Tola et al.

BMC Public Health (2019) 19:1658

https://doi.org/10.1186/s12889-019-7980-x

RESEARCH ARTICLE Open Access

Treatment Outcome of Tuberculosis and

Associated Factors among TB-HIV Co-
Infected Patients at Public Hospitals of
Harar Town, Eastern Ethiopia. A five-year
retrospective study
Assefa Tola1* , Kirubel Minsamo Mishore2, Yohanes Ayele2, Abraham Nigussie Mekuria3 and Nanati Legese4

Abstract
Background: The bidirectional relationship between the twin epidemics of Tuberculosis (TB) and Human
Immunodeficiency Virus (HIV) causes major global health challenges in the twenty-first century. TB-HIV co-infected
people are facing multifaceted problems like high lost to follow up rates, poor treatment adherence, high TB
recurrence rate, and high mortality risk. Our objective was to assess the outcomes of TB treatment and associated
factors among TB-HIV co-infected patients in Harar town, Eastern part of Ethiopia, 2018.
Methods: A retrospective study was conducted among systematically selected 349 TB/HIV co-infected patients who
registered from 2012 to 2017 in two public hospitals in Harar town. The data were collected through document
review by using a pre-tested structured data extraction checklist. The data were analyzed using SPSS Version 21.
Bivariate and multivariate logistic regression were determined at 95% confidence intervals.
Results: Among the 349 TB/HIV co-infected patients included in the study, 30.1% were cured, 56.7% had completed their
treatment, 7.7% died, 1.7% were lost to follow up, and 3.7% were treatment failure. Overall, 86.8% of the TB-HIV co-
infected patients had successful TB treatment outcomes. The patients who were on re-treatment category (AOR = 2.91,
95% CI: 1.17–7.28), who had a history of opportunistic infection (AOR = 3.68, 95% CI: 1.62–8.33), and who did not take co-
trimoxazole prophylaxis (AOR = 3.54, 95% CI: 1.59–7.89) had 2.91, 3.68, and 3.54 times higher odds of having unsuccessful
TB treatment outcome than their counterparties, respectively. The chance of unsuccessful TB treatment outcome was 4.46
(95% CI: 1.24–16.02), 5.94 (95% CI: 1.87–18.85), and 3.01 (95% CI: 1.15–7.91) times higher among TB/HIV patients in stage 2,
3 and 4 than those in stage 1, respectively.
Conclusions: The overall rate of the success of the TB treatment among TB-HIV co-infected patients in this study was
higher compared with many previous studies. TB/HIV patients with a history of previous TB treatment, smear-positive
pulmonary TB, late HIV stage, history of opportunistic infection and not being on co-trimoxazole prophylaxis therapy were
at a high risk of getting poor treatment outcomes.
Keywords: TB- HIV co-infected, TB treatment outcome, Harar

* Correspondence: [email protected]
1
Department of Epidemiology and Biostatistics, School of public health,
College of Health and Medical Sciences, Haramaya University, Harar, Ethiopia
Full list of author information is available at the end of the article

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Tola et al. BMC Public Health (2019) 19:1658 Page 2 of 12

Background According to Ethiopia national guidelines, all HIV

The twin epidemics of Tuberculosis (TB) and Human positive patients should be evaluated for TB before ART
Immunodeficiency Virus (HIV) are the major global is initiated and then at every visit. Similarly, all TB pa-
health challenges of the twenty-first century [1, 2]. These tients should be offered HIV testing services in TB
two infectious diseases have a bidirectional relationship clinics [4, 6, 15]. Regarding the therapeutic management,
that poses a dual public health burden to resource- ART should be provided to all TB/HIV co-infected pa-
limited countries [2]. TB-HIV co-infected people are ex- tients regardless of CD4 count or WHO stage. However,
periencing “double trouble” that puts them at high risk ant-tuberculosis treatment should be provided first,
of mortality, rapid disease progression, and development followed by ART within the first 8 weeks of treatment
of other opportunistic infection [3, 4]. [6, 15]. Likewise, Co-trimoxazole prophylaxis (CPT)
HIV, because of its immunosuppressive nature, is a should be provided to all TB-HIV co-infected patients,
strong risk factor for the development of TB [5] and re- regardless of their CD4 count [4, 6, 9, 15]. In order to
activation of latent TB [4, 6]. In 2017, globally about prevent the reactivation of latent TB, after excluding the
920,000 people who were infected with HIV developed presence of active TB infection, IPT should be given to
TB [7]. Reports also indicate that people living with HIV all people with HIV at least for six months [6, 10].
are 20 to 21times more likely to develop active TB than The percentage of patients treated successfully is a key
those HIV negative people [7, 8]. On the other hand, TB indicator for monitoring and evaluating the effectiveness
is the most frequent life-threatening opportunistic infec- of the TB Directly Observed Therapy (DOT) program
tion and the leading cause of death among HIV positive [10]. This is particularly necessary for patients with TB-
people [1, 3, 6–9]. About one in four deaths among HIV HIV co-morbidities where the treatment outcome could
positive people is caused by TB [3]. TB speeds up the be affected by many factors. Hence, it is important to con-
viral replication and load in HIV infected people [4, 6]. duct a periodic evaluation of the treatment outcome for
In general, TB affects about one-third of the 36.7 million this segment of the population to assess the level of quality
people living with HIV worldwide [10]. of care and to imply possible directions for improvement.
Ethiopia is one of the 14 countries with high-burden of Unfortunately, there are only a few studies that have
all TB, TB/HIV, and Multi-drug Resistant TB (MDR-TB) assessed TB treatment outcomes among TB-HIV co-
in the world [5]. According to studies and reports, the infected people in Eastern Ethiopia. Hence, our objective
prevalence of TB-HIV co-infection in the country is so high was to assess the outcomes of TB treatment and the asso-
that it has ranged from 6.1 to 40.4% [2, 4, 5, 11, 12]. The ciated factors among TB-HIV co-infected patients in pub-
global 2018 TB report showed that in Ethiopia an estimated lic hospitals in Harar town, Eastern Ethiopia.
3600 (95% CI: 2500–5000) TB-HIV co-infected people died
while on TB treatment in the year 2017 alone [5]. Methods
In TB-HIV co-infected patients, HIV affects the effect- Study setting
iveness and success of TB treatment in many ways [5]. This study was done in Harar town, which is the capital
For example, the co-infected patients are exposed to town of Harari Regional State. In the town, there are five
many regimens including Antiretroviral Therapy (ART), hospitals: Hiwot Fana Specialized University Hospital
anti-TB therapy, and preventive therapy of HIV-related (HFSUH), Jugal Hospital, Yimaj Hospital, Harar General
co-morbidities which in turn is associated with an in- Hospital, and Federal Police Hospital. Only the first two
creased incidence of adverse drug reaction, poor adher- are public hospitals.
ence often due to pill burden, and decreased drug
effectiveness. Consequently, the patient may experience Study design and period
a high default rate leading to TB recurrence and in- An institution-based retrospective study was conducted
creased risk of death [4, 13]. Hence, well-coordinated from April 1–10, 2018 to determine TB treatment out-
therapeutic management is needed to ensure optimum come and the associated factors among the TB-HIV co-
treatment outcomes in terms of response and prevention infected patients in the two public hospitals in Harar
of drug resistance. town, Eastern Ethiopia.
Collaborative TB/HIV activities and management of co-
morbidities are the key components of the ‘end TB strategy’ Populations
[5, 13, 14]. To reduce the dual burden of TB/HIV among All the TB/HIV co-infected patients who registered in
people living with HIV, it is recommended to scale up the the public hospitals in Harar town were the source
three I’s, which are intensified TB case-finding followed by population. The study population was TB/HIV patients
high-quality TB treatment, isoniazid preventive therapy who systematically selected from those TB/HIV patients
(IPT), and infection control for TB in all congregate set- registered in the hospitals from 1st January 2012 to 31st
tings and health facilities providing HIV care [6, 9, 10]. December 2017. All the complete medical records of the
Tola et al. BMC Public Health (2019) 19:1658 Page 3 of 12

TB/HIV co-infected patients who registered in the hos- calculated at 95% confidence intervals to evaluate the
pitals from1st January 2012 to 31st December 2017 were crude association between each exposure variable and
included. The records of the patients with missing values outcome variable. Multi-collinearity was checked among
on the variable’s interest were excluded. In addition, independent variables like baseline CD4 count, WHO
medical records of the transferred out patients were ex- staging, functional status and history of opportunistic in-
cluded since the TB treatment outcomes were unknown. fections through variance inflation factor (VIF). No col-
linearity was found among the variables. Then to control
Sample size determination and sampling technique the effect of confounding factors, a variable that had P-
The sample size was calculated using a single population value ≤ 0.2 in the bivariate analysis was entered into the
formula with 95% confidence interval (CI), a 5% margin multivariate logistic regression model, as the independ-
of error and taking the proportion of successful treat- ent variable and TB treatment outcome status being a
ment outcomes among TB/HIV co-infected patients dependent variable. Multivariate logistic regression ana-
from previous studies [16–19]. Then we took the largest lysis was employed to assess the independent association
sample size (324), which was based on a study con- of each exposure variable with TB treatment outcome.
ducted in Mizan Tepi of Ethiopia that found TB Treat- P-value < 0.05 was considered statistically significant in
ment success rate of 30.3% [18]. After adding a 10% the final model. Hosmer and Lemeshow test was used to
non-response rate for missing data, we found the final check the assumption on the fitness of goodness of the
sample size to be 356. final model and it was found fit.
This sample size was then allocated proportionally to
the two selected public hospitals based on the number
Ethical consideration
of registered TB cases in each hospital. Then in each se-
Ethical clearance was secured from the research review
lected hospital, the profiles of all TB/HIV co-infected
technical committee of Harar Health Science College. A
patients were evaluated to check the fulfillment of the
legal permission letter was taken from Harar Health Sci-
inclusion criteria. Finally, study subjects who met the in-
ence College to the selected public hospitals and official
clusion criteria were selected from the TB registration
permission was obtained from the administration of the
Book through a systematic random sampling technique
hospitals. Furthermore, before reviewing medical records
using calculated Kth value. We calculated the K-value by
of the TB/HIV co-infected patients, permission was ob-
dividing expected patients during the study period to the
tained from the TB treatment unit heads.
number of samples and took every Kth patient in the
The information obtained from the study was used
registration book.
only for the purpose of the study and is kept confiden-
tial. Since the data were collected through review of
Data collection tools and procedure
medical records, there is no harm to the patients and
The data were collected through reviewing all the neces-
their relatives provided confidentiality is maintained.
sary documents (TB treatment registry, monthly cohort
Moreover, no personal identifier was used on the data
form, and follow up form) of the TB patients using a
collection form. The recorded data were not accessed by
pre-tested structured data extraction format which was
a third person except the principal investigators.
developed by considering the variables to be studied.
The format contained all the important socio-
demographic, baseline clinical and laboratory, and follow Definitions of treatment outcome
up data. Four public health officers who had training on According to the standard definitions of the National
comprehensive TB-HIV care and experience in collect- Tuberculosis and Leprosy Control Program guideline
ing data in similar situations gathered the data. The data (NTLCP) [20] and the WHO Definitions and reporting
collectors were trained and the filled form was checked framework for tuberculosis 2013 revision [21], the fol-
for completeness on a daily basis during the data collec- lowing treatment outcome definitions were used:
tion. The whole process was supervised by the principal
investigators. ✓ “Cured: A pulmonary TB patient with
bacteriologically confirmed TB at the beginning of
Data processing and analysis treatment who was smear- or culture-negative in the
The collected data were edited, cleaned, coded, entered last month of treatment and on at least one previous
and analyzed by using SPSS [Statistical package for so- occasion.
cial science] Version 21 for windows. Frequencies, pro- ✓ Treatment completed: A TB patient who
portions, and summary statistics were used to describe completed treatment without evidence of failure BUT
the study population in relation to socio-demographic with no record to show that sputum smear or culture
and clinical characteristics. Binary logistic regression was results in the last month of treatment and on at least
Tola et al. BMC Public Health (2019) 19:1658 Page 4 of 12

one previous occasion were negative, either because Table 1 Socio-demographic characteristics of TB/HIV co-
tests were not done or because results are unavailable. infected patients at Public Hospitals of Harar town, Eastern
✓ Treatment failure: A TB patient whose sputum Ethiopia, 2018
smear or culture is positive at 5 months or later during Variables Frequency Percent
treatment. Or Patients found to harbor a multidrug- Sex
resistant (MDR) strain at any point of time during the Male 161 46.1
treatment, whether they are smear-negative or Female 188 53.9
-positive.
Age
✓ Died: A TB patient who dies for any reason before
starting or during the course of treatment. 1–15 years 10 2.9
✓ Lost to follow-up: A TB patient who has been on 16–30 years 146 41.8
treatment for at least four weeks and whose treatment 31–45 years 137 39.3
was interrupted for eight or more consecutive weeks > 45 years 56 16.0
✓ Not evaluated: A TB patient for whom no Place of residence
treatment outcome is assigned. This includes cases
Urban 284 81.4
“transferred out” to another treatment unit as well as
cases for whom the treatment outcome is unknown to Rural 65 18.6
the reporting unit”. Pretreatment Weight
20–29 Kg 5 1.4
In line with WHO criteria (21), treatment outcome 30–39 Kg 22 6.3
is categorized into: 40–54 Kg 156 44.7
✓ “Successful outcome- if TB patients are cured (i.e.,
> 54 Kg 166 47.6
negative smear microscopy at the end of treatment and
on at least one previous follow-up test) or completed Pretreatment BMI (Kg/m2)
treatment with resolution of symptoms”. < 18.5 98 28.1
✓ “Poor outcome – if treatment of TB patients 18.5–24.9 201 57.6
resulted in treatment failure (i.e., remaining smear- ≥ 25 50 14.3
positive after 5 months of treatment), lost to follow-up Year of TB treatment initiated
(i.e., patients who interrupted their treatment for two
2012 41 11.7
consecutive months or more after registration), or
death”. 2013 69 19.8
2014 70 20.1
Results 2015 71 20.3
Socio-demographic characteristics 2016 64 18.3
A total of 349 TB/HIV co-infected patients under TB 2017 34 9.7
treatment were included. The charts three TB/HIV co-
infected patients were excluded due to missing variables.
Of the 349 patients, 188 (53.9%) were female and 284 TB (34.1%), and the remaining 119 had Extra-
(81.4%) were urban residents. Their ages ranged from 8 pulmonary TB (34.1%). About two-thirds of patients
years to 65 years with a mean of 33.6 (SD ± 9.6) years); have been on working functional status (67%) and
their baseline weight was from 20 to 100 Kg with a mean 128 (36.7%) were categorized to WHO stage 3 HIV
of 55. 0 (standard deviation (SD) ± 12.3) Kg; and their disease during the initiation of their TB treatment.
pretreatment body mass index (BMI) from 11.9 to 30 More than half (54.4%) of the patients had experi-
with a mean of 20.9 (SD ± 3.9). Most of the study partic- enced opportunistic infections like pneumonia (37%),
ipants (81.1%) were in the productive age group (16–45 candidiasis (26%) and sexually transmitted infections
years). Nearly half of the patients (47.6%) weighed more (11%). The majority (84.8% with 95% CI: 80.8–88.5)
than 54Kg. The BMI of the 201 (57.6%) patients was of study participants were on CPT. The Cluster of
within 18.5–24.9 (Table 1). Differentiation 4 (CD4) count of the patients ranged
from 43 to 1058 cells/μl, with median (Inter Quartile
Clinical characteristics Range (IQR)) of 298 (187.5–460.5) cells/μl; and only
Majority (88.8%) of the study participants were new 75 (21.5%) patients had CD4 count more than 500
TB cases (95% CI: 85.7–92.0%). One hundred eleven cells/μl (95% CI: 17.2–25.8). At the time of TB diag-
of the study participants had smear-positive Pulmon- nosis, the patients have spent 1 to 15 years on ART
ary TB (31.8%), 119 had smear-negative Pulmonary with a median duration of 4.00 and IQR of 3–6 years.
Tola et al. BMC Public Health (2019) 19:1658 Page 5 of 12

About two-thirds of the patients (64.8%) have spent Table 2 Clinical characteristics of TB/HIV co-infected patients at
less than 5-years on ART (Table 2). Public Hospitals of Harar town, Eastern Ethiopia, 2018
Variables Frequency Percent
Treatment outcome of TB- HIV co-infected patients on TB clinic
anti-TB therapy Jugal hospital 180 51.6
Among the TB-HIV co-infected patients, 105 (30.1% with Hiwot Fana Specialized University Hospital 169 48.4
95% CI: 25.5–34.9) were cured, 198 (56.7% with 95% CI: Category of patient
51.6–61.6) had completed their treatment, 27 (7.7% with New 310 88.8
95% CI: 4.9–10.6) died, 6 (1.7% with 95% CI: 0.6–3.2) were Retreatment 39 11.2
lost to follow up from their treatment and the remaining 13 Types of TB
(3.7% with 95% CI: 1.7–5.7) were treatment failure (Fig. 1). Smear positive PTB (SPP TB) 111 31.8
Overall, 303 (86.8%) (95% CI: 83.1–90.3) of the TB-HIV co-
Extra-pulmonary TB (EP TB) 119 34.1
infected patients had successful TB treatment outcome
Smear negative PTB (SNP TB) 119 34.1
whereas, the remaining 46 (13.2%) (95% CI: 9.7–16.9) pa- a
Functional status
tients had unsuccessful TB treatment outcome.
Working 234 67.0
Ambulatory 106 30.4
Factors associated with successful TB treatment outcome
The association between successful TB treatment out- Bedridden 9 2.6

come and the socio-demographic and clinical character- WHO staging

istics of the study participants was assessed through stage 1 79 22.6

bivariate and multivariate analyses. stage 2 102 29.2
The treatment success rate was similar between male stage 3 128 36.7
(87.6%) and female (86.2%) patients. The urban study par- stage 4 40 11.5
ticipants had slightly higher (87.7%) successful TB treat- History of opportunistic infectionc
ment outcome than the rural ones (83.1%). TB treatment No 159 45.6
success rate was 82.7, 87.6 and 92% in the patients with Yes 190 54.4
baseline BMI less than 18.5, 18.5 to 24.9, and greater than Types of Opportunistic infections (n = 190)
25, respectively. However, the increment in the success
Pneumonia 71 37.4
rate was not statistically significant (p. value =0.236). The
Candidiasis 49 25.8
bivariate analysis showed that successful TB treatment
Sexually transmitted infections, 21 11.1
outcome was not statistically associated with sex, age,
Intestinal parasites, 19 10.0
place of residence, baseline BMI, and year of treatment of
Urinary tract infections 17 8.9
the TB/HIV co-infected patients (Table 3).
On the other hand, types of TB, functional status, WHO Others 13 6.8

staging, history of opportunistic infection, and CPT initiated CPT initiated

showed statistically significant association with TB treatment Yes 296 84.8

outcome among the TB/HIV co-infected patients. Those on No 53 15.2
the retreatment category had a higher unsuccessful treatment Baseline CD4 count
rate (23.1%) than the new TB cases (11.9%). In this study, < 200 cells/μL 102 29.2
smear-positive PTB patients had higher unsuccessful treat- 201–499 cells/μL 172 49.3
ment outcomes (18.9%) than extrapulmonary TB (14.3%) > 500 cells/μL 75 21.5
and smear-negative PTB (6.7%) cases and this difference was Duration on ART
statistically significant (p. value = 0.028). More than half of Less than 5 years 226 64.8
the patients under the category of bedridden functional sta-
More than 5 Years 123 35.2
tus had unsuccessful treatment outcome but only 11.5% of
Positive smear result (n = 230)
the patients with working functional status had unsuccessful
After 2 month 26 11.3
treatment (P. value = 0.005). Similarly, about 30% of the TB/
After 5 month 13 5.7
HIV co-infected patients with WHO stage 4 diseases had an
After 7 month 13 5.7
unsuccessful treatment outcome whereas 8.9% of the patients
with stage 1 disease had unsuccessful treatment. (P. value =
a
Working: “able to perform usual work in or out of the house, harvest, go to
school or, for children, normal activities or playing”, Ambulatory: “able to
0.003). A higher proportion of the TB/HIV co-infected pa- perform activities of daily living but not able to work or play” and Bedridden:
“not able to perform activities of daily living” (22)
tients with a history of opportunistic infection (18.9%) had b
including co-infection
unsuccessful TB treatment outcomes than those who had no
history of opportunistic infection (6.3%) (Table 4).
Tola et al. BMC Public Health (2019) 19:1658 Page 6 of 12

Fig. 1 TB treatment outcome of TB-HIV co-infected patients attending public hospitals of Harar town, Eastern Ethiopia, 2018

Multivariate logistic regression revealed that patient cat- evidence for evaluating the performance of the TB control
egory, types of TB, WHO staging, history of opportunistic program in the country and forward future direction.
infection, and CPT initiated were significantly associated According to WHO 2017 Global Tuberculosis Report,
with TB treatment outcome among the TB/HIV co-infected the global treatment success rate for HIV-associated
patients. Patients on the retreatment category had 2.91 times TB cases among the 2015 cohort was 78% and in the
(AOR = 2.91, 95% CI: 1.17–7.28) higher odds of unsuccess- WHO Africa region, it was 80% [23]. In this study, the
ful treatment outcome compared with the new cases. Those overall TB treatment success rate among the TB-HIV
patients with extrapulmonary TB and smear-negative pul- co-infected patients was 86.8%, which is similar to the
monary TB had 76.5% (AOR = 0.235, 95% CI: 0.090–0.617) results reported from studies carried out in rural South
and 65.7% (AOR = 0.343, 95% CI: 0.129–0.911) lower odds Africa (82.2%) [24], Ahmedabad (84.2%) [25] and Addis
of having unsuccessful TB treatment outcome, respectively, Ababa, Ethiopia (88.2%) [26]. Our finding is lower than
than the patients with smear-positive pulmonary TB. that of the Chandigarh study which reported a treat-
The odds of having unsuccessful TB treatment outcome ment success rate of 93.1% [27]. On the other hand, TB
was 4.46 (AOR =4.46, 95% CI: 1.24–16.02), 5.94 (AOR = treatment success rate of our study is higher than sev-
5.94, 95% CI: 1.87–18.85), and 3.01 (AOR = 3.01, 95% CI: eral studies conducted in Gondar (22.6%) [28], Mizan
1.15–7.91) times higher among the patients in clinical stage Aman (28.5%) [17], Mizan Tepi (30.3%) [18], Malaysia
2, clinical stage 3, and clinical stage 4 HIV disease respect- (53.4%) [29], Brazil (55%) [30], Western Ethiopia
ively than those in clinical stage 1 disease. The patients with (60.7%) [31], Cameroon (60.8%) [32], Iran (64%) [33],
opportunistic infection history had 3.68 (AOR = 3.68, 95% Ebonyi State Nigeria (65.8%) [34], Karnataka India
CI: 1.62–8.33) times higher odds of having unsuccessful TB (66.1%) [35], Tigray (70.8%) [16], Viet Nam (73%) [36],
treatment outcomes than those without opportunistic in- Arsi Negele (73%) [37], Yavatmal India (75%) [38],
fection history. Similarly, the odds of unsuccessful TB treat- South India (75%) [39] and Free State province of South
ment outcome was 3.54 (AOR = 3.54, 95% CI: 1.59–7.89) Africa (75.5%) [40].
times higher among the patients who were not put on CPT This observed variation might be due to the difference
compared with patients on CPT (Table 5). in the quality of service in the TB/HIV clinic; proper
counseling, health education, and appropriate follow up
by the clinician. Another possible explanation might be
Discussion the inclusion of transferred out patients in the final ana-
Tuberculosis and HIV/AIDS constitute the main burden lysis by some of the previous studies. The proportion of
of infectious disease in resource-limited countries [1, 2]. transferred out TB/HIV co-infected patients were ran-
Determining the TB treatment outcome among TB-HIV ging from 3.8% in the Ebonyi State of Nigeria [34] to
co-infected patients in different settings may provide 64.2% in Mizan Aman of Ethiopia [17]. But, in our study,
Tola et al. BMC Public Health (2019) 19:1658 Page 7 of 12

Table 3 Bivariate analysis of TB Treatment outcome with socio-demographic of TB/HIV co-infected patients attended Public
hospitals of Harar town, Eastern Ethiopia, 2018
Characteristics Treatment outcome P- value Crude Odds
Ratio
Successful Treatment Unsuccessful Treatment
(COR)(95%CI)
Sex
Male 141 (87.6%) 20 (12.4%) 0.699 1.00
Female 162 (86.2%) 26 (13.8%) 0.88 (0.44–1.57)
Age group
1–15 years 9 (90.0%) 1 (10.0%) 0.741 1.43 (0.17–11.89)
16–30 years 126 (86.3%) 20 (13.7%) 1.00
31–45 years 122 (89.1%) 15 (10.9%) 0.483 1.29 (0.63–2.64)
> 45 years 46 (82.1%) 10 (17.9%) 0.458 0.73 (0.32–1.68)
Place of residence
Urban 249 (87.7%) 35 (12.3%) 0.325 1.00
Rural 54 (83.1%) 11 (16.9%) 0.69 (0.33–1.44)
Pretreatment weight
20–29 Kg 4 (80.0%) 1 (20.0%) 0.529 0.49 (0.05–4.59)
30–39 Kg 16 (72.7%) 6 (27.3%) 0.037 0.32 (0.11–0.93)
40–54 Kg 135 (86.5%) 21 (13.5%) 0.473 0.78 (0.40–1.53)
> 54 Kg 148 (89.2%) 18 (10.8%) 1.00
Pretreatment BMI (kg/m2)
< 18.5 81 (82.7%) 17 (17.3%) 1.00
18.5–24.9 176 (87.6%) 25 (12.4%) 0.251 1.48 (0.76–2.89)
≥ 25 46 (92.0%) 4 (8.0%) 0.132 2.41 (0.77–7.61)
Year of TB treatment initiated
2012 36 (87.8%) 5 (12.2%) 1.00
2013 61 (88.4%) 8 (11.6%) 0.925 1.06 (0.32–3.48)
2014 60 (85.7%) 10 (14.3%) 0.756 0.83 (0.26–2.63)
2015 62 (87.3%) 9 (12.7%) 0.941 0.96 (0.30–3.08)
2016 57 (89.1%) 7 (10.9%) 0.843 1.13 (0.33–3.84)
2017 27 (79.4%) 7 (20.6%) 0.328 0.54 (0.15–1.87)
Significant values are set in bold

we did not include transferred patients because their re- death rate among the TB/HIV co-infected patients during
cords were not available and their treatment final out- TB treatment was 7.7%. Also, similar death rates were
comes were unknown. Furthermore, the high success seen in study carried out in Mizan Aman (6%) [17], Au-
rate observed in our study could be due to the initiation rangabad city, India (8.2%) [41], Addis Ababa, Ethiopia
of ART for all co-infected patients. As it has been re- (8.3%) [26], South India (9%) [39], WHO African Region
ported in previous studies [16, 24, 29, 31–33, 35, 36, 39], (9%) [23], Gondar (9.6%) [28], Viet Nam (10%) [36] and
patients who had ART initiated were found to have high South Africa (10.5%) [24]. The death rate in our study was
success rate compared with their counterpart. For in- lower than the rate reported in deferent part of Ethiopia
stance, studies conducted in Tigray Ethiopia [16] and which reported a death rate ranging from 12.8 to 20.2%
Malaysian [29] stated that TB/HIV co-infected patients [16, 18, 31, 37, 42–44], Cameroon 29.4% [32], Karnataka
who were not receiving ART were 3.42 and 5.10 times India 15.7% [45], Yavatimal India 16% [38], Malaysia 21%
more likely to have unsuccessful TB treatment outcomes [29], Ebonyi State Nigeria 19% [34], Free State province
than those who received ART, respectively. South Africa 17.4% [40], Iran 18.9% [33]. Moreover, the
TB/HIV co-infected patients have increased mortality WHO reported that TB associated death rate among
due to rapid disease progression, late diagnosis and other HIV-positive TB patients was 11% [23]. Nonetheless,
opportunistic infections [4, 6]. In the current study, the smaller death rates were reported from Chandigarh India
Tola et al. BMC Public Health (2019) 19:1658 Page 8 of 12

Table 4 Bivariate analysis of TB Treatment outcome with Clinical characteristics of TB/HIV co-infected patients attended Public
hospitals of Harar town, Eastern Ethiopia, 2018
Characteristics Treatment outcome P- value COR (95%CI)
Successful Treatment Unsuccessful Treatment
Patient category
New cases 273 (88.1%) 37 (11.9%) 0.058 1.00
Re treatment 30 (76.9%) 9 (23.1%) 0.45 (0.20–1.03)
Types of TB
SPP TB 90 (81.1%) 21 (18.9%) 1.00
EP TB 102 (85.7%) 17 (14.3%) 0.346 1.40 (0.70–2.82)
SNP TB 111 (93.3%) 8 (6.7%) 0.007 3.24 (1.37–7.65)
Functional status
Working 207 (88.5%) 27 (11.5%) 1.00
Ambulatory 92 (86.8%) 14 (13.2%) 0.662 0.86 (0.43–1.71)
Bedridden 4 (44.4%) 5 (55.6%) 0.001 0.10 (0.03–0.41)
WHO staging
stage 1 72 (91.1%) 7 (8.9%) 1.00
stage 2 94 (92.2%) 8 (7.8%) 0.806 1.14 (0.39–3.30)
stage 3 109 (85.2%) 19 (14.8%) 0.212 0.56 (0.22–1.39)
stage 4 28 (70.0%) 12 (30.0%) 0.005 0.23 (0.08–0.64)
History of opportunistic infection
No 149 (93.7%) 10 (6.3%) 0.001 1.00
Yes 154 (81.1%) 36 (18.9%) 0.29 (0.14–0.60)
CPT initiated
Yes 264 (89.2%) 32 (10.8%) 0.003 1.00
No 39 (73.6%) 14 (26.4%) 0.34 (0.17–0.69)
CD4 count
< 200 cells/μL 83 (81.4%) 19 (18.6%) 1.00
201–499 cells/μL 155 (90.1%) 17 (9.9%) 0.041 2.09 (1.03–4.23)
± 500 cells/μL 65 (86.7%) 10 (13.3%) 0.349 1.49 (0.65–3.42)
Duration on ART
< 5 years 199 (88.1%) 27 (11.9%) 0.357 1.00
> 5 Years 104 (84.6%) 19 (15.4%) 0.74 (0.39–1.39)

1.14% [27], Ahmedabad 4.17% [25] and Brazil 4.3% [30]. conducted in Ethiopia [18, 31], Africa [24, 32, 34, 40]
The lower death rate in our study could be explained by and Asia [29, 35, 36, 39, 41]. The possible explanation
the fact that all our study participants had been on ART. might be a good implementation of the DOT program.
ART is protective against mortality treatment outcomes In addition, an increased patient awareness, and treat-
during TB treatment. as demonstrated by previously con- ment adherence, and service accessibility might have
ducted in Tigray [16], Karnataka India [35], south India contributed to this smaller proportion of lost for fol-
[39] and North West Ethiopia [43]. low up in our settings.
Lost for follow up from TB treatment program is a The management of TB and HIV co-infected indi-
major public health problem that can be associated viduals is challenging because of the high pill bur-
with adverse drug reactions, social stigma and lack of den, increased the adverse effect and drug-drug
awareness of the disease [17]. The proportion of lost interaction [6]. In addition, several factors play a
for follow up found in this study (1.7%) is comparable significant role in determining the TB treatment out-
Mizan Aman 1.3% [17], Tigray 2% [16] and Gondar come among TB/HIV co-infected patients. In this re-
2.2% [28] studies. This lost for follow up of TB/HIV gard, our study revealed that patient category, types
co-infected patients is lower than many studies of TB, WHO staging, history of opportunistic
Tola et al. BMC Public Health (2019) 19:1658 Page 9 of 12

Table 5 Multivariate analysis of TB Treatment outcome with socio-demographic and Clinical characteristics of TB/HIV co-infected
patients attended Public hospitals of Harar town, Eastern Ethiopia, 2018
Characteristics Treatment outcome P- value Adjusted Odds Ratio
(AOR) (95% CI)
Successful Treatment Unsuccessful treatment
Pretreatment Weight
20–29 Kg 4 (80.0%) 1 (20.0%) 0. 843 0.92 (0.38–2.19)
30–39 Kg 16 (72.7%) 6 (27.3%) 0.244 0.22 (0.02–2.79)
40–54 Kg 135 (86.5%) 21 (13.5%) 0.150 0.40 (0.12–1.39)
> 54 Kg 148 (89.2%) 18 (10.8%) 1.00
Pretreatment BMI (kg/m2)
< 18.5 81 (82.7%) 17 (17.3%) 1.00
18.5–24.9 176 (87.6%) 25 (12.4%) 0.834 0.85 (0.18–4.01)
≥ 25 46 (92.0%) 4 (8.0%) 0.497 0.64 (0.17–2.34)
Patient category
New cases 273 (88.1%) 37 (11.9%) 0.022 1.00
Re treatment 30 (76.9%) 9 (23.1%) 2.91 (1.17–7.28)
Types of TB
SPP TB 90 (81.1%) 21 (18.9%) 1.00
EP TB 102 (85.7%) 17 (14.3%) 0.003 0.23 (0.09–0.62)
SNP TB 111 (93.3%) 8 (6.7%) 0.032 0.34 (0.13–0.91)
WHO staging
stage 1 72 (91.1%) 7 (8.9%) 1.00
stage 2 94 (92.2%) 8 (7.8%) 0.022 4.46 (1.24–16.02)
stage 3 109 (85.2%) 19 (14.8%) 0.003 5.94 (1.87–18.85)
stage 4 28 (70.0%) 12 (30.0%) 0.025 3.01 (1.15–7.91)
History of opportunistic infection
No 149 (93.7%) 10 (6.3%) 0.002 1.00
Yes 154 (81.1%) 36 (18.9%) 3.68 (1.62–8.33)
CPT initiated
Yes 264 (89.2%) 32 (10.8%) 0.002 1.00
No 39 (73.6%) 14 (26.4%) 3.54 (1.59–7.89)
CD4 count
< 200 cells/μL 83 (81.4%) 19 (18.6%) 1.00
201–499 cells/μL 155 (90.1%) 17 (9.9%) 0.451 1.50 (0.52–4.33)
≥ 500 cells/μL 65 (86.7%) 10 (13.3%) 0.433 1.47 (0.56–3.82)
Significant values are set in bold

infection and CPT initiation significantly associated of unsuccessful treatment was lower among the TB/HIV
with TB treatment outcome. co-infected patients with extrapulmonary TB and smear-
The TB/HIV co-infected patients on the retreatment cat- negative pulmonary TB compared with the smear-
egory had a higher chance of unsuccessful treatment out- positive pulmonary TB. A similar, finding was reported
come than the new patients. A similar result was reported by a study conducted in Addis Ababa Ethiopia [26],
from South India [39], Brazil [30], Yavatmal India [38] and South India [39] and Yavatmal India [38].
Viet Nam [36] studies. The poor treatment outcome ob- In contrary to this, a study conducted in Mizan Tepi
served in the retreatment category could be due to drug re- [18] demonstrated TB/HIV co-infected patients with
sistance because of re-exposure to the drugs [30]. Smear positive PTB had a higher chance of successful
Type of TB infection was identified by different previ- TB treatment outcome. In addition, studies conducted
ous studies [18, 26, 29, 30, 37–39, 43, 44] as a determin- in many places [29, 37, 43, 44] reported that TB/HIV
ant of TB treatment outcome. In our study, the chance co-infection patients with extrapulmonary TB had a
Tola et al. BMC Public Health (2019) 19:1658 Page 10 of 12

higher chance of mortality during TB treatment than apply every effort to make sure that all HIV patients
pulmonary TB patients. In general, a high successful took CPT.
treatment outcome is expected among patients with
smear-positive PTB compared with smear-negative PTB
Limitation of the study
and extrapulmonary TB. However, the higher successful
The major limitation of this study is related to the use of
treatment outcome observed in our study among smear-
retrospective secondary data. Some important variables
negative PTB and extrapulmonary TB patients might be
which might have impact on treatment outcome of TB/
due to the close supervision and attention given by
HIV co-infected patients, like socioeconomic character-
health care provides to these groups of patients. This
istics (income, family size, educational status, living con-
close supervision might be associated with considering
dition, social support, distance to the health facility),
these patients are at high risk for unsuccessful treatment
treatment and disease-related variables (adherence level,
outcomes.
viral load, drug resistance), as well as behavioral factors
In the present study, similar to studies conducted
(knowledge and attitude about the diseases, alcohol
elsewhere (MizanTepi [18] and Tigray [16]), the chance
abuse, cigarette smoking, illicit drug use) were not re-
of unsuccessful TB treatment outcome was higher
corded. Moreover, some patients were transferred to
among the TB/HIV co-infected patients in stage 2,
other health facilities where it is difficult to track what
stage 3 and stage 4 HIV diseases than those in stage 1
happened thereafter. Exclusion of medical records of a
disease. This might be due to the development of dif-
patient who were transferred out and/or found to be in-
ferent opportunistic infections among the patients with
complete may have also slightly affected our results. This
advanced stages of HIV which can directly affect TB
study also involved only patients in public hospitals. Its
treatment outcomes.
result may not be applicable to the patients in the pri-
The association between opportunistic infections
vate setting.
and TB treatment outcomes was also demonstrated
by our study. Those TB/HIV co-infected patients with
opportunistic infection history had a higher chance of Conclusion
unsuccessful TB treatment outcome than the patients The overall TB treatment success rate among the TB-
without opportunistic infection history. This is in line HIV co-infected patients in this study was higher com-
with a finding reported in Cameroon [32]. The pos- pared with many previous studies. TB/HIV patients with
sible reason could be poor adherence to anti TB a history of previous TB treatment, smear-positive pul-
drugs due to the high pill burden and associated monary TB, advanced HIV stage, history of opportunis-
drugs adverse effect. tic infection and no CPT initiated were at a high risk of
Similarly, the chance of unsuccessful TB treatment getting poor treatment outcomes.
outcome was higher among the TB/HIV co-infected Therefore, TB treatment facilities should give special
patients who did not take CPT compared with those attention to those TB-HIV co-infected patients with a
who took the prophylaxis. Studies conducted in North higher risk of unsuccessful TB treatment outcome.
West Ethiopia [43], Cameroon [32], South Africa Free
State [40], Yavatmal India [38] and Viet Nam [36] also Abbreviations
reported similar findings. The administration of CPT AOR: Adjusted Odds Ratio; ART: Antiretroviral Therapy; BMI: Body mass index;
CD4: Cluster of Differentiation 4; CI: Confidence interval; COR: Crude Odds
can reduce the chance of co-morbidities such as pneu- Ratio (); CPT: Co-trimoxazole prophylaxis; DOT: Directly Observed Therapy; EP
monia with pneumocystis, toxoplasmosis and other TB: Extra-pulmonary TB; HFSUH: Hiwot Fana Specialized University Hospital;
bacterial infections, which may then be related to a HHSC: Harar Health Science College; HIV: Human Immunodeficiency Virus;
IPT: Isoniazid preventive therapy; IQR: Inter Quartile Range; MDR: Multi-drug
higher rate of cure and treatment completion and a Resistant; PTB: Pulmonary TB; SD: Standard deviation; SNP TB: Smear negative
lower rate of deaths. PTB; SPP TB: Smear positive PTB; SPSS: Statistical package for social science;
According to the Ethiopia National Guidelines, CPT TB: Tuberculosis; TB/HIV: Tuberculosis/ Human Immunodeficiency Virus;
WHO: World health organization
should be provided for all HIV positive TB patients as
one component of the national TB/HIV collaborative
activity [6]. In spite of this, only 84.8% of TB/HIV co- Acknowledgments
The authors would like to thank the data collectors, hospital staff working at
infected patients were on CPT. This finding was con- TB clinics and heads of hospitals for their willingness and unreserved
sistent with studies from western Ethiopia (80%) [31], contribution in this study. We would like also to acknowledge Mr. Asfachew
Yavatmal India (80%) [38], Gondar (88.9%) [19] and Balcha for editing our manuscript.

Addis Ababa Ethiopia (77%) [26]. However, this figure

was higher than a report from Ebonyi State Nigeria Authors’ contributions
All authors were involved in the conception of the study, design, data
(55.3%) [34] and Free State province South Africa acquisition, data analysis and interpretation. The manuscript was also
(42.1%) [40]. Hence, health care providers should developed through active participation of all authors.
Tola et al. BMC Public Health (2019) 19:1658 Page 11 of 12

Authors’ information 10. Kebede A. Technical guideline for tuberculosis (TB) and TB-HIV program
AT is a lecturer in the Department of Epidemiology and Biostatistics, School implementation. World Vis. 2017.
of public health, College of Health and Medical Sciences, Haramaya 11. Belay M, Bjune G, Abebe F. Prevalence of tuberculosis, HIV, and TB-HIV co-
University. infection among pulmonary tuberculosis suspects in a predominantly pastoralist
KMM is a lecturer in the Department of Clinical Pharmacy, School of area, northeast Ethiopia. Glob Health Action. 2015;8(1).
Pharmacy, College of Health and Medical Sciences, Haramaya University. 12. MOH E-. National TB/HIV Sentinel Surveillance. One year Report (July 2011–June
YA is an assistant professor and Head of School of Pharmacy, College of 2012). Ethiopian Health and nutrition research institute (EHNRI)/ Ministry of
Health and Medical Sciences, Haramaya University. Health (MOH) with technical and financial support of PAPFAR trough the center
ANM is a lecturer in the Department of Pharmacology, School of Pharmacy, for disease control-Ethiopia 2013.
College of Health and Medical Sciences, Haramaya University and currently 13. Frontières MS, Partnership ST. OUT OF STEP 2017. TB policies in 29
he is a Ph.D. candidate. countries. A survey of prevention, testing and treatment policies and
NL is also a lecturer and a researcher in the department of pharmaceutics practices July 2017.
and social pharmacy, School of Pharmacy, College of Health and Medical 14. WHO. THE END TB STRATEGY: Global strategy and targets for tuberculosis
Sciences, Haramaya University. prevention, care, and control after 2015. Geneva World Health Organization 2014.
15. Health FMo. National Consolidated Guidelines for comprehensive HIV
Funding prevention, care and treatment. Addis Ababa, Ethiopia August 2018.
No funding from any source was obtained for this study. 16. Belayneh M, Giday K, Lemma H. Treatment outcome of human
immunodeficiency virus and tuberculosis co-infected patients in public
Availability of data and materials hospitals of the eastern and southern zone of Tigray region, Ethiopia. Bras J
The datasets of the study are available on reasonable request from the Infect Dis Poverty. 2015;19(1):47–51.
corresponding author. 17. Fiseha T, Gebru T. H G, Y D. tuberculosis treatment outcome among HIV co-
infected patients at Mizan- Aman general hospital, Southwest Ethiopia: a
Ethics approval and consent to participate retrospective study. J Bioeng Biomed Sci. 2015;5(139):1–4.
Ethical clearance was secured from the research review technical committee 18. Kefale AT, Anagaw YK. The outcome of tuberculosis treatment and its
of Harar Health Science College. A legal permission letter was taken from predictors among HIV infected patients in Southwest Ethiopia. Int J Gen
Harar Health Science College to the selected public hospitals and permission Med. 2017;10:161–70.
was obtained from the administration of the hospitals. Furthermore, before 19. Sinshaw Y, Alemu S, Fekadu A, Gizachew M. Successful TB treatment
reviewing medical records of the TB/HIV co-infected patients, permission was outcome and its associated factors among TB/HIV co-infected patients
obtained from the TB treatment unit heads of the hospital. Patient records/ attending Gondar University Referral Hospital, Northwest Ethiopia: an
information was anonymized and de-identified prior to analysis. Since the institution-based cross-sectional study. BMC Infect Dis. 2017;17:1–9.
data were collected through a review of medical records and confidentiality 20. The Federal democratic republic of Ethiopia MoH. Guidelines for clinical and
is maintained, there is no harm to the patients and their relatives. programmatic management of TB, Leprosy and TB/HIV in Ethiopia, fifth
edition Addis Ababa; April 2012.
Consent for publication 21. WHO. Definitions and reporting framework for tuberculosis – 2013 revision.
Not applicable. Geneva 27, Switzerland: World Health Organization December 2014.
22. WHO . Patient monitoring Guidelines for HIV care and Antiretroviral Therapy
Competing interests (ART). Geneva 27, Switzerland: World Health Organization 2006.
The authors declare that they have no competing interests. 23. WHO. Global tuberculosis report 2017. Switzerland, Geneva: World Health
Organization; 2017.
Author details 24. Jacobson KB, Moll AP, Friedland GH, Shenoi SV. Successful tuberculosis
1 treatment outcomes among HIV/TB Coinfected patients Down- referred
Department of Epidemiology and Biostatistics, School of public health,
College of Health and Medical Sciences, Haramaya University, Harar, Ethiopia. from a district hospital to primary Health clinics in rural South Africa. PLoS
2 One. 2015;10(5):1–11.
Department of Clinical Pharmacy, School of Pharmacy, College of Health
and Medical Sciences, Haramaya University, Harar, Ethiopia. 3Department of 25. Tripathi SB, Kapadia VK. Treatment outcome of tuberculosis in HIV
Pharmacology, School of Pharmacy, College of Health and Medical Sciences, seropositive patients: an experience of the southeast region of Ahmedabad.
Haramaya University, Harar, Ethiopia. 4Department of pharmaceutics and Natl J Commu Med. 2015;6(4):462–5.
social pharmacy, School of Pharmacy, College of Health and Medical 26. Ali SA, Thandisizwe R, Mavundla FR. Awoke T Outcomes of TB treatment in
Sciences, Haramaya University, Harar, Ethiopia. HIV co-infected TB patients in Ethiopia: a cross-sectional analytic study. BMC
Infect Dis. 2016;16:9.
Received: 11 June 2019 Accepted: 20 November 2019 27. Saini S, Singh M, Garg A. A retrospective cohort study of treatment
outcome among HIV positive and HIV negative TB patients in Chandigarh,
India. IND J COMMUN HEALTH JUN. 2016;28(02):145–50.
References 28. Cheru F, Mekonen D, Girma T, Belyhun Y, Unakal C, Endris M, et al.
1. Bruchfeld J, Correia-Neves M. Llenius GK. Tuberculosis and HIV Coinfection. Comparison of treatment outcomes of tuberculosis patients with and
Cold Spring Harb Perspect Med. 2015;5:1–16. without HIV in Gondar University Hospital: a retrospective study. J Pharm
2. Tesfaye B, Alebel A, Gebrie A, Zegeye A, Tesema C, Kassie B. The twin Biomed Sci. 2013;34:1606–13.
epidemics: Prevalence of TB/HIV coinfection and its associated factors in 29. Ismail I, Bulgiba A. Determinants of unsuccessful tuberculosis treatment
Ethiopia. Syst Rev Metaanal PLoS ONE. 2018;13(10). outcomes in Malaysian HIV-infected patients. Prev Med. 2013;57:S27–30.
3. Mburu G, Richardson DA. Community-based TB and HIV integration. Good 30. Prado TNd, Rajan JV, Miranda AE, Dias EdS, Cosme LB, Possuelo L Ga, et al.
Practice Guide. In: Hove. IHAAaP, editor. The first edition ed2013. Clinical and epidemiological characteristics associated with unfavorable
4. The Federal Democratic Republic of Ethiopia MoH. National comprehensive tuberculosis treatment outcomes in TB-HIV co-infected patients in Brazil: a
tuberculosis, leprosy and TB/HIV training manual for health care workers. hierarchical polytomous analysis. Braz J Infect Dis 2017;21(2):162–170.
Addis Ababa November 2016. 31. Ejeta E, Birhanu T, Wolde T. Tuberculosis treatment outcomes among
5. WHO. Global tuberculosis report 2018 Geneva: World Health Organization; 2018. tuberculosis/human immunodeficiency co-infected cases treated under
6. Health FMO. National guidelines for comprehensive HIV prevention, care, directly observed treatment of short course in Western Ethiopia. J AIDS HIV
and treatment. Addis Ababa: Federal Ministry Of Health; 2017. Res September 2014;6(8):164–170.
7. WHO. TB/HIV: HIV-Associated Tuberculosis World Health Organization; 2018. 32. Ako A, Agbor J, Bigna JR, Billong SC, Tejiokem MC, Ekali GL, Plottel CS, et al.
8. WHO. Latent tuberculosis infection: updated and consolidated guidelines factors associated with death during tuberculosis treatment of patients co-
for programmatic management. Geneva: World Health Organization; 2018. infected with HIV at the Yaounde´ central hospital, Cameroon: an 8-year
9. WHO. WHO policy on collaborative TB/HIV activities: Guidelines for national hospital-based retrospective cohort study (2006–2013). PLoS One. 2014;
programs and other stakeholders. World Health Organization 2012. 9(12):1–22.
Tola et al. BMC Public Health (2019) 19:1658 Page 12 of 12

33. Tabarsi P, Chitsaz E, Moradi A, Baghaei P, Marjani M, Mansouri D. Treatment

outcome and mortality: their predictors among HIV/TB co-infected patients
from Iran. Int J MycobacterIology. 2012;1:82–6.
34. Oshi DC, Oshi SN, Alobu I, Ukwaja KN. Profile, outcomes, and determinants
of unsuccessful tuberculosis treatment outcomes among HIV-infected
tuberculosis patients in a Nigerian state. Hindawi Publishing Corporation
Tuberculosis Res Treat 2014;2014:1–8.
35. Shastri S, Nagaraja SB, Tripathy JP, Satyanarayana S, Rewari BB. Predictors
and timing of ATT initiation among HIV-TB patients at ART centers of
Karnataka. India: Two Year Follow-Up PLoS ONE. 2015;10(09):1–11.
36. Huyen TTT, Nhung NV, Shewade HD, Hoa NB, Harries AD. Collaborative
activities and treatment outcomes in patients with HIV-associated
tuberculosis in Viet Nam. Public Health Action. March 2016;6(1):8 - `15.
37. Gebremariam G, Asmamaw G, Hussen M, Hailemariam MZ, Asegu D,
Astatkie A, et al. Impact of HIV status on treatment outcome of tuberculosis
patients registered at Arsi Negele Health center, southern Ethiopia: a six-
year retrospective study. PLoS One. 2016;11:1–11.
38. Ambadekar NN, Zodpey SP, Soni RN, Lanjewar SP. Treatment outcome and
its attributes in TB-HIV co-infected patients registered under the revised
national TB control program: a retrospective cohort analysis. Public Health.
2015;129:783–8.
39. Vijay S, Kumar P, Chauhan LS, Rao SVN, Vaidyanathan P. Treatment outcome
and mortality at one and half year follow-up of HIV infected TB patients
under TB control Programme in a district of South India. PLoS One. July
2011;6(7):1–8.
40. Engelbrecht MC, Kigozi NG, Chikobvu P, Botha S. Rensburg HCJv.
Unsuccessful TB treatment outcomes with a focus on HIV co-infected cases:
a cross-sectional retrospective record review in a high-burdened province
of South Africa. BMC Health Serv Res. 2017;17(470):1–10.
41. Warkari PD, Nakel MP, Mahajan SM, Adchitre SA. Study of treatment
outcome of tuberculosis among HIV co-infected patients: a cross-sectional
study in Aurangabad city, Maharashtra. Int J Commun Med Public Health.
2017;4(12):4466–71.
42. Refera H, Wencheko E. Survival of HIV-TB co-infected adult patients under
ART in ambo referral hospital. Ethiop Ethiop J Health Dev. 2013;27(2):88–93.
43. Sileshi B, Deyessa N, Girma B, Melese M, Suarez P. Predictors of mortality
among TB-HIV co-infected patients being treated for tuberculosis in
Northwest Ethiopia: a retrospective cohort study. BMC Infect Dis. 2013;
13(297):1–10.
44. Gesesew H, Tsehayneh B, Massa D, Gebremedhin A, Kahsay H, Mwanri L.
Predictors of mortality in a cohort of tuberculosis/HIV co-infected patients
in Southwest Ethiopia. Infect Dis Poverty. 2016;5(109):1–9.
45. Shastri S, Naik B, Shet A, Rewari B, Costa AD. TB treatment outcomes among
TB-HIV co-infections in Karnataka, India: how do these compare with non-
HIV tuberculosis outcomes in the province? BMC Public Health. 2013;
13(838):1–6.

Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.