Adem et al., 2014; Ciechanowski, Katon, & Russo,
2000; Patel, Simon, Chowdhary, Kaaya, & Araya, 2009). The findings of the
current study indicate that most often, patients experience constant pressure as
a result of having TB. Such feeling may cause them to feel guilty and blame
themselves for having the disease (Nyblade et al., 2003). Adem et al. (2014)
suggested that TB treatment and care should incorporate psychiatric comorbidities to achieve the treatments primary outcomes. Another study also
suggested that treatment, such as antidepressants and psychotherapy are
effective and should be integrated into primary care in managing depression in
low and middle income countries (Adem et al., 2014). Furthermore, Peltzer et
al. (2012) suggested training of providers to improve necessary skills in
screening adult patients with sign of depression, anxiety, and able to assist with
referral and treatment intervention is important. To provide a conceptual model
for health behaviour, a number of psychosocial constructs have also been
Discussion
145
proposed (Ajzen & Fishbein, 1980; Centers of Disease Control and Prevention,
1993; Dick & Lombard, 1997; Dick, Van der Walt, Hoogendoorn, & Tobias,
1996; Glanz, Rimer, & Viswanath, 2008).
Currently, in Timor-Leste, the number of available professional counsellors is

unknown. To date there appears to be very few health workers in TB services
who are qualified in counselling. The MoH in Timor-Leste should encourage
medical students to undertake counselling studies. Meanwhile, capacity building
through adequate training on basic counselling services should be provided to
the current health workers. Adequate treatment of depression will potentially
enhance patients overall health quality, and hence improve treatment
outcomes.
A public health approach to reducing stigma is to work with self-help, advocacy
and support groups. According to a number of studies, access to support
groups is likely to reduce the effects of stigma on patients undergoing treatment
(Demissie, Getahun, & Lindtjorn, 2003; Lyon & Woodward, 2003). Support
groups can enable patients to exchange information and receive mutual
support, which can help improve their self-esteem (Heijnders & van der Meij,
2006).
Social support from family members, friends, and health professionals can be
important for optimal treatment adherence. Social connectedness has been
reported in the field of HIV/AIDS to be an influencing factor on adherence, as
can help with reinforcement, reassurance or encouragement (Do, 2011). Given
that TB patients are likely to interact and receive support from different groups
during treatment, establishing a strong relationship is likely to be vital. Often,
sick patients are cared for by their family members, community health
volunteers and health care professionals, and it has been reported in a case
with HIV/AIDS in other settings that such care can cause the experience of
stigma (Nyblade et al., 2003). An important key aspect to be addressed is to
educate those caring for patients, especially with their knowledge about the
disease, and the appropriate language used to communicate to patients during
146
Discussion
their daily interactions (Kidd & Clay, 2003). Because many TB patients reside in
remote areas of the country, Kidd and Clay (2003) suggested that improved
education can help empowering community members, especially patients
family members who could benefit patients with their treatment. Thus, educating
different groups of people who are involved directly in patients TB treatment
could also directly influence the success of treatment completion.
Strong beliefs associated with traditional healers in Timor-Leste are part of the
culture and their existence has been historically embedded in the community for
generations. Due to the strong beliefs of the elders about traditional healers,
many people can easily be influenced by such beliefs. In Timor-Leste many TB
patients visit traditional healers before seeking proper treatment, and such
behaviour is found to be consistent with patients in other settings (Colvin et al.,
2014). Given the history and background of traditional healers, to address
issues related to TB treatment and to improve overall health outcome of the
population, effective interventions are required. Other societies which have
encountered the same issues have proposed solutions to address such pivotal
concerns. A potential solution to the problem is to blend conventional services
and traditional healers in terms of supervision of TB treatment, provide possible
community TB education, and engage in collaborative case findings and
diagnosis (Wilkinson et al., 1999). Additionally, a study conducted in Zulu-Natal,
South Africa, suggested that traditional healers are potentially an important
resource to be integrated into TB control programs (Colvin, Gumede,
Grimwade, Maher, & Wilkinson, 2003).
In order to achieve a better community-based TB intervention with the

involvement of traditional healers, first, the MoH Timor-Leste could work more
collaboratively with the local health professionals and community members to
identify potential traditional healers. Second, improved training, such as general
knowledge and understanding in identifying and screening individuals with TB
symptoms is important. In addition, it is vital to train and monitor traditional
healers to independently refer patients, conduct DOTS and assist patients with
ongoing treatment (Colvin et al., 2014).
Discussion
147
The current study found no association between adherence and social support.
Also, disclosure of treatment status was not significantly associated with
adherence. However, the present study found a significant interaction of these
variables with psychological distress, which suggested that individuals who
received adequate support were likely to have a better psychological well-being,
and thus be more likely to adhere successfully to treatment. This result
suggests that social support can be an important element which may not
directly impact adherence, but it can reinforce, reassures and buffers the effects
of variables that directly influence adherence. Hence, efforts to strengthen
treatment and care, and the participation of family members and social support
to maintain social connectedness may help improve treatment adherence.
CONCLUSION
Medication adherence for TB is of utmost importance in helping to cure patients,
prevent further spread of the disease and avoid the development of drug
resistance.
Hence,
identifying
patients
prone
to
non-adherence,
and
determining the level of adherence and its influencing factors is crucial. In doing
so, it allows health carers to recognize individuals with special characteristics
that should be carefully supported to maximise treatment adherence. The result
of this study suggested that it is vital to ensure that assistance is given to those
who struggle financially, and attention is needed to screen for sign of
depression and alcohol use. Evidence also suggests that TB care and treatment
services should integrate effective treatment for depression, counselling for
risky behaviours, traditional healers, and the belief that chance or luck
determines health outcomes. Furthermore, to empower patients, support should
be given to maintain social connectedness with family and the community.
148
Discussion
References
Abadia, E., Zhang, J., dos Vultos, T., Ritacco, V., Kremer, K., Aktas, E., . . . Sola, C.
(2010). Resolving lineage assignation on Mycobacterium tuberculosis clinical
isolates classified by spoligotyping with a new high-throughput 3R SNPs based
method.
Infect
Genet
Evol,
10(7),
1066-1074.
doi:
10.1016/j.meegid.2010.07.006
Adem, A., Tesfaye, M., & Mohammed, M. A. (2014). The Prevalence and Pattern of
Depression in Patients with Tuberculosis on Follow up at Jimma University
Specialized Hospital and Jimma Health Center. Med-Science, 3(1), 955-968.
Ahmad, R. A., Mahendradhata, Y., Utarini, A., & de Vlas, S. J. (2011). Diagnostic delay
amongst tuberculosis patients in Jogjakarta Province, Indonesia is related to the
quality of services in DOTS facilities. Tropical Medicine & International
Health, 16(4), 412-423. doi: 10.1111/j.1365-3156.2010.02713.x
Ahmad, S. A., Wray, C. J., Rilo, H. R., Choe, K. A., Gelrud, A., Howington, J., . . .
Matthews, J. B. (2006). Chronic Pancreatitis: Recent Advances and Ongoing
Challenges.
Current
Problems
in
Surgery,
43(3),
135-238.
doi:
http://dx.doi.org/10.1067/j.cpsurg.2005.12.005
Ailinger, R. L., Black, P., Nguyen, N., & Lasus, H. (2007). Predictors of adherence to
Latent Tuberculosis Infection therapy in Latino immigrants. Journal of
Community Health Nursing, 24(3), 191-198.
Ailinger, R. L., Martyn, D., Lasus, H., & Lima Garcia, N. (2010). The effect of a
cultural intervention on adherence to latent tuberculosis infection therapy in
Latino immigrants. Public Health Nursing, 27(2), 115-120. doi: 10.1111/j.15251446.2010.00834.x
Ailinger, R. L., Moore, J. B., Nguyen, N., & Lasus, H. (2006). Adherence to latent
tuberculosis infection therapy among Latino immigrants. Public Health Nursing,
23(4), 307-313.
Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social
behavior. Englewood Cliffs, N.J.: Prentice-Hall.
References
149
Allen, P., & Bennett, K. (2012). SPSS statistics: a practical guide version 20. South
Melbourne, Vic: Cengage Learning Australia.
Amico, K. R., Fisher, W. A., Cornman, D. H., Shuper, P. A., Redding, C. G., KonkleParker, D. J., . . . Fisher, J. D. (2006). Visual analog scale of ART adherence:
association with 3-day self-report and adherence barriers. J Acquir Immune
Defic Syndr, 42(4), 455-459. doi: 10.1097/01.qai.0000225020.73760.c2
Amuha, M. G., Kutyabami, P., Kitutu, F. E., Odoi-Adome, R., & Kalyango, J. N. (2009).
Non-adherence to anti-TB drugs among TB/HIV co-infected patients in Mbarara
Hospital Uganda: Prevalence and associated factors. African Health Sciences,
9(Suppl 1), S8-S15.
Anderson, L. F., Tamne, S., Watson, J. P., Cohen, T., Mitnick, C., Brown, T., . . .
Abubakar, I. (2013). Treatment outcome of multi-drug resistant tuberculosis in
the United Kingdom: retrospective-prospective cohort study from 2004 to 2007.
Euro Surveill, 18(40).
Aparicio, J. P., Capurro, A. F., & Castillo-Chavez, C. (2002). Markers of disease
evolution: the case of tuberculosis. Journal of Theoretical Biology, 215(2), 227237.
Arlt, S., Lindner, R., Rsler, A., & von Renteln-Kruse, W. (2008). Adherence to
Medication in Patients with Dementia: Predictors and Strategies for
Improvement. Drugs & Aging, 25(12), 1033-1047.
Armijos, R. X., Weigel, M. M., Qincha, M., & Ulloa, B. (2008). The meaning and
consequences of tuberculosis for an at-risk urban group in Ecuador. Significado
y consecuencias de la tuberculosis para un grupo urbano de riesgo en Ecuador.,
23(3), 188-197.
Aronson, J. K. (2007). Compliance, concordance, adherence. British Journal of Clinical
Pharmacology, 63(4), 383-384. doi: 10.1111/j.1365-2125.2007.02893.x
Awofeso, N. (2008). Anti-tuberculosis medication side-effects constitute major factor
for poor adherence to tuberculosis treatment... Garner P, Smith H, Munro S,
Volmink J. Promoting adherence to tuberculosis treatment. Bull World Health
Organ 2007;85:404-6. Bulletin of the World Health Organization, 86(3), B-d.
Ayisi, J. G., van't, Hoog, A. H., Agaya, J. A., McHembere, W., Nyamthimba, P. O., . . .
Marston, B. J. (2011). Care seeking and attitudes towards treatment compliance
by newly enrolled tuberculosis patients in the district treatment programme in
150
References
rural western Kenya: a qualitative study. BMC Public Health, 11(1), 515-524.
doi: 10.1186/1471-2458-11-515
Baldwin, M. R., Yori, P. P., Ford, C., Moore, D. A. J., Gilman, R. H., Vidal, C., . . .
Evans, C. A. (2004). Tuberculosis and nutrition: disease perceptions and health
seeking behavior of household contacts in the Peruvian Amazon. The
international journal of tuberculosis and lung disease : the official journal of the
International Union against Tuberculosis and Lung Disease, 8(12), 1484-1491.
Baral, S. C., Karki, D. K., & Newell, J. N. (2007). Causes of stigma and discrimination
associated with tuberculosis in Nepal: a qualitative study. BMC Public Health, 7,
211-211.
Barnhoorn, F., & Adriaanse, H. (1992). In search of factors responsible for
noncompliance among tuberculosis patients in Wardha District, India. Social
Science & Medicine (1982), 34(3), 291-306.
Baylan, O. (2011). [Extensively drug resistant and extremely drug resistant tuberculosis
forms after multi-drug resistant tuberculosis: new faces of the old disease].
Mikrobiyoloji Blteni, 45(1), 181-195.
Behera, D. (2008). Global tuberculosis control 2008: Surveillance, planning, financing.
Indian Journal of Medical Research, 128(1), 89-90.
Behera, D. (2009). Implementing the WHO stop TB strategy: A handbook for national
tuberculosis control programmes. Indian Journal of Medical Research, 130(1),
95-96.
Belilovsky, E. M., Borisov, S. E., Cook, E. F., Shaykevich, S., Jakubowiak, W. M., &
Kourbatova, E. V. (2010). Treatment interruptions among patients with
tuberculosis in Russian TB hospitals. International Journal of Infectious
Diseases, 14(8), e698-e703. doi: http://dx.doi.org/10.1016/j.ijid.2010.03.001
Bell, J., Airaksinen, M. S., Lyles, A., Chen, T. F., & Aslani, P. ( 2007). Concordance is
not synonymous with compliance or adherence. Br J Clin Pharmacol, 64(5),
710-711.
Berkwits, M., & Inui, T. S. (1998). Making use of qualitative research techniques. J Gen
Intern Med, 13(3), 195-199.
Berridge, V. (2005). Making health policy: networks in research and policy after 1945.
Clio Medica (Amsterdam, Netherlands), 75, 5-37.
Blackwell, B. (1979). Treatment compliance. Journal Of Medical Education, 54(5),
443-444.
References
151
Bobrik, A., Danishevski, K., Eroshina, K., & McKee, M. (2005). Prison health in
Russia: the larger picture. Journal Of Public Health Policy, 26(1), 30-59.
Bock, N. N., Sterling, T. R., Hamilton, C. D., Pachucki, C., Wang, Y.-C., Conwell, D. S.,
. . . Vernon, A. (2002). A prospective, randomized, double-blind study of the
tolerability of rifapentine 600, 900, and 1,200 mg plus isoniazid in the
continuation phase of tuberculosis treatment. American Journal Of Respiratory
And Critical Care Medicine, 165(11), 1526-1530.
Brannon, L., & Feist, J. (2000). Health psychology. An introduction to behaviour and
health.
Brassard, P., Anderson, K. K., Menzies, D., Schwartzman, K., & Macdonald, M. E.
(2008). Knowledge and perceptions of tuberculosis among a sample of urban
Aboriginal people. Journal of Community Health, 33(4), 192-198.
Brimnes, N. (2007). Vikings against Tuberculosis: The International Tuberculosis
Campaign in India, 1948-1951. Bulletin of the History of Medicine, 81(2), 407430.
Brouwer, J. A., Boeree, M. J., Kager, P., Varkevisser, C. M., & Harries, A. D. (1998).
Traditional healers and pulmonary tuberculosis in Malawi. The International
Journal Of Tuberculosis And Lung Disease: The Official Journal Of The
International Union Against Tuberculosis And Lung Disease, 2(3), 231-234.
Brown, M. T., & Bussell, J. K. (2011). Medication Adherence: WHO Cares? Mayo
Clinic Proceedings, 86(4), 304-314. doi: 10.4065/mcp.2010.0575
Bumburidi, E., Ajeilat, S., Dadu, A., Aitmagambetova, I., Ershova, J., Fagan, R., &
Favorov, M. O. (2006). Progress Toward Tuberculosis Control and Determinants
of Treatment Outcomes -- Kazakhstan, 2000-2002. MMWR: Morbidity &
Mortality Weekly Report, 55, 11-15.
Buskin, S. E., Gale, J. L., Weiss, N. S., & Nolan, C. M. (1994). Tuberculosis risk factors
in adults in King County, Washington, 1988 through 1990. The American
Journal of Public Health, 84(11), 1750-1756.
Cambanis, A., Yassin, M., Ramsay A, B., Squire, S., Arbide, I., & Cuevas, L. (2005).
Rural poverty and delayed presentation to tuberculosis services in Ethiopia.
Tropical Medicine & International Health, 10(4), 300-305.
Centers fo Disease Control and Prevention. (2008). Trends in tuberculosis -- United
States, 2007 MMWR: Morbidity & Mortality Weekly Report (Vol. 57, pp. 281285).
152
References
Centers of Disease Control and Prevention. (1993). Approaches to Improving

Adherence to Antituberculosis Therapy 74-75.
Chang, K. C., Leung, C. C., & Tam, C. M. (2004). Risk factors for defaulting from antituberculosis treatment under directly observed treatment in Hong Kong. Int J
Tuberc Lung Dis, 8(12), 1492-1498.
Chaulk, C. P., Kazandjian, V. A., & Panel, f. t. P. H. T. G. (1998). Directly Observed
Therapy for Treatment Completion of Pulmonary Tuberculosis: Consensus
Statement of the Public Health Tuberculosis Guidelines Panel. Journal of the
American Medical Association, 279(12), 943-948.
Chiang, Y.-C., Lin, Y.-M., Lee, J.-A., Lee, C.-N., & Chen, H.-Y. (2012). Tobacco
consumption is a reversible risk factor associated with reduced successful
treatment outcomes of anti-tuberculosis therapy. International Journal of
Infectious
Diseases,
16(2),
e130-e135.
doi:
http://dx.doi.org/10.1016/j.ijid.2011.10.007
Ciechanowski, P. S., Katon, W. J., & Russo, J. E. (2000). Depression and diabetes:
Impact of depressive symptoms on adherence, function, and costs. Archives Of
Internal Medicine, 160(21), 3278-3285. doi: 10.1001/archinte.160.21.3278
Clark, P. M., Karagoz, T., Apikoglu-Rabus, S., & Izzettin, F. V. (2007). Effect of
pharmacist-led patient education on adherence to tuberculosis treatment.
American Journal
of
Health-System Pharmacy, 64(5), 497-506. doi:
10.2146/ajhp050543
Colvin, C., Mugyabuso, J., Munuo, G., Lyimo, J., Oren, E., Mkomwa, Z., . . .
Richardson, D. A. (2014). Evaluation of community-based interventions to
improve TB case detection in a rural district of Tanzania. Global Health: Science
and Practice, 2(2), 219-225. doi: 10.9745/ghsp-d-14-00026
Colvin, M., Gumede, L., Grimwade, K., Maher, D., & Wilkinson, D. (2003).
Contribution of traditional healers to a rural tuberculosis control programme in
Hlabisa, South Africa. Int J Tuberc Lung Dis, 7(9 Suppl 1), S86-91.
Connolly, C., Davies, G. R., & Wilkinson, D. (1999). Who fails to complete tuberculosis
treatment? Temporal trends and risk factors for treatment interruption in a
community-based directly observed therapy programme in a rural district of
South Africa. The International Journal Of Tuberculosis And Lung Disease: The
Official Journal Of The International Union Against Tuberculosis And Lung
Disease, 3(12), 1081-1087.
References
153
Coreil, J., Lazardo, M., & Heurtelou, M. (2004). Cultural feasibility assessment of
tuberculosis prevention among persons of Haitian origin in South Florida. J
Immig Health, 6(2), 63-69.
Corless, I. B., Wantland, D., Bhengu, B., McInerney, P., Ncama, B., Nicholas, P. K., . . .
Davis, S. M. (2009). HIV and tuberculosis in Durban, South Africa: adherence
to
two
medication
regimens.
AIDS
Care,
21(9),
1106-1113.
doi:
10.1080/09540120902729932
Courtwright, A., & Turner, A. N. (2010). Tuberculosis and Stigmatization: Pathways
and Interventions. Public Health Reports, 125(Suppl 4), 34-42.
Craig, G. M., Booth, H., Story, A., Hayward, A., Hall, J., Goodburn, A., & Zumla, A.
(2007). The impact of social factors on tuberculosis management. Journal of
Advanced Nursing, 58(5), 418-424. doi: 10.1111/j.1365-2648.2007.04257.x
Creswell, J. W., & Plano Clark, V. L. (2011). Designing and conducting mixed methods
research. Los Angeles: SAGE Publications.
Curry, L. A., Nembhard, I. M., & Bradley, E. H. (2009). Qualitative and mixed methods
provide unique contributions to outcomes research. Circulation, 119(10), 14421452. doi: 10.1161/circulationaha.107.742775
Daniel, W. W. (2009). Biostatistics: a foundation for analysis in the health sciences.
Hoboken, N.J: John Wiley & Sons.
Demissie, M., Getahun, H., & Lindtjorn, B. (2003). Community tuberculosis care
through TB clubs in rural North Ethiopia. Social Sci Med, 56, 2009-2018.
Demissie, M., & Kebede, D. (1994). Defaulting from tuberculosis treatment at the
Addis Abeba Tuberculosis Centre and factors associated with it. Ethiop Med J,
32(2), 97-106.
Dick, J., Jaramillo, E., Maher, D., & Volmink, J. (2003). World Health Organisation:
Adherence to long-term therapies: Evidence for action. Retrieved from:
http://www.who.int/chp/knowledge/publications/adherence_introduction.pdf
Dick, J., & Lombard, C. (1997). Shared vision--a health education project designed to
enhance adherence to anti-tuberculosis treatment. Int J Tuberc Lung Dis, 1(2),
181-186.
Dick, J., Van der Walt, H., Hoogendoorn, L., & Tobias, B. (1996). Development of a
health education booklet to enhance adherence to tuberculosis treatment. Tuber
Lung Dis, 77(2), 173-177.
DiMatteo, M. R., & Martin, L. R. (2002). Health psychology. Boston: Allyn and Bacon.
154
References
Do, H. M. (2011). Antiretroviral therapy (ART) adherence among people living with
HIV/AIDS (PLHIV) in the north of Vietnam : a multi-method approach
Do, H. M., Dunne, M. P., Kato, M., Pham, C. V., & Nguyen, K. V. (2013). Factors
associated with suboptimal adherence to antiretroviral therapy in Viet Nam: a
cross-sectional study using audio computer-assisted self-interview (ACASI).
BMC Infectious Diseases, 13(1).
Dodge, J. A., & Clark, N. M. (1999). Exploring self-efficacy as a predictor of disease
management. Health Education & Behaviour. 26(72-89).
Dodor, E. A., & Kelly, S. J. (2010). Manifestations of tuberculosis stigma within the
healthcare system: The case of Sekondi-Takoradi Metropolitan district in Ghana.
Health
Policy,
98(23),
195-202.
doi:
http://dx.doi.org/10.1016/j.healthpol.2010.06.017
Dodor, E. A., Neal, K., & Kelly, S. (2008). An exploration of the causes of tuberculosis
stigma in an urban district in Ghana. The International Journal Of Tuberculosis
And Lung Disease: The Official Journal Of The International Union Against
Tuberculosis And Lung Disease, 12(9), 1048-1054.
Eastwood, S. V., & Hill, P. C. (2004). A gender-focused qualitative study of barriers to
accessing tuberculosis treatment in The Gambia, West Africa. The International
Journal Of Tuberculosis And Lung Disease: The Official Journal Of The
International Union Against Tuberculosis And Lung Disease, 8(1), 70-75.
Elliott RA, M. J. ( 2009). Standardised assessment of patients' capacity to manage
medications: a systematic review of published instruments. BMC Geriatr, 9(27).
Eticha, T., & EKassa, E. (2014). Non-Adherence to Anti-TB Drugs and Its Predictors
among TB/HIV Co- Infected Patients in Mekelle, Ethiopia. Journal of
Bioanalysis & Biomedicine.
Farmer, P., Robin, S., Ramilus, S., & Kim, J. (1991). Tuberculosis, poverty, and
compliance: lessons from rural Haiti. Semin Respir Infect, 6(4), 254-260.
Figueroa-Munoz, J., Palmer, K., Poz, M., Blanc, L., Bergstrm, K., & Raviglione, M.
(2005). The health workforce crisis in TB control: a report from high-burden
countries. Human Resources for Health, 3(1), 2.
Finnie, R. K. C., Khoza, L. B., van den Borne, B., Mabunda, T., Abotchie, P., & Mullen,
P. D. (2011). Factors associated with patient and health care system delay in
diagnosis and treatment for TB in sub-Saharan African countries with high
References
155
burdens of TB and HIV. Tropical Medicine & International Health, 16(4), 394411. doi: 10.1111/j.1365-3156.2010.02718.x
Fourie, B. (2001). The burden of tuberculosis in South Africa. Tygerberg Medical
Research Council, National Tuberculosis Research Programme.
Galassi, J. P., Schanberg, R., & Ware, W. B. (1992). The Patient Reactions Assessment:
A brief measure of the quality of the patient-provider medical relationship.
Psychological Assessment, 4(3), 346-351. doi: 10.1037/1040-3590.4.3.346
Garner, P., Smith, H., Munro, S., & Volmink, J. (2007). Promoting adherence to
tuberculosis treatment. Bulletin of the World Health Organization, 85(5), 404406. doi: 10.2471/blt.06.035568
Gebremariam, M. K., Bjune, G. A., & Frich, J. C. (2010). Barriers and facilitators of
adherence to TB treatment in patients on concomitant TB and HIV treatment: a
qualitative study. BMC Public Health, 10, 651-659. doi: 10.1186/1471-2458-10651
Getahun, H., & Aragaw, D. (2001). Tuberculosis in rural northwest Ethiopia:
community perspective. Ethiop Med J, 39(4), 283-291.
Gibson, N., Cave, A., Doering, D., Ortiz, L., & Harms, P. (2005). Socio-cultural factors
influencing prevention and treatment of tuberculosis in immigrant and
Aboriginal communities in Canada. Social Science & Medicine (1982), 61(5),
931-942.
Glanz, K., Rimer, B. K., & Viswanath, K. (2008). Health behavior and health
education: theory, research, and practice (4th ed. ed.). San Francisco, CA: John
Wiley & Sons.
Golden, J. (2004). When the Diaspora Returns: Language Choices in Post-Independence
Timor Lorosa'e. 2, p. 118.
Gonzalez-Angulo, Y., Geldenhuys, H., Van As, D., Buckerfield, N., Shea, J., Mahomed,
H., . . . Hatherill, M. (2013). Knowledge and acceptability of patient-specific
infection control measures for pulmonary tuberculosis. American Journal of
Infection Control(0). doi: http://dx.doi.org/10.1016/j.ajic.2012.10.003
Gough, A., & Kaufman, G. (2011). Pulmonary tuberculosis: clinical features and patient
management. Nursing Standard, 25(47), 48-56.
Gross, P. A., Barrett, T. L., Dellinger, E. P., Krause, P. J., Martone, W. J., McGowan, J.
E., Jr., . . . Wenzel, R. P. (1994). Purpose of quality standards for infectious
diseases. Infectious Diseases Society of America. Clinical Infectious Diseases:
156
References
An Official Publication Of The Infectious Diseases Society Of America, 18(3),

421-421.
Gupta, K. B., Gupta, R., Atreja, A., Verma, M., & Vishvkarma, S. (2009). Tuberculosis
and nutrition. Lung India : Official Organ of Indian Chest Society, 26(1), 9-16.
doi: 10.4103/0970-2113.45198
Hand, C. H., & Bradley, C. (1996). Health beliefs of adults with asthma: towards an
understanding of the difference between symptomatic and preventive use of
inhaler treatment. J Asthma, 33, 331338.
Harling, G., Ehrlich, R., & Myer, L. (2008). The social epidemiology of tuberculosis in
South Africa: A multilevel analysis. Social Science & Medicine, 66(2), 492-505.
doi: http://dx.doi.org/10.1016/j.socscimed.2007.08.026
Harper, D. C. (1987). Review of 'Social stigmaThe psychology of marked
relationships'. Rehabilitation Psychology, 32(1), 62-64. doi: 10.1037/h0092827
Harper, M., Ahmadu, F. A., Ogden, J. A., McAdam, K. P., & Lienhardt, C. (2003).
Identifying the determinants of tuberculosis control in resource-poor countries:
insights from a qualitative study in The Gambia. Transactions of the Royal
Society
of
Tropical
Medicine
and
Hygiene,
97(5),
506-510.
doi:
http://dx.doi.org/10.1016/S0035-9203(03)80007-X
Haynes, R. B. (1979). A critical review of the determinants of patient compliance with
therapeutic regimens. In: Haynes R B, Taylor D W, Sackett D L, eds.
Compliance in health care. Baltimore: John Hopkins University Press, 2639.
Heijnders, M., & van der Meij, S. (2006). The fight against stigma: an overview of
stigma-reduction strategies and interventions. Psychology, Health & Medicine,
11(3), 353-363.
Hill, H. (2001). Tiny, Poor and War-Torn: Development Policy Challenges for East
Timor.
World
Development,
29(7),
1137-1156.
doi:
http://dx.doi.org/10.1016/S0305-750X(01)00035-3
Hinman, A. R., Judd, J. M., Kolnik, J. P., & Daitch, P. B. (1976). Changing risks in
tuberculosis. American Journal Of Epidemiology, 103(5), 486-497.
Hoa, N. P., Chuc, N. T. K., & Thorson, A. (2009). Knowledge, attitudes, and practices
about tuberculosis and choice of communication channels in a rural community
in
Vietnam.
Health
Policy,
90(1),
8-12.
doi:
http://dx.doi.org/10.1016/j.healthpol.2008.08.006
References
157
Holtgrave, D. R., & Crosby, R. A. (2004). Social determinants of tuberculosis case rates
in the United States. American Journal of Preventive Medicine, 26(2), 159-162.
doi: http://dx.doi.org/10.1016/j.amepre.2003.10.014
Horsburgh, C. R., Jr., Feldman, S., & Ridzon, R. (2000). Practice guidelines for the
treatment of tuberculosis. Clinical Infectious Diseases: An Official Publication
Of The Infectious Diseases Society Of America, 31(3), 633-639.
Hsieh, C., Lin, L., Kuo, B. I., Chiang, C., Su, W., & Shih, J. (2008). Exploring the
efficacy of a case management model using DOTS in the adherence of patients
with pulmonary tuberculosis. Journal of Clinical Nursing, 17(7), 869-875. doi:
10.1111/j.1365-2702.2006.01924.x
Inui, T. S. (1996). The virtue of qualitative and quantitative research. Ann Intern Med,
125(9), 770-771.
Jackson, S., Sleigh, A., Wang, G., & Liu, X. (2006). Poverty and the economic effects of
TB in rural China. Int J Tuberc Lung Dis, 10(10), 11041110.
Jakubowiak, W. M., Bogorodskaya, E. M., Borisov, S. E., Danilova, I. D., &
Kourbatova, E. V. (2007). Risk factors associated with default among new
pulmonary TB patients and social support in six Russian regions. Int J Tuberc
Lung Dis, 11(1), 46-53.
Jaramillo, E. (1998). Pulmonary tuberculosis and health-seeking behaviour: how to get
a delayed diagnosis in Cali, Colombia. Tropical Medicine & International
Health: TM & IH, 3(2), 138-144.
Jaramillo, E. (1999). Tuberculosis and Stigma: Predictors of Prejudice Against People
with Tuberculosis. Journal of Health Psychology, 4(1), 71-79. doi:
10.1177/135910539900400101
Johansson, E., Long, N. H., Diwan, V. K., & Winkvist, A. (1999). Attitudes to
compliance with tuberculosis treatment among women and men in Vietnam. The
International Journal Of Tuberculosis And Lung Disease: The Official Journal
Of The International Union Against Tuberculosis And Lung Disease, 3(10), 862868.
Johansson, E., Long, N. H., Diwan, V. K., & Winkvist, A. (2000). Gender and
tuberculosis control: perspectives on health seeking behaviour among men and
women in Vietnam. Health Policy (Amsterdam, Netherlands), 52(1), 33-51.
158
References
Johansson, E., & Winkvist, A. (2002). Trust and transparency in human encounters in
tuberculosis control: lessons learned from Vietnam. Qualitative Health
Research, 12(4), 473-491.
Kaona, F. A., Tuba, M., Siziya, S., & Sikaona, L. (2004). An assessment of factors
contributing to treatment adherence and knowledge of TB transmission among
patients on TB treatment. BMC Public Health, 4, 68. doi: 10.1186/1471-2458-468
Karim, F., Chowdhury, A. M. R., Islam, A., & Weiss, M. G. (2007). Stigma, gender, and
their impact on patients with tuberculosis in rural Bangladesh. Anthropology &
Medicine, 14(2), 139-151.
Kebede, A., & Wabe, N. T. (2012). Medication adherence and its determinants among
patients on concomitant tuberculosis and antiretroviral therapy in South west
ethiopia. N Am J Med Sci, 4(2), 67-71. doi: 10.4103/1947-2714.93376
Kelly, H. H., & Michela, L. (1980). Attribution theory and research. Annual Review of
Psychology. 31, 457-501.
Khan, A., Walley, J., Newell, J., & Imdad, N. (2000). Tuberculosis in Pakistan:
sociocultural constraints and opportunities in treatment. Soc Sci Med, 50, 247254.
Kidd, R., & Clay, S. (2003). Understanding and challenging HIV stigma: Toolkit for
action.
King, K. P. (2009). Workplace Performance-PLUS: Empowerment and Voice through
Professional Development and Democratic Processes in Health Care Training.
Performance Improvement Quarterly, 21(4), 55-74.
Kipp, A. M., Pungrassami, P., Nilmanat, K., Sengupta, S., Poole, C., Strauss, R. P., . . .
Van Rie, A. (2011a). Socio-demographic and AIDS-related factors associated
with tuberculosis stigma in southern Thailand: a quantitative, cross-sectional
study of stigma among patients with TB and healthy community members. BMC
Public Health, 11, 675-675.
Lachenbruch, P. A. (1991). Sample Size Determination in Health Studies: A Practical
Manual. Journal of the American Statistical Association, 86(416), 1149-1149.
Lee, W. Y., Ahn, J., Kim, J. H., Hong, Y. P., Hong, S. K., Kim, Y. T., . . . Morisky, D. E.
(2013). Reliability and validity of a self-reported measure of medication
adherence in patients with type 2 diabetes mellitus in Korea. J Int Med Res,
41(4), 1098-1110. doi: 10.1177/0300060513484433
References
159
Leimane, V., Riekstina, V., Holtz, T. H., Zarovska, E., Skripconoka, V., Thorpe, L. E., . .
. Wells, C. D. (2005). Clinical outcome of individualised treatment of multidrugresistant tuberculosis in Latvia: a retrospective cohort study. Lancet, 365(9456),
318-326. doi: 10.1016/s0140-6736(05)17786-1
Leventhal, H., & Cameron, L. (1987). Behavioral theories and the problem of
compliance. Patient Education and Counseling, 10(2), 117-138. doi:
http://dx.doi.org/10.1016/0738-3991(87)90093-0
Liam, C. K., Lim, K. H., Wong, C. M., & Tang, B. G. (1999). Attitudes and knowledge
of newly diagnosed tuberculosis patients regarding the disease, and factors
affecting treatment compliance. The International Journal Of Tuberculosis And
Lung Disease: The Official Journal Of The International Union Against
Tuberculosis And Lung Disease, 3(4), 300-309.
Liefooghe, R., Baliddawa, J., Kipruto, E., Vermeire, C., & De Munynck, A. (1997).
From their own perspective: a Kenyan communitys perception of tuberculosis.
Tropical Medicine & International Health, 2(8), 809-821.
Lienhardt, C., Glaziou, P., Uplekar, M., Lnnroth, K., Getahun, H., & Raviglione, M.
(2012). Global tuberculosis control: lessons learnt and future prospects. Nature
Reviews. Microbiology, 10(6), 407-416. doi: 10.1038/nrmicro2797
Long, N. H., Johansson, E., Diwan, V. K., & Winkvist, A. (1999). Different tuberculosis
in men and women: beliefs from focus groups in Vietnam. Social Science &
Medicine, 49(6), 815-822.
Lnnroth, K., Jaramillo, E., Williams, B. G., Dye, C., & Raviglione, M. (2009). Drivers
of tuberculosis epidemics: The role of risk factors and social determinants.
Social
Science
&
Medicine,
68(12),
2240-2246.
doi:
http://dx.doi.org/10.1016/j.socscimed.2009.03.041
Lnnroth, K., Tran, T.-U., Thuong, L. M., Quy, H. T., & Diwan, V. (2001). Can I afford
free treatment?: Perceived consequences of health care provider choices among
people with tuberculosis in Ho Chi Minh City, Vietnam. Social Science &
Medicine, 52(6), 935-948. doi: http://dx.doi.org/10.1016/S0277-9536(00)001957
Lyon, M., & Woodward, K. (2003). Nonstigmatizing ways to engage HIV-positive
African-American teens in mental health and support services: A commentary. J
Natl Med Assoc, 95, 196-200.
160
References
Ma, Z., Lienhardt, C., McIlleron, H., Nunn, A. J., & Wang, X. (2010). Global
tuberculosis drug development pipeline: the need and the reality. The Lancet,
375(9731), 2100-2109. doi: http://dx.doi.org/10.1016/S0140-6736(10)60359-9
Malcarne, V. L., Fernandez, S., & Flores, L. (2005). Multidimensional Health Locus of
Control Scales--Shortened Version. doi: 10.1037/t15552-000
Full; Text; 999915552_full_001.pdf
Malterud, K. (2001). The art and science of clinical knowledge: evidence beyond
measures and numbers. Lancet, 358(9279), 397-400. doi: 10.1016/s01406736(01)05548-9
Martins, N., Grace, J., & Kelly, P. (2008). An ethnographic study of barriers to and
enabling factors for tuberculosis treatment adherence in Timor Leste.
International Journal of Tuberculosis and Lung Disease, 12(5), pp. 532-537.
Martins, N., Heldal, E., Sarmento, J., Araujo, R. M., Rolandsen, E. B., & Kelly, P. M.
(2006). Tuberculosis control in conflict-affected East Timor, 1996-2004. Int J
Tuberc Lung Dis, 10(9), 975-981.
Martins, N., Kelly, P. M., Grace, J. A., & Zwi, A. B. (2006). Reconstructing
Tuberculosis Services after Major Conflict: Experiences and Lessons Learned in
East Timor. PLoS Med, 3(10), e383. doi: 10.1371/journal.pmed.0030383
Martins, N., Morris, P., & Kelly, P. M. (2009). Food incentives to improve completion
of tuberculosis treatment: randomised controlled trial in Dili, Timor-Leste. BMJ:
British Medical Journal (Overseas & Retired Doctors Edition), 339(7729),
1131-1131.
Mavhu, W., Dauya, E., Bandason, T., Munyati, S., Cowan, F. M., Hart, G., . . .
Chikovore, J. (2010). Chronic cough and its association with TBHIV coinfection: factors affecting help-seeking behaviour in Harare, Zimbabwe.
Tropical Medicine & International Health, 15(5), 574-579. doi: 10.1111/j.13653156.2010.02493.x
McGowan, J. E., Jr., Chesney, P. J., Crossley, K. B., & LaForce, F. M. (1992).
Guidelines for the use of systemic glucocorticosteroids in the management of
selected infections. Working Group on Steroid Use, Antimicrobial Agents
Committee, Infectious Diseases Society of America. J Infect Dis, 165(1), 1-13.
Mesfin, M. M., Newell, J. N., Walley, J. D., Gessessew, A., & Madeley, R. J. (2009).
Delayed consultation among pulmonary tuberculosis patients: a cross sectional
References
161
study of 10 DOTS districts of Ethiopia. BMC Public Health, 9, 1-10. doi:

10.1186/1471-2458-9-53
Mills, E. J., Nachega, J. B., Bangsberg, D. R., Singh, S., Rachlis, B., Wu, P., . . . Cooper,
C. (2006). Adherence to HAART: A Systematic Review of Developed and
Developing Nation Patient-Reported Barriers and Facilitators. PLoS Med, 3(11),
e438. doi: 10.1371/journal.pmed.0030438
Ministery of Health Timor Leste. (2011). Annual Report for TB case finding 2011. Dili:
Ministry of Health Timor Leste.
Ministry of Health Timor Leste. (2007). Health Sector Strategic Plan 2008 - 2012 Dili:
Ministry of Health Timor Leste.
Ministry of Health Timor Leste. (2008). Manual - National Tuberculosis Program
Control Dili: Ministry of Health Timor Leste.
Ministry of Health Timor Leste. (2010). Timor Leste National Stop TB Strategy Plan
2011 - 2015 Dili: Ministry of Health Timor Leste.
Ministry of Health Timor Leste. (2011). Knowledge, Attitude and Practice (KAP)
towards Tuberculosis and Socio-economic Impact of the Disease in Timor-Leste.
Mishra, P., Hansen, E. H., Sabroe, S., & Kafle, K. K. (2005). Socio-economic status and
adherence to TB treatment: A case-control study in a district of Nepal. Journal of
TB and Lung Disease, 9, 1134-1139.
Mohamed, A. I., Pharm, M., Yousif, M. A., Ottoa, P., & Bayoumi, A. (2007).
Knowledge of Tuberculosis: A Survey among Tuberculosis Patients in
Omdurman, Sudan Sudanese Journal of Public Health, 2, 21-28.
Morisky, D. E., Green, L. W., & Levine, D. M. (1986). Concurrent and predictive
validity of a self-reported measure of medication adherence. Medical Care,
24(1), 67-74.
Mulenga, C., Mwakazanga, D., Vereecken, K., Khondowe, S., Kapata, N., Shamputa, I.
C., . . . Rigouts, L. (2010). Management of pulmonary tuberculosis patients in an
urban setting in Zambia: a patients perspective. BMC Public Health, 10, 756763. doi: 10.1186/1471-2458-10-756
Munro, S., Lewin, S., Swart, T., & Volmink, J. (2007). A review of health behaviour
theories: how useful are these for developing interventions to promote long-term
medication adherence for TB and HIV/AIDS? BMC Public Health, 7(1), 104.
Munro, S. A., Lewin, S. A., Smith, H. J., Engel, M. E., Fretheim, A., & Volmink, J.
(2007). Patient Adherence to Tuberculosis Treatment: A Systematic Review of
162
References
Qualitative
Research.
PLoS
Med,
4(7),
e238.
doi:
10.1371/journal.pmed.0040238
Myo Su, K., Win Myint Oo, & Khay Mar, M. (2015 ). Factors influencing adherence to
TB treatment among TB/HIV co-infected patients attending TB clinic at
Mingalardon Specialist Hospital
Nackers, F., Huerga, H., Espi, E., Aloo, A. O., Bastard, M., Etard, J.-F., . . . Bonnet, M.
(2012). Adherence to Self-Administered Tuberculosis Treatment in a High HIVPrevalence Setting: A Cross-Sectional Survey in Homa Bay, Kenya. Plos One,
7(3), e32140. doi: 10.1371/journal.pone.0032140
Naidoo, P., Dick, J., & Cooper, D. (2009). Exploring Tuberculosis Patients' Adherence
to Treatment Regimens and Prevention Programs at a Public Health Site.
Qualitative Health Research, 19(1), 55-70.
Naidoo, P., & Mwaba, K. (2010). Helplessness, Depression, and Social Support Among
People Being Treated For TUberculosis in South Africa Social Behavior &
Personality:
An
International
Journal,
38(10),
1323-1333.
doi:
10.2224/sbp.2010.38.10.1323
Naidoo, P., Peltzer, K., Louw, J., Matseke, G., McHunu, G., & Tutshana, B. (2013).
Predictors of tuberculosis (TB) and antiretroviral (ARV) medication nonadherence in public primary care patients in South Africa: a cross sectional
study. BMC Public Health, 13(1), 1-10. doi: 10.1186/1471-2458-13-396
National Institute for Health and Clinical Excellence. (2008). Medicines Concordance
(Involving Patients in Decisions about Prescribed Medicines).
Needham, D., Godfrey-Faussett, P., & Foster, S. (1998). Barriers to tuberculosis control
in urban Zambia: the economic impact and burden on patientsprior to diagnosis.
Int J Tuberc Lung Dis, 2(10), 811-817.
Needham, D. M., Bowman, D., Foster, S. D., & Godfrey-Faussett, P. (2004). Patient
care seeking barriers and tuberculosis programme reform: a qualitative study.
Health
Policy,
67(1),
93-106.
doi:
http://dx.doi.org/10.1016/S0168-
8510(03)00065-4
Nezenega, Z. S., Gacho, Y. H. M., & Tafere, T. E. (2013). Patient satisfaction on
tuberculosis treatment service and adherence to treatment in public health
facilities of Sidama zone, South Ethiopia. BMC Health Services Research, 13(1),
110-110.
References
163
Ngamvithayapong, J., Winkvist, A., & Diwan, V. (2000). High AIDS awareness may
cause tuberculosis patient delay: Results from an HIV epidemic area, Thailand.
AIDS, 14(10), 1413-1419. doi: 10.1097/00002030-200007070-00015
Nguyen, D., Proulx, J., Westley, J., Thibert, L., Dery, S., & Behr, M. A. (2003).
Tuberculosis in the Inuit community of Quebec, Canada. American Journal of
Respiratory & Critical Care Medicine, 168(11), 1353-1357.
Nouwen, A., Urquhart Law, G., Hussain, S., McGovern, S., & Napier, H. (2009).
Comparison of the role of self-efficacy and illness representations in relation to
dietary self-care and diabetes distress in adolescents with type 1 diabetes.
Psychology and Health, 24(9), 1071-1084.
Noyes, J., & Popay, J. (2007). Directly observed therapy and tuberculosis: how can a
systematic review of qualitative research contribute to improving services? A
qualitative meta-synthesis. Journal of Advanced Nursing, 57(3), 227-243. doi:
10.1111/j.1365-2648.2006.04092.x
Nyblade, L., Pande, R., Mathur, S., MacQuarrie, K., Kidd, R., Banteyerga, H., & Bond,
V. (2003). Disentangling HIVand AIDS stigma in Ethiopia, Tanzania and
Zambia.
O'Donnell, M. R., Wolf, A., Werner, L., Horsburgh, C. R., & Padayatchi, N. (2014).
Adherence in the treatment of patients with extensively drug-resistant
tuberculosis and HIV in South Africa: a prospective cohort study. J Acquir
Immune Defic Syndr, 67(1), 22-29. doi: 10.1097/qai.0000000000000221
Oliveira-Filho, A. D., Barreto-Filho, J. A., Neves, S. J. F., & Lyra Junior, D. P. d.
(2012). Association between the 8-item Morisky Medication Adherence Scale
(MMAS-8) and blood pressure control. Arquivos Brasileiros De Cardiologia,
99(1), 649-658.
Orr, P. (2011). Adherence to tuberculosis care in Canadian Aboriginal populations, Part
1: definition, measurement, responsibility, barriers. International Journal Of
Circumpolar Health, 70(2), 113-127.
Osterberg, L., & Blaschke, T. (2005). Adherence to Medication. N Engl J Med, 353(5),
487-497.
Pan, Y., & Puente, M., de La. . (2005). Census Bureau Guideline for the Translation of
Data Collection Instruments and Supporting Materials: Documentation on how
the Guideline Was Developed. Statistical Research Division.
164
References
Patel, V., Simon, G., Chowdhary, N., Kaaya, S., & Araya, R. (2009). Packages of Care
for Depression in Low- and Middle-Income Countries. PLoS Med, 6(10),
e1000159. doi: 10.1371/journal.pmed.1000159
Pednekar, M. S., & Gupta, P. C. (2007). Prospective study of smoking and tuberculosis
in India. Preventive Medicine, 44(6), 496-498.
Peloquin, C. (2003). Treatment of Tuberculosis. MMWR: Morbidity & Mortality Weekly
Report, 52(RR-11), 1.
Peltzer, K. (2014). Conjoint alcohol and tobacco use among tuberculosis patients in
public primary healthcare in South Africa. South African Journal of Psychiatry,
20, 21-26. doi: http://dx.doi.org/10.7196/sajp.482
Peltzer, K., Naidoo, P., Matseke, G., Louw, J., McHunu, G., & Tutshana, B. (2012).
Prevalence of psychological distress and associated factors in tuberculosis
patients in public primary care clinics in South Africa. BMC Psychiatry, 12, 8989. doi: 10.1186/1471-244X-12-89
Porter, M., Alder, B., & Abraham, C. (1999). Psychology and sociology applied to
medicine: An illustrated colour text.: Churchill Livingstone.
Qian, L., Smith, H., Tuohong, Z., Shenglan, T., & Garner, P. (2011). Patient medical
costs for tuberculosis treatment and impact on adherence in China: a systematic
review. BMC Public Health, 11(Suppl 4), 393-401. doi: 10.1186/1471-2458-11393
Ramakrishnan, C. V., Rajendran, K., Jacob, P. G., Fox, W., & Radhakrishna, S. (1961).
The role of diet in the treatment of pulmonary tuberculosis: an evaluation in a
controlled chemotherapy study in home and sanatorium patients in South India.
Bulletin of the World Health Organization, 25(3), 339-359.
Revenson, T. A., & Schiaffino, K. M. (1990). Development of a contextual social
support measure for use with arthritis populations. Paper presented at the Paper
presented at the Annual Convention of the Arthritis Health Professions
Association, Seattle, WA.
Rieder, H. (1999). Epidemiologic Basis of Tuberculosis Control. Paris: International
Union Against Tuberculosis and ung Disease.
Rotter, J. B. (1966). Generalized expectancies for internal versus external control of
reinforcement. Psychological Monographs: General and Applied, 80(1), 1-28.
doi: 10.1037/h0092976
References
165
Rubel, A. J., & Garro, L. C. (1992). Social and cultural factors in the successful control
of tuberculosis. Public Health Reports, 107(6), 626-636.
Sagbakken, M., Frich, J., & Bjune, G. (2008a). Barriers and enablers in the management
of tuberculosis treatment in Addis Ababa, Ethiopia: a qualitative study. BMC
Public Health, 8(11).
Sagbakken, M., Frich, J. C., & Bjune, G. (2008b). Barriers and enablers in the
management of tuberculosis treatment in Addis Ababa, Ethiopia: a qualitative
study. BMC Public Health, 8, 11. doi: 10.1186/1471-2458-8-11
Salaniponi, F. M., Harries, A. D., Banda, H. T., Kang'ombe, C., Mphasa, N., Mwale, A.,
. . . Boeree, M. J. (2000). Care seeking behaviour and diagnostic processes in
patients with smear-positive pulmonary tuberculosis in Malawi. Int J Tuberc
Lung Dis, 4(4), 327-332.
Schaberg, T., Rebhan, K., & Lode, H. (1996). Risk factors for side-effects of isoniazid,
rifampin and pyrazinamide in patients hospitalized for pulmonary tuberculosis.
Eur Respir J, 9(10), 2026-2030.
Seligman, M. E. P. (1975). Helplessness: On depression, development, and death: W.H.
Freeman & Company
Sengupta,
S.,
Pungrassami,
P.,
Balthip,
Q.,
Strauss,
R.,
Kasetjaroen, Y.,
Chongsuvivatwong, V., & Van Rie, A. (2006). Social impact of tuberculosis in

southern Thailand: views from patients, care providers and the community. The
International Journal Of Tuberculosis And Lung Disease: The Official Journal
Of The International Union Against Tuberculosis And Lung Disease, 10(9),
1008-1012.
Shao, Y., Yang, D., Xu, W., Lu, W., Song, H., Dai, Y., . . . Wang, J. (2011).
Epidemiology of anti-tuberculosis drug resistance in a chinese population:
current situation and challenges ahead. BMC Public Health, 11(1), 110.
Shargie, E. B., & Lindtjorn, B. (2007). Determinants of treatment adherence among
smear-positive pulmonary tuberculosis patients in Southern Ethiopia. PLoS Med,
4(2), e37. doi: 10.1371/journal.pmed.0040037
Shargie, E. B., & Lindtjrn, B. (2007). Determinants of Treatment Adherence Among
Smear-Positive Pulmonary Tuberculosis Patients in Southern Ethiopia. Plos
Medicine, 4(2), e37. doi: 10.1371/journal.pmed.0040037
http://dx.doi.org/10.1371/journal.pmed.0040037
166
References
Shimao, T. (2005). [Tuberculosis and its control--lessons from the past and future
prospect]. Kekkaku: [Tuberculosis], 80(6), 481-489.
Slade, T., Grove, R., & Burgess, P. (2011). Kessler Psychological Distress Scale:
normative data from the 2007 Australian National Survey of Mental Health and
Wellbeing. The Australian And New Zealand Journal Of Psychiatry, 45(4), 308316.
Sockrider, M. M., & Wolle, J. M. (1996). Helping patients better adhere to treatment
regimen. J Respir Dis, 17, 204-216.
Somma, D., Thomas, B. E., Karim, F., Kemp, J., Arias, N., Auer, C., . . . Weiss, M. G.
(2008). Gender and socio-cultural determinants of TB-related stigma in
Bangladesh, India, Malawi and Colombia. The International Journal Of
Tuberculosis And Lung Disease: The Official Journal Of The International
Union Against Tuberculosis And Lung Disease, 12(7), 856-866.
Stop
TB
Partnership.
(2015).
Fast
Facts
on
Tuberculosis
(TB).
from
http://www.stoptb.org/resources/factsheets/fastfacts.asp
Sumartojo, E. (1993). When tuberculosis treatment fails: A social behavioural account
of patient adherence. Am Rev Respir Dis, 147, pp. 1311-1320.
Sumartojo, E. (2000). Structural factors in HIV prevention: Concepts, examples and
implications for research. AIDS, 14, S3S10.
Tachfouti, N., Slama, K., Berraho, M., & Nejjari, C. (2012). The impact of knowledge
and attitudes on adherence to tuberculosis treatment: a case-control study in a
Moroccan region. Pan Afr Med J, 12, 52.
Tahir, M., Sharma, S. K., Rohrberg, D.-S., Gupta, D., Singh, U. B., & Sinha, P. K.
(2006). DOTS at a tertiary care center in northern India: successes, challenges
and the next steps in tuberculosis control. The Indian Journal Of Medical
Research, 123(5), 702-706.
Tekle, B., Mariam, D. H., & Ali, A. (2002). Defaulting from DOTS and its determinants
in three districts of Arsi Zone in Ethiopia. Int J Tuberc Lung Dis, 6(7), 573-579.
Thiam, S., LeFevre, A. M., Hane, F., Ndiaye, A., Ba, F., & Fielding, K. L. (2007).
Effectiveness of a strategy to improve adherence to Tuberculosis treatment in a
resource-poor setting. Journal of the American Medical Association, 297, 380386.
References
167
Thompson, K., Kulkarni, J., & Sergejew, A. (2000a). Reliability and validity of a new
Medication Adherence Rating Scale (MARS) for the psychoses. Schizophrenia
Research, 42(3), 241-247.
Thompson, K., Kulkarni, J., & Sergejew, A. A. (2000b). Reliability and validity of a
new Medication Adherence Rating Scale (MARS) for the psychoses.
Schizophrenia Research, 42(3), 241-247. doi: http://dx.doi.org/10.1016/S09209964(99)00130-9
Thompson, L., & McCabe, R. (2012). The effect of clinician-patient alliance and
communication on treatment adherence in mental health care: a systematic
review. BMC Psychiatry, 12(1), 87.
Tilson, H. H. (2004). Adherence or compliance? Changes in terminology. Ann
Pharmacother, 38(1), 161-162.
Torun, T., Gungor, G., Ozmen, I., Bolukbasi, Y., Maden, E., Bicakci, B., . . . Tahaoglu,
K. (2005). Side effects associated with the treatment of multidrug-resistant
tuberculosis. Int J Tuberc Lung Dis, 9(12), 1373-1377.
United Nation Development Programme. (2014). Human Development Report 2014.
Van Rie, A., Sengupta, S., Pungrassami, P., Balthip, Q., Choonuan, S., Kasetjaroen, Y., .
. . Chongsuvivatwong, V. (2008). Measuring stigma associated with tuberculosis
and HIV/AIDS in southern Thailand: exploratory and confirmatory factor
analyses of two new scales. Tropical Medicine & International Health, 13(1),
21-30. doi: 10.1111/j.1365-3156.2007.01971.x
Wallston, K. A., Wallston, B. S., & DeVellis, R. (1978). Development of the
Multidimensional Health Locus of Control (MHLC) Scales. Health Education
Monographs, 6(2), 160-170.
Weiguo, X., Wei, L., Yang, Z., Limei, Z., Hongbing, S., & Jianming, W. (2009).
Adherence to anti-tuberculosis treatment among pulmonary tuberculosis
patients: a qualitative and quantitative study. BMC Health Services Research, 9,
169-176.
WHO. (1982). Efficacy of various durations of isoniazid preventive therapy for
tuberculosis: five years of follow-up in the IUAT trial. International Union
Against Tuberculosis Committee on Prophylaxis. Bull World Health Organ,
60(4), 555-564.
WHO. (2003b). The DOTS strategy for controlling TB. (04 May 2012). Retrieved from:
http://www.who.int/tb/publications/manual_for_participants_pp51_98.pdf
168
References
WHO. (2006). The Patients' Charter for Tuberculosis Care. Retrieved from:
http://www.stoptb.org/assets/documents/global/plan/IP_OMS_Charte_GB_Epre
uve.pdf
WHO. (2008a). Implementing the Stop TB Strategy: A handbook for national
tuberculosis
control
programmes.
Retrieved
from:
WHO. (2008b). Anti-Tuberculosis Drug Resistance in the World.
WHO. (2009). Management of MDR-TB: A field guide
Retrieved from
WHO. (2010). Global tuberculosis control: key findings from the December 2009 WHO
report Weekly Epidemiological Record (Vol. 85, pp. 69-79). Geneva WHO.
WHO. (2011). The Global Plan to Stop TB 2011-2015. Retrieved from:
http://www.stoptb.org/assets/documents/global/plan/TB_GlobalPlanToStopTB2
011-2015.pdf
WHO. (2012a). Measurement of substance abuse: process of translation and adaptation
of instruments. Retrieved from
WHO. (2012c). Global Tuberculosis Report Geneva WHO.
WHO.
(2013a).
Timor
Leste
Retrieved
30
November,
2015,
from
http://www.who.int/countries/tls/en/
WHO. (2013b). Tuberculosis Financing and Funding Gaps.
WHO. (2014). Noncommunicable diseases country profiles 2014.
WHO. (2014b). Global Tuberculosis Report Geneva
WHO. (2015). WHO Report on the Global Tobacco Epidemic, 2015. Retrieved from
WHO. (2015a). Trade, foreign policy, diplomacy and health: Tuberculosis (TB).
WHO. (2015b). Multidrug-Resistant Tuberculosis (MDR).
WHO. (2015c). Timor-Leste Tuberculosis profile.
Wild, D., Grove, A., Martin, M., Eremenco, S., McElroy, S., Verjee-Lorenz, A., &
Erikson, P. (2005). Principles of Good Practice for the Translation and Cultural
Adaptation Process for Patient-Reported Outcomes (PRO) Measures: report of
the ISPOR Task Force for Translation and Cultural Adaptation. Value In Health:
The Journal Of The International Society For Pharmacoeconomics And
Outcomes Research, 8(2), 94-104.
References
169
Wilkinson, D., Gcabashe, L., & Lurie, M. (1999). Traditional healers as tuberculosis
treatment supervisors: precedent and potential. Int J Tuberc Lung Dis, 3(9), 838842.
Wilkinson, R., & Marmot, M. (2003). Social determinants of health: The solid facts
(2nd Edition ed.): World Health Organization, Copenhagen
Woith, W. M., & Larson, J. L. (2008). Delay in seeking treatment and adherence to
tuberculosis medications in Russia: a survey of patients from two clinics.
International Journal of Nursing Studies, 45(8), 1163-1174.
Yang, L. H., Wonpat-Borja, A. J., Opler, M. G., & Corcoran, C. M. (2010). Potential
stigma associated with inclusion of the psychosis risk syndrome in the DSM-V:
An empirical question. Schizophrenia Research, 120(13), 42-48. doi:
http://dx.doi.org/10.1016/j.schres.2010.03.012
Zhang, T., Liu, X., Bromley, H., & Tang, S. (2007). Perceptions of tuberculosis and
health seeking behaviour in rural Inner Mongolia, China. Health Policy, 81(2
3), 155-165. doi: http://dx.doi.org/10.1016/j.healthpol.2005.12.009
Zierski, M. (1976). Prospects of retreatment of chronic resistant pulmonary tuberculosis
patients. A critical review. Lung, 154(1), 91-102.
Zvavamwe, S., & Ehlers, V. J. (2008). Implementing a community-based tuberculosis
programme in the Omaheke region of Namibia: nurses' perceived challenges.
Health SA Gesondheid, 13(3), 54-68.
Zwi, A. B., Blignault, I., Glazebrook, D., Correia, V., Bateman Steel, C. R., Ferreira, E.,
& Pinto, B. M. (2009). Timor-Leste Health Care Seeking Behaviour Study.
Retrieved from
170
References
Appendices
Appendix A
Worldwide Qualitative Studies on Tuberculosis treatment
adherence
Appendices
171
Appendix A
Table 27 Summary of Worldwide Qualitative Studies on TB treatment
Factors
Categories
Country/Author
Theme
Service provider & health care
professional
Health Service Delivery

Factors
Accessibility: cost, time &
Treatment related cost, transport cost
distance
and economic burden
Availably: Information, presence
of health care worker and quality
of services.
Treatment accessibility
172
Kenya (Ayisi et al., 2011)

Systematic review of qualitative research (Munro, Lewin,
Smith, et al., 2007)
Timor-Leste (Martins et al., 2008)
Nepal (Baral et al., 2007)
22 TB high-burden countries (Figueroa-Munoz et al.,
2005)
Sub-Saharan African Countries (Finnie et al., 2011),
(Needham et al., 2004)
Gambia (Eastwood & Hill, 2004), (Harper et al., 2003)
Vietnam (Hoa et al., 2009), (Lnnroth et al., 2001)
Thailand (Sengupta et al., 2006)
Ho Chi Minh City, Vietnam (Lnnroth et al., 2001)
Ethiopia (Gebremariam et al., 2010)
Russia (Belilovsky et al., 2010)
Hlabisa, South Africa (Connolly et al., 1999)
Thailand (Kipp et al., 2011a), (Sengupta et al., 2006)
Ho Chi Minh City, Vietnam (Lnnroth et al., 2001)
Gambia (Eastwood & Hill, 2004)
Appendices
Directly Observed Treatment ShortCourse, Directly Observed Therapy
Education and knowledge
Gender
Social and Economic Factors
occupation and marital status
Social motivation: family/friend
support & health workers and
social support
Appendices
Family, friends and relative support

Sub-Saharan African Countries (Finnie et al., 2011)
WHO Geneva (Garner et al., 2007)
Gambia (Harper et al., 2003)
Qualitative meta-analysis (Noyes & Popay, 2007)
Gambia, West Africa (Eastwood & Hill, 2004), (Harper et
al., 2003)
WHO Geneva (Garner et al., 2007)
Kenya (Ayisi et al., 2011),
Ecuador (Armijos et al., 2008)
Nepal(Baral et al., 2007)
Vietnam (Hoa et al., 2009)
Canada (Brassard et al., 2008), (Gibson et al., 2005)
Ghana (Dodor et al., 2008)
Vietnam (Johansson et al., 1999), (Johansson et al.,
2000), (Long et al., 1999)
Bangladesh (Karim et al., 2007)
Thailand (Kipp et al., 2011a), (Sengupta et al., 2006)
Bangladesh, India, Malawi and Colombia (Somma et al.,
2008), (Karim et al., 2007)
173
Stigma (individual and community)
Traditional healer, belief, perception,

attitude, behaviour and cultural
factors
Personal and disease related

factors
Treatment, risk behaviour,
knowledge, belief and attitude
towards treatment, mental
health and locus of control
Side-effect, diagnosis, depression
174
Sub-Saharan, South Africa (Naidoo & Mwaba, 2010)

Thailand (Sengupta et al., 2006), (Van Rie et al., 2008)
Bangladesh (Karim et al., 2007)
Sub-Saharan African Countries (Finnie et al., 2011)
Thailand (Sengupta et al., 2006)
Malawi (Brouwer, Boeree, Kager, Varkevisser, & Harries,
1998)
Hlabisa, South Africa (Connolly et al., 1999)
Vietnam (Johansson et al., 1999), (Hoa et al., 2009)
Canada (Gibson et al., 2005),
Colombia (Jaramillo, 1998)
Colombia (Jaramillo, 1998)
Sub-Saharan, South Africa (Naidoo & Mwaba, 2010)
Appendices
Appendix B
Summary of Quantitative Studies
Appendices
175
Appendix B
Summary of quantitative studies on TB

Table 28 Summary of Quantitative Studies on TB
Source
Country
Study Design
(Weiguo et
al., 2009)
China
Mixed method,
multi-stage
sampling
strategy
Research
aim(s)/research
question(s)
To gain insights into

the factors that could
contribute to
adherence.
Measurement
Setting and sample size
Results
Quantitative: structured
questionnaire, highlighting basic
characteristics, SES, treatment
history, adherence to antituberculosis treatment. Nonadherent participants were
presented 16 reasons of nonadherent. Observed treatment was
divided into groups accordingly with
types of group support.
Jiangsu Province, eastern

cost of China. Thirteen
municipalities and 106
counties (districts), with a
total population of 74
million.
Overall proportion of non-adherent was 12.2%

(82/670). Almost half of non-adherent patients
interrupted their treatment intermittently (missed a
total of more than 10% of all doses).
Illiterate, divorced/widowed, lacked health
insurance or were migrants, were more likely to be
non-adherent. The crude ORs (95%CIs) were
2.38(1.37-4.13), 2.42(1.30-4.52), 1.89(1.07-3.32),
1.98(1.03-3.83), respectively.
Qualitative methods: in-depth

interview. A semi-schematized guide
covering general as well as specific
questions was used. Interviews
focused on healthcare seeking
history, knowledge and attitude
towards TB, and their adherence on
treatment.
(Ailinger et
al., 2007)
USA
Descriptive
Study
Examine the
predictors of Latent
Tuberculosis Infection
(LTBI) therapy in
Latino immigrants at a
public health clinic.
176
160 records were randomly

chosen, 2 clients had been
treated with Rifampin and five
records were incomplete.
Therefore, total of 153 for data
analysis.
Adherence was defined as the
number of doses of INH taken
as reported by the client. Full
adherence is defined by the
CDC as completion of 270
doses in 12 months.
Participants were considered
Appendices
A total of 670 patients

were successfully enrolled
for quantitative study.
Nineteen patients were
conveniently selected for
in-depth interview.
Population of Latino
immigrant, majority came
from El Salvador, Bolivia,
or Guatemala. Total
sample of 153 records
were included for analysis
Risk of non-adherent was lower among patients

under direct observations by village doctors or
home visit by health workers with ORs (95%CIs) of
0.19 (0.10-0.36) and 0.23 (0.10-0.51),
respectively.
Main reasons for non-adherent listed by patients
were adverse reaction to drugs (37.8%), relieved
symptoms (26.8%), long course of regimen and
large dose of drugs (15.9%) and worry about
danger of drugs (15.9%)
A total 52 patients (34%) completed the 8 months
of LTBI therapy and were given final month supply.
Seventy two participants (47%) completed 6
months of LTBI therapy. Fifty eight clients (38%)
reported side effect and of those only four cases
(3%) were discontinued with medical advice. Sideeffects occurred most often in the first month of
therapy in 24 patients (16%).
Source
Country
Study Design
Research
aim(s)/research
question(s)
Measurement
Results
non-adherent if they did not

show up for their clinic
appointment and did not
respond to the public health
nurse's phone calls or letter
(Ailinger et
al., 2010)
USA
PreExperimental
design
Examine the effect of

a cultural intervention
(CI) on increasing
adherence to latent
tuberculosis infection
(LTBI) therapy among
Latino immigrants.
Adherence was measured by

the number of pills reported by
the clients, which was
documented in the medical
record.
The total number of pills for 8
months was recorded for this
comparison between the
experimental and the historical
groups.
Public health chest clinic

in an urban country of
Virginia with over 20% of
Latino population. A
random sample of 131
records was drawn from
over 600 records
The intervention group took greater number of pills

than the comparison group and the difference is
statistically significant t (85) = 1.94, p=0.028. The
mean of doses taken by the intervention group
was 157 (5.2 months) and 129 (4.3 months) for the
comparison group. This comes to a difference of
28 doses, which is approximately a month's
treatment.
Almost 70% of the participants were male. Many

participants held misconceptions or lacked of
knowledge, which lead to delaying diagnoses and
treatment non-adherent, as a consequence lead
the disease to spread further. Patients with a prior
personal or family TB history were more likely to
report familiarity with TB (89.3% vs 66%; 2=11;
p=0.001). Furthermore, those with more than nine
years of formal schooling have better recognition
and understanding compare to less educated
counterparts (78.2% vs. 61%; 2=7; p=0.008).
Education was the single most important of
knowledge, beliefs, perceptions and attitudes.
Participants who ever smoke had more negative
treatment outcome compare to those who never
smoke (60% vs. 43.2%, p=0.005). High tobacco
consumption (>20 cigarettes per day) was
significantly associated with decreased odds of
achieving a positive treatment outcome (odds ratio
(OR) 0.23, 95% confidence interval (CI) 0.050.98,
p = 0.047), and even moderate smokers (120
cigarettes per day) registered borderline
insignificance (OR 0.61, 95% CI 0.351.05, p =
0.073) compared to those who did not consume
tobacco.
(Armijos et
al., 2008)
Ecuador
Cross sectional
study
To explore knowledge,
beliefs, perceptions,
and attitudes about
Tuberculosis in a highrisk group in Ecuador.
Convenient sampling. Face to face

interview using structured
questionnaire containing close and
open-ended questions.
A national institute of
hygiene - Public health
facility. There were 212
adults, 18 years or over
who did not have any
known conditions that that
may impede their ability to
respond to questions.
(Chiang et
al., 2012)
China
Retrospective
observational
case control
study
To evaluate the
reversible factors
that could possibly
affect outcomes of
anti-tuberculosis
anti-TB treatment.
Data collected from both electronic

record and medical charts.
Laboratory data including albumin,
aspartate aminotransferase (AST),
alanine aminotransferase (ALT),
total bilirubin, conjugated bilirubin,
and serum alkaline phosphate (ALP)
were obtained from the
computerized laboratory reporting
system.
The study involved 302

newly diagnosed TB
patients in Taipei.
Appendices
177
Source
(Corless et
al., 2009)
(Nackers et
al., 2012)
(Kipp et al.,
2011a)
178
Country
South
Africa
Study Design
Descriptive
Exploratory
Design
Research
aim(s)/research
question(s)
Measurement
Examined sense of
coherence, social
support, symptom
status, quality of life,
and adherence to
medications in two
samples of individuals
being treated either for
TB or human
immunodeficiency
virus/acquired
immunodeficiency
syndrome (HIV/AIDS).
Self-report questionnaire was used

to collect data highlighting
demographic items, measures of
sense of coherence (SOC), social
support, symptom status, QoL,
adherence to ARV therapy or TB
medication and clinic visit
Two independent groups diagnosed
with HIV (n=149) and TB (n=159).
Results
Clinics across different

settings included semiprivate hospital-based HIV
clinic, freestanding clinic
conducting HIV research
located in a shopping
center in an area with lowincome housing in Durban,
South Africa.
Sense of Coherence (SOC) measuring an

individuals sense of social belonging was
significantly lower for individuals with TB compare
to the HIV sample (mean TB=67.9, SD=24.4,
mean HIV=77.4, SD=23.6, p
0.001). In addition, social support was lower in TB
sample compare to the HIV sample (mean
TB=55.5, SD=11.1, mean HIV=64.4, SD=14.7,
p0.001).
Kenya,
Africa
Cross sectional
survey
To measure
adherence to TB
treatment among
patients receiving 6
months of standard
TB chemotherapy with
FDC under SAT in a
limited resource, high
TB-HIV burden
setting.
Recent adherence (last 4 days) and

adherence in the last month
measure using two subjective
(questionnaire and VAS) and two
objective adherence monitoring tools
(urine test for INH and a pill count)
The study was conducted

in Homa Bay district,
Nyanza Province in
Western Kenya. 212 of
279 eligible patients were
assessed for adherence.
Southern
Thailand
Cross sectional
survey
Use formally validated

tools to measure the
level of TB stigma
among patients with
Data were collected from two

different groups of participants: TB
patients and healthy community
members. Questionnaire on socio-
The study was conducted

in Southern Thailand.
Participants included 480
TB patients and 300
Appendices
For both diseases, there were no significant

differences in the average of missed appointments
(mean HIV=1.8, SD=1.6, mean TB=1.85, SD=1.7,
p=0.76). While, the Morisky medication adherence
scale scores were equivalently high in both groups
(mean HIV=3.7, SD=0.69, mean TB=3.8,
SD=0.55, p=0.49).
The estimated level of adherence was high using
all the adherence measurement tools. Overall,
95.2% [95%CI: 91.3 to 97.7] of the patients
reported not having missed a pill in the last 4 days.
On the VAS, complete adherence was estimated
at 92.5% [95%CI: 88.0 to 95.6]. Urine INH test was
positive for 97.6% [95%CI: 94.6 to 99.2] of the
patients. Pill count could only be assessed among
70% of the interviewed patients. The proportion of
pills taken by the patient was classified
ascomplete for 82.3% [95%CI: 75.1 to 88.1].
Except for the VAS, main adherence results
tended to be better during the intensive phase
than during the continuation phase.
The mean community TB stigma score was higher
than the mean perceived TB stigma score reported
by patients (20.13 vs.18.43) or 18.25 after
standardization) and exclusion of patients with
Source
Country
Study Design
Research
aim(s)/research
question(s)
Measurement
TB and healthy
community members.
demographic characteristics, and

available tools to measure stigma
was used.
healthy community
members
known HIV infection). Believing that TB increases

the chance of getting AIDS was associated with
experienced/felt TB stigma among patients (MD
2.16; 95% CI: 1.38, 2.94) and knowing someone
who died of TB was associated with community TB
stigma among community members (2.59; 95% CI:
0.96, 4.22). Older age (0.66; 0.22, 1.11 per 10
year increase) and Muslim religion (-1.37; 95% CI:
-2.32, -0.42) were both associated with perceived
TB stigma, while being female (-0.80; 95% CI: 1.67, 0.07) was only associated with
experienced/felt TB stigma. Patients with some or
no primary education reported higher scores on
both TB stigma scales, especially the
experienced/felt stigma scale (1.22; 95% CI: 0.17,
2.28), compared to patients who completed
secondary education.
Using the medical records

from six clinics, a total of
736 samples were
selected and 400 were
recruited.
Compliance of patients
over time to prescribed TB
drug regimens identified
from medical records.
While TB patients were
also asked directly if they
had ever stopped taking
the drugs since the
starting of treatment.
3 Primary clinics in Dili,
Timor Leste.
A total of 114 (29.8%) out of 382 patients stopped

taking TB drugs at some point during the
treatment.
There were 39.1% females and 29.8% males
reported to have stopped taking medication within
the first 2 months of commencing treatment.
Demographic factors such as sex, education,
marital status were not significantly associated
with non-adherence. Other factors such as sharing
room, friend suffered from TB, number of times
suffered from TB and use of traditional healer was
also not significant.
(Kaona et
al., 2004)
Ndola,
Zambia
Cross sectional
study
Factors contributing to
treatment adherence
and knowledge of TB
transmission among
patients on TB
treatment
Compliance refers to patients who

took their TB drugs daily for 8
months. Patients are considered
non-adherent if they stopped taking
TB drug during the treatment period.
(Martins et
al., 2009)
Timor
Leste
Randomized
Control Trial
Determine the
effectiveness of the
provision of whole
food to enhance
completion of
treatment for
tuberculosis
Primary measure: Confirmation of

clearance of acid fast bacilli from the
sputum after completion or 8 months
of treatment.
Secondary measure: clinic
attendance and pill counts.
Appendices
Results
179
270 previously untreated

newly diagnosis
pulmonary TB, 18 years or
older.
Food as incentive improved outcome with TB

treatment with higher weight gain, but no
significant correlation to treatment adherence or
successful completion of treatment.
Appendix C
Semi-structured Questionnaire for Health Professional
180
Appendices
Appendix C
Qualitative Interview Questions for Health Professionals
Date:
Interviewer:
Health Professional: 001
Data Source: Health Professionals
Method: Qualitative Semi-Structured Interview
Introductions: First the interviewer should introduce himself and then read the following
script below, describing the purpose of the session.
Why we are here: We are here to ask you about your perceptions about on
Tuberculosis treatment. The interview will highlight demographic characteristics, lifestyle,
education, beliefs and social and family support given to an individual patient receiving
treatment.
Your consent is important: We will share the information you provide us today only with
your consent. You are required to provide consent by signing a consent form. Any input
you provide will not be associated with you personally, but rather as group input. If you
are in any way uncomfortable with this arrangement, we will delete your names
altogether from this discussions transcripts and not include them at all in the document.
Length of interview: We will take no longer than an hour
1. How do you think patients demographic characteristic is associated with completion of TB
treatment? (Examples of demographic characteristics: age, education, marital status, employment, income, and
living condition).
2. How does TB patients lifestyle affect TB treatment adherence? (Examples of lifestyle: smoking,
alcohol and drug use)
3. How does quality of life affect (mental disorder, health symptoms and illness characteristics)
patients treatment adherence?

4. Do knowledge, beliefs and attitude of a patient impact treatment adherence? Probing how and
why
5. What is the common adherent factors affecting medication treatment?
6. What system and structures are in place in our current health system to support TB patients for
their treatment? (I.e. diagnosis, medical availability, staff etc.)
7. What is the formal and non-formal support provided towards a TB patient during his/her
treatment?
8. What adherence aids do patients usually use to remind them to take medication? (I.e. family
members, using clock, alarm, mobile alarm or calendar ).
Appendices
181
Appendix D
Semi-structured Questionnaire for TB patient
182
Appendices
Appendix D
Qualitative Interview Questions for TB Patients
Name of Interviewer:
Date:
Data Source: Tuberculosis Patients
Patient number:
Method: Qualitative Semi-Structured Interview
Introductions: First the interviewer should introduce himself and then read the
following script below, describing the purpose of the session.
Why we are here: We are here to ask you about your perceptions about on
Tuberculosis treatment. The interview will highlight demographic characteristics,
lifestyle, education, beliefs and social and family support given to an individual patient
receiving treatment.
Your consent is important: We will share the information you provide us today only
with your consent. Any input you provide will not be associated with you but rather as
group input. If you are in any way uncomfortable with this arrangement, we will delete
your names altogether from this discussions transcripts and not include them at all in
the document.
1. Has there been any time that you miss taking your medication? Why? Why not? Probing
If the patient says no continue to ask (has there been any time at all that you nearly forget to take
your medication?)
2. What makes you so organized that you do not forget to take your medication?
3. What do you normally use to remind yourself of the time to take your medication?
We heard that some people stop their treatment due to some specific reasons? Would you
like to explain it to us?
4. Based on your knowledge, how important do you think it is for a patient to follow instruction
and adhere to their medication regimen? Why? Why not? Probing
5. Do you think peoples age, level of education, their job and distance from the health
services will have impact on their treatment adherence? Explain
6. Do you think having a good relationship with them will help with your treatment progress? If
so, why and why not? Probing
7. Do you family, friends and colleagues from work know about your TB treatment?
8. How comfortable do feel to share about your treatment? Why? Why not? Probing
9. What kind of support do you usually get from:
Family members
Friends
Peer educators
Social Organizations
10. Can you tell us what kind of information you were given by the health care providers when
you start diagnosis and treatment?
Appendices
183
Appendix E
Map of Timor-Leste
184
Appendices
Appendix E
Map of Timor-Leste
Figure 7 Map of Timor Leste
Data was collected in Dili, Liquica and Bobonaro (Maliana) Distircts.
Appendices
185
Appendix F
Human Ethics Approval Certificate
186
Appendices
Appendices
187
Appendix F
188
Appendices
Appendix G
Human Ethics Approval Certificate
Approval Letter
Research Cabinet and Health Development
Ministry of Health, Timor Leste
Appendices
189
Appendix G
190
Appendices
Appendices
191
Appendix H
Quantitative Survey Questionnaire
English Version
192
Appendices
Appendix H
Survey Questionnaire
Instruction: The answers you give on this form will be treated confidentially and will only be used by the
researcher to develop a plan to help people who have Tuberculosis to better follow their regimens. Please
do the best you can to answer all the questions. If you do not wish to answer a question, please leave it
blank or circle refused option where available. If a question is not clear, please ask the researcher to clarify
for you. Thank you for participating in this important study!
Your answers will be kept confidential since they will be used for the purpose of the study only.
Section I. Questions about You and Your Health

A.
Question about yourself: Demographic Information
Please place () or tick a number which best represents your answer.
Question
Response
Male
Female
A1
Sex
A2
What is your date of birth?
A3
How old are you?
Years
A4
In total, how many years have you spent

at school or in full-time study (excluding
pre-school)?
Years
A5
Which district & sub-district are you

from?
1
2
y y y y
District___________
Sub-district___________
EXPANDED: Demographic Information

A6
What is the highest level of

education you have completed?
A7
What is your marital status?
A8
Which of the following best

describes your main work status
over the past 12 months?
Select responses that are
appropriate
A9
What is the main subsistence

(financial support) in current time?
appropriate
A10
What is the monthly total average

income of your family (including all
people who share house and meal
with you?
Select one response
Appendices
No formal schooling
Primary school
Secondary School
1
2
3
High school
College/University
3
4
Never married
Currently married
1
2
Separated/ Divorced
Widowed
3
4
No job/jobless
Farmer
Labor work
Government staff
Student or pupil
1
2
3
4
5
From your income(s)

Support from husband/wife
Support from family
Support from friend
1
2
3
4
< 30USD/month
31USD-60USD/month
61USD-90USD/month
91USD-120USD/month
1
2
3
4
Housewife
Unemployed (able to work)
Unemployed (unable to
work)
Other (please
specify)_______
Others (please
specify)______
None
6
7
8
9
121USD-150USD/month
151USD-180USD/month
181USD-210USD/month
>210USD/month
5
6
7
8
193
5
6
A11
A12
A13
A14
A15
A16
Does your family have these items?

appropriate
Who do you live and share main

meals with?
appropriate
How many people whom are you
living with and sharing the main
meal with?
Select one response
Who is the person taking care of
you and reminding you to take your
medication frequently?
appropriate
About how long would it take you to
walk from your home to your
nearest clinic/Health
Post/CHC/Referral Hospital?
For all journeys combined, in the
past 12 months, which types of
transport were used more
frequently? Select responses that
are appropriate
B.
Solid roof
Solid floor
SolidSolid
bathbathS
room
Standard toilet
Water pumping machine
Wardrobe
Stay alone
Husband/wife
Parent(s)
Children
1
2
3
4
5
6
1
2
3
4
TV
Mobile phone
Bicycle
Fridge
Motorbike
Car
Friends
Co-worker(s)
7
8
9
10
11
12
5
6
7
Only myself
Two of us
Three people including
myself
Four people including
myself
Husband/wife
Parent
Children
Friend(s)
15 minutes or less
16-30 minutes
31-45 minutes
46-60 minutes
Local mini bus
Local bus/truck
Bicycle
Scooter/motorbike
1
2
R
3
e
4
1f
u
2
s
2
3
e
4
d
5
1
2
3
4
1
2
3
4
More than four people
Co-worker(s)
PSF (Family Health
Promoter)
Community Health
Centerstaff
None/take care myself
6
7
8
9
>1 hour but < 2 hours

2 hours or longer
Dont know
5
6
7
Walking to/from a destination
Lifestyle - Alcohol Use

Question
Response
B1
Do you drink alcohol?
B2
How often have you had a drink

containing alcohol a glass of beer,
wine, local wine, a mixed drink or any
kind of alcohol beverage?
B3
How many UNITS of alcohol do you

drink on a typical day when you are
drinking?
B4
How often do you have 6 or more

UNITS of alcohol on one occasion?
194
Yes
No, continue to B5
Daily or nearly daily

(> 4 times a week)
Some/month (1 3 times/month)
Some week
(1-4 times/week)
Drink when available
1 to 2
3 to 4
5 to 6
Never
Less than monthly
Monthly
1
2
3
1
2
3
7 to 8
9+
4
5
Weekly
Daily or almost daily
4
5
Appendices
Lifestyle: Smoking
Please place (x) or tick () a number which best represents your answer.
Question
B5
B6
B7
B8
Response
Currently smoking
Used to smoke
Stopped smoking started of
Within 5treatment
minutes
Do you smoke tobacco?

How soon after wake up do
you smoke your first
cigarette?
Normally, how many
cigarettes do you smoke
each day?
How many cigarettes did you
smoke in the last 24 hours?
6-30 minutes
31 60 minutes
10 or less
11-20
1
2
3
Never
If Never, continue to section B9
After 60 minutes
21-30
31 or more
3
4
1
2
3
1
2
_________________
______________
Lifestyle Illicit Drug Use

The following are questions asking about your drug use. Once again, your answer will be kept confidential since
they will be used for study purpose only.
Question
B9
B10
Response
Never
Continue to the next section
Some/month (1 3 times/month)
Use only when available
Yes and I currently use it
Yes, but I used it in the past
Have you ever used illicit

drug?
In the last month, how

often have you used
drug?
Daily or nearly daily

(> 4 times a week)
Some week
(1-4 times/week)
Section II
C.
Psychological Distress
C1
.C2
.C3
.C4
.C5
.
C6
.C7
.
C8
.
C9
.
C1
0.
Response
During the past 30 days, about how often did you feel tired out for no good
reasons?
During the past 30 days, about how often did you feel nervous?
During the past 30, how often did you feel so nervous that nothing could calm
you
down?
During
the past 30 days, about how often did you feel hopeless?
During the past 30 days, about how often did you feel restless or fidgety?
During the past 30 days, about how often did you feel so restless you could not
sit still?the past 30 days, about how often did you feel depressed?
During
During the past 30 days, about how often did you feel that everything is an effort?
During the past 30 days, how often did you feel so sad that nothing would cheer
you
up?the past 30 days, about how often do you feel worthless?
During
Appendices
None
of the
time
Question
A little
of the
time
Some
of the
time
Most
of the
time
All of
the time
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
195
D.
Multidimensional Health Locus of Control Scales - Shortened Version

Strongl
y Agree
The main thing that affects health is what I myself do.
If I take care of myself, I can avoid illness.
Having regular contact with my doctor is the best way to avoid illness.
When I don't feel well, I should consult a medical professional.
When I recover from an illness, it's because other people have been taking care of me
Luck plays a big part in how soon I will recover from an illness.
My good health is largely a matter of good fortune.
If it's meant to be, I will stay healthy.
Agree
D1
D2.
D3.
D4.
D5.
D6.
D7.
D8.
D9.
Response
Disagree
Question
Strongly
Disagree
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
Session III: We would like to know how much you understand about Tuberculosis.
E. Your TB Knowledge about Tuberculosis
E1.
How is Tuberculosis transmitted from one person to

another?
E2.
Would a person with TB eat together with other family

members?
E3.
Do family contacts need to be screened for TB?
E4.
Is Tuberculosis curable with proper treatment?
E5.
Can Tuberculosis be cured with a week course of

treatment?
E6.
What is the length of treatment required to cure TB?
E7.
Can treatment be discontinued once you feel better,

even the prescribed treatment duration has not been
reached?
Infectious droplet
spread
Sharing eating utensils
Punishment of bad
behavior
1
2
3
No
Yes
1
2
No
Yes
No
Yes
No
Yes
1
2
1
2
1
2
1-2 months
6-8 months
No
Yes
1
2
Rest of life your life

Varies according to
individual
Section IV: Treatment Adherence

We would like you to answer these questions to understand how people with TB are really doing with their pills, but not
evaluating your medication taking. Dont worry about telling us that you dont take some of your pills or forget to take
them. We need to know what is really happening, to improve your and others TB adherence. No one involved in your
medical care will hear about what you have told us.
F. Taking Your Medication

F1
Do you ever forget to take your medication?
196
Yes
No
Appendices
Skip to
G
3
4
F2
F3
Are you careless at times at taking medication?

When you feel better do you sometimes stop taking your medication?
Sometimes if you feel worse when you take the medication do you stop
F4.
taking it?
F5. Do you take medication only when you feel sick?
F6. By staying on medication, can you prevent getting sick?
F7
Does medication make you feel sick?
F8
Do you feel weird on medication?
Compliant = code 2 in all questions
Yes
Yes
1
1
No
No
2
2
Yes
No
Yes
Yes
Yes
Yes
1
2
1
1
No
No
No
No
2
1
2
2
For the following question please circle (O) or tick () a percentage which best represents your answer.
We would like to get your best guess about how much of your TB medications you have managed to take during
the last month. We would be surprised if this was 100% for many people. Please make a mark on the
line to show how many of your TB pills you think you managed to take in the last month (e.g. 0 means you have
taken no medication; 50% means you have taken half of our medication, 100% means you have taken every
single dose of medication
F9
How often did you follow the medication instruction in the last month?
10%
20%
30%
40%
50
%
60%
70%
80%
90%
100%
Reason for missing dose

People may miss taking their medications for various reasons. Here is a list of possible
reasons why you may miss taking your medications during the last month.
G.
Adherent Factors
Reasons for missing a dose: Each item below is a statement about your health state
Did you miss taking your medications during the last month because you:
No
Yes
G1.
Simply forgot
G2.
Felt good
G3.
Were away from home
G4.
G5.
G6.
G7.

Clinic is too far
1
1
1
2
2
2
G8.
G9.
G1
0.
G1
1.
G1
2.
G1
3.
G1
4.
G1
5.
G1
6.
G1
7.
Felt like the drug was toxic/ harmful

Felt asleep /slept through dose time
Felt sick
Felt depressed/ overwhelmed
Had problems taking pills at specified times (with meals, on empty stomach, before bed time etc.)
Use drugs
Drink alcohol
Took them, then get nausea, but did not take the replacement
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
Appendices
197
G1
8.
G1
9.
G2
0.
G2
1.
Dont have money for eating and to cover transport cost

Ran out of pills but could not go to the clinic to get them
Dont believe in the usefulness of the drug
1
1
1
2
2
2
Section V Stigma
H.
Tuberculosis related stigma scales
Yes
No
H1.
H2.
Do you think people may not want to eat or drink with friends who have TB?
Do you think people feel uncomfortable about being near those with TB?
H3.
H4.
Do people keep their distance from others with TB?
H5.
H6.
H7.
Do you think people may not want to eat or drink with relatives who have TB?
Yes
No
Community perspectives toward tuberculosis
Do you think people who have TB feel hurt of how others react to knowing they have TB
H9.
Do people who have TB feel alone?
H10.
Do you think people who have TB are afraid to tell those outside their family members?
Do you think people who have TB feel guilty because their family has the burden of caring for
them?
Do you think people who have TB are afraid to tell their family that they have TB?
Patient perspectives toward tuberculosis

H8.
H11.
H12.
Traditional Healers
These people are including elderly or those claims to have healing power to treat sick
people in the community. Each item below is a statement about your health state.
Please circle the number that response for indicating your level of
agreement/disagreement to the following reasons of why you might visit a traditional
healer.
I.
Traditional Healers
Somewha
t Agree
Disagree
2
Neither
Agree nor
Somewhat
Disagree
Why do you visit a traditional healer?
Strongly
Disagree
Strongly
Agree
Do you ever visit a traditional healer? (No, continue to next section)
No
I13.
Yes
I14.
I want a traditional medicine
I15.
I could not be cured at the clinic
I16.
I17.
There are no clinics close to where I live

Cost of transport is too expensive
1
1
2
2
3
3
4
4
5
5
198
Appendices
Section VI: HCP- Patient Communication
J.
These people including doctors, nurses, DTC/DTA, lab technician & SISCa/TB volunteers.
Somewh
at Agree
Strongly
Agree
HCP told me what the possible side-effects of each of the TB drug are?
J2.
HCP told me what treatment would do
J3.
J4.
J5.
HCP told me how the TB treatment would be

HCP told me the changes to expect in my health when taking TB medication
Treatment procedure was clearly explained by the HCP
1
1
1
2
2
2
3
3
3
4
4
4
5
5
5
Very
Satisfied
Somewha
t satisfied
In general, how satisfied are you with the overall support you get from the HCP?
Somewhat
dissatisfied
J6.
Circle a number which represent the appropriate

response
Very
Dissatisfie
d
Question
Response
Disagree
Somewhat
Disagree
J1.
Question
Neither
Agree nor
Strongly
Disagree
Section VII: Social and Family Support:

The following questions ask you about the support you receive from your family and community.
In order to help you have better idea of who they are, we divided them in to four groups:
a. Family: Just count your extended family members whom you are living with and sharing
meals with. Those people also can include people who do not live with you but regularly
communicate with you, and you feel they are closed to you.
b. Friends: People whom you make friends by yourself, but not through any organizations
and social groups. For example, whom you know from classes, workplaces, relatives
but not, home-based care, SISCa/TB Volunteers.
c. Peer educators:, people who come to check you up regularly and report to CHC your
treatment progress(e.g. SISCa Volunteers, SISCa TB volunteers), home-based care,
peer educator, groups/clubs.
d. Health Care Support and Social Organization: staff who are working for mass
organizations (MSS or staff who work for mass organizations (committee,) or health
clinics.
K.
Social and Family Support
A Lot
Somewhat
A Little
Question: Do you tell some of the following people that you are having TB treatment?
Not At All
K1.
Family
K2.
Friends
K3
Neighbor(s)
Appendices
199
2
2
3
3
A Lot
1
1
Somewhat
Question: To what extent do the following people help you to remember to take TB medication?
0
0
A Little
Peer educators (SISCa volunteers)

MSS (Ministry Solidarity Social)
Not At All
K4.
K5.
K7.
K8
K9.
K10.
Friends
Neighbor(s)
0
0
0
0
1
1
1
1
2
2
2
2
3
3
3
3
Question: To what extent do the following people provide you emotional support (e.g.
acceptance, freedom to talk openly about your health, care, ability to confide in, ability to share
grief, and the like)?
A Lot
Somewhat
A Little
Family
Not At All
K6.
K12.
K13
K14.
Friends
Neighbor(s)
0
0
0
1
1
1
2
2
2
3
3
3
K15.
Question: To what extent do the following people provide you tangible support (e.g. finances,
transportation, house chore, housing, clothing, food supplies, medical supplies, children
education and the like)?
A Lot
Somewhat
A Little
Family
Not At All
K11.
K17.
K18
Friends
Neighbor(s)
0
0
1
1
2
2
3
3
K19.
K20.
Question: To what extent do the following people provide you information support (e.g. advise,
guidance, feedback or information on a variety of issues and the like)?
A Lot
Somewh
at
A Little
Family
Not At All
K16.
K22.
K23
K24.
Friends
Neighbor(s)
0
0
0
1
1
1
2
2
2
3
3
3
K25.
Question: In general, how satisfied are you with the overall support you get from the following?
K26.
K27.
K28
K29.
K30.
A Lot
Somewhat
A Little
Family
Not At All
K21.
Family
Friends
Neighbor(s)
0
0
0
0
1
1
1
1
2
2
2
2
3
3
3
3
200
Appendices
Thank you very much for you participation!
Appendices
201
Appendix I
Quantitative Survey Questionnaire
Tetum Version
202
Appendices
Appendix I
Survey Questionnaire
Instrusaun: Ita nia resposta iha formulariu Ida ne sei konsidera hanesan segredu, no sei uja deit
atu dezenvolve planu ruma hodi tulun maluk sira neebe infetaduTuberculoze hodi kumpri
tratamentu ho diak. Favor responde pergunta hirak nee tuir ita nia kapasidade. Se pergunta
ruma mak la klaru favor husu klarifikasaun husi peskisador nebe asisti ita boot. Brigado wain
ba ita nia participasaun iha estudu importante ida ne!
Ita nia resposta ne segredu e sei uja deit para completu estudu ne.
Section I. Perguntas Kona ba Ita Nia An

A.
Informasaun demografiku
Favor tau sinal () ou marka numeru nebe representa diak liu ita nia resposta.
Pergunta
Resposta
Mane
Feto
A1
Seksu
A2
Ita moris sa tinan?
A3
Ita tinan hira?
Tinan
A4
Iha total, tinan hira ita gasta iha eskola (la inklui
TK)
Tinan
Note 0 = Nunka eskola
A5
Ita husi distritu no sub-distritu nebe?
exemplu: 1983
t t t t
Distritu:
Sub-distritu:
EXPANDED: Informasaun demografiku

Fahe malu
Laen/fen mate
3
4
Laiha servisu
Dona de casa
Toos nain
Swasta/buru kasar
Funsionariu publiku
Estudante
2
3
4
5
Desempregado (bele
servs)
Desemprego (la
bele
servs)
Seluk (spesifika)______
7
8
Apoiu finanseiru saida mak bain-bain ita

hetan?
Hili liu husi opsaun ida, ba sira neebe
apropria
Salario rasik
Suporta husi kaben
Suporta husi familia
1
2
3
Suporta husi kolega

Seluk (spesifika)_______
4
5
Total salario kada fulan iha ita nia familia

(inklui ema neebe hela ho ita iha uma)
maiz ou menuz hira?
Hili opsaun ida nebe apropria
< 30USD/fulan
31USD-60USD/fulan
61USD-90USD/fulan
91USD-120USD/fulan
121USD-150USD/fulan
151USD-180USD/fulan
181USD-210USD/fulan
>210USD/fulan
5
6
7
8
Estadu sivil?
A10
klosan
Kaben nain
1
2
A7
A9
3
4
Ita nia nivel edukasaun ikus liu mak

saida?
A8
La iha edukasaun formal

Eskola primaria (SD)
Eskola primaria (SMP)
Sekundaria - SMA
Colegio/Universidade
A6
Opsaun sira ne, ida nebe mak esplika diak

liu kona ba ita nia servisu iha fulan 12
liuba?
apropria
Appendices
2
3
1
2
3
4
203
A11
A12
Iha ita nia familia laran, sasan hirak tuir mai nee
iha glae?
apropria
Ita hela hamutuk ho se?

apropria
A13
Ema nain hira mak hela hamutuk ho ita?

Hili opsaun ida neebe apropria
A14
Semak bain-bain fo hanoin ita atu bele hemu

aimoruk tuir orariu nebe pesoal saude sira fo
ona?
apropria
A15
Husi uma, maiz ou menuz oras/minitu hira ita lao

ba klinik/Posto Saude/CHC/Hispital Referral
nebe besik liu?
A16
Iha kombinasaun fulan 12 nia laran dala hira ita

uja trasporte nebe hanesan? Hili liu husi
opsaun ida, ba sira neebe apropria
B.
Uma kalen
Uma laran sementi
Haris fatin nebe ho
estandar
Saneamentu estandar
1
2
3
Televizaun
Telemovel
Bisikaleta
7
8
9
Jaleira
10
Motabomba
Motor
11
Armari ropa
Hela mesak
Stay aman
alone
Inan
6
1
2
Kareta
Membru familia seluk
Kolegas
12
5
6
Ho oan
Kolega servisu
Hela ho fen/laen
Hau mesak
Ami nain rua
Ema nain tolu
hela ho ita
Ita nia laen/fen
Inan aman
Oan Sira
1
2
R
3
ef
us1
ed
2
3
Ema nain hat

Liu husi ema nain hat
4
5
Kolega servisu(s)
PSF
Peasoal Saude
6
7
8
Membru familia seluk
Hau hemu rasik
Kolegas
Minute 15 ou menuz
Minute 16-30
Minute 31-45
5
1
2
3
Minute 46-60
Oras 1 - 2
Oras 2 ou liu
4
5
6
Scooter/Motor
Lao deit
tratamentu
/from a destination
4
5
Mikrolet/Anguna
Bis/Trek
Taxi
Biskaleta
Vida moris Hemu tua manas

Pergunta
B1
B2
B3
1
2
3
Resposta
Ita hemu tua manas?
Sim
Para hemu wainhira hahu tratamentu
Dala hira ita hemu tua manas

hanesan beer, tinto, tua mutin/tua
sabu ou tua manas ruma kahur ho
alkohol?
Alkohol ho quantidade hira mak ita
boot bain-bain konsume?
(refere ba figura iha leten)
204
Loro-loron ou de vez
enkuandu
(> dala 4 kada
semana)
Semana-semana
(dala 1 - 3 kada
semana)
1 to 2
3 to 4
5 to 6
Lae, continua ba B5
Dala balun kada fulan

(dala 1-3 kada fulan)
Hemu wainhira iha
1
2
3
7 to 8
9+
Appendices
4
5
B4
Ita hemu tua manas dala hira ho

quantidade 6 ou liu iha tempu ita
hemu?
Nunka
Kadavez
Kada fulan
1
2
3
Kada semana
Loron-loron
4
5
Vida moris: Fuma Sigaru/Tobacco

Pergunta
B5
Resposta
Agora sei fuma hela
Uluk mak fuma
Para fuma wainhira hahu

tratamentu
Minitu 5 nia laran
Minitu 6 - 30
Minitu 31-60
10 ou menus
21-30
11-20
31 ou liu
Ita fuma sigaru?
Wainhira hader iha dader, minitu hira

tia mak ita hahu fuma sigaru lolon
primeiru?
B6
B7
Bain-bain sigaru lolon hira mak ita

fuma kada loron ida?
B8
Iha oras 24 nia laran liu ba, ita fuma

sigaru lolon hira?
Nunka
Se nunka, continua ba B9
Depois de oras ida nia laran
Lolon_________________
______________
Vida moris Droga

Perguntas hirak tuir mai husu ita kona ba uja droga. Dala ida tan, ita nia resposta sei tau hanesan segredu
no sei uja deit para atinji objetivu ba estudu nee.
Question
B9
B10
Response
Nunka
Se nunka, continua ba sesaun C.
Kadavez iha fulan ida (1 3

kada/fulan)
Usa wainhira iha
Sim, agora sei uja hela
Simi, maibe iha pasado
Loro-loron ou dala ruma

(> dala 4 semana ida)
Kadavez iha samena ida
(1-3 kada/semana)
Ita uja droga?
Iha fulan liu ba nia laran,

dala hira ita uja droga?
Appendices
205
Sempre
Dala
barak
Tempu
balun
Dala
ruma
Iha loron 30 nia laran liu ba:
Nunka
Sesaun II
C.
Psychological Distress
C1
Mais ou menus dala hira ona ita sente kolen ho laiha razaun?
C2
Mais ou menus dala hira mak ita sente nakdedar?
C3
Dala hira ona mak ita sente nakdedar los no laiha buat ida atu hakalma?
C4
Mais ou menus dala hira ona mak ita sente laiha esperansa ba moris?
Mais ou menus dala hira ita la bele deskansa ho diak?

Dala hira ita sente la bele los atu deskansa ho diak no tur la metin?
Mais ou menus dala hira ona mak ita sente depresaun (tekanan)?
Dala hira mak ita sente katak halo buat hotu ne esforsu ida ba ita?
Dala hira mak ita sente triste liu e laiha buat ida mak bele halo kontente?
Mais ou menus dala hira ona mak ita sente moris nee la vale?
1
1
1
1
1
1
2
2
2
2
2
2
3
3
3
3
3
3
4
4
4
4
4
4
5
5
5
5
5
5
C5
C6
C7
C8
C9
C10
.
D.
Multidimensional Health Locus of Control Scales - Shortened Version

La aseita liu
La aseita
Aseita
Aseita liu
D1
Hau rasik mak kontrola hau nia saude
D2.
Hau rasik responsavel ba buat nebe afeita ba hau nia saude
D3.
Se hau kuidadu hau nia an, hau bele hadok an husi moras.
D4.
Iha kontaktu regularmente ho doutor maka dalan diak atu hadook an husi moras.
D5.
Wainhira hau laran baruk, hau sei halo konsultasaun ho pesoal saude sira.
D6.
Wainhira hau diak husi moras, nee tanba iha ema seluk nebe tau matan ba hau
D7.
Sorti sai hanesan parte ida oinsa hau bele hetan diak lalais husi moras.
D8.
Hau nia saude diak sai nudar sorti diak ida mos ba hau.
D9.
Se ida ne mak nune duni ona, hau hakarak moris saudavel.
Pergutans
Resposta
Session III: Ami hakarak hatene ita nia komprensaun kona ba moras tuberculoze
E. Your TB Knowledge about Tuberculosis
E1
Oin sa Tuberkuloze bele hadaet husi ema ida ba ema

seluk?
Mear la taka ibun
Han hamutuk
Causa husi
espiritu aat
E2
Ita bele han hamutuk ho membru familia sira?
Sim
Lae
E3
Familia direita presisa halo screening/teste ba tuberculoze?
Sim
Lae
E4
Tuberculoze bele kura ho tratamentu nebe propria?
Sim
Lae
E5
Tuberculoze bele kura durante tratamentu iha semana ida

nia laran?
Sim
Lae
206
Appendices
E6
E7
Atu kura moras Tuberculoze, presiza tuir tratamentu too

wainhira?
Fulan 1 2
Iha moris tomak
Fulan 6-8 nia laran
La hanesan entre
pasiente
Sim
Lae
Tratamentu bele hakotu klae, wainhira laiha ona simtomas,

maske durasaun ba tratamentu nebe determina seidauk
remata?
Section IV: Kumpri Tratamentu

Ami husu ita atu responde perguntas hirak tuir mai atu komprende lolos realidade nebe akontese ba
ita hemu ita nia aimoruk, maibe sei la evalua ita hemu ita nia aimoruk ou lae. La bele tauk atu
informa ami katak ita hemu ita nia aimoruk ou haluha atu hemu. Ami presisa atu hatene lolos saida
mak akontese atu bele ajuda sira seluk nebe la kumpri ba sira nia tratamentu. Laiha doutor ou
profesaun saude sira nebe tau matan ba ita mak sei hatene ou rona kona ba saida mak ita hatete
mai ami.
F. Hemu Ita Nia Aimoruk
F1
Kadavez ita haluha hemu ita nia aimoruk glae?
F2
Karik dala balun mak ita la iha interese wainhira atu hemu aimoruk?
F3
Wainhira ita sente diak ona, ita la kontinua (para) atu hemu ita nia aimoruk?
F4
Kadavez se ita sente moras liu tan wainhira hemu aimoruk, ita para hemu?
F5
Ita hemu ita nia aimoruk deit wainhira ita sente moras?
F6
Liu husi hemu aimoruk, ita bele hadook an husi moras?
F7
Hemu aimoruk nee halo ita bele moras?
F8
Ita sente aneh wainhira hemu aimoruk?
Si
m
Si
m
Si
m
Si
m
Si
m
Si
m
Si
m
Si
m
1
1
1
1
1
2
1
1
La
e
La
e
La
e
La
e
La
e
La
e
La
e
La
e
2
2
2
2
2
1
2
2
Compliant = code 2 in all questions

Ba pergunta tuir mai favor sirkulu (O) or marka () pursentu nebe representa ita nia responde diak liu
Ami hakarak ita boot atu sik kona ba aimoruk hira ona mak ita konsege hemu iha fulan liu ba. Ami sei supresa se
ema barak hemu 100%. Favor marka linha iha kraik atu representa aimoruk hira maka ita konsege hemu iha fulan
liu ba (exemplu 0 indika katak ita la hemu aimoruk ruma, 50% indika katak ita hemu sorin balun no 100% indika ita
hemu aimoruk ho completu).
Dala hira ona ita halo tuir instrusaun hemu aimoruk iha fulan kotuk liu ba?
F9
0
10%
20%
30%
40%
50
%
60
%
70%
80%
90
%
100
%
Rasaun tanba sa la hemu aimoruk
Appendices
207
Ema barak bele la konsege hemu sira nia aimoruk ho rasaun oin-oin. Refere ba lista
tuir mai, karik ita bele identifika rasaun ruma nebe karik impede ita hemu ita nia
aimoruk iha fulan liu ba.
G.
Fator kompromisiu
Razaun la hemu aimoruk: Deklarasaun hirak tuir mai sei aprezenta ita nia estatutu saude.
Sim
Ita haluha hemu ita nia aimoruk iha fulan ida liu ba nia laran tanba:
Lae
G1
G2
G3
G4
G5
G6
G7
G8
G9
G10
G11
Haluha deit?
Sente diak ona?
Dook husi uma?
Tanba iha aimoruk barak liu?
Hakarak evita efeitu koleteral?

Postu Saude/Klinika/CHC/Ospital Referal dook liu?
1
1
2
2
Troka rutina loro-loron?
Sente aimoruk nee hanesan veneno/perigozu?
Sente matan dukur bebeik tanba doze aimoruk?
Sente moras?
Sente depresaun (tertekan)?
1
1
2
2
G12
Hetan problema atu hemu tuir tempu espesifikadu (han ho hahan, kabun mamuk, antes toba sst)?
G13
G14
G15
G16
G17
G18
G19
G20
G21
Postu Saude/Klinika/CHC/Ospital Referal la fo aimoruk mai hau?

Okupadu ho atividade barak seluk?
Uja droga?
Hemu tua?
1
1
1
1
2
2
2
2
Hemu sente laran sae no la hemu tan?
Laiha osan ba hahan no selu transporte?
Aimoruk remata no la konsege ba foti iha Postu Saude/Klinika/CHC/Ospital Referal?
Lakohi ema seluk hatene katak hau hemu aimoruk?
La fiar ba funsaun aimoruk ne?
Section V Stigma
H.
Tuberculosis related stigma scales
Sim
Lae
Tuir ita nia hanoin ema lakohi atu han ou hemu hamutuk ho kolega sira nebe sofre moras TB?
Tuir ita nia hanoin ema sente la seguru wainhira besik ema sira nebe ho moras tuberculoze?
Tuir ita nia hanoin ema hadook an husi sira nebe sofre moras tuberculoze?
Tuir ita nia hanoin ema lakohi atu koalia ho sira nebe sofre moras tuberculoze?
Tuir ita nia hanoin ema tauk ho sira nebe sofre moras tuberculoze?
Ita sente ema tauk atu kaer liman ho sira nebe sofre moras tuberculoze?
Ita sente ema lakohi atu han e hemu hamutuk ho membru familia nebe sofre moras tuberculoze?
Pespetivo komunidade hasoru moras tuberculoze

H1.
H2.
H3.
H4.
H5.
H6.
H7.
208
Appendices
Sim
Lae
Ita hanoin ema nebe sofre moras tuberculoze sente laran moras tanba ema nia reasaun hatene katak
sira moras TB?
Ita hanoin ema nebe sofre moras tuberculoze sente mesak-mesak (maluk laiha)?
Ita hanoin ema nebe sofre moras tuberculoze tauk atu hatete ba ema seluk nebe laos nia familia?
Ita hanoin ema nebe sofre moras TB sente sala tanba hatene katak sira nia familia lakon tempu tau
matan ba sira?
Ita hanoin ema nebe sofre moras TB tauk atu hatete ba nia familia kona ba sira nia status moras?
Perspectivo pasiente hasoru moras tuberculoze

H8.
H9.
H10
.H11
.
H12
.
Matan dook sira

Ema hirak nee inklui katuas ou ferik sira nebe rekonhese sira nia an katak sira iha
abilidade atu kura ema moras iha komunidade. Pontus hirak tuir mai nudar
deklarasaun kona ba ita nia saude. Favor tau sinal ruma ba numeru nebe indika se ita
aseita ou la aseita ho rasaun sira tuir mai tanba sa ita hakarak visita matan dook sira.
I.
Traditional Healers
Ita visita matan dook sira? (Se lae, kontunua ba seasaun tuir mai)
Netral
Aseita
I14
I15
I16
I17
La
aseita
liu
La
aseita
Tanba sa ita visita matan dook sira?
Hau hakarak atu hemu aimoruk tradisional
Hau la bele hetan kura iha klinik/CHC/Postu saude?
Laiha klinika/CHC/Postu saude ruma mak besik hau nia hela fatin?
Kostu ba transporte karun liu
1
1
2
2
3
3
4
4
Aseita
liu
La
e
I13
Si
m
5
5
5
5
Section VI: Komunikasaun entre pesoal saude ho pasiente
J.
Ema hirak ne inklui doutor, Infermeira, DTC/DTA, lab technician & PSF
Aseita
Aseita liu
J2
Pesoal saude dehan saida mak tratamentu TB sei fo mai hau
J3
Pesoal saude dehan mai hau oinsa klalaok tratamentu nian
J4
Pesoal saude dehan kona ba mudanca ba hau nia saude wainhira hemu aimoruk
TB
Pesoal saude esplika klaru los kona ba lalaok tratamentu
J5
Pergunta
Appendices
Respsta
209
Satisfaz
liu
Netral
Stisfaz
Pesoal saude fo hatene mai hau saida mak efeitu kolateral husi aimoruk TB nian
Resposta
La satisfaz
J1
Pergunta
La satisfaz
liu
La aseita
La aseita
liu
Iha jeral, ita satisfaz ho suporta neebe ita hetan husi pesoal saude ka lae?
J6
Section VII: Suporta familia no suporta sosial

Pergunta sira tuir mai sei husu kona ba suporta sira mak ita simu husi ita nia familia no husi
komunidade. Atu fo idea mak diak kona ba se make ma sira nee, sei fahe ba grupu 4 hanesan
tuir mai nee:
a. Familia: sura deit membru familia rasik mak ita hela ho no han hamutuk. Ema sira nee
mos bele inklui sira neebe laos hela hamutuk maibe komunka regular ho ita, no ita
sente besik ba sira.
b. Kolega: ema sira neebe ita halo kolega ho maibe laos liu husi organizasaun ou grupu
sosial ruma. Ezemplu: kolega eskola, kolega servisu, parenti maibe laos husi pesoal
no voluntariu saude sira.
c. Peer edukador sira: ema sira nee be mak regulamente mai trata/fo konsulta ita no
relata ba CHC kona ba ita nia tratamentu nia progresu. Ezemplu mak voluntariu SISCa
no pesoal saude husi CHC.
d. Organizasaun sosial komum (Ministeriu Solidaridade Social ou funsinariu komite nebe
servisu ba organizasaun o klinika saude.
K.
Suporta Sosial ho Suporta Familia
Pergunta: Ita bele hatete ba ema hirak tuir mai katak ita tuir hela tratamentu
Tuberculoze?
Lae
Oitoan
Barak
oitoan
Barak
K1
Familia
K2
Kolegas
K3
Vizinho
K4
Voluntariu SISCa & pesoal saude
K5
Membru Ministeriu Solidaridade Sosial no organizasaun social seluk
4
Barak
Kolegas
K8
Vizinho
K9
K10
Pergunta: Husi lista ema sira tuir mai nee, se mak fo suporta emosional? (eg.
Simu realidade kona ba moras nee, libre atu koalia kona ba ita nua estutu
moras nian, kuidadu no abilidade atu rai konfidensia, abilidade atu bele fahe
tristeza)
K11 Familia
K12 Kolegas
Barak
Barak
oitoan
Barak oitoan
Oitoan
Familia
Oitoan
K6
K7
Lae
Lae
Pergunta: Husi ema sira tuir mai nee, se mak fo hanoin ita bebeik ita atu
hemu ita nia aimoruk?
K13
Vizinho
K14
210
Appendices
Pergunta: ema hirak tuir mai, se mak bele fo suporta bebeik (Ezemplu:
finanseiru, transporte, hela fatin, roupa, hahan, mediku, labarik sira nia eskola,
no suporta sira seluk).
Lae
Oitoan
Barak
oitoan
Barak
K16
Familia
K17
Kolegas
K18
Vizinho
K19
K20
Pergunta: ema hirak tuir mai, se mak fornese suporta informasaun (Ezemplu :
sujestaun, mata dalan, lian menon ou informasaun)?
Oitoan
K21
Familia
K22
Kolegas
K23
K24
Vizinho
1
1
2
2
3
3
4
4
K25
Stisfaz
K26
Familia
K27
K28
K29
Kolegas
Vizinho
1
1
1
2
2
2
3
3
3
4
4
4
K30
Brigado wain tebes ba ita nia partisipasaun!
Appendices
211
Satisfaz
liu
La satisfaz
Pergunta: Iha jeral, ita sinte satisfaz husi suporta neebe ita hetan husi grupu
ema sira tuir mai ka lae?
Barak
Barak
oitoa
n
La satisfaz
liu
Lae
K15
Appendix J
Management of TB Treatment Interruption
212
Appendices
Source: (2008; Peloquin, 2003)

Figure 8 Management of TB Treatment Interruption
Appendices
213
Appendix K
In-depth Univariate Analysis
214
Appendices
Appendix K
Sociodemographic
factors
Number patient (%)
Table 29 In-depth Univariate Analysis

Adherence Scale
MARS (16 of 16 =
adherent)
Nonadherent
(N, %)
Min:0, max:8,
Mean: 5.921.45
Adherent
(N, %)
Lifestyle
VAS
(90% = adherent)
Nonadherent
(N, %)
Min: 10, max: 100
Mean: 76.2419.3
Adherent
(N, %))
Alcohol consumption (%)
Tobacco Smoking
Illicit drug use
Current
drinker
Stop at
start of
treatment
Never
Currently
smoking
Used to
smoke
Stop
start of
t'ment
Never
Yes, in
the
past
Never
Age
18-29
30-39
40-49
50-59
60+
119
(34.3)
63
(18.2)
38
(11.0)
35
(10.0)
92
(26.5)
20
(14.9)
5
(9.4)
2
(6.1)
3
(9.1)
8
(10.0)
114
(85.1)
48
(90.6)
31
(93.9)
30
(90.9)
72
(90.0)
24
(25.5)
13
(14.8)
4
(14.8)
12
(16.7)
11
(19.6)
70
(74.5)
75
(85.2)
23
(85.2)
60
(63.3)
45
(80.4)
18
(13.1)
8
(14.8)
4
(11.8)
6
(16.2)
8
(9.6)
47
(34.3)
27
(50.0)
17
(50.0)
16
(43.2)
47
(56.6)
72
(52.6)
19
(35.2)
13
(38.2)
15
(40.5)
28
(33.7)
5
(3.6)
2
(3.7)
0
(0.0)
2
(5.4)
3
(3.6)
11
(8.0)
2
(3.7)
6
(17.6)
1
(2.7)
12
(14.5)
41
(29.7)
32
(59.3)
13
(38.2)
11
(29.7)
26
(31.3)
81
(58.7)
18
(33.3)
15
(44.1)
23
(62.2)
42
(50.6)
2
(1.5)
1
(1.9)
1
(2.9)
0
(0.0)
0
(0.0)
133
(98.5)
51
(98.1)
33
(97.1)
37
(100)
83
(100)
191
(55.0)
156
(45.0)
18
(9.8)
20
(13.4)
166
(90.2)
129
(86.6)
28
(15.2)
36
(23.4)
156
(84.8)
118
(76.6)
32
(16.9)
12
(7.7)
121
(64.0)
33
(21.2)
36
(19.0)
111
(71.2)
11
(5.8)
1
(0.6)
27
(14.2)
5
(3.2)
113
(59.5)
10
(6.4)
39
(20.5)
140
(89.7)
4
(2.2)
0
(0.0)
181
(97.8)
156
(100)
118
(34.2)
212
(61.4)
15
(4.4)
13
(11.2)
25
(12.4)
0
(0.0)
103
(88.8)
177
(87.6)
14
(100)
17
(14.4)
43
(21.1)
4
(26.7)
101
(85.6)
161
(78.9)
11
(73.3)
19
(16.0)
23
(11)
1
(6.7)
50
(42.0)
100
(47.6)
4
(26.7)
50
(42.0)
87
(41.4)
10
(66.7)
4
(3.4)
8
(3.8)
0
(0.0)
11
(9.2)
16
(7.6)
5
(33.3)
46
(38.7)
73
(34.6)
4
(26.7)
58
(48.7)
114
(54.0)
6
(40.0)
2
(1.7)
2
(1.0)
0
(0.0)
115
(98.3)
206
(99.0)
15
(100)
Gender
Male
Female
Marital
Status
Never married
Married
Div/sep/wid
Appendices
215
Number patient
(%)
Sociodemographic
factors
Adherence Scale
MARS (16 of 16 =
adherent)
Lifestyle
VAS
(90% = adherent)
Alcohol consumption (%)
Tobacco Smoking
Illicit drug use
Adherent
(N, %)
Nonadherent
(N, %)
Adherent
(N, %))
Nonadherent
(N, %)
Current
drinker
Stop at
start of
treatment
Never
Currently
smoking
Used to
smoke
Stop
start of
t'ment
Never
Yes, in
the
past
Never
98
(28.3)
92
(26.6)
28
(8.1)
72
(20.7)
56
(16.1)
8
(8.8)
10
(11.2)
3
(11.1)
8
(11.4)
9
(16.4)
83
(91.2)
79
(88.8)
24
(88.9)
62
(88.6)
46
(83.6)
24
(25.5)
13
(14.8)
4
(14.8)
12
(16.7)
11
(19.6)
70
(74.5)
75
(85.2)
23
(85.2)
60
(83.3)
45
(80.4)
8
(8.2)
10
(11.0)
6
(22.2)
9
(12.3)
11
(19.6)
48
(49.5)
48
(52.7)
8
(29.6)
30
(41.1)
20
(35.7)
41
(42.3)
33
(36.3)
13
(48.1)
34
(36.6)
25
(44.6)
4
(4.1)
3
(3.3)
1
(3.7)
3
(4.1)
1
(1.8)
8
(8.2)
9
(9.8)
1
(3.7)
8
(11.0)
6
(10.7)
31
(32.0)
35
(35.0)
9
(33.3)
31
(42.5)
17
(30.4)
54
(55.7)
48
(48.9)
16
(59.3)
31
(42.5)
32
(57.1)
0
(0.0)
1
(1.1)
0
(0.0)
1
(1.5)
2
(3.6)
97
(100)
91
(98.9)
27
(100)
67
(98.5)
54
(96.4)
146
(42.1)
47
(13.5)
18
(5.2)
37
(10.7)
22
(6.3)
7
(2.0)
29
(8.4)
41
(11.8)
12
(8.5)
6
(13.3)
2
(11.8)
4
(11.4)
3
(14.3)
2
(28.6)
4
(14.3)
5
(12.8)
129
(91.5)
39
(86.7)
15
(88.2)
31
(88.6)
18
(85.7)
5
(71.4)
24
(85.7)
34
(87.2)
24
(16.8)
4
(8.5)
4
(23.5)
8
(22.9)
7
(35.0)
2
(28.6)
5
(17.9)
10
(24.4)
119
(83.2)
43
(91.5)
13
(76.5)
27
(77.1)
13
(65.0)
5
(71.4)
23
(82.1)
31
(75.6)
21
(14.4)
12
(26.1)
1
(5.9)
3
(8.1)
3
(13.6)
0
(0.0)
4
(13.8)
0
(0.0)
65
(45.5)
23
(50.0)
10
(58.8)
14
(37.8)
5
(22.7)
2
(28.6)
15
(51.7)
20
(48.8)
60
(41.1)
11
(23.9)
6
(35.3)
20
(54.1)
14
(63.6)
5
(71.4)
10
(34.5)
21
(51.2)
6
(4.1)
3
(6.4)
0
(0.0)
1
(2.7)
0
(0.0)
0
(0.0)
2
(6.9)
0
(0.0)
15
(10.3)
6
(12.8)
2
(11.8)
3
(8.1)
3
(13.6)
0
(0.0)
2
(6.9)
1
(2.4)
49
(33.6)
19
(40.4)
8
(47.1)
14
(37.8)
5
(22.7)
2
(28.6)
12
(41.4)
14
(34.1)
76
(52.1)
19
(40.4)
7
(41.2)
19
(51.4)
14
(63.6)
5
(71.4)
13
(44.8)
26
(63.4)
2
(1.4)
1
(2.2)
0
(0.0)
1
(2.8)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
142
(98.6)
45
(97.8)
17
(100)
35
(97.2)
22
(100)
6
(100)
29
(100)
41
(100)
Education
No education
Primary
Secondary
High school
College/Uni
Income
USD/month
<30
31-60
61-90
91-120
121-150
151-210
>210
Do not know
216
Appendices
Patients' TB Knowledge
N, %
N, %
N, %
N, %
No
TB curable
Yes
No
TB cured in one
week treatment
Yes
No
Discontinue
treatment after
feeling better
Yes
No
Varies
between
patients
Screen family
with direct
contact
Rest of life
Eat with family

members
Length of time required to cure

TB
6-8
months
Severe
mental
disorder
Moderate
mental
disorder
Mild mental
disorder
Likely to
be well
Sociodemographic
factors
Min: 10, max: 48, Mean: 26.838.0
1-2
months
Psychological distress K10
Yes
No
Yes
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
Age
18-29
30-39
40-49
50-59
60+
41
(33.1)
7
(13.5)
15
(12.1)
8
(15.4)
33
(26.6)
16
(30.8)
35
(28.2)
21
(40.4)
80
(58.8)
31
(57.4)
56
(41.2)
23
(42.6)
76
(56.3)
27
(50.9)
59
(43.7)
26
(49.1)
124
(90.5)
52
(98.1)
13
(9.5)
1
(1.9)
22
(16.8)
5
(9.3)
109
(83.2)
49
(90.7)
25
(18.4)
3
(5.6)
111
(81.6)
51
(94.4)
3
(2.2)
2
(3.7)
122
(89.7)
48
(88.9)
2
(1.5)
0
(0.0)
9
(6.6)
4
(7.4)
5
(14.7)
8
(22.9)
12
(15.2)
8
(23.5)
7
(20.0)
14
(17.7)
10
(29.4)
5
(14.3)
20
(25.3)
11
(32.4)
15
(42.9)
33
(41.8)
21
(63.6)
24
(64.9)
52
(64.2)
12
(36.4)
13
(35.1)
29
(35.8)
15
(42.9)
14
(37.8)
31
(37.8)
20
(57.1)
23
(62.2)
51
(62.2)
32
(91.4)
36
(97.3)
78
(95.1)
3
(8.6)
1
(2.7)
4
(4.9)
4
(11.8)
5
(13.9)
7
(8.9)
30
(88.2)
31
(86.1)
72
(91.1)
8
(22.9)
5
(13.5)
6
(7.2)
27
(77.1)
32
(86.5)
77
(92.8)
1
(2.9)
2
(5.4)
5
(6.1)
32
(91.4)
31
(83.8)
68
(82.9)
1
(2.9)
0
(0.0)
1
(1.2)
1
(2.9)
4
(10.8)
8
(9.8)
38
(21.3)
35
(24.0)
30
(16.9)
22
(15.1)
51
(28.7)
33
(22.6)
59
(33.1)
56
(38.4)
116
(61.7)
92
(60.1)
72
(38.3)
61
(39.9)
87
(46.3)
76
(49.4)
101
(53.7)
78
(50.6)
176
(92.6)
146
(94.8)
14
(7.4)
8
(5.2)
20
(10.9)
23
(15.3)
164
(89.1)
127
(84.7)
26
(13.8)
21
(13.5)
163
(86.2)
135
(86.5)
6
(3.2)
7
(4.5)
167
(88.8)
134
(85.9)
3
(1.6)
1
(0.6)
12
(6.4)
14
(9.0)
32
(29.6)
37
(18.4)
3
(21.4)
14
(13.0)
36
(17.9)
2
(14.3)
36
(33.3)
45
(22.4)
3
(21.4)
26
(24.1)
83
(41.3)
6
(42.9)
65
(55.1)
133
(64.3)
9
(60.0)
53
(44.9)
74
(35.7)
6
(40.0)
65
55.1)
133
(64.3)
9
(60.0)
53
(44.9)
74
(35.7)
6
(40.0)
108
(91.5)
199
(94.8)
14
(93.3)
10
(8.5)
11
(5.2)
1
(6.7)
14
(12.2)
27
(13.3)
2
(13.3)
101
(87.8)
176
(86.7)
13
(86.7)
21
(17.6)
25
(11.9)
1
(6.7)
98
(82.4)
185
(88.1)
14
(93.3)
3
(2.5)
9
(4.3)
1
(6.7)
107
(89.9)
180
(86.1)
13
(86.7)
1
(0.8)
3
(1.4)
0
(0.0)
8
(6.7)
17
(8.1)
1
(6.7)
Gender
Male
Female
Marital
Status
Never married
Married
Div/sep/wid
Education
Appendices
217
Patients' TB Knowledge
Psychological distress K10
College/Uni
Income
USD/month
<30
31-60
61-90
91-120
121-150
151-210
>210
Do not know
218
TB cured in one
week treatment
Discontinue
treatment after
feeling better
24
(26.4)
14
(16.3)
9
(33.3)
20
(30.3)
17
(32.1)
36
(39.6)
44
(51.2)
11
(40.7)
15
(22.7)
9
(17.0)
69
(73.4)
52
(56.5)
15
(62.5)
43
(58.1)
29
(51.8)
25
(26.6)
40
(43.5)
9
(37.5)
31
(41.9)
27
(48.2)
69
(73.4)
52
(56.5)
15
(62.5)
31
(58.1)
29
(51.8)
25
(26.6)
40
(43.5)
9
(37.5)
41
(41.9)
27
(48.2)
91
(95.8)
87
(94.6)
25
(92.6)
66
(90.4)
52
(92.9)
4
(4.2)
5
(5.4)
2
(7.4)
7
(9.6)
4
(7.1)
10
(11.0)
14
(15.6)
2
(8.0)
11
(15.5)
6
(10.7)
81
(89.0)
76
(84.4)
23
(92.0)
60
(84.5)
50
(89.3)
8
(8.3)
12
(13.2)
3
(11.1)
16
(21.6)
8
(14.3)
88
(91.7)
79
(86.8)
24
(88.9)
58
(78.4)
48
(85.7)
4
(4.2)
5
(5.5)
0
(0.0)
4
(5.4)
0
(0.0)
87
(91.6)
73
(80.2)
25
(92.6)
64
(86.5)
51
(91.1)
0
(0.0)
2
(2.2)
1
(3.7)
1
(1.4)
0
(0.0)
4
(4.2)
11
(12.1)
1
(3.7)
5
(6.8)
5
(8.9)
22
(16.5)
7
(15.9)
6
(33.3)
8
(23.5)
7
(31.8)
4
(57.1)
8
(29.6)
11
(28.2)
20
(15.0)
9
(20.5)
3
(16.7)
7
(20.6)
3
(13.6)
0
(0.0)
2
(7.4)
8
(20.5)
35
(26.3)
9
(20.5)
4
(22.2)
7
(20.6
6
(27.3)
3
(42.9)
9
(33.3)
11
(28.2
56
(42.1)
19
(43.2)
5
(27.8)
12
(35.3)
6
(27.3)
0
(0.0)
8
(29.6
9
(23.1)
78
(53.8)
23
(48.9)
13
(76.5)
24
(68.6)
12
(54.5)
5
(71.4)
18
(66.7)
35
(85.4)
67
(46.2)
24
(51.1)
4
(23.5)
11
(31.4)
10
(45.5)
2
(28.6)
9
(33.3)
6
(14.6)
78
(53.8)
23
(48.9)
13
(76.5)
24
(68.6)
12
(54.5)
5
(71.4)
18
(66.7)
35
(85.4)
67
(46.2)
24
(51.1)
4
(23.5)
11
(31.4)
10
(45.5)
2
(28.6)
9
(33.3)
6
(14.6)
137
(94.5)
44
(93.6)
18
(100.)
34
(94.4)
21
(95.6)
6
(85.7)
26
(89.7)
36
(90.0)
8
(5.5)
3
(6.4)
0
(0.0)
2
(5.6)
1
(4.5)
1
(14.3)
3
(10.3)
4
(10.0)
16
(11.3)
8
(17.0)
2
(11.8)
4
(12.1)
1
(4.5)
2
(28.6)
5
(17.2)
5
(13.5)
126
(88.7)
396
(83.0)
15
(88.4)
29
(87.9)
21
(95.5)
5
(71.4)
24
(82.8)
32
(86.5)
18
(12.4)
7
(14.9)
1
(5.6)
6
(16.7)
3
(13.6)
0
(0.0)
6
(20.7)
6
(14.6)
127
(87.6)
40
(85.1)
17
(94.4)
30
(83.3)
19
(86.4)
7
(100)
23
(79.3)
35
(85.4)
8
(5.5)
1
(2.1)
0
(0.0)
1
(2.8)
0
(0.0)
0
(0.0)
3
(10.7)
0
(0.0)
120
(82.8)
39
(83.0)
16
(88.9)
34
(94.4)
22
(100)
5
(71.4)
25
(89.3)
40
(97.6)
2
(1.4)
1
(2.1)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
1
(2.4)
15
(10.3)
6
(12.8)
2
(11.2)
1
(2.8)
0
(0.0)
2
(28.6)
0
(0.0)
0
(0.0)
No
Yes
No
Yes
No
Appendices
Yes
No
Yes
No
Varies
between
patients
18
(19.8)
13
(15.1)
6
(22.2)
9
(13.6)
6
(11.3)
Yes
1-2
months
13
(14.3)
15
(17.4)
1
(3.7)
22
(33.3)
21
(39.6)
Mild mental
disorder
Rest of life
High school
TB curable
6-8
months
Secondary
Screen family
with direct
contact
Severe
mental
disorder
Primary
Eat with family

members
Length of time required to cure

TB
Moderate
mental
disorder
No education
Likely to
be well
Sociodemographic
factors
Min: 10, max: 48, Mean: 26.838.0
Multidimensional Health Locus of control Scale (MHLCS)
N, %
N, %
N, %
N, %
N, %
N,%
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
Strongly
agree
Strongly
Disagree/dis
agree
N, %
Agree
Strongly
Disagree/dis
agree
Strongly
agree
Get better because other

people's care
Agree
Do not feel well consult

with medical
professionals
Strongly
agree
Regular contact with

doctor will avoid illness
Agree
Strongly
agree
Agree
If I take care myself I

will avoid illness
Strongly
Disagree/dis
agree
Strongly
agree
Agree
Main thing affect health

is what I do myself
Strongly
Disagree/dis
agree
Strongly
agree
Agree
Strongly
Disagree/dis
agree
Sociodemographic
factors
Powerful Others (3,5,14)

Min: 3, max: 12, Mean:9.81.52
Strongly
Disagree/dis
agree
Internal (6,12,13)
Min: 7, max: 12, Mean: 9.961.1
Age
18-29
30-39
40-49
50-59
60+
4
(2.9)
1
(1.9)
1
(2.9)
2
(5.4)
1
(1.2)
88
(64.7)
34
(63.0)
26
(70.3)
26
(70.3)
53
(64.6)
44
(32.4)
19
(35.2)
8
(22.9)
9
(24.3)
28
(34.1)
7
(5.1)
3
(5.6)
1
(2.9)
0
(0.0)
3
(3.7)
81
(59.6)
31
(57.4)
23
(65.7)
23
(62.2)
53
(64.4)
48
(35.3)
20
(37.0)
11
(31.4)
14
(37.8)
26
(31.7)
2
(1.5)
1
(1.9)
0
(0.0)
2
(5.4)
1
(1.2)
82
(59.9)
27
(50.0)
22
(62.9)
21
(56.8)
49
(59.0)
53
(38.7)
26
(48.1)
13
(37.1)
14
(37.8)
33
(39.8)
1
(0.7)
0
(0.0)
0
(0.0)
0
(0.0)
1
(1.2)
92
(66.7)
30
(55.6)
23
(65.7)
27
(73.0)
52
(62.7)
45
(32.6)
25
(44.4)
12
(34.3)
10
(27.0)
30
(36.1)
13
(9.5)
6
(11.1)
4
(11.4)
5
(13.5)
5
(6.0)
83
(60.6)
26
(48.1)
24
(68.6)
24
(16.9)
53
(63.9)
41
(29.9)
22
(40.7)
7
(20.0)
8
(21.6)
25
(30.1)
13
(9.4)
2
(3.7)
4
(11.4)
5
(13.5)
5
(6.0)
79
(57.2)
29
(53.7)
20
(57.1)
19
(51.4)
47
(56.6)
46
(33.3)
23
(42.6)
11
(31.4)
13
(35.1)
31
(37.3)
5
(2.7)
4
(2.6)
121
(64.4)
121
(64.4)
62
(33.0)
46
(29.5)
7
(3.7)
7
(4.5)
120
(63.8)
91
(58.3)
61
(32.4)
58
(37.3)
4
(2.1)
2
(1.3)
106
(55.8)
95
(60.9)
80
(42.1)
59
(37.8)
2
(1.0)
0
(0.0)
119
(62.3
105
(67.3)
70
(36.6
51
(32.7)
19
(9.9)
14
(9.0)
113
(59.2)
113
(59.2)
59
(30.9)
59
(30.9)
16
(8.4)
13
(8.3)
107
(56.0)
87
(55.8)
68
(35.6)
56
(35.9)
2
(1.7)
7
(3.3)
0
(0.0)
80
(67.2)
138
(66.0)
9
(60.0)
37
(31.1)
6
(40.0)
6
(40.0)
6
(5.0)
6
(2.9)
2
(13.3)
68
(57.1)
131
(62.7)
11
(73.3)
45
(37.8)
72
(34.4)
2
(13.3)
1
(0.8)
5
(2.4)
0
(0.0)
65
(55.1)
131
(61.8)
5
(33.3)
52
(44.1)
76
(35.8)
10
(66.7)
1
(0.8)
1
(0.5)
0
(0.0)
74
(62.2)
143
(67.5)
6
(40.0)
44
(37.0)
68
(32.1)
9
(60.0)
12
(10.2)
20
(9.4)
1
(6.7)
69
(58.5)
131
(61.8)
9
(60.0)
37
(31.4)
61
(28.8)
5
(33.3)
9
(7.6)
18
(8.5)
2
(13.3)
68
(57.1)
120
(56.6)
6
(40.0)
42
(35.3)
74
(34.9)
7
(46.7)
Gender
Male
Female
Marital Status
Never married
Married
Div/sep/wid
Education
Appendices
219
No education
Primary
Secondary
High school
College/Uni
Income
USD/month
<30
31-60
61-90
91-120
121-150
151-210
>210
Do not know
220
Strongly
agree
Agree
Get better because other

people's care
Strongly
Disagree/dis
agree
Strongly
agree
Do not feel well consult

with medical
professionals
Agree
Strongly
agree
Agree
Regular contact with

doctor will avoid illness
Strongly
Disagree/dis
agree
Strongly
agree
Agree
If I take care myself I

will avoid illness
Strongly
Disagree/dis
agree
Strongly
agree
Agree
Main thing affect health

is what I do myself
Strongly
Disagree/dis
agree
Strongly
agree
Agree
Strongly
Disagree/dis
agree
Sociodemographic
factors
Powerful Others (3,5,14)

Min: 3, max: 12, Mean:9.81.52
Strongly
Disagree/dis
agree
Internal (6,12,13)
Min: 7, max: 12, Mean: 9.961.1
4
(4.2)
1
(1.1)
1
(3.7)
2
(2.7)
1
(1.8)
67
(70.5)
60
(65.9)
19
(70.4)
45
(60.8)
35
(62.5)
24
(25.3)
30
(33.0)
7
(25.9)
27
(36.5)
30
(35.7)
1
(1.1)
4
(4.4)
0
(0.0)
5
(6.8)
4
(7.1)
64
(67.4)
50
(54.9)
22
(81.5)
41
(55.5)
34
(60.7)
30
(31.6)
37
(40.7)
5
(18.5)
28
(37.8)
18
(32.1)
3
(3.1)
1
(1.1)
1
(3.7)
1
(1.4)
0
(0.0)
63
(64.9)
53
(57.6)
13
(48.1)
44
(59.5)
27
(49.1)
31
(32.0)
38
(41.3)
13
(48.1)
29
(39.2)
28
(50.9)
1
(1.0)
0
(0.0)
1
(3.7)
0
(0.0)
0
(0.0)
71
(73.2)
58
(63.0)
13
(48.1)
45
(60.8)
36
(64.3)
25
(25.8)
34
(37.0)
13
(48.1)
29
(39.2)
20
(35.7)
7
(7.2)
14
(15.2)
4
(14.8)
7
(9.5)
1
(1.8)
70
(72.2)
49
(53.3)
12
(44.4)
43
(58.1)
35
(63.6)
20
(20.6)
29
(31.5)
11
(40.7)
24
(32.4)
19
(34.5)
9
(9.3)
12
(13.0)
2
(7.4)
6
(8.1)
0
(0.0)
60
(61.9)
44
(47.8)
14
(51.9)
39
(52.7)
36
(64.3)
28
(28.9)
36
(39.1)
11
(40.7)
29
(39.2)
20
(35.7)
6
(4.1)
0
(0.0)
0
(0.0)
0
(0.0)
1
(4.5)
0
(0.0)
0
(0.0)
2
(4.9)
83
(57.2)
28
(59.6)
10
(55.6)
28
(77.8)
14
(63.6)
5
(71.4)
23
(82.1)
36
(87.8)
56
(38.6)
19
(40.4)
8
(44.4)
8
(22.2)
7
(31.8)
2
(28.6)
5
(17.9)
3
(7.3)
8
(5.5)
2
(4.3)
0
(0.0)
2
(5.6)
1
(4.5)
0
(0.0)
1
(3.6)
0
(0.0)
88
(60.7)
30
(63.8)
13
(72.2)
25
(69.4)
12
(54.5)
5
(71.4)
15
(53.6)
23
(56.1)
49
(33.8)
15
(31.9)
5
(27.8)
9
(25.0)
9
(40.9)
2
(28.6)
12
(42.9)
18
(43.9)
4
(2.7)
1
(2.1)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
1
(2.4)
63
(43.2)
22
(46.8)
9
(50.0)
27
(73.0)
15
(68.2)
5
(83.3)
25
(86.2)
35
(85.4)
79
(54.1)
24
(51.1)
9
(50.0)
10
(27.0)
7
(31.8)
1
(16.7)
4
(13.8)
5
(12.2)
1
(0.7)
0
(0.0)
0
(0.0)
1
(2.7)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
72
(49.3)
25
(53.2)
10
(55.6)
29
(78.4)
16
(72.7)
6
(85.7)
22
(89.7)
40
(97.6)
73
(50.0)
22
(46.8)
8
(44.4)
7
(18.9)
6
(27.3)
1
(14.3)
3
(10.3)
1
(2.4)
19
(13.0)
3
(6.4)
0
(0.0)
3
(8.1)
2
(9.1)
0
(0.0)
4
(13.8)
2
(5.0)
67
(45.9)
23
(48.9)
12
(66.7)
27
(73.0)
17
(77.3)
5
(71.4)
23
(79.3)
36
(90.0)
60
(41.1)
21
(44.7)
6
(33.3)
7
(18.9)
3
(13.6)
2
(28.6)
2
(6.9)
2
(5.0)
17
(11.6)
1
(2.1)
0
(0.0)
4
(10.8)
3
(13.6)
0
(0.0)
3
(10.3)
1
(2.4)
61
(41.8)
22
(46.8)
9
(50.0)
26
(70.3)
13
(59.1)
6
(85.7)
21
(72.4)
36
(87.8)
68
(46.6)
24
(51.1)
9
(50.0)
7
(18.9)
6
(27.3)
1
(14.3)
5
(17.2)
4
(9.8)
Appendices

Chance
Min: 7, max: 12, Mean: 10.21.45
N, %
Strongly
Disagree/disa
gree
If it meant to be I will stay

healthy
Min: 23, max:36, Mean: 29.9833.443
N, %
N, %
N, %
N, %
Strongly
agree
N, %
Strongly
disagree
N, %
Strongly
agree
N,%
Agree
N, %
Strongly
agree
N, %
Agree
Agree
N, %
Strongly
Disagree/disa
gree
Agree
N, %
My good health is largely a matter

of good fortune
Disagree
N, %
Strongly
agree
Luck plays big part in how soon I will

recover from an illness
Strongly
Disagree/disa
gree
ocio-demographic
factors
Sum MHLC
Age
18-29
30-39
40-49
50-59
60+
4
(2.9)
1
(1.9)
2
(5.7)
0
(0.0)
0
(0.0)
89
(64.5)
27
(50.0)
22
(62.9)
24
(64.9)
45
(54.9)
45
(32.6)
26
(48.1)
11
(31.4)
13
(35.1)
37
(45.1)
1
(0.7)
1
(1.9)
2
(5.7)
1
(2.7)
0
(0.0)
79
(57.7)
27
(50.0)
18
(51.4)
22
(59.5)
43
(51.8)
57
(41.6)
26
(48.1)
15
(42.9)
14
(37.8)
40
(48.2)
3
(2.2)
0
(0.0)
1
(2.9)
1
(2.7)
0
(0.0)
76
(55.1)
26
(48.1)
19
(54.3)
22
(59.5)
45
(54.2)
59
(42.8)
28
(51.9)
15
(42.9)
14
(37.8)
38
(45.8)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
6
(4.4)
2
(3.7)
2
(5.7)
3
(8.1)
2
(2.4)
121
(89.6)
49
(90.7)
31
(88.6)
33
(89.2)
74
(90.2)
8
(5.9)
3
(5.6)
2
(5.7)
1
(2.7)
6
(7.3)
4
(2.1)
3
(1.9)
114
(60.0)
93
(59.6)
72
(37.9)
60
(38.5)
2
(1.1)
3
(1.9)
100
(52.6)
89
(57.1)
88
(46.3)
64
(41.0)
0
(0.0)
5
(3.2)
105
(55.0)
83
(53.2)
86
(45.0)
68
(43.6)
0
(0.0)
0
(0.0)
9
(4.8)
6
(3.9)
168
(89.4)
140
(90.3)
11
(5.9)
9
(5.8)
4
(3.4)
3
(1.4)
70
(58.8)
131
(62.1)
45
(37.8)
77
(36.5)
1
(0.8)
3
(1.4)
66
(55.9)
119
(56.1)
51
(43.2)
90
(42.5)
1
(0.8)
3
(1.4)
63
(52.9)
122
(57.5)
55
(46.2)
87
(41.0)
0
(0.0)
0
(0.0)
4
(3.4)
10
(4.7)
104
(89.7)
190
(90.0)
8
(6.9)
11
(5.2)
Gender
Male
Female
Marital Status
Never married
Married
Appendices
221

Chance
Min: 7, max: 12, Mean: 10.21.45
Div/sep/wid
Strongly
agree
Agree
Disagree
Min: 23, max:36, Mean: 29.9833.443
Strongly
disagree
Strongly
agree
Agree

healthy
Strongly
Disagree/disa
gree
Strongly
agree
Agree

of good fortune
Strongly
Disagree/disa
gree
Strongly
agree
Agree

Strongly
Disagree/disa
gree
ocio-demographic
factors
Sum MHLC
0
(0.0)
6
(40.0)
9
(60.0)
1
(6.7)
4
(26.7)
10
(66.7)
1
(6.7)
3
(20.0)
11
(73.3)
0
(0.0)
1
(6.7)
13
(86.7)
1
(6.7)
0
(0.0)
2
(2.2)
0
(0.0)
2
(2.7)
3
(5.4)
63
(65.6)
51
(55.4)
14
(51.9)
45
(60.8)
33
(58.9)
33
(34.4)
39
(42.4)
13
(48.1)
27
(36.5)
20
(35.7)
1
(1.0)
2
(2.2)
0
(0.0)
1
(1.4)
1
(1.8)
61
(62.9)
45
(48.9)
14
(51.9)
39
(53.4)
30
(53.6)
35
(36.1)
45
(48.9)
13
(48.1)
33
(45.2)
25
(44.6)
1
(1.0)
3
(3.3)
0
(0.0)
1
(1.4)
0
(0.0)
62
(63.9)
47
(51.1)
14
(51.9)
38
(51.4)
27
(48.2)
34
(35.1)
42
(45.7)
13
(48.1)
35
(47.3)
29
(51.8)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
5
(5.2)
5
(5.4)
1
(3.7)
3
(4.1)
1
(1.9)
88
(91.7)
78
(84.8)
25
(92.6)
65
(89.0)
51
(94.4)
3
(3.1)
9
(9.8)
1
(3.7)
5
(6.8)
2
(3.7)
3
(2.1)
0
(0.0)
1
(5.6)
0
(0.0)
0
(0.0)
0
(0.0)
67
(46.2)
25
(53.2)
9
(50.0)
27
(73.0)
17
(77.3)
6
(85.7)
75
(51.7)
22
(46.8)
8
(44.4)
10
(27.0)
5
(22.7)
1
(14.3)
3
(2.1)
2
(4.3)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
62
(42.5)
20
(43.5)
8
(44.4)
24
(64.9)
12
(54.5)
5
(71.4)
81
(55.5)
24
(52.4)
10
(55.6)
13
(35.1)
10
(45.5)
2
(28.6)
2
(1.4)
1
(2.1)
0
(0.0)
1
(2.7)
0
(0.0)
0
(0.0)
64
(43.8)
20
(42.6)
7
(38.9)
23
(62.2)
11
(50.0)
6
(85.7)
80
(54.8)
26
(55.3)
11
(61.1)
13
(35.1)
11
(50.0)
1
(14.3)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
11
(7.6)
1
(2.2)
0
(0.0)
1
(2.7)
0
(0.0)
0
(0.0)
119
(82.1)
41
(89.1)
17
(94.4)
36
(97.3)
22
(100)
6
(100)
15
(10.3)
4
(8.7)
1
(5.6)
0
(0.0)
0
(0.0)
0
(0.0)
Education
No education
Primary
Secondary
High school
College/Uni
Income USD/month
<30
31-60
61-90
91-120
121-150
151-210
222
Appendices

Chance
Min: 7, max: 12, Mean: 10.21.45
>210
Do not know
24
(82.8)
32
(78.0)
5
(17.2)
6
(14.6)
0
(0.0)
0
(0.0)
23
(79.3)
35
(85.4)
6
(20.7)
6
(14.6)
1
(3.4)
0
(0.0)
22
(75.9)
35
(85.4)
6
(20.7)
6
(14.6)
0
(0.0)
0
(0.0)
1
(3.4)
1
(2.5)
Strongly
agree
Agree
Strongly
disagree
Disagree
Min: 23, max:36, Mean: 29.9833.443
Strongly
agree
Agree

healthy
Strongly
Disagree/disa
gree
Strongly
agree
Agree
Strongly
agree
Agree
0
(0.0)
3
(7.3)

of good fortune
Strongly
Disagree/disa
gree

Strongly
Disagree/disa
gree
ocio-demographic
factors
Sum MHLC
28
(96.6)
39
(97.5)
0
(0.0)
0
(0.0)
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N,
%
10
(7.2)
6
(11)
2
(5.7)
22
(16.3)
2
(3.7)
4
(11.4)
40
(29.2)
14
(25.9)
9
(25.7)
40
(29.4)
15
(27.8)
8
(22.9)
16
(11.6)
9
(16.7)
5
(14.7)
33
(23.9)
6
(11.1)
10
(29.4)
14
(10.1)
4
(7.5)
6
(17.1)
9
(6.7)
3
(5.7)
0
(0.0)
28
(20.7)
6
(11.1)
4
(11.4)
9
(6.6)
4
(7.4)
1
(2.9)
70
(51.1)
31
(57.4)
16
(48.5)
2
(1.5)
0
(0.0)
0
(0.0)
10
(7.3)
1
(1.9)
2
(5.7)
18
(13.1)
6
(11.1)
6
(17.1)
35
(25.5)
9
(17.0)
8
(22.9)
34
(24.8)
16
(29.6)
7
(20.0)
23
(16.9)
5
(9.3)
2
(5.7)
8
(5.8)
2
(3.7)
0
(0.0)
The drug is
not useful
Prevent
others from
noticing
N, %
Consume
alcohol
N, %
Use drugs
N, %
Busy with
other things
N, %
Felt sick
N, %
Slept
through
doses
N,%
Drug
toxic/harmful
N, %
Change of
daily routine
N, %
Clinic too far
N, %
Avoid side
effect
Run out of
pills and did
not get them
50
(36.2)
21
(38.9)
16
(45.7)
No money
for care
seeking
36
(26.1)
10
(18.5)
8
(22.9)
Get nausea &

take no
replacement
82
(59.4)
38
(70.4)
20
(57.1)
Health clinic
provide no
drug
N, %
Problem take
pills at
specific time
N, %
Depressed/
Overwhelme
d
N, %
Too many
pills
Away from
home
Felt good
Sociodemographic
factors
Simply
forgot
Reasons for missing does
Age
18-29
30-39
40-49
Appendices
223
Problem take
pills at
specific time
Health clinic
provide no
drug
Busy with
other things
Use drugs
Get nausea &

take no
replacement
No money
for care
seeking
Run out of
pills and did
not get them
Prevent
others from
noticing
The drug is
not useful
12
(32.4)
18
(21.7)
5
(13.5)
14
(16.9)
2
(5.4)
13
(15.9)
1
(2.7)
6
(7.3)
2
(5.4)
10
(12.0)
9
(24.3)
11
(13.3)
0
(0.0)
4
(4.8)
27
(73.0)
41
(50.0)
0
(0.0)
1
(1.2)
0
(0.0)
1
(1.2)
9
(24.3)
15
(18.1)
11
(29.7)
18
(21.7)
7
(18.9)
22
(26.5)
1
(2.7)
5
(6.0)
1
(2.7)
3
(3.6)
118
(62.1)
89
(57.1)
46
(24.1)
37
(23.7)
77
(40.3)
67
(43.2)
15
(7.9)
12
(7.7)
18
(9.5)
15
(9.7)
49
(25.7)
47
(30.3)
47
(24.7)
46
(29.7)
27
(14.2)
46
(29.7)
36
(18.6)
28
(18.2)
18
(19.4)
13
(8.4)
11
(5.9)
13
(8.4)
28
(14.8)
30
(19.4)
12
(6.3)
6
(3.9)
103
(54.5)
82
(53.2)
3
(1.6)
0
(0.0)
12
(6.3)
2
(1.3)
30
(15.7)
24
(15.5)
43
(22.5)
38
(24.7)
47
(24.6)
39
(25.2)
17
(8.9)
19
(12.3)
11
(5.8)
3
(1.9)
73
(61.3)
126
(59.7)
7
(46.7)
33
(27.7)
46
(21.7)
4
(26.7)
52
(43.7)
83
(39.3)
9
(60.0)
10
(8.4)
16
(7.5)
1
(6.7)
21
(17.9)
12
(5.7)
0
(0.0)
42
(35.6)
51
(24.1)
3
(20.0)
39
(33.1)
50
(23.7)
4
(26.7)
16
(13.4)
27
(12.8)
5
(33.3)
33
(27.7)
26
(12.4)
4
(26.7)
15
(12.7)
15
(7.1)
1
(6.7)
9
(7.8)
13
(6.2)
2
(13.3)
23
(19.5)
33
(15.7)
2
(13.3)
11
(9.4)
5
(2.4)
2
(13.3)
66
(56.4)
111
(52.9)
7
(46.7)
2
(1.7)
1
(0.5)
0
(0.0)
8
(6.8)
5
(2.4)
1
(6.7)
19
(16.1)
27
(12.7)
7
(46.7)
32
(27.1)
43
(20.4)
6
(40.0)
34
(28.8)
45
(21.2)
7
(46.7)
20
(17.1)
14
(6.6)
2
(13.3)
11
(9.3)
2
(0.9)
1
(6.7)
57
(58.8)
58
(63.7)
21
(77.8)
42
(56.8)
29
(51.8)
14
(14.4)
31
(33.7)
7
(25.9)
16
(21.6)
15
(26.8)
37
(38.5)
48
(52.2)
14
(51.9)
27
(36.5)
18
(32.1)
4
(4.1)
8
(8.7)
1
(3.7)
9
(12.2)
5
(8.9)
3
(3.1)
5
(5.4)
0
(0.0)
14
(19.4)
11
(20.0)
27
(27.8)
22
(23.9)
9
(33.3)
22
(30.1)
16
(28.6)
22
(22.7)
28
(30.4)
7
(25.9)
16
(21.9)
20
(36.4)
12
(12.4)
16
(17.4)
4
(15.4)
9
(12.2)
8
(14.3)
11
(11.3)
12
(13.2)
4
(15.4)
20
(27.0)
16
(28.6)
5
(5.2)
4
(4.3)
3
(11.5)
13
(17.6)
6
(10.7)
6
(6.2)
7
(7.8)
0
(0.0)
6
(8.5)
5
(8.9)
12
(12.4)
16
(17.4)
6
(23.1)
16
(22.2)
8
(14.3)
4
(4.1)
1
(1.1)
1
(3.7)
8
(11.0)
4
(7.3)
51
(53.1)
56
(60.9)
18
(69.2)
32
(43.8)
28
(50.9)
1
(1.0)
0
(0.0)
0
(0.0)
1
(1.4)
1
(1.8)
1
(1.0)
2
(2.2)
2
(7.4)
5
(6.8)
4
(7.3)
18
(18.6)
15
(16.3)
6
(22.2)
10
(13.5)
5
(9.1)
20
(20.6)
25
(27.2)
5
(19.2)
19
(25.7)
12
(21.8)
16
(16.5)
28
(30.4)
8
(29.6)
19
(25.7)
15
(27.3)
3
(3.1)
8
(8.7)
1
(3.7)
13
(17.6)
10
(18.5)
2
(2.1)
1
(1.1)
1
(3.7)
7
(9.5)
3
(5.5)
91
(62.8)
29
(61.7)
43
(29.5)
15
(31.9)
71
(49.0)
28
(59.6)
11
(7.5)
5
(10.6)
14
(9.8)
7
(15.2)
56
(38.6)
16
(34.0)
37
(25.5)
14
(30.4)
29
(19.9)
9
(19.1)
32
(22.1)
8
(17.0)
9
(6.2)
7
(15.2)
14
(9.7)
3
(6.4)
27
(18.5)
5
(10.9)
6
(4.1)
7
(15.6)
75
(51.7)
28
(60.9)
1
(0.7)
1
(2.2)
4
(2.7)
6
(13.0)
30
(20.5)
10
(21.7)
43
(29.5)
17
(37.8)
41
(28.1)
18
(39.1)
13
(8.9)
12
(26.1)
5
(3.4)
6
(13.)
60+
Consume
alcohol
Depressed/
Overwhelme
d
12
(32.4)
21
(25.3)
Felt sick
Drug
toxic/harmful
0
(0.0)
5
(6.1)
Slept
through
doses
Change of
daily routine
1
(2.7)
8
(9.6)
Clinic too far
22
(59.5)
35
(42.7)
Avoid side
effect
8
(21.6)
21
(25.3)
Too many
pills
20
(54.1)
47
(57.3)
Away from
home
50-59
Felt good
Sociodemographic
factors
Simply
forgot
Gender
Male
Female
Marital
Status
Never married
Married
Div/sep/wid
Education
No education
Primary
Secondary
High school
College/Uni
Income
USD/month
<30
31-60
224
Appendices
The drug is
not useful
0
(0.0)
2
(5.4)
0
(0.0)
0
(0.0)
0
(0.0)
2
(4.9)
Prevent
others from
noticing
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
1
(2.4)
Run out of
pills and did
not get them
8
(44.4)
22
(61.1)
7
(31.8)
3
(42.9)
14
(50.0)
28
(68.3)
No money
for care
seeking
1
(5.6)
0
(0.0)
0
(0.0)
0
(0.0)
3
(10.3)
1
(2.4)
Get nausea &

take no
replacement
1
(5.6)
7
(19.4)
5
(22.7)
1
(14.3)
7
(24.1)
5
(12.5)
Consume
alcohol
0
(0.0)
2
(5.4)
3
(13.6)
1
(14.3)
1
(3.7)
0
(0.0)
Use drugs
2
(11.1)
3
(8.3)
3
(13.6)
0
(0.0)
2
(6.9)
5
(12.2)
Busy with
other things
6
(33.3)
2
(5.4)
6
(27.3)
2
(28.6)
4
(14.3)
4
(9.8)
Health clinic
provide no
drug
3
(16.7)
2
(5.6)
1
(4.5)
0
(0.0)
2
(6.9)
3
(7.3)
Felt sick
4
(22.2)
11
(29.7)
4
(18.2)
2
(28.6)
7
(24.1)
14
(34.1)
Slept
through
doses
Clinic too far

6
(33.3)
8
(21.6)
1
(4.5)
1
(14.3)
2
(6.9)
6
(14.6)
Problem take
pills at
specific time
Do not know
1
(5.6)
1
(2.7)
2
(9.1)
2
(28.6)
2
(6.9)
4
(9.8)
Depressed/
Overwhelme
d
>210
2
(11.1)
3
(8.1)
2
(9.1)
0
(0.0)
1
(3.4)
3
(7.3)
Drug
toxic/harmful
151-210
6
(33.3)
9
(24.3)
4
(18.2)
1
(14.3)
11
(37.9)
14
(34.1)
Change of
daily routine
121-150
Avoid side
effect
91-120
5
(27.8)
6
(16.2)
3
(13.6)
2
(28.6)
4
(13.8)
5
(12.2)
Too many
pills
12
(66.7)
22
(59.5)
9
(40.9)
2
(28.6)
14
(48.3)
28
(68.3)
Away from
home
61-90
Felt good
Sociodemographic
factors
Simply
forgot
1
(5.6)
6
(16.2)
2
(9.1)
0
(0.0)
3
(10.3)
2
(4.9)
2
(11.1)
6
(16.2)
3
(13.6)
2
(28.6)
2
(6.9)
6
(14.6)
4
(22.2)
8
(21.6)
3
(13.6)
1
(14.3)
7
(24.1)
4
(9.8)
2
(11.1)
2
(5.4)
4
(18.2)
0
(0.0)
2
(6.9)
1
(2.4)
0
(0.0)
0
(0.0)
1
(4.5)
0
(0.0)
1
(3.4)
1
(2.4)
Tuberculosis Related Stigma Scale
N, %
N, %
N, %
N, %
People who have TB

are afraid to disclose
TB status with their
family
People who have TB

feel guilty of their
family having to look
after them
People who have TB

afraid to tell others
outside their family
members
People who have TB

feel hurt of others'
reactions knowing
they have TB
People may not eat or

drink with relatives
who have TB
People would not try

to touch others with
TB
People afraid of those

with TB
People do not want to

talk to others with TB
People keep distance

from others with TB
People feel
uncomfortable being
near others with TB

drink with friends who
have TB
Socio-demographic
factors
People who have TB

feel alone
Community perspectives towards tuberculosis
N, %
N, %
N,%
N, %
N, %
N, %
N, %
N, %
100
(73.5)
54
(39.4)
85
(62.0)
109
(79.6)
79
(57.7)
67
(48.9)
70
(51.1)
57
(41.9)
Age
18-29
Appendices
110
(80.3)
113
(82.5)
102
(74.5)
81
(59.1)
225

37
(68.5)
28
(80.0)
30
(81.1)
64
(78.0)
27
(50.0)
24
(68.6)
23
(62.2)
46
(56.1)
38
(70.4)
26
(74.3)
26
(70.3)
61
(74.4)
42
(77.8)
31
(88.6)
29
(78.4)
71
(86.6)
36
(66.7)
20
(57.1)
25
(67.6)
54
(65.9)
31
(59.6)
19
(55.9)
20
(54.1)
46
(56.1)
25
(46.3)
19
(55.9)
19
(51.4)
44
(53.7)
22
(40.7)
19
(54.3)
16
(43.2)
44
(57.3)
160
(84.2)
129
(82.7)
164
(86.3)
126
(81.3)
149
(78.4)
121
(78.1)
123
(64.7)
104
(67.1)
140
(74.1)
119
(78.8)
91
(47.9)
83
(53.5)
128
(67.4)
108
(69.7)
152
(80.0)
130
(83.9)
110
(57.9)
104
(67.1)
96
(51.1)
87
(56.5)
91
(48.1)
86
(55.5)
83
(43.7)
75
(48.7)
98
(83.1)
181
(85.4)
9
(60.0)
100
(84.7)
178
(84.0)
11
(78.6)
85
(72.0)
175
(82.5)
9
(64.3)
66
(55.9)
150
(70.8)
10
(71.4)
81
(74.8)
166
(78.3)
8
(57.1)
45
(38.1)
128
(58.5)
5
(35.7)
74
(62.7)
154
(72.6)
8
(57.1)
92
(78.0)
176
(83.0)
13
(92.9)
62
(52.5)
143
(67.5)
8
(57.1.)
57
(48.7)
116
(55.2)
9
(64.3)
59
(50.0)
109
(51.7)
8
(57.1)
44
(37.6)
106
(50.0)
7
(50.0)
81
(83.5)
82
(89.1)
22
(84.6)
82
(85.4)
81
(88.0)
22
(84.6)
83
(86.5.)
74
(80.4)
20
(76.9)
68
(70.8)
73
(79.3)
16
(61.5)
75
(78.1)
72
(78.3)
20
(76.9)
56
(58.3)
52
(56.5)
13
(50.0)
73
(76.0)
64
(69.6)
16
(61.5)
82
(85.4)
80
(87.0)
23
(88.5)
67
(69.8)
64
(69.6)
17
(65.4)
53
(55.8)
54
(59.3)
18
(69.2)
49
(51.0)
51
(56.0)
16
(61.5)
48
(50.0)
42
(46.2)
15
(57.7)
People who have TB

family
31
(57.4)
27
(77.1)
29
(78.4)
59
(72.0)
People who have TB

after them
38
(70.4)
30
(85.7)
32
(86.5)
68
(82.9)
People who have TB

members
43
(79.6)
31
(88.6)
33
(89.2)
70
(85.4)
People who have TB

feel alone

who have TB
44
(81.5)
30
(85.7)
32
(86.5)
73
(88.0)
People who have TB

reactions knowing
they have TB

TB
60+

with TB
50-59

40-49

from others with TB
30-39
People feel
uncomfortable being
near others with TB
Socio-demographic
factors

have TB
Gender
Male
Female
Marital Status
Never married
Married
Div/sep/wid
Education
No education
Primary
Secondary
226
Appendices

55
(74.3)
36
(65.5)
32
(43.2)
21
(37.5)
52
(70.3)
30
(53.6)
57
(77.0)
39
(69.6)
39
(52.7)
26
(46.4
33
(45.2)
24
(42.9)
33
(44.6)
27
(48.2)
30
(40.5)
22
(39.3)
130
(89.0)
38
(80.9)
16
(88.9)
24
(66.7)
17
(77.3)
7
(100)
24
(82.8)
33
(80.5)
129
(89.0)
42
(89.4)
15
(83.3)
23
(63.9)
17
(77.3)
6
(85.7)
24
(82.8)
34
(82.9)
114
(78.6)
38
(80.9)
14
(77.8)
28
(72.2)
18
(81.8)
7
(100)
23
(79.3)
30
(73.2)
93
(64.1)
27
(57.4)
15
(83.3)
22
(61.1)
15
(68.2)
5
(71.4)
20
(69.0)
30
(73.2)
112
(77.2)
35
(74.5)
11
(61.1)
26
(72.2)
16
(72.7)
5
(83.3)
21
(72.4)
33
(80.5)
69
(47.6)
21
(44.7)
9
(50.0)
16
(44.4)
12
(54.5)
5
(71.4)
18
(62.1)
24
(58.5)
105
(72.4)
32
(68.1)
13
(72.2)
21
(58.3)
14
(63.6)
6
(85.7)
20
(69.0)
25
(61.0)
122
(84.1)
36
(76.6)
14
(77.8)
29
(80.6)
18
(81.8)
7
(100)
21
(72.4)
35
(85.4)
104
(71.7)
25
(53.2)
11
(61.1)
23
(63.9)
12
(54.5)
3
(42.9)
12
(41.4)
24
(58.5)
95
(65.5)
21
(46.7)
11
(61.1)
16
(45.7)
10
(45.5)
3
(42.9)
10
(34.5)
17
(41.5)
93
(64.1)
26
(56.5)
10
(55.6)
15
(41.7)
11
(50.0)
4
(57.1)
10
(34.5)
8
(19.5)
79
(54.5)
26
(55.3)
9
(50.0)
13
(36.1)
9
(40.9)
2
(28.6)
10
(34.5)
10
(25.0)
People who have TB

family
39
(52.7)
30
(53.6)
People who have TB

after them

who have TB
53
(71.6)
39
(69.6)
People who have TB

members

TB
61
(82.4)
43
(76.8)
People who have TB

feel alone

with TB
59
(79.7)
44
(78.6)
People who have TB

reactions knowing
they have TB

College/Uni

from others with TB
High school
People feel
uncomfortable being
near others with TB
Socio-demographic
factors

have TB
Income USD/month
<30
31-60
61-90
91-120
121-150
151-210
>210
Do not know
Appendices
227
Sociodemographic
factors
Health Care Professional-Patients Communication
Disagree
Neither
Agree
N, %
N, %
N,%
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
Very
satisfied
Agree
Somewhat
satisfied
Neither
Somewhat
Dissatisfied
Disagree
Very
dissatisfied
Overall support satisfaction
Agree
HCP explained treatment

procedures clearly
Neither
N, %
HCP explained changes

expect in health when
taking TB medication
Disagree
Disagree
Agree
N, %
HCP explained how

treatment would be
Agree
N, %
HCP explained what

treatment would be
Neither
N, %
Neither
Disagree
HCP explained possible side

effect of TB drug
Age
18-29
30-39
40-49
50-59
60+
30
(21.7)
13
(24.1)
7
(20.0)
9
(24.3)
19
(22.9)
12
(8.7)
3
(5.6)
2
(5.7)
2
(5.4)
14
(16.9)
96
(69.6)
38
(70.4)
26
(74.3)
26
(70.3)
50
(60.2)
7
(5.1)
3
(5.6)
2
(5.7)
2
(5.4)
7
(8.4)
11
(8.0)
2
(3.7)
3
(8.6)
4
(10.8)
5
(6.0)
119
(86.9)
49
(90.7)
30
(85.7)
31
(83.8)
71
(85.5)
6
(4.4)
2
(3.7)
0
(0.0)
1
(2.7)
7
(8.4)
9
(6.6)
4
(7.4)
5
(14.3)
4
(10.8)
5
(6.0)
122
(89.1)
48
(88.9)
30
(85.7)
32
(86.5)
71
(85.5)
7
(5.1)
2
(3.7)
1
(2.9)
2
(5.4)
6
(7.2)
5
(3.6)
0
(0.0)
2
(5.7)
4
(10.8)
9
(10.8)
126
(91.3)
52
(96.3)
32
(91.4)
31
(83.8)
68
(81.9)
7
(5.1)
2
(3.7)
1
(2.9)
1
(2.7)
4
(4.9)
6
(4.3)
3
(5.6)
2
(5.7)
2
(5.4)
9
(11.0)
125
(90.6)
49
(90.7)
32
(91.4)
32
(91.9)
69
(84.1)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
8
(5.9)
2
(3.7)
0
(0.0)
1
(2.7)
5
(6.0)
82
(60.3)
24
(44.4)
27
(81.8)
27
(73.0)
57
(68.7)
46
(33.8)
28
(51.9)
6
(18.2)
9
(24.3)
21
(25.3)
36
(18.8)
42
(26.9)
16
(8.4)
17
(10.9)
139
(72.8)
97
(62.2)
11
(5.8)
10
(6.4)
9
(4.7)
16
(10.3)
170
(89.5)
130
(83.3)
7
(3.7)
9
(5.8)
13
(6.8)
14
(9.0)
171
(89.5)
132
(85.2)
7
(3.7)
11
(7.1)
10
(5.2)
10
(6.4)
174
(91.1)
135
(86.5)
7
(3.7)
8
(5.2)
12
(6.3)
10
(6.5)
172
(90.1)
137
(88.4)
0
(0.0)
0
(0.0)
8
(4.3)
8
(5.2)
118
(62.8)
99
(63.9)
62
(33.0)
48
(31.0)
24
(20.2)
51
(24.1)
3
(20.0)
13
(10.9)
18
(8.5)
1
(6.7)
82
(68.9)
143
(67.5)
11
(73.3)
4
(3.4)
16
(7.5)
1
(6.7)
11
(9.3)
12
(5.7)
1
(6.7)
103
(87.3)
184
(86.8)
13
(86.7)
3
(2.5)
12
(5.7)
1
(6.7)
11
(9.3)
14
(6.6)
1
(6.7)
104
(88.1)
186
(87.7)
13
(86.7)
3
(2.5)
12
(5.7)
1
(6.7)
11
(9.3)
14
(6.6)
1
(6.7)
104
(88.1)
186
(87.7)
13
(86.7)
4
(3.4)
10
(4.7)
1
(6.7)
4
(3.4)
16
(7.6)
1
(6.7)
111
(93.3)
185
(87.7)
13
(86.7)
0
(0.0)
5
(4.2)
10
(4.8)
1
(6.7)
72
(61.0)
132
(63.2)
12
(80.0)
41
(34.7)
67
(32.1)
2
(13.3)
Gender
Male
Female
Marital Status
Never married
Married
Div/sep/wid
Education
228
Appendices
0
(0.0)
Sociodemographic
factors
Health Care Professional-Patients Communication
Secondary
High school
College/Uni
Income
USD/month
<30
31-60
61-90
91-120
121-150
151-210
>210
Do not know
Appendices
Somewhat
satisfied
Somewhat
Dissatisfied
Very
dissatisfied
Overall support satisfaction
8
(8.2)
6
(6.5)
1
(3.7)
2
(2.7)
1
(1.8)
8
(8.2)
6
(6.5)
2
(7.4)
2
(2.7)
2
(3.6)
81
(83.5)
80
(87.0)
24
(88.9)
70
(94.6)
53
(94.6)
5
(5.2)
5
(5.4)
3
(11.1)
1
(1.4)
1
(1.8)
7
(7.3)
6
(6.5)
2
(7.4)
1
(1.4)
5
(8.9)
84
(87.5)
21
(88.0)
22
(81.5)
72
(97.3)
50
(89.3)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
4
(4.1)
7
(7.6)
2
(8.0)
2
(2.8)
1
(1.8)
72
(74.2)
52
(56.5)
14
(56.0)
48
(66.7)
30
(53.6)
21
(21.6)
33
(35.9)
9
(36.0)
22
(30.6)
25
(44.6)
30
(20.5)
8
(17.0)
4
(22.2)
10
(27.0)
4
(18.2)
1
(14.3)
7
(24.1)
14
(34.1)
10
(6.8)
7
(14.9)
0
(0.0)
4
(10.8)
1
(4.5)
0
(0.0)
1
(3.4)
10
(24.4)
106
(72.6)
32
(68.1)
14
(77.8)
23
(62.2)
17
(77.3)
6
(85.7)
21
(72.4)
17
(41.5)
12
(8.3)
1
(2.1)
1
(5.6)
1
(2.7)
0
(0.0)
0
(0.0)
4
(13.8)
2
(4.9)
10
(6.9)
9
(19.1)
2
(11.1)
2
(5.4)
0
(0.0)
0
(0.0)
1
(3.4)
1
(2.4)
123
(84.8)
37
(78.7)
15
(83.3)
34
(91.9)
22
(100)
7
(100)
24
(82.8)
38
(92.7)
12
(8.3)
0
(0.0)
1
(5.6)
0
(0.0)
0
(0.0)
0
(0.0)
2
(6.9)
1
(2.4)
13
(9.0)
8
(17.0)
0
(0.0)
3
(8.1)
1
(4.5)
0
(0.0)
2
(6.9)
0
(0.0)
120
(82.8)
39
(83.0)
17
(94.4)
34
(91.9)
21
(95.5)
7
(100)
25
(86.2)
40
(97.6)
12
(8.2)
0
(0.0)
0
(0.0)
2
(5.4)
0
(0.0)
0
(0.0)
2
(6.9)
2
(4.9)
14
(9.6)
5
(10.6)
1
(5.6)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
120
(82.2)
42
(89.4)
17
(94.4)
35
(94.6)
22
(100)
7
(100)
27
(93.1)
39
(95.1)
13
(8.9)
0
(0.0)
0
(0.0)
1
(2.8)
0
(0.0)
0
(0.0)
1
(3.4)
0
(0.0)
12
(8.2)
5
(10.6)
1
(5.6)
1
(2.8)
1
(4.5)
0
(0.0)
1
(3.4)
1
(2.4)
121
(82.9)
42
(89.4)
17
(94.4)
34
(94.4)
21
(95.5)
7
(100)
27
(93.1)
40
(97.6)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
11
(7.6)
1
(2.2)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
3
(10.3)
1
(2.4)
80
(55.2)
23
(51.1)
10
(55.6)
28
(77.8)
17
(77.3)
3
(42.9)
19
(65.5)
37
(90.2)
54
(37.2)
21
(46.7)
8
(44.4)
8
(22.2)
5
(22.7)
4
(57.1)
7
(24.1)
3
(7.3)
Very
satisfied
83
(85.6)
82
(90.1)
23
(85.2)
65
(87.8)
49
(87.5)
Agree
8
(8.2)
3
(3.3)
3
(11.1)
7
(9.5)
6
(10.7)
Neither
6
(6.2)
6
(6.6)
1
(3.7)
2
(2.7)
1
(1.8)
Agree
86
(88.7)
78
(84.8)
21
(77.8)
65
(89.0)
49
(87.5)
Agree
4
(4.1)
5
(5.4)
5
(18.5)
4
(5.5)
7
(12.5)
Neither
7
(7.2)
9
(9.8)
1
(3.7)
4
(5.5)
0
(0.0)
Agree
56
(57.7)
67
(72.8)
13
(48.1)
59
(79.7)
40
(71.4)
229
Disagree
HCP explained treatment

procedures clearly
13
(13.4)
5
(5.4)
4
(14.8)
5
(6.8)
6
(10.7)
Neither
Disagree
HCP explained changes

expect in health when
taking TB medication
28
(28.9)
20
(21.7)
10
(37.0)
10
(13.5)
10
(17.9)
Agree
Disagree
HCP explained how

treatment would be
Neither
Primary
HCP explained what

treatment would be
Disagree
No education
Neither
Disagree
HCP explained possible side

effect of TB drug
Traditional Healer
Strongly Agree
Somewhat
Agree
Neither
Somewhat
Disagree
Strongly
disagree
Strongly Agree
Neither
Strongly
disagree
Neither
Strongly
disagree
Unaffordable cost of transport
Somewhat
Agree
68
(35.8)
56
(35.9)
Somewhat
Disagree
84
(61.3)
32
(59.3)
25
(71.4)
21
(56.8)
60
(72.3)
Strongly Agree
53
(38.7)
22
(40.7)
10
(28.6)
16
(43.2)
23
(27.7)
N, %
Somewhat
Agree
N, %
Somewhat
Disagree
N, %
Clinics too far
Strongly Agree
No
Somewhat
Agree
Yes
Neither
Strongly
disagree
Visit traditional
healer
Somewhat
Disagree
Traditional medicine
Socio-demographic
factors
N, %
N, %
N,%
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
N, %
3
(5.7)
0
(0.0)
0
(0.0)
0
(0.0)
2
(8.7)
6
(11.3)
1
(4.5)
0
(0.0)
0
(0.0)
2
(8.7)
17
(32.1)
6
(27.3)
2
(20.0)
5
(31.2)
5
(21.7)
25
(47.2)
15
(68.2)
7
(70.0)
11
(68.8)
13
(56.5)
2
(3.8)
0
(0.0)
1
(10.0)
0
(0.0)
1
(4.3)
3
(5.9)
2
(9.5)
1
(10.0)
0
(0.0)
1
(4.3)
22
(43.1)
8
(38.1)
2
(20.0)
5
(31.3)
6
(26.1)
13
(25.5)
7
(33.3)
4
(40.0)
7
(43.8)
10
(43.5)
12
(23.5)
4
(19.0)
3
(30.0)
4
(25.0)
6
(26.1)
1
(2.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
1
(1.9)
3
(14.3)
0
(0.0)
0
(0.0)
1
(4.3)
21
(40.4)
10
(47.6)
3
(30.0)
8
(50.0)
7
(30.4)
15
(28.8)
5
(23.8)
4
(40.0)
4
(25.0)
10
(43.5)
13
(25.0)
3
(14.3)
3
(30.0)
4
(25.0)
5
(21.7)
2
(3.8)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
1
(1.9)
4
(19.0)
0
(0.0)
0
(0.0)
1
(4.3)
21
(40.4)
10
(47.6)
4
(40.0)
9
(56.3)
5
(21.7)
14
(26.9)
5
(23.8)
5
(50.0)
5
(31.3)
12
(52.2)
12
(23.1)
2
(9.5)
1
(10.0)
2
(12.5)
5
(21.7)
4
(7.7)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
122
(64.2)
100
(64.1)
4
(5.9)
1
(1.8)
5
(7.4)
4
(7.1)
19
(27.9)
16
(28.6)
39
(57.4)
32
(57.1)
1
(1.5)
3
(5.4)
6
(9.1)
1
(1.8)
25
(37.9)
18
(32.7)
18
(27.3)
23
(41.8)
16
(24.2)
13
(23.6)
1
(1.5)
0
(0.0)
4
(6.1)
1
(1.8)
29
(43.9)
20
(35.7)
19
(28.8)
19
(33.9)
14
(21.2)
14
(25.0)
0
(0.0)
2
(3.6)
4
(6.1)
2
(3.6)
29
(43.9)
20
(35.7)
20
(30.3)
21
(37.5)
11
(16.7)
11
(19.6)
2
(3.0)
2
(3.6)
44
(37.3)
74
(34.9)
5
(33.3)
74
(62.7)
138
(65.1)
10
(66.7)
4
(9.1)
1
(1.4)
0
(0.0)
4
(9.1)
4
(5.4)
1
(20.0)
17
(38.6)
15
(20.3)
2
(40.0)
18
(40.9)
51
(68.9)
2
(40.0)
1
(2.3)
3
(4.1)
0
(0.0)
3
(7.0)
4
(5.6)
0
(0.0)
15
(34.9)
25
(34.7)
2
(40.0)
12
(27.9)
27
(37.5)
2
(40.0)
12
(27.9)
16
(22.2)
1
(20.0)
1
(2.3)
0
(0.0)
0
(0.0)
1
(2.3)
4
(5.6)
0
(0.0)
17
(38.6)
28
(38.9)
3
(60.0)
14
(31.8)
22
(30.6)
2
(40.0)
12
(27.3)
16
(22.2)
0
(0.0)
0
(0.0)
2
(2.8)
0
(0.0)
1
(2.3)
5
(6.9)
0
(0.0)
15
(34.1)
30
(41.7)
3
(60.0)
13
(29.5)
26
(36.1)
2
(40.0)
11
(25.0)
11
(15.3)
0
(0.0)
4
(9.1)
0
(0.0)
0
(0.0)
26
(26.8)
44
(47.8)
7
(26.9)
71
(73.2)
48
(52.2)
19
(73.1)
2
(7.7)
0
(0.0)
1
(14.3)
2
(7.7)
3
(6.8)
0
(0.0)
7
(26.9)
7
(15.9)
1
(14.3)
15
(57.7)
31
(70.5)
5
(71.4)
0
(0.0)
3
(6.8)
0
(0.0)
1
(3.8)
2
(4.5)
1
(14.3)
9
(34.6)
14
(31.8)
2
(28.6)
8
(30.8)
20
(45.5)
2
(28.6)
8
(30.8)
8
(18.2)
1
(14.3)
0
(0.0)
0
(0.0)
1
(14.3)
1
(3.8)
2
(4.5)
1
(14.3)
10
(38.5)
16
(36.4)
4
(57.1)
8
(30.8)
15
(34.1)
1
(14.3)
7
(26.9)
10
(22.7)
1
(14.7)
0
(0.0)
1
(2.3)
0
(0.0)
1
(3.8)
3
(6.8)
1
(14.3)
11
(42.3)
18
(40.9)
4
(57.1)
10
(38.5)
17
(38.6)
0
(0.0)
4
(15.4)
6
(13.6)
1
(14.3)
0
(0.0)
0
(0.0)
1
(14.3)
Age
18-29
30-39
40-49
50-59
60+
Gender
Male
Female
Marital Status
Never married
Married
Div/sep/wid
Education
No education
Primary
Secondary
230
Appendices
Traditional Healer
Traditional medicine
Somewhat
Agree
Strongly Agree
1
(4.3)
0
(0.0)
2
(10.0)
1
(4.2)
6
(30.0)
12
(50.0)
4
(20.0)
7
(29.2)
8
(40.0)
4
(16.7)
0
(0.0)
0
(0.0)
1
(4.8)
0
(0.0)
8
(38.1)
11
(45.8)
6
(28.6)
8
(33.3)
5
(23.8)
5
(20.8)
1
(4.8)
0
(0.0)
1
(4.8)
0
(0.0)
5
(23.8)
11
(45.8)
7
(33.3)
7
(29.2)
7
(33.3)
4
(16.7)
1
(4.8)
2
(8.3)
51
(34.9)
25
(53.2)
9
(50.0)
13
(36.1)
8
(36.4)
2
(28.6)
6
(20.7)
10
(24.4)
95
(65.1)
22
(46.8)
9
(50.0)
23
(63.9)
14
(63.6)
5
(71.4)
23
(79.3)
31
(75.6)
1
(2.0)
1
(4.0)
0
(0.0)
1
(7.7)
0
(0.0)
0
(0.0)
0
(0.0)
2
(20.0)
3
(5.9)
1
(4.0)
1
(11.1)
3
(23.1)
0
(0.0)
0
(0.0)
0
(0.0)
1
(10.0)
15
(29.4)
7
(28.0)
3
(33.3)
1
(7.7)
4
(50.0)
1
(50.0)
2
(33.3)
2
(20.0)
30
(58.8)
15
(60.0)
5
(55.6)
7
(53.8)
4
(50.0)
1
(50.0)
4
(66.7)
5
(50.0)
2
(3.9)
1
(4.0)
0
(0.0)
1
(7.7)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
4
(8.2)
2
(8.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
1
(10.0)
13
(26.5)
4
(16.0)
4
(44.4)
7
(58.7)
4
(50.0)
2
(100)
5
(83.3)
4
(40.0)
23
(46.9)
12
(48.0)
2
(22.2)
1
(8.3)
2
(25.0)
0
(0.0)
0
(0.0)
1
(10.0)
9
(18.4)
7
(28.0)
3
(33.3)
4
(33.3)
2
(25.0)
0
(0.0)
1
(16.7)
3
(30.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
1
(10.0)
3
(6.0)
1
(4.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
1
(10.0)
17
(34.0)
4
(16.0)
6
(66.7)
7
(58.3)
3
(37.5)
2
(100)
6
(100)
4
(40.0)
19
(38.0)
12
(48.0)
0
(0.0)
2
(16.7)
3
(37.5)
0
(0.0)
0
(0.0)
2
(20.0)
11
(22.0)
7
(28.0)
3
(33.3)
2
(16.7)
2
(25.0)
0
(0.0)
0
(0.0)
3
(30.0)
0
(0.0)
1
(4.0)
0
(0.0)
1
(8.3)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
3
(6.0)
1
(4.0)
1
(11.1)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
1
(10.0)
17
(34.0)
3
(12.0)
6
(66.7)
9
(75.0)
3
(37.5)
1
(50.0)
6
(100)
4
(40.0)
22
(44.0)
10
(40.0)
0
(0.0)
2
(16.7)
4
(50.0)
1
(50.0)
0
(0.0)
2
(20.0)
7
(14.0)
9
(36.0)
2
(22.2)
1
(8.3)
1
(12.5)
0
(0.0)
0
(0.0)
2
(20.0)
1
(2.0)
2
(8.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
0
(0.0)
1
(10.0)
Neither
9
(39.1)
11
(45.8)
Somewhat
Disagree
Strongly Agree
10
(43.5)
10
(41.7)
Strongly
disagree
Somewhat
Agree
2
(8.7)
2
(8.3)
Neither
Strongly Agree
1
(4.3)
1
(4.2)
Somewhat
Disagree
Somewhat
Agree
51
(68.1)
32
(57.1)
Strongly
disagree
Somewhat
Disagree
23
(31.1)
24
(42.9)
Neither
No
Neither
Yes
Somewhat
Disagree
Strongly
disagree
College/Uni
Unaffordable cost of transport
Strongly Agree
High school
Clinics too far
Somewhat
Agree
Visit traditional
healer
Strongly
disagree
Socio-demographic
factors
Income USD/month
<30
31-60
61-90
91-120
121-150
151-210
>210
Do not know
Appendices
231
Appendix L
Health Professional Qualitative Interview Transcript
232
Appendices
Appendix L
Table 30 Health Professional Interview Transcripts
Questions/Health
Professional
How do you think

patients'
demographic
characteristic is
associated with
completion of TB
treatment?
How does TB
patients' life style
affect TB treatment
adherence? (i.e.
lifestyle, smoking,
alcohol and drug)
Appendices
HP 1
Women tend to put priority for

their family over themselves.
People that are employed tend
to have frequent access
regardless the distance to
accessible health service.
Those that are employed most
commonly have better living
conditions.
All TB cases who are male are

smokers. Those who drink
tend to not follow instruction
given by HW nor taking advice
from family members.
HP 2
Illiterate rate is high, chances of
completing treatment is low. Better
understand by patients about the
expectation if they are educated.
Illiteracy rate contributes towards
people completing treatment.
Distance and transport cost is an
issue especially if they have to
travel to treatment centers.
Employment status is not relevant.
Those who live below poverty line
have more chances of suffering
from TB. Because of poverty, there
is lack of food, malnutrition,
overcrowding. Poverty leads to
ignorance of symptoms, duration
of treatment, types of treatment,
unhygienic conditions, and no
sanitation.
It is proved that smoking is
common cause of death in TB
patient.
TB patients who smoke have
higher death rates. Chronic alcohol
raise liver problems (liver cirocis
known factors to develop TB).
Those defaulters are chronic
alcoholics. Consume more alcohol
and not bother to continue
medication.
HP 3
Those who have good education

may be likely to adhere to
treatment. However, if they are
concerned with income loss, lack
of support, reside far from health
services while facing other direct
factors, they are most likely to not
adhere to treatment regardless of
their knowledge of the treatment.
Living condition that is below

standard.
Poor hygiene etc. Drink and
smoke are not affecting TB,
however it helps facilitating a
person to get TB and worse.
233
HP 4
Employment does not
really matter as it does
not make people
understand about TB,
especially those that do
not work in health areas.
Income may not be an
issue also due to the fact
that everything is free of
access. Well transport
cost may be an issue
though. Health education
is the main factors for
everything. Patients
would stop treatment
after feeling better if they
have lack of knowledge
on TB.
HP 5
Living conditions: city or

remote areas, there is an
association, yet there is
no study to be certain
about it. Marital status
may not really have
connection. Yet level of
education may have some
connection
Alcohol addictive person

that is under treatment,
within 2 months there may
be no complains of not
drinking. Yet, once they
feel better they are less
likely to adhere and
getting back to the drink
routine. This is due to the
fact that he or she is in the
community living with
people that smoke or
drink.
How does mental

health affect
(mental disorder,
health symptoms
and illness
characteristic)
patient's treatment
adherence?
How do the
symptoms of TB
affect level of
adherence to
treatment?
How does side

effect of the drug
affect treatment
adherence of an
individual patient?
Do
knowledge, beliefs
and attitude of a
patient impact
treatment
adherence Probing
how and why?
234
People that have mental

health disorder would not want
to follow treatment. When the
sickness becomes chronic
they go away to hide or to run
away.
All patients take medication for
2 months and once feel better
and the symptoms are gone
they would decide to not take
medicine. Even worst if the
patient live in distance or busy
with daily life.
People that already have

mental health problems
would not be able to get
treatment as they do not
follow suggestions for
treatment well.
Very few incidents in that problem.

It is not too relevant.
If we were to measure anxiety
depression maybe relevant.
Severity of disease, there here is a

tendency to complete.
How severity is measured is
important. Severity is measured
through (1+, 2+ and 3+). It is
evident that those who complete
successfully are those who are
chronically ill. Delay of getting
treatment. Those smokers more
likely to think that coughing due to
smoking. Therefore presenting
themselves later.
Those that have lack of knowledge

and have light symptoms of TB
tend to feel frustrated when
informed of having TB.
Those that are having heavy level
of symptoms tend to adhere as
they have experienced the
sickness and wanting to get better,
therefore wanting to be treated
and adhere. This can also be due
to the fact that patients do not
receive sufficient information when
diagnosed.
Side effect that is not too much

to be handled is kind of ok.
Yet, if it is getting worst and
involve injection, many
patients would not be able to
handle the pain.
Identifying side effect timely and

treating them is important.
Boosting moral and strengthening
knowledge helps adhere to
treatment.
People discontinue their

medication due to the fact that
health workers do not explain
clearly what would happen in
terms of side effect when
consuming the medicine.
Many TB cases would

prioritize traditional healers
before going to the clinics.
Patients think that traditional
healer and treatment are
related. Therefore they need to
do some cultural ceremony
before going to clinic for
diagnosis. People that have
Knowledge: pt have no knowledge

will have lack of access to
services, diagnosis, don't have the
sickness and therefore stopping
treatment. Beliefs: besides
following treatment pt also have
their own things which they believe
in. Therefore they will stil spend
money to do cultural beliefs.
Lack of knowledge also because

lack of information given by the
health worker at the time of
diagnosis.
If information in relation to
treatment and basic health
education is clearly given in the
first place, it would be much easier
for the patient to follow treatment
Appendices
When the symptoms are

not too serious (light)
he/she would rather take
anti-biotic.
Patients are supposed to

be informed side effect of
drugs at the time of
diagnosis.
Low level of knowledge

would affect patient to
follow up treatment.
If a person is too
depressed is really hard to
be control therefore will
likely not being able to
complete treatment.
There is a huge difference
between those who have
suffered from TB and not
able to get treated earlier
to those who are early
detected. Those that are
too sick are afraid of
death, therefore they are
happy to follow treatment.
Yet, those that early
diagnosed, without good
counselling and lack of
information and
knowledge will be hard.
If health professionals do
not provide sufficient
information, patient will
less likely to adhere to
treatment.
Good level of knowledge:
Positive idea about
treatment and otherwise.
Traditional beliefs, people
that believe in traditional
stuff is due to lack of
knowledge. Those that
have bad attitude towards
other life style (drinking,
lack of education tend to be

more honest about their status
of sickness compare to those
who have good level of
education. Those that are
educated are afraid of being
discriminated.
Overall, what
factors do you think
may affect
medication
treatment?
Appendices
Nutrition, knowledge on good

nutrition, distance and
relationship between
volunteers and patients.
Family abandoned the sick ones

due to lack of knowledge of the
family members. Some cases,
ladies are not allowed to cook
once identified as TB. Stop
breastfeeding as they are afraid it
will transmit through the milk.
Unfavourable outcome on
treatment at the end. Attitude, after
two months of treatment they will
stop treatment because they feel
better and not wanting to listen to
directions. Attitude also related to
other responsibilities which they
may think is more prioritizing than
their treatment.
Co-morbidity: along with TB

patients have other diseases also.
So if their disease is not
managed, there will be
unfavourable outcome. I.e. HIV or
Diabetes while having TB.
Whatever a person eats and the
medicine is sufficient. Yet patient
tends to have some special diets
which cost them more. Distance
between DOT providers and
patients' resident is also important
towards completion of treatment.
Time and place: agriculture person
would live early and come home
late. But DOT providers may not
be able to provide medicine. 3As =
Accessible, Acceptable and
Accountable to the health system.
without any doubt. Attitude: if we

make patient realize for all
information we need to give, they
should not have any problem and
eventually would change their
attitude.
Lack of social support and there is

discrimination against people
affected with TB. This is also due
to lack of health education in the
community.
235
smoking) may also have

problems.
Lack of knowledge would

make people not
knowing what sickness
they are experiencing.
Therefore, not following
their treatment. Lack of
human resources.
Those that have family,

especially a male patient
that have many
dependents while having
low socio-economic status
is a challenge. This is due
to the fact that he is
responsible for many
while having no money to
travel. Poor living
conditions, lack of food
and water is also a
challenge for a patient to
follow treatment. Lack of
family support is
important. Poor family
support can also affect so
much on their treatment.
Health professionals have
to have good
responsibility to provide
enough information and
provide good support
towards patients'
treatment.
What system and

structures are in
place in our current
healthy system to
support TB patients
for their treatment?
(diagnosis, medical
availability, staff
etc)
236
All staff have multiple roles

and responsibilities.
Quality of work is poor.
People that work for TB
program also have to
contribute their time to support
other programs (lack of human
resources)
Heath workers in village or CHC

level need further training to be
able to do their job. Increase
volunteers' capacity, provide more
supervision and monitoring.
Lack of health staff, lack of training
Appendices
Diagnosis is great,
medical availability,
distribution of drugs,
examination. Lack of
human resources due to
multi-function of jobs.
From level National TB

Program, District TB
Coordinator and TB
assistance and volunteer.
Medical availability and
structure is also in place.
DTA is not only
responsible for TB
program. He or she is
multi-function therefore
affect the quality of work
which he/she does. Lack
of follow up, not able to
get medicine on time,
stock out etc.
What is the informal

support for a TB
patient during
his/her treatment?
What adherence
aid do patients
usually use to
remind them to take
their medication?
(i.e. family
members, using
clock, alarm, mobile
alarm or calendar)
Support in terms of
information: suggestion,
motivation, information on side
effect and information on the
consequences of not
completing treatment. Work
with local leaders and
volunteers if the patient live
distance from available health
services.
We do not have informal support,

generally.
There is no informal support
Family members' support is

vital.
Patients need to come to the clinic

daily for two months.
Continuation phase for two days.
Not necessary to investigate.
PSF in the community normally

will have
agreement whether pt will go to
PSFs house or otherwise.
Questions/Health
Professional
How do you think patients'

demographic characteristic is
associated with completion of
TB treatment?
Appendices
HP 6
Distance, access to
transport, malnutrition as a
risk factor, majority of TB
patients are farmers or
having labour work, more
males of approximately 80%
are having TB out of all
registered TB patients
because they are mobilizing
everywhere. Majority within
HP 7
Completing treatment does
not depend on education.
In some cases, those that
are older, even have lack of
education, they are likely to
seek for treatment and follow
their treatment successfully.
The young ones are likely to
not follow treatment and
prioritize work or other
HP 8
Those who are educated
normally would less likely
to follow treatment.
Whereas, those who are
old more likely to follow
treatment completely.
This is due to the fact
that those who are
educated more likely to
feel shame of themselves
237
We have local food

which we grow yet, we
do not have the
knowledge to prepare it
well.
Family support,
supposedly people that
have low SES should be
supported by Ministry
Social Solidarity should
support. In the past there
was, but not anymore.
There is no law in place, it
was a temporary decision.
Counselling support to
encourage is important is
already happening, yet it
is not systematic. DTA
and DTC need to work
hard to help patients.
Family member and

health professionals
Family and health

professional. Yet number
of health professional is
limited. Geographically it
is impossible to travel to
receive DOTS.
HP 9
Those who have lack of
education are less likely
to feel
to be discriminated.
Whereas those who are
educated, they are more
likely to feel that they
may be discriminated if
others know that he/she
has TB. Those that do
HP 10
TB is everywhere and East
Timor is the worst.
If you eliminated all the easy
bacteria, you end up with the
hard one and it spreads all
over. The prevalence data
currently used is not from
Timor. TB is overwhelming in
the country. Using wrong
database to make everything
the age range of 15-24 years

of age. Houses lack of
ventilation systems, poor
living conditions.
How does TB patients' life

style affect TB treatment
adherence? (i.e. lifestyle,
smoking, alcohol and drug)
238
Smoking and drinking are

not directly causing TB but,
somewhat it becomes
facilitating factors for a
patient to get TB.
activities than their

treatment. Also those who
work are likely to prioritize
their work rather than
treatment. Those that have
no job have lack of money to
travel, yet there is family and
social support to provide
their needs.
Those that have smoked

normally have high impact
on their
treatment. There are some
that are committed to not
smoke or drink. However,
after getting sick again,
normally their family will
reveal that he/she had
started smoking or drinking
again. Much worst for the
youth group, where normally
they would drink for 2 weeks
or so and started drinking
and/smoking again. This
could be due to the fact that
for getting the disease.

Therefore, it is very
important if health
workers provide sufficient
and clear information
regarding TB treatment.
For instance, side effect,
duration of treatment and
consequences of not
following the treatment
until the end of treatment
duration. Cost of
transport can be a
problem. Therefore,
those who do not have
transport will have
support from the health
workers by giving them
medicine up to 3 days
instead of getting it from
the clinic every single
day.
not work or do not have a

job are likely to adhere to
their treatment. However,
those who are employed
most of the time they
would worry about losing
their job. Therefore they
prioritize their work rather
than their treatment.
There are some patients

that promise that they will
not
smoke or drink.
However, most cases
when the families of the
patients are visited, they
often say that patients
still do drink and/smoke.
Therefore it is sometimes
important to give
counselling to provide
support during the
treatment.
Normally, they would be

encouraged to not drink
and smoke. They are
given thorough
explanation regarding the
impact of smoking and
drinking on treatment.
However, it seems that
some family members
report that those who are
under treatment
sometimes still drink or
smoke. But not all of
them though.
Appendices
looks good. Giving the

estimated TB is taken from
other country that has similar
economic standard.
145/100.000 is very low and it
is not the real thing. TB is
everywhere and it is number
one problem in East Timor.
From Bairopite Clinic in 2011,
296 cases positive sputum.
85% success rate is the
biggest mistake miscalculation
ever. If suggestion is given to
MoH, they would refuse and
may say "our number looks
good to be spread around the
world" so we don't want to
hear anything more.
Popular participation,
decentralization.
Housing in East Timor

contributes to TB treatment.
GASPORTO
use of PCR, nucleic acid of
sputum.
they have lack of

understanding about the
treatment.
How does mental health affect

(mental disorder, health
symptoms and illness
characteristic) patient's
treatment adherence?
How do the symptoms of TB

affect level of adherence to
treatment?
Appendices
Severe TB patients more

likely to adhere. Different
perceptions depend on the
patient's knowledge. Some
get better after taking
medication would continue
treatment, but some would
stop.
There are some that all they

care is to get better
regardless
of their symptoms. In the
past experience, where there
are some patients who got
really sick and cannot get
themselves to the clinic.
However, now we have the
half -way house (alberque)
where people can stay
temporarily up to 2 months.
Some patients that are

mentally affected. Even
the family
and relatives do not want
to look after them,
because the patients do
not want to take their
medicine. Some cases
we suggest that the
patients to stay in half
way home (alberque) to
follow the treatment until
he/she finishes
treatment.
There are some patients
that stop in the middle of
treatment. This is due to
the length of treatment
and that they have to
come back and forth to
clinic to receive medicine.
This can also be because
the health workers in the
clinic did not provide
sufficient information
regarding the treatment.
239
Some patients feel really

stressful when they take
the medicine and get side
effect. They would feel
really stress and
complain about the side
effect. There are some
that would report their
problem and get support.
So far those patients

under our care follow
instruction given.
They are to receive their
medicine in the clinic.
However, sometimes
they are given medicine
for up to 2 or 3 days
because they complain of
travelling long distance
and no money for
transport cost. Those that
There is no link and people in

the community not even part
of what is going on. Our
training is not very good. Our
attitude is no good. We are
the most under-utilized system
everywhere. People prefer to
go to see traditional healer
rather than to see a doctor. It
is because people are not part
of the people.
By the end of 2 months they

normally will be able to move
around freely and so will be
able to go back home and
receive their medicine from
nearby clinics or health
centers.
How does side effect of the

drug affect treatment
adherence of an individual
patient?
240
Side effect is a common

challenge for patients to
adhere to their treatments.
However, due to lack of
information provided at the
time of diagnosis by the
health workers. There should
at least be a 15 minutes
session explaining about the
treatment process prior to
patients enrols to treatment.
Therefore, to follow up on
patients having this type
of behaviour, it is best to
check their monthly
medication record and
visit them to ensure that
they can get back to their
normal routine of taking
medication to get better.
Some patients that even get

side effect from the drug,
they are still willing to follow
their treatment.
This is due to the fact that
they were provided sufficient
information regarding the
side effect of the drug.
Therefore, they know that
such reaction of side effect
may take place. But there
are many that complain a lot
about the side effect, and if
they are not convinced well
they are more likely to go to
other clinic for treatment or
stop treatment after all. For
instance, there were some
that have headache,
stomach-ache etc will
complain previously they do
not feel that, yet after taking
the medicine they start to
feel sick.
normally take medicine

home, we will not be able
to comment whether they
drink or not. When they
are in the clinic and we
ask to drink they will
drink. But once they are
at home and they have to
drink by themselves we
would not know if they
do.
There are some that

have itchy skin, fever or
headache.
Once they report to us,
we will be able to tell
them that it is normal as
a consequence of
drinking the medicine.
So, once they know the
reason why than they are
likely to adhere and
continue their treatment.
Appendices
Do knowledge, beliefs and

attitude of a patient impact
treatment adherence Probing
how and why?
Lack of knowledge still links

to side effect. If people have
lack of knowledge about TB
and its treatment process,
this will become a
determinant factor to adhere
to treatment or not.
When a patient experiences
side effect, they more likely
to seek for traditional
healers. Attitude links with
ignorance of not seeking for
treatment, even worst when
the health service is distance
from where the patient lives.
Overall, what factors do you

think may affect medication
treatment?
Appendices
Cultural beliefs and beliefs of

ancestors connection
Some people take traditional

medicine.
So, in many cases they say
that because they have
taken traditional medicine, it
is ok to drink or smoke.
Many that so adhere to their
treatment, yet they complain
about cost of transport to
travel every day. Beliefs of
traditional healers still a big
thing in the country. In a
case in one of the subdistricts where there were no
patient admitted to the health
center, but there were nearly
50 patients that were looked
after by a traditional healer.
Out of the 50 patients were
diagnosed and found that 37
patients had positive TB.
Geographic location is a big
issue in the country. Due to
distance people have to
travel from their homes to a
clinic or health centers, they
are more likely to seek for a
nearby traditional healer
instead. People are also
likely to drop out from
treatment if they are not
committed to travel distance
over a period of time.
In regards to belief and

attitude, in some cases,
even though patients
have been diagnosed
with TB many family
members would not
believe it. They
sometimes will not accept
it easily as they think that
their family will have no
way to have got TB.
241
What system and structures

are in place in our current
healthy system to support TB
patients for their treatment?
(diagnosis, medical availability,
staff etc)
We do have system that is

established from National to
village levels. However,
there is still lack of human
resources and lack of
knowledge of some health
workers. There is sufficient
diagnosis equipment, though
we still need more.
There is system and

structure in place for our
clinic,
yet I do not know if any
other clinics or health
centers throughout the
country would have
things that we have in
this clinic.
Informal support from SISCa

(Community Integrated
Health System) in the
community.
What is the informal support

for a TB patient during his/her
treatment?
There are some informal

supports provided by
Ministry of Social
Solidarity. Yet, it is only
for those who are really
in need. This support can
be monetary to support
transport cost or food.
What adherence aid do

patients usually use to remind
them to take their medication?
(i.e. family members, using
clock, alarm, mobile alarm or
calendar)
No other adherence aid

that any patient would
normally use. But, most
common aid patients get
is their family members.
Normally their family
members would remind
them to drink their
medicine.
242
Appendices
Appendix M
Draft Paper
Appendices
243
Appendix M Draft paper

Factors Associated with Medication Adherence Among Tuberculosis Patients in Timor-Leste: A
Cross-Sectional Study
Juliao dos Reis1, Xiang-Yu Hou1, Michael P Dunne1
Abstract
Background: Adhering to tuberculosis (TB) treatment is one of the most important components for
treatment success. This study assessed potential barriers and enabling factors associated with optimal TB
medication adherence among patients in Timor-Leste (also known as East Timor).
Methods: The study used a cross-sectional design. In-person survey was conducted to collect data from 347
pulmonary TB patients at six community health centres and three specialist clinics in three districts of
Timor-Leste
Results: The visual analogue scale (VAS) was used to measure suboptimal adherence over a one-month
period and adherence prevalence was 39.6%. After taking into account socio-demographic factors, the
optimal adherence impediments were poor health services received (p=0.021), use of a traditional healer as a
form of health service (p=0.006), an external health locus of control (p=<0.001), and perceived stigma
(p=0.049). Enabling factors for TB medication adherence were the patients TB knowledge (p=0.018),
reasonable income or sufficient family financial support (p=0.048) and the availability of a caregiver to
remind the patient of his/her medication time (p=0.009).
Conclusions: This study provides important evidence on specific factors that may influence treatment
adherence direct or indirectly. Furthermore, the study contributes towards a limited number of quantitative
studies on TB medication adherence in the Southeast Asia region The evidence strongly suggests that TB
programs in Timor-Leste should focus on strengthening health education to address stigma prevention,
ongoing individual deleterious health behaviours such as smoking and drinking, counselling to support
patients' psychological well-being, beliefs that chance or luck are determinants of health outcomes, and
appropriate interventions to work cooperatively with local traditional healers.
Keywords: medication adherence, stigma, Timor-Leste, tuberculosis, health services, patients knowledge,
financial support.
Background
Tuberculosis (TB) is a major contributor to the global burden of disease, predominantly in low- and midincome countries [1]. it is ranked the second most fatal, worldwide communicable disease after HIV/AIDS
[2].
_______________________________
Author details
*
1
Correspondence: [email protected],
Queensland University of Technology, School of Public Health and Social Work, Brisbane, Australia
244
Appendices
A full list of author information is available at the end of the article.
There are approximately 9 million new cases of TB per year [3], and TB itself contributes to 2.8% of the
global disability burden [4]. Over 95% of TB deaths occur in low and middle-income countries and
approximately 10 million children worldwide have been orphaned due to the death of a parent from TB [3,
5]. In 2013 alone, 1.5 million people died from the disease, an estimated 550,000 children became ill, and of
these, 80,000 HIV-negative children died of TB [2]. Southeast Asia and the Western Pacific regions
collectively accounted for 56% of the cases worldwide in 2013 [2].
In developing nations, especially East Timor, where poverty is widespread, people are facing high
incidences of infectious diseases, such as TB, malaria, and measles. Timor-Leste is a small nation located in
Southeast Asia, with an estimated population of 1.17 million in 2014 and a population density of 76.2 people
per square kilometre. The World Bank estimated that in 2015, almost 53% of people lived on less than
US$1.25 per day. Poor TB treatment is likely to be affected by many challenges including high population
density (especially in some districts), in combination with high rates of illiteracy, low community awareness,
poor health-seeking with relatively poor use of health facilities, mountainous terrain, long distances between
villages and health centres, bad roads, and nascent health systems [6].
The Ministry of Health (MoH) of Timor-Leste has identified TB as one of the major public health problems
facing the country [6]. While a large-scale survey of TB prevalence and incidence have not yet been
undertaken, the estimated incidence rate of TB is 115 - 175 new smear positive cases (NSP/100,000/year)
[6]. Furthermore, as a consequence of civil war, many senior health officers left the country or lost their
lives, and much of the countrys physical infrastructures were destroyed, resulting in an inadequate supply of
public utilities [7].
Several studies examined factors associated with TB medication adherence and have found that lack
treatment knowledge, beliefs, poor attitudes and negative perceptions, unemployment, poor health services
and long distance to the nearest health service are associated with poor medication adherence [4, 8-14]. A
study in Ecuador involving a sample of 212 adults undergoing diagnostic TB testing found that
misconceptions or a lack of key knowledge could adversely affect early diagnosis, treatment adherence and
the further spread of the disease. It was also highlighted that education was the most important predictor of
knowledge, beliefs, perceptions and attitudes [4]. Furthermore, a mixed-methods survey conducted in Jingsu
Province in China found that patients being illiterate, divorced/widowed and having no health insurance were
more likely to be non-adherent.
Weiguo et als findings showed that patients who experienced social stigma, an adverse drug reaction and a
heavy financial burden were more likely to unsuccessfully complete treatment. The study also found that
patients who received direct and regular home visits by health workers appeared to reduce the risk of nonAppendices
245
adherence [11]. Additionally, in Kenya, a cross-sectional survey was conducted among new TB patients
receiving six months of standard TB treatment. The study reported the reasons provided by the non-adherent
patients were that they ran out of pills, which was likely to be explained by missed appointments, a loss of
pills, and being away from home, mostly due to economic activities [9].
Similarly, a case control study of 384 TB patients conducted in Ndola, Zambia found an overall of 29.8% of
the sample had stopped taking TB drugs at some point during the treatment. The most common reasons
given for stopping treatment by both the adherent and non-adherent patients were that they felt better (45.1%
and 38.6%, respectively). Additionally, the most frequently described reasons were including a lack of
knowledge of the benefit of completing treatment (25.7%), the TB drugs being too strong (20.1%) and a lack
of food at home (11. 4%). While, non-complaint patients described the most frequent reasons for stopping
treatment as that they ran out of drugs at home (25.4%) and that the TB drugs were too strong (20.2%) [8]
Another case control study was conducted in Morocco, North Africa that involved 290 TB patients. The
study found non-adherence was significantly associated with education, distance to the nearest health
services, treatment knowledge, health-professional and patient communication, health service provider, and
attitudes towards treatment [10]. Some of the findings were consistent with a cross-sectional study in China
which showed that a lack of knowledge and longer travel to the nearest health services were significantly
associated with poor treatment adherence [15].
Thus far, there has been little research conducted in Timor-Leste to examine the ways that TB patients
comply with treatment or to investigate the factors influencing treatment adherence. One ethnographic study
in Timor-Leste qualitatively addressed issues associated with TB treatment, such as local cultural practices,
knowledge and socio-economic factors associated with treatment adherence [16]. A survey was also
conducted by the Ministry of Health Timor Leste [17] to determine the level of knowledge, attitudes and
practices among community members and TB patients, and an RCT study investigated the impact of
nutritional support on TB treatment [18]. To date, there is no quantitative, peer-reviewed study reporting the
level of TB treatment adherence and factors impeding optimal adherence in Timor-Leste [15].
246
Appendices
Methods
Study sites
Participants were recruited from the patient lists of six community health centres (CHCs) and three specialist
clinics in three districts of Timor Leste. Of the six CHCs, three were located in areas of Dili, the capital city
of Timor-Leste and six in rural areas. All six CHCs are operated by Ministry of Health staff, while local
clinics are run by church organization or local NGOs,
Study participants
Participants were selected from the TB registry and those above 18 years of age were approached. Prior to
the data collection, three survey administrators who had previously conducted surveys for the ministry of
health of Timor Leste were recruited and trained. Patients who spoke Tetum (Timor Lestes national
language) fluently or their local language could be translated by the survey administrators and agreed to
complete the survey were selected. Participants who were too ill, refused to participate or were
geographically out of reach were excluded. Patients from the health services in rural and remote areas were
recruited when they came to receive their medication at the health centres, and those that were not able to
come to the clinic were contacted by CHC staff to organize an interview time in their homes. Due to
participants poor understanding and limited experience in completing a survey, the questionnaire was
administered by survey administrators to patients who were with high school education or lower. Participants
with a university education were asked to complete the questionnaire independently under supervision.
Measurement of tuberculosis medication adherence

The study used two measures for medication adherence. A visual analogue scale (VAS) was adopted from
Nackers et al. [9] as it is the most commonly used tool to measure adherence for both TB and HIV/AIDS
studies [9, 19]. Second, the Medication Adherence Rating Scale (MARS) was used to measure adherence in
the past month, and was adopted with modification to cross-validate the VAS adherence estimate [20]. The
MARS scale has been used to measure the behaviours and attitudes of patents towards medication adherence
[20]. Both measurements used a cut-off point of < 80% classified as suboptimal adherent and 80% as
optimal adherence [9, 21]. Studies using the VAS and other measurement methods to determine adherence to
HIV treatment have reported good agreement [19, 22]
Measurement of factors related to medication adherence

The variables were selected based on a global literature review [16, 23, 24] and findings from a pilot study
conducted by the research team prior to the main survey. The variables on medication adherence were
classified into three categories of health service delivery, social and economic, and personal and diseaserelated factors. Health service delivery factors were comprised of health services (urban or rural), patienthealth professional communication, distance from home to clinic, and patients satisfaction [25]. Social and
economic factors involved gender, age, marital status, education, income, lifestyle choice (alcohol, smoking
Appendices
247
and drug intake), social support (by family, friends, and neighbours, community volunteer and social
organization) and involvement of traditional healer [25]. The final category was personal and disease-related
factors, comprising psychological distress, stigma (of the patients or from the community), TB knowledge
and chance health locus of control (i.e. the degree to which individuals believed that their health status was
controlled by chance or luck) [25].
The Kessler Psychological Distress Scale (K10) was used to measure anxiety and depressive symptoms
experienced by the participants in the past month [26]. The internal consistency (Cronbachs ) for the K10
was 0.806. Chance health locus of control was measured with a subscale of the multidimensional Health
Locus of Control Scale (Cronbachs =0.911) [27]. The Tuberculosis-Related Stigma Scale measures both
community and patient perspectives toward TB with a Cronbachs of 0.841) [28]. A patient assessment
tool with Cronbachs =0.895 was used to measure the effectiveness of communication between patient and
care provider [19]. A traditional healer scale (Cronbachs =0.623) to measure the influence of the
Traditional healers on TB patients and social support (family, friend, neighbour, peer educator and social
organization) (Cronbachs =0.761) was also used to determine the level of support received by TB patients
throughout their treatments [19, 29].
Questionnaire translation
The translation process was carried out by adapting the World Health Organization Translation Guideline
[30] as well as a number of translational validity methods, including Principles of Good Practice for the
Translation and Cultural Adaptation process for Patient-Reported Outcomes by Wild, Grove [31]. The tool
was translated from the original language of English to Tetum (Timor Lestes national language) by a
translator with background in health. The translation was further reviewed by two translators with
backgrounds in health and by the principal researcher. The final version in Tetum was back-translated by a
professional language translator and the same process was repeated. Although the guidelines for the
translation validity process were closely followed, due to the incompleteness of the Tetum vocabulary [32],
some translations were refined by adding further justification and explanation to ensure that the accuracy of
the content and meaning of the English version was preserved. As part of the data collection quality control,
the three survey administrators were provided an intensive three days of training on the content of the
questionnaire and its administration process.
Interview process
For every participant, the survey administrator introduced him/herself and highlighted the purpose of their
presence and that we did not represent any government body, health professional or any other organization
directly involved with the participant's treatment. The participants were encouraged to provide accurate and
honest responses and acknowledged the importance of their participation in the survey to help identify ways
in which TB treatment may be improved. The questionnaire took an average of 45 minutes to an hour to
complete and the majority of the questionnaire was completed at clinics, community health centers and
248
Appendices
patients homes. The research administrators tasks were to supervise, answered questions, checked
understanding and completion of the survey, and read the items to the illiterate participants.
Data Analysis
The survey data were entered in EpiData version 3.1 and converted to SPSS version 21 to be analysed.
Univariate logistic regression analysis was used to estimate the associations between each independent
variable and medication adherence. Multivariate logistic regression analysis was conducted by including all
independent variables associated with adherence at p 0.1, while controlling for the demographic variables.
Independent variables that were not considered as confounders were simultaneously removed from
multivariate logistic regression analysis to determine the relative contributions of each variable to treatment
adherence at a significance level of p < 0.05.
Results
Characteristics of study participants
According to the database of the National TB Program in Timor Leste, between 1st January, 2013 and 1st
September, 2013 there were 1160 patients, 18 years and older registered in six community health centres and
three private clinics selected across three districts in Timor Leste. The majority of these patients were
scattered across remote and difficult-to-reach areas that had no contact. Of those registered in the National
TB Program database, 29.9% (359) were eligible, but 1% (4) were too unwell to participate, 5% (1.4%)
could not be contacted, 0.56% (2) were willing to participate but no translator could assist, 0.28% (1) was
taking a trip to the countryside. 96.7% (347) patients were able to complete the questionnaire.
From the 347 participants, 16.2% (56) who had a university/college education completed the questionnaire
independently while closely supervised, and 83.8% (291) who had trouble reading the items for themselves
received support from a research administrator. The participants demographic, behavioural and clinical
characteristics are summarized in Table , indicating that 55% were male and the ages ranged from 18 to 84
years (Mean: 38.76, 17.14). Over 60% were married, almost one third of the sample had no formal
education, and over 50% reported income or family financial support of less than USD$60 per month. More
than half of the sample reported that they currently or previously drank alcohol and almost 50% reported
current or previous cigarette smoking.
Tuberculosis medication adherence

The prevalence of a one-month, suboptimal adherence measured by the VAS was 39.6%, while the past
month suboptimal adherence for MARS is 25.3%. The observed agreement of both measurements was
69.4%, with a kappa coefficient of 0.323 (standard error=0.052, p<0.0001). This suggests that the majority of
individuals were classified as either adherent or as non-adherent by the two measures, while 30.6% had a
discordant classification (Table 2)
Appendices
249
Factors associated with tuberculosis medication adherence

Table 3 presents the results of the univariate and multivariate logistic regression analysis of the factors
associated with tuberculosis medication adherence, measured by the VAS.
The univariate analysis showed a significant association of adherence with low income or limited family
financial support, alcohol consumption, quality of health service received, and caregivers to remind
medication time, patients poor TB knowledge, psychological distress, the patients perception of stigma, use
of traditional healers and external health locus of control.
After controlling the socio-demographic variables, the risk factors that remained significantly associated with
non-adherence in the multivariate logistic regression were quality of health service provided, low income or
lack of family financial support, lack of a caregiver or people to remind the participants to take their
medication, low treatment knowledge, use of a traditional healer, perceived stigmatization and external
health locus of control.
When the other variables were controlled for in the model, for every unit increase on the income scale, the
odds of adherence was increased by 1.87 suggesting that financial support is directly associated with
treatment adherence. Poor health service provision was significantly associated with non-adherence (OR =
0.72; p < 0.05). Similarly, patients who lacked caregivers or people to remind them of their medication time
were associated with non-adherence (OR = 0.70; p = 0.009).
For every unit increase on the knowledge scale, the odds of adherence were multiplied by 1.35 indicating
that better TB treatment knowledge is associated with optimal adherence. In contrast, for stigma, traditional
healer and health locus of control, for every unit increase in each of these scales, the odds of being nonadherent to treatment were multiplied by 0.91, 0.94 & 0.86, respectively.
Factors associated with non-adherence listed by the participants

There were 21 common factors listed by the participants as affecting adherence (see Table 4). Among the
most common factors cited by over 70% of the participants simply forgot to take medication (72.4%), busy
with other things (64.7%), away from home (50.3%), health service too far (33.6%) and had a change in
daily routine (32.5%). Furthermore, almost one in every three people cited to have no money for eating and
health care seeking (28.3%). Illicit drug use was the least reported reason for suboptimal adherence (1.0%).
Discussion
The study using the VAS to measure adherence in the previous month found a suboptimal adherence of
39.6%. In comparison with other studies, the percentage of non-adherence varies. A study conducted in
Ndola, Zambia measured adherence of patients taking TB drugs daily for eight months and reported that
250
Appendices
29.8% patients stopped taking TB drug at some point during the treatment [8]. In Kenya, a cross-sectional
survey assessed adherence with numerous tools, in which VAS for patients missed medication in the past
four days and found 7.5% of non-adherence [9]. In Jiangsu Province, China, a study measured adherence
using a structured questionnaire, with a 90% cut-off point and found a 12.2% non-adherence rate [11].
According to the International Union Agent Tuberculosis Committee on Prophylaxis, the recommended cutoff point for TB adherence is 80% [21]. These studies indicate lower percentages of non-adherence
compared to the results of the current study may be due to the time reference such as adherence in the past
four days rather than the past month, or perfect adherence to treatment regimens.
The study conducted by Weiguo et al. [11] in Jingsu Province, China, recruited 780 sputum-smear positive
patients in 13 counties (districts), and patients who missed 10% of their medication were deemed to be nonadherent. The study found that financial burden was associated with medication non-adherence, despite the
Chinese government's free TB service policy. Similarly, treatment provided by public health services in
Timor-Leste is free of charge. In addition, the Timor-Leste National Stop TB Strategy Plan 2011-2015 also
highlighted that poor patients and those living in remote areas are compensated for the indirect cost of TB
treatment and food supplements where applicable [6]. However, the present study revealed that those who
received support or had an income of less than
USD60 per month were associated with suboptimal
adherence, which is consistent with Weiguo et al.s finding in China [11]. Therefore, the results of the
current study and the study in China suggest that free treatment alone is not likely to improve treatment
adherence, and implementation of the free-treatment policy warrants further research.
Research on the association between quality of health service delivery and TB medication adherence is
limited [25]. In Timor Leste, community-based TB services are responsible for the provision and training of
peer educators to provide directly observed treatment short courses (DOTS) to patients. But some urban
health services and private clinics have no community-based integrated TB program. This may be a reason
that the quality of service provided contributes to the overall level of treatment adherence.
A systematic review of qualitative research by Munro et al. [1] suggested that family and community support
strongly influence treatment adherence. Several studies have confirmed types of support, such as financial,
emotional, counselling, sharing of knowledge and reminder of medication time can be enabling factors
associated with medication adherence [24, 33, 34]. However, an HIV/AIDS antiretroviral medication
adherence study by Do et al. [19] did not find a significant relationship between medication adherence and
support received among these individual groups. The present studys results were consistent with Do et al.s
finding and suggested that family, friends and peer educators support were not strongly associated with
treatment adherence.
Several studies have described the impact of stigma on tuberculosis medication adherence [35-37]. A
qualitative study in Nepal investigated the causes of stigma and discrimination associated with TB. The
Appendices
251
study found that, if disclosure of treatment status caused supportive behaviour from others, it may be useful
for increasing treatment adherence. However, disclosing the treatment status to family and friends may also
facilitate further disclosure, which may result in patients feeling discriminated. In the Nepal study, although
TB patients were reminded routinely by health workers that TB stops being infectious after two weeks of
commencing treatment, self-discrimination remained high throughout the eight months of treatment [35].
Additionally, some of the participants perceived that disclosure of information might cause the risks such as
loss of business, loss of job, emotional injury and stigmatization from others [38]. This was confirmed in our
study, where the perception of stigma by the participants was significantly associated with medication nonadherence. Patients who encountered stigmatization had severe suffering, which in turn was likely to result in
non-compliance [36]. Stigma could cause people to become socially isolated from family, health
professionals and people outside their family to avoid becoming a subject of gossip [35]. Disclosure of
treatment status to may gain social support or causing discrimination depends on the level of knowledge and
understanding of those who receive the information.
In addition to support received, the current study showed that participants who have people to remind them
about the medication time are more likely to be adherent compared to those who did not have such support.
This finding agrees with Weiguo et al.s study in China, where the risk of non-adherence was lower among
patients under direct observation by village doctors or receiving home visits by health workers [11].
In relation to risky behaviours, a multi-level analysis of a nationally representative sample of the South
African population conducted in Cape Town showed that cigarette smoking and alcohol consumption were
significant predictors of TB risk factors [39], while a systematic review and meta-analysis conducted by
Hsien-Ho Lin et al. [40] also found consistent evidence that tobacco smoking was associated with an
increased risk of TB. Cigarette smoking may delay treatment completion, due to a deficiency in the function
of the respiratory system [41]. However, cigarette smoking itself may not directly contribute to sub-optimal
adherence, as confirmed by the current study which found no significant relationship between smoking and
medication non-adherence.
An antiretroviral HIV/AIDS study investigating factors associated with optimal adherence of HIV patients
in Vietnam found that alcohol use was significantly associated with medication non-adherence [19]. Alcohol
was also frequently mentioned as an influencing factor in the context of TB treatment non-adherence by a
study conducted in Russia [40]. The finding of the current study is consistent with both studies, suggesting
that alcohol consumption is significantly associated with suboptimal adherence. Therefore, future
interventions to identify and address patients with alcohol issue while under treatment is necessary for
improving level of treatment adherence in Timor Leste.
252
Appendices
Traditional healers in certain countries have much influence on tuberculosis treatment completion. It is
reported that patients who prefer a traditional healer as an alternative source of treatment are likely to delay
seeking early treatment [42]. A survey conducted across five districts (urban & rural) in Timor-Leste
indicated that 14.5% of the total population cited traditional healers as some form of health facility and
34.6% considered traditional healers as an option for health care [17]. The current study found that those
who considered traditional healers as a form of treatment were less likely to adhere to medication regimens.
In order to improve patients treatment adherence, it is crucial for the Ministry of Health in Timor-Leste to
work collaboratively with the traditional healers. In addition, provision of general training and understanding
in identifying and screening people with TB symptoms is important. Moreover, monitor traditional healers to
independently refer patients and assist them with ongoing treatment would improve level of adherence and
treatment completion [43].
Many studies have showed general agreement that knowledge is likely to be an important factor to determine
the successfulness of TB treatment [4, 15, 18, 33, 34, 44]. Our study found that patients' knowledge was
positively significantly associated with adherence. Both quantitative and qualitative studies worldwide have
shown that improved levels of knowledge are likely to reduce stigma and negative attitudes towards
treatment. Additionally, those who have accurate knowledge and understanding of the treatment process are
less likely to have treatment interruptions and not considering untested treatment such as a traditional healer
[4, 16, 33, 34, 44]. Often, sick patients are cared for by their family members, community health volunteers
and health care professionals, and it has been reported in a case with HIV/AIDS in other settings that such
care can cause the experience of stigma [45]. An important key aspect to be addressed is to educate those
caring for patients, especially with their knowledge about the disease, and the appropriate language used to
communicate to patients during their daily interactions [46]. Because many TB patients reside in remote
areas of the country in Timor Leste, Kidd and Clay [46] suggested that improved education can help
empowering community members, especially patients family members who could benefit patients with their
treatment. Thus, educating different groups of people who are involved directly in patients TB treatment
could also directly influence the success of treatment completion.
Adherence research generally finds that psychological distress affects compliance with the TB medication
regimen [24]. This study confirmed that psychological distress was significantly associated with TB
medication non-adherence. One of the psychological factors investigated in our study was the dimensional
health locus of control that assessed the belief that chance or luck significantly influenced one's destiny,
including the outcome of an individual's health. After adjusting for demographic variables, the multivariate
analysis indicated that chance health locus of control remained a significant predictor of TB medication
adherence, where participants who believed in chance, luck or powerful control by others were less likely to
be adherent. This finding is useful to guide health professionals in Timor-Leste to discuss fatalistic beliefs as
barriers to self-care and healthy behaviours.
Appendices
253
Conclusion
This study provides much needed evidence on TB treatment adherence in a developing country and of the
specific enabling and impeding factors direct or inderectly influencing the TB overall treatment. Participants
were from diverse health services include private, public, rural and urban health settings in Timor Leste..
Factors impeding treatment adherence in Timor-Leste include poor health services received, preference for
traditional healers as alternative theraphists, an external health locus of control, and percieved stigma of the
patients or from the community. Enabling factors include patients accurate knowledge on TB treatment, the
availability of caregivers to remind patients of their medication time and satisfactory income or adequate
family financial support. The findings of this study could be used as scientific evidence for the Ministry of
Health of Timor-Leste to implement more effective TB intervention program to optimize the level of
treatment adherence. The findings may also provide a helpful guide for decision makers in other countries
with similar settings.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
JDR was involved in the study design, data collection, data analysis and interpretation, and drafted the
manuscript. Both XH and MD were involved in the study conceptualization, design, supervision, guided the
statistical analysis and critically reviewed drafts of the manuscript.
Acknowledments
We would like to thank all the staff at the Global Fund Division, Ministry of Health of Timor Leste, and
clinics, the I/LNGO staff in the selected sites and the survey administrators for their support. We would also
like to express our sincere appreciation to the Terrace Timor Network (TTN) & the Faculty of Health,
Queensland University of Technology (QUT) for providing financial support for the data collection.
References
1.
Munro, S.A., et al., Patient Adherence to Tuberculosis Treatment: A Systematic Review of

Qualitative Research. PLoS Med, 2007. 4(7): p. e238.
2.
WHO, Global Tuberculosis Report 2014b: Geneva
3.
Stop
TB
Partnership.
Fast
Facts
on
Tuberculosis
(TB).
2015;
Available
from:
http://www.stoptb.org/resources/factsheets/fastfacts.asp.
4.
Armijos, R.X., et al., The meaning and consequences of tuberculosis for an at-risk urban group in
Ecuador. Significado y consecuencias de la tuberculosis para un grupo urbano de riesgo en Ecuador.,
2008. 23(3): p. 188-197.
5.
254
WHO, Noncommunicable diseases country profiles 2014. 2014.

Appendices
6.
Ministry of Health Timor Leste, Timor Leste National Stop TB Strategy Plan 2011 - 2015 2010,
Ministry of Health Timor Leste: Dili.
7.
Hill, H., Tiny, Poor and War-Torn: Development Policy Challenges for East Timor. World
Development, 2001. 29(7): p. 1137-1156.
8.
Kaona, F.A., et al., An assessment of factors contributing to treatment adherence and knowledge of
TB transmission among patients on TB treatment. BMC Public Health, 2004. 4: p. 68.
9.
Nackers, F., et al., Adherence to Self-Administered Tuberculosis Treatment in a High HIVPrevalence Setting: A Cross-Sectional Survey in Homa Bay, Kenya. PLoS ONE, 2012. 7(3): p.
e32140.
10.
Tachfouti, N., et al., The impact of knowledge and attitudes on adherence to tuberculosis treatment:
a case-control study in a Moroccan region. Pan Afr Med J, 2012. 12: p. 52.
11.
Weiguo, X., et al., Adherence to anti-tuberculosis treatment among pulmonary tuberculosis patients:
a qualitative and quantitative study. BMC Health Services Research, 2009. 9: p. 169-176.
12.
Peltzer, K., Conjoint alcohol and tobacco use among tuberculosis patients in public primary
healthcare in South Africa. South African Journal of Psychiatry 2014. 20: p. 21-26.
13.
Peltzer, K., et al., Hazardous and harmful alcohol use and associated factors in tuberculosis public
primary care patients in South Africa. International Journal Of Environmental Research And Public
Health, 2012. 9(9): p. 3245-3257.
14.
Yao, S., et al., Treatment adherence among sputum smear-positive pulmonary tuberculosis patients
in mountainous areas in China. BMC Health Serv Res, 2011. 11: p. 341.
15.
Tang, Y., et al., Non-adherence to anti-tuberculosis treatment among internal migrants with
pulmonary tuberculosis in Shenzhen, China: a cross-sectional study. BMC Public Health, 2015.
15(1): p. 474.
16.
Martins, N., J. Grace, and P. Kelly, An ethnographic study of barriers to and enabling factors for
tuberculosis treatment adherence in Timor Leste. International Journal of Tuberculosis and Lung
Disease, 2008. 12(5): p. pp. 532-537.
17.
Ministry of Health Timor Leste, Knowledge, Attitude and Practice (KAP) towards Tuberculosis and
Socio-economic Impact of the Disease in Timor-Leste. 2011.
18.
Martins, N., P. Morris, and P.M. Kelly, Food incentives to improve completion of tuberculosis
treatment: randomised controlled trial in Dili, Timor-Leste. BMJ: British Medical Journal (Overseas
& Retired Doctors Edition), 2009. 339(7729): p. 1131-1131.
19.
Do, H.M., et al., Factors associated with suboptimal adherence to antiretroviral therapy in Viet
Nam: a cross-sectional study using audio computer-assisted self-interview (ACASI). BMC Infectious
Diseases, 2013. 13(1).
20.
Thompson, K., J. Kulkarni, and A. Sergejew, Reliability and validity of a new Medication Adherence
Rating Scale (MARS) for the psychoses. Schizophrenia research, 2000. 42(3): p. 241-247.
Appendices
255
21.
WHO, Efficacy of various durations of isoniazid preventive therapy for tuberculosis: five years of
follow-up in the IUAT trial. International Union Against Tuberculosis Committee on Prophylaxis.
Bull World Health Organ, 1982. 60(4): p. 555-64.
22.
Amico, K.R., et al., Visual analog scale of ART adherence: association with 3-day self-report and
adherence barriers. J Acquir Immune Defic Syndr, 2006. 42(4): p. 455-9.
23.
Clark, P.M., et al., Effect of pharmacist-led patient education on adherence to tuberculosis

treatment. American Journal of Health-System Pharmacy, 2007. 64(5): p. 497-506.
24.
Naidoo, P. and K. Mwaba, Helplessness, Depression, and Social Support Among People Being
Treated For TUberculosis in South Africa Social Behavior & Personality: An International Journal,
2010. 38(10): p. 1323-1333.
25.
Dick, J., et al. World Health Organisation: Adherence to long-term therapies: Evidence for action.
2003.
26.
Slade, T., R. Grove, and P. Burgess, Kessler Psychological Distress Scale: normative data from the
2007 Australian National Survey of Mental Health and Wellbeing. The Australian And New Zealand
Journal Of Psychiatry, 2011. 45(4): p. 308-316.
27.
Malcarne, V.L., S. Fernandez, and L. Flores, Multidimensional Health Locus of Control Scales-Shortened Version. 2005.
28.
Van Rie, A., et al., Measuring stigma associated with tuberculosis and HIV/AIDS in southern
Thailand: exploratory and confirmatory factor analyses of two new scales. Tropical Medicine &
International Health, 2008. 13(1): p. 21-30.
29.
Munro, S., et al., A review of health behaviour theories: how useful are these for developing
interventions to promote long-term medication adherence for TB and HIV/AIDS? BMC Public
Health, 2007. 7(1): p. 104.
30.
WHO Measurement of substance abuse: process of translation and adaptation of instruments.

2012a.
31.
Wild, D., et al., Principles of Good Practice for the Translation and Cultural Adaptation Process for
Patient-Reported Outcomes (PRO) Measures: report of the ISPOR Task Force for Translation and
Cultural Adaptation. Value In Health: The Journal Of The International Society For
Pharmacoeconomics And Outcomes Research, 2005. 8(2): p. 94-104.
32.
Golden, J., When the Diaspora Returns: Language Choices in Post-Independence Timor Lorosa'e.
2004. 2: p. p. 118.
33.
Ayisi, J.G., et al., Care seeking and attitudes towards treatment compliance by newly enrolled
tuberculosis patients in the district treatment programme in rural western Kenya: a qualitative
study. BMC Public Health, 2011. 11(1): p. 515-524.
34.
Gebremariam, M.K., G.A. Bjune, and J.C. Frich, Barriers and facilitators of adherence to TB
treatment in patients on concomitant TB and HIV treatment: a qualitative study. BMC Public Health,
2010. 10: p. 651-659.
256
Appendices
35.
Baral, S.C., D.K. Karki, and J.N. Newell, Causes of stigma and discrimination associated with
tuberculosis in Nepal: a qualitative study. BMC Public Health, 2007. 7: p. 211-211.
36.
Dodor, E.A., K. Neal, and S. Kelly, An exploration of the causes of tuberculosis stigma in an urban
district in Ghana. The International Journal Of Tuberculosis And Lung Disease: The Official Journal
Of The International Union Against Tuberculosis And Lung Disease, 2008. 12(9): p. 1048-1054.
37.
Karim, F., et al., Stigma, gender, and their impact on patients with tuberculosis in rural Bangladesh.
Anthropology & Medicine, 2007. 14(2): p. 139-151.
38.
Do, H.M., Antiretroviral therapy (ART) adherence among people living with HIV/AIDS (PLHIV) in
the north of Vietnam : a multi-method approach 2011.
39.
Harling, G., R. Ehrlich, and L. Myer, The social epidemiology of tuberculosis in South Africa: A
multilevel analysis. Social Science & Medicine, 2008. 66(2): p. 492-505.
40.
Jakubowiak, W.M., et al., Risk factors associated with default among new pulmonary TB patients
and social support in six Russian regions. Int J Tuberc Lung Dis, 2007. 11(1): p. 46-53.
41.
Chiang, Y.-C., et al., Tobacco consumption is a reversible risk factor associated with reduced
successful treatment outcomes of anti-tuberculosis therapy. International Journal of Infectious
Diseases, 2012. 16(2): p. e130-e135.
42.
Finnie, R.K.C., et al., Factors associated with patient and health care system delay in diagnosis and
treatment for TB in sub-Saharan African countries with high burdens of TB and HIV. Tropical
Medicine & International Health, 2011. 16(4): p. 394-411.
43.
Colvin, C., et al., Evaluation of community-based interventions to improve TB case detection in a

rural district of Tanzania. Global Health: Science and Practice, 2014. 2(2): p. 219-225.
44.
Hoa, N.P., N.T.K. Chuc, and A. Thorson, Knowledge, attitudes, and practices about tuberculosis
and choice of communication channels in a rural community in Vietnam. Health Policy, 2009. 90(1):
p. 8-12.
45.
Nyblade, L., et al., Disentangling HIVand AIDS stigma in Ethiopia, Tanzania and Zambia. 2003.
46.
Kidd, R. and S. Clay, Understanding and challenging HIV stigma: Toolkit for action. 2003.
Appendices
257
Table 1 Participants socio-demographics, behaviours and clinical characteristics (N=347)

Socio-demographic
Behaviour and clinical characteristics
191
55

Never drink
147
42.6
characteristics
Gender (male)
Age (Mean: 38.7617.14)
18-29
138
39.8
Used to drink
154
44.6
30-49
89
25.7
Currently drinking
44
12.7
50+
120
34.6
Never smoke
179
51.7
Marital status
Cigarette smoking status (last month)
Unmarried
134
38.7
Used to smoke
155
44.8
Married
212
61.3
Currently smoking
12
3.5
337
98.8
1.2
Public health services with CBTC
34
9.8
Private clinics
134
38.6
Education
Illicit drug use
No formal education
97
28
Never
Primary school
92
26.6
Yes, in the past
101
29.2
College/University
56
16.2
Wages/family support
Health Services
<60USD/month
193
55.6
Public health services
69
19.9
>60USD/month
154
44.4
Home visit patients
110
31.7
Less than 30 minutes
197
58.6
More than 30 minutes
139
40.4
Time to nearest health service
CBTC, Community Based TB Care Program
258
Appendices
Table 2 Agreement between two measures of TB medication adherence

MARS Score1
Non-adherence
Adherence
Total
Non-adherence
58(17.6%)
75(22.7%)
133(40.3%)
Adherence
26(7.9%)
171(51.8%)
197(59.7%)
Total
84(25.5%)
246(74.5%)
330(100%)
Non-adherence
58(17.6%)
75(22.7%)
133(40.3%)
Adherence
26(7.9%)
171(51.8%)
197(59.7%)
Total
84(25.5%)
246(74.5%)
330(100%)
VAS Score 2
MARS, Medication Adherence Rating Scale, 2VAS, Visual Analogue Scale.
Appendices
259
Table 3 Factors associated with TB medication adherence measured by Visual Analogue Scale
(n=347)
Univariate
Multivariate
Optimal
Suboptimal
adherence
adherence
Unadjusted OR
Adjusted OR
frequency
frequency
(95%CI)
value
(95%CI
value
(%)
(%)
Male
110(59.8)
74(40.2)
Female
94(61.0)
60(39.0)
1.05(0.68-1.03)
18-29
82(60.3)
54(39.7)
30-49
53(60.2)
35(39.8)
0.99(0.58-1.73)
0.992
1.10(0.48-2.21)
0.805
50+
69(60.5)
45(39.5)
1.01(0.61-1.68)
0.970
1.09(0.39-3.04)
0.871
Single
63(53.8)
54(46.2)
Married
139(63.5)
80(36.5)
1.45(0.94-2.35)
No formal education
63(66.3)
32(33.7)
Primary school
49(55.7)
39(44.3)
0.64(0.35-1.16)
0.141
1.32(0.55-3.15)
0.538
57(58.2)
41(41.8)
0.71(0.39-1.28)
0.244
0.91(0.31-2.64)
0.854
College/University
34(60.7)
22(39.3)
0.78(0.40-1.56)
0.488
1.63(0.48-5.47)
0.431
<60USD/month
94(49.5)
96(50.5)
>60USD/month
110(74.3)
38(25.7)
2.96(1.86-4.71)
Never drink
92(64.3)
51(35.7)
Used to drink
97(64.7)
53(35.3)
1.02(0.63-1.64)
0.953
1.01(0.51-2.33)
0.975
Currently drinking
13(13.2)
30(69.8)
0.24(0.12-0.50)
<0.001
0.41(0.15-1.12)
0.083
Non-smoker
107(61.8)
66(38.2)
Ever smoke
81(59.1)
56(39.4)
0.89(0.56-1.41)
0.626
197(60.1)
131(39.9)
Factors
Gender
1
0.814
1.10(0.60-1.98)
0.778
Age
1
Marital status
1
0.087
1.84(0.81-4.15)
0.143
Education
1
1
<0.001
1.87(1.01-3.47)
0.048
Behavioural factor
1
Tobacco use (last

month)
Illicit drug use

Never
260
Appendices
Yes, in the past
2(50.0)
Mean of
optimal
adherent
(n=204)
2(50.0s)
Mean of
0.67(0.09-4.78)
0.685
Univariate analysis
Multivariate analysis
subopti
mal
adherent
Unadjusted OR
(95%CI)
p value
Adjusted OR
(95%CI)
value
(n=134)
25.70
28.52
0.96(0.93-0.98)
0.003
0.99(0.95-1.04)
0.815
Health Locus Control
28.13
30.26
0.83(0.78-0.89)
<0.001
0.86(0.79-0.93)
<0.001
Perceived stigma
7.49
8.74
0.88(0.82-0.95)
0.001
0.91(0.83-1.00)
0.049
Traditional healer
3.36
6.08
0.93(0.90-0.98)
<0.001
0.94(0.90-0.98)
0.006
Patients TB knowledge
5.94
5.50
1.34(0.10-1.64)
0.004
1.35(1.05-1.72)
0.018
22.91
23.41
0.96(0.91-1.02)
0.223
Service provider
2.66
2.89
0.79(0.63-0.99)
0.037
0.72(0.54-0.95)
0.021
12.61
12.72
0.98(0.91-1.07)
0.726
0.88
1.48
0.60(0.49-0.74)
<0.001
0.70(0.54-0.92)
0.009
Family support
12.73
12.45
1.03(0.96-1.11)
0.415
Friend support
8.79
9.30
0.95(0.88-1.02)
0.144
Neighbour support
8.37
8.70
0.96(0.89-1.04)
0.963
Peer educator support
9.60
9.52
1.01(0.94-1.08)
0.809
Social organization support
5.22
5.41
0.95(0.86-1.06)
0.402
People remind medication

time
currently smoking and used to smoke were combined due to small cell sizes
Appendices
261
Table 4 Suboptimal adherent factors listed by patients

Items
262
Responses
n
Simply forgot
207
72.4
Felt good
83
29.0
were away from home
144
50.3
27
9.4
33
11.5
Clinic is too far
96
33.6
93
32.5
Felt like the drug was toxic/harmful
49
17.1
Felt asleep/slept through dose time
64
22.4
Felt sick
31
10.8
Felt depressed/overwhelmed
24
8.4
Had problems taking pills at specific times
58
20.3
18
6.3
185
64.7
Use drugs
1.0
Drink alcohol
14
4.9
Took them, then get nausea, but did not take the replacement
54
18.9
Don't have money for eating and health care seeking
81
28.3
Ran out of pills but could not go to the clinic to get them
86
30.1%
36
12.6%
Don't believe in the usefulness of the drug
14
4.9%
Appendices

Factors Associated With Medication Adherence Among Tuberculosis Patients in Timor Leste

Uploaded by

Copyright:

Available Formats

Factors Associated With Medication Adherence Among Tuberculosis Patients in Timor Leste

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Factors Associated With Medication Adherence Among Tuberculosis Patients in Timor Leste

Uploaded by

Copyright:

Available Formats

FACTORS ASSOCIATED WITH MEDICATION

ADHERENCE AMONG TUBERCULOSIS

Juliao dos Reis

A thesis submitted in fulfilment of the requirements for the degree of

School of Public Health and Social Work

Tuberculosis (TB) is an infectious disease that damages lungs and other

factors associated with medication adherence among tuberculosis patients in timor-leste

The cultural appropriateness of the qualitative and quantitative instruments

The qualitative study revealed themes that were common to health

and patients suggested that non-adherence to TB treatment was relatively

The quantitative survey used two measures of non-adherence, with a cut-off

To date, this is one of the largest studies in Southeast Asia to examine TB

treatment. In addition, incentives should be provided to those who struggle

Adherence, adherence knowledge, medication adherence, social stigma, traditional

factors associated with medication adherence among tuberculosis patients in timor-leste

Abstract ...................................................................................................................................................... iii

Aim and Research Questions .................................................................................................5

Significance of the Study ..........................................................................................................5

Structure of the Thesis ..............................................................................................................6

Chapter 2: Literature Review ............................................................................... 9

factors associated with medication adherence among tuberculosis patients in timor-leste

Chapter 3: Methodology ...................................................................................... 53

Chapter 4: Qualitative Study Results ............................................................... 87

4.4.6 Cigarette Smoking and Alcohol Consumption ................................................... 95

Chapter 5: Main Survey Results ...................................................................... 103

Chapter 6: Discussion ....................................................................................... 129

factors associated with medication adherence among tuberculosis patients in timor-leste

References ............................................................................................................ 149

Figure 1. Literature mapping of TB treatment adherence................................................. 21

factors associated with medication adherence among tuberculosis patients in timor-leste

Table 1 Evidence-Based Rating System ................................................................................. 10

Table 21 Agreement between the VAS and MARS on TB medication

factors associated with medication adherence among tuberculosis patients in timor-leste

Acquired Immune Deficiency Syndrome

Community Health Center

District Health Service

Directly Observed Treatment

Directly Observed Treatment Short-course

District Tuberculosis Assistance

District Tuberculosis Coordinator

Fixed Dose Combination

Human Immunodeficiency Virus

International Nongovernmental organization

Knowledge Attitude and Practice

Local Non-government Organization

Medication Adherence Rating Scale

New Smear Positive

National Tuberculosis Program

Population Attributable Fraction

People Living with HIV

Queensland University of Technology

Randomized Control Trial

Visual Analogue Scale

World Health Organization

factors associated with medication adherence among tuberculosis patients in timor-leste

Statement of Original Authorship

QUT Verified Signature

Juliao dos Reis