Das et al.

BMC Public Health (2017) 17:203

DOI 10.1186/s12889-017-4097-y

RESEARCH ARTICLE Open Access

Health literacy in a community with low

levels of education: findings from Chakaria,
a rural area of Bangladesh
Susmita Das1*, Mohammad Nahid Mia1, Syed Manzoor Ahmed Hanifi1, Shahidul Hoque1 and Abbas Bhuiya2

Abstract
Background: Health literacy (HL) helps individuals to make effective use of available health services. In low-income
countries such as Bangladesh, the less than optimum use of services could be due to low levels of HL. Bangladesh’s
health service delivery is pluralistic with a mix of public, private and informally trained healthcare providers.
Emphasis on HL has been inadequate. Thus, it is important to assess the levels of HL and service utilization patterns.
The findings from this study aim to bridge the knowledge gap.
Materials and Methods: The data for this study came from a cross-sectional survey carried out in September 2014, in
Chakaria, a rural area in Bangladesh. A total of 1500 respondents were randomly selected from the population
of 80,000 living in the Chakaria study area of icddr, b (International Centre for Diarrhoeal Disease Research, Bangladesh).
HL was assessed in terms of knowledge of existing health facilities and sources of information on health care,
immunization, diabetes and hypertension. Descriptive and cross-tabular analyses were carried out.
Results: Chambers of the rural practitioners of allopathic medicine, commonly known as ‘village doctors’, were
mentioned by 86% of the respondents as a known health service facility in their area, followed by two public sector
community clinics (54.6%) and Union Health and Family Welfare Centres (28.6%). Major sources of information on
childhood immunization were government health workers. Almost all of the respondents had heard about diabetes
and hypertension (97.4% and 95.4%, respectively). The top three sources of information for diabetes were neighbours
(85.7%), followed by relatives (27.9%) and MBBS (Bachelor of Medicine and Bachelor of Surgery) doctors (20.4%). For
hypertension, the sources were neighbours (78.0%), followed by village doctors (38.2%), MBBS doctors (23.2%) and
relatives (15%). The proportions of respondents who knew diabetes and hypertension control measures were 40.9%
and 28.0%, respectively. More females knew about the control of diabetes (44.4% to 36.6%) and hypertension
(31.1% to 24.2%) than males.
Conclusions: A low level of HL in terms of modern health service facilities, diabetes and hypertension clearly
indicated the need for a systematic HL programme. The relatively high levels of literacy concerning immunization
show that it is possible to enhance HL in areas with low levels of education through systematic awareness-raising
programmes, which could result in higher service coverage.
Keywords: Health literacy, Hypertension, Diabetes, Village doctor

* Correspondence: [email protected]
1
Health Systems and Population Studies Division, International Centre for
Diarrhoeal Disease Research, Bangladesh (icddr,b), Dhaka, Bangladesh
Full list of author information is available at the end of the article

© The Author(s). 2017 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.
Das et al. BMC Public Health (2017) 17:203 Page 2 of 10

Background network of public health facilities and providers arranged

Very simply put, Health Literacy (HL) is defined as the according to administrative tiers, informal providers are
“ability to engage with health information and services” trusted health care providers for the many rural people of
[1]. The phrase “Health Literacy” has evolved as a dynamic Bangladesh for health services and information [11].
multi-layered concept since its appearance in the litera- There have not been adequate systematic explorations
ture in 1974 [2]. HL refers to multiple factors that individ- of HL in Bangladesh. In the context of rural Bangladesh,
uals or communities need to access, understand, appraise this study attempts to fill this knowledge gap. In our
and use health-related information and services for mak- study, we considered HL in terms of knowledge of exist-
ing the best health-related decisions [1, 3]. These factors ing health facilities, care-seeking patterns, immunization,
include individual characteristics, general literacy, experi- and knowledge of diabetes and hypertension.
ence with diseases and health systems, and social, cultural,
and psychological factors [1, 3]. In today’s societies, HL is Materials and Methods
gaining more attention than before for many reasons. Study Area
Modern health systems are complex and can be difficult The study was conducted in Chakaria Upazila (a sub dis-
to navigate and understand. The education systems might trict) under the Cox’s Bazar District situated in the south-
not equip people with necessary skills that will help them east coastal area of Bangladesh [12, 13]. Like other parts
effectively interact with modern health systems and infor- of Bangladesh, agricultural farming is one of the major
mation for improving their health [4, 5]. Differences in HL economic activities of the villagers, along with woodcut-
contribute to health inequities and health outcomes [1]. ting and fishing, due to the proximity of the villages to the
There are mainly three types of HL: basic/functional, sea and forest. Thirty percent of the households were
interactive/communicative and critical HL [6]. Basic HL landless and about half of the households were dependent
refers to the minimum amount of reading and writing on income from menial labour [14]. The study site has
skills necessary for everyday functioning, whereas inter- been one of the most conservative areas in terms of reli-
active HL includes higher-level cognitive, literacy and so- gion and openness to modern ideas, with low levels of
cial skills that enable people to extract, interpret and use secular education. The literacy rate among the population
information accordingly [6]. Critical literacy is associated aged 7 years or older was 47.0% and was similar among
with higher autonomy and empowerment as it allows for males and females [15]. More than 90% percent of the
the critical analysis of information and greater control population were Muslim, while the remainder were either
over situations [6]. Low HL leads to a detrimental lifestyle Hindu or Buddhist [14, 16]. At the time of this survey,
and behaviour; prevents the uptake of disease prevention there was one Upazila Health Complex at the sub-district
and detection services; hampers self-management of level, 10 Union Health and Family Welfare Centres
chronic disease, compliance with medications and un- (UHFWCs) at the union level, 254 satellite clinics for
derstanding of provider communication; raises health immunization run by government health workers, 23
care costs; and worsens existing inequities [4]. Non- Community Clinics at the lowest level and one rural dis-
communicable diseases (NCDs) are one of the leading pensary for first-level services run by the government. In
causes of deaths globally and are associated with mul- addition, there were 7 community-established and icddr,b
tiple modifiable behavioural risk factors [4, 7]. Low facilitated Village Health Posts for primary care consult-
HL adversely affects the behavioural risk factors of ation services, and 3 private diagnostic centres [13]. Apart
NCDs and is often worse for vulnerable groups such from these, 342 village doctors (informal) provided pri-
as the elderly, individuals with low levels of educa- mary care and dispensed both allopathic and homeopathic
tion, minorities and immigrants, who have inadequate drugs [13]. In 2012, the life expectancy of the population
HL as well as a high risk for developing NCDs [4]. at birth in the study area was 68 years for males and
NCDs are responsible for 59% of the proportional 70 years for females [13]. In Chakaria, pulmonary tubercu-
mortality in Bangladesh (% of the total deaths in both losis, acute respiratory infection, stroke, diabetes and
sexes and all age groups) [8]. In 2011, the BDHS chronic obstructive pulmonary diseases were the leading
(Bangladesh Demographic and Health Survey) report causes of death in all age groups [14].
stated that 32% of women and 19% of men had elevated
blood pressure (BP) and an additional 28% of women and Study design and data collection
men were pre-hypertensive; 11% of men and women were This study is a cross-sectional survey carried out in
diabetic and 25% of women and men were pre-diabetic some villages of Chakaria, where icddr,b has been run-
[9]. The struggling health system is inadequately prepared ning a Health and Demographic Surveillance System
to face this challenge [10]. Bangladesh already has a crit- (HDSS) comprising a population of 80,000. In Septem-
ical shortage of health workers with urban clustering [11]. ber 2014, a total of 1500 respondents, with equal pro-
Despite having a public health delivery system with a vast portions of males and females, aged more than 18 years
Das et al. BMC Public Health (2017) 17:203 Page 3 of 10

were randomly selected from the Chakaria HDSS data- Data analysis
base. Of the 1500 respondents, 283 could not be inter- Descriptive and cross-tabular analyses were carried out
viewed due to their non-availability; 23 migrated out of using STATA software (STATA/SE version14; StataCorp,
the area, 22 died, 226 were absent at the time of the 4905 Lakeway Drive, College Station, Texas 77845 USA).
visit, 5 were deaf, 3 were sick, and 4 respondents refused Chi–square and Fisher’s exact tests were performed to
to be interviewed. Thirteen local female interviewers assess the statistical significance of associations between
with at least 10 years of schooling were recruited for the outcome variables and the background characteris-
data collection. The interviewers received a two-day tics of the respondents.
training on interviewing before data collection. The
questionnaire was drafted in Bangla and developed on Results
the basis of previous literature and available information The survey team interviewed 1217 respondents out of
in the context of the health system, e.g., the existing 1500 sampled respondents through repeated visits. Three
health facilities and the existing health care system of visits were made to minimize the rate of non-response.
Bangladesh. The questionnaire had some listed answers Men were less frequently available for the interview during
with provisions for recording other answers. For some daytime due to their external engagements. In total, 72.1%
questions, provisions for recording multiple answers (541) males and 90.1% (676) females were interviewed.
were made. Two experienced supervisors supervised the
data collection operation. The supervisors revisited 5% Background characteristics of the respondents
of the randomly chosen respondents within 2 days of Forty-four percent of the 1217 people interviewed were
data collection by the field workers to ensure the data males. Approximately 8% of the respondents were be-
quality. Later, the supervisors and the field workers tween 20 and 29 years of age, and 24% were 60 years old
sorted out any inconsistencies in the data collected dur- or older. One-fourth of the respondents were between
ing the re-interviews. All completed questionnaires were 30 and 39 years of age, and 22% and 20% were between
manually checked for completeness and for any incon- 40 and 49 years and 50 and 59 years of age, respectively.
sistencies, which were identified by computer-based Fifty-two percent of the respondents had no formal
ranges and consistency checks. schooling, while 29% had 1–5 years of schooling; only
19% had 6 or more years of schooling.

Definition of the study variables Knowledge of existing health service facilities

HL was assessed by the knowledge of existing health facil- Table 1 presents the respondents’ level of knowledge of
ities, immunization, diabetes and hypertension detection existing health facilities in the study area. The village
and control measures and the common consultation of doctor’s chamber was the most widely known service fa-
health care providers. The respondents were asked the fol- cility (86.0% of respondents and 46.0% of responses),
lowing questions: “Please name the health facilities that followed by the community clinic (54.6% of respondents
exist in your union?”, “Whom do you usually consult and 29.2% of responses), the Union Health and Family
when you need any information about health-related mat- Welfare Centre (28.6% of respondents and 15.3% of re-
ters? Please mention the first three people in order of pri- sponses), and icddr,b-run village health posts (12.3% of
ority.”; “Has s/he (child less than five years of age) been respondents and 6.6% of responses).
vaccinated?”; “How did you know for the first time
whether you are diabetic or not?” “How can diabetes be Table 1 Knowledge of existing health facilities among the
controlled? (Multiple answers possible)”; “How did you respondentsa
know for the first time whether you are hypertensive or Facilities % of the % of
responses respondents
not?”; “How can hypertension be controlled? (Multiple
answers possible)”. The mode of identifying diabetes was Village doctor's chamber 45.9 86.0
categorized as urine and/or blood test and others, and Community clinic 29.2 54.6
mode of detecting hypertension were categorized as a Union Health and Family Welfare 15.3 28.6
blood-pressure measurement by a health care provider or Centre (UHFWC)
by other means. Where appropriate, multiple responses icddr,b health centre 6.6 12.3
were recorded. The respondents’ background characteris- SACMO’s (Sub assistant community 0.9 1.7
tics, including age, number of completed years of school- medical officer) chamber
ing, and a list of household items owned by any member Others 2.1 3.9
of the household were recorded. Items owned by the N 2279 1217
household were used to calculate a household asset score a
Multiple responses were reported by a given respondent, therefore the
using Principal Component Analysis (PCA) [12, 17]. number of responses are higher than number of respondents
Das et al. BMC Public Health (2017) 17:203 Page 4 of 10

Preferred sources of consultation for health matters Immunization

and use Out of the total of 550 under-five children residing in
Graduate physicians or formally trained registered the respondents’ households, 97.8% were reported to be
medical practitioners with MBBS degrees were the most vaccinated. Out of the total respondents, 45.4% learned
preferred source for consultations on health matters about vaccination from EPI (Expanded Programme on
(41.6%), followed by village doctors (37.5%) and family Immunization) workers, 40.7% from non-EPI health
members (15.1%). More females than males preferred to workers, 19.5% from neighbours and 13.0% from loud-
consult family members for health-related matters. speaker announcements by EPI workers (Table 4).
Males were more likely to prefer consulting sources
outside of the family, e.g., MBBS doctors, SACMO and Diabetes and hypertension
village doctors, for health-related matters. With the im- The majority of the respondents had heard about dia-
provement of socioeconomic status and education levels, betes (97.4%) and hypertension (95.4%). Slightly higher
more respondents preferred to consult MBBS doctors proportions of males compared to females knew about
for health matters and the preference for village doctors diabetes (99.1% versus 96.0%) and hypertension (97.2%
declined (Table 2). versus 93.9%). A total of 1989 responses were recorded
Table 3 presents the source of consultation in the case for diabetes and 1976 for hypertension.
of sickness during the last twelve months. In as many as The most frequently mentioned source of informa-
51.6% of the cases, the respondents contacted a village tion for diabetes was neighbours (85.7%), followed by
doctor followed by MBBS doctors (39.4%) and SACMO relatives (27.9%), MBBS doctors (20.4%) and village
(4.9%). Males and females consulted MBBS doctors doctors (18.5%) (Table 5). Fewer respondents in the
equally, while more males than females sought consult- low socioeconomic status group (12.5%), the illiterate
ation from village doctors and SACMO. Respondents group (14.7%), and females (20.0%) reported that for-
with higher education levels and socioeconomic status mal healthcare providers such as MBBS doctors were
were more likely to seek consultations from MBBS doc- their source of information compared to respondents
tors, and this trend was statistically significant. The op- from the high socioeconomic status group (34.3%),
posite was observed for village doctors. those with 6 or more years of education (30.4%), and

Table 2 First preference health care information providers by socio-demographic characteristic

Socio-demographic Characteristics No. of respondents Family Neighbour Drug seller/Village doctor MBBS doctor SACMO Others P-value
All 1217 15.1 0.8 37.5 41.6 3.6 1.4
Sex 0.013
Male 541 11.3 0.9 38.5 44.6 4.3 0.9
Female 676 18.2 0.7 36.7 39.2 3.1 1.8
Age 0.436
20–29 102 21.6 0.0 40.2 32.4 2.9 2.9
30–39 306 12.8 0.7 41.2 38.2 5.6 1.3
40–49 273 14.3 1.1 37.4 42.1 3.7 1.8
50–59 245 17.1 0.8 31.8 46.9 2.5 0.4
60+ 291 14.4 1.0 37.5 43.3 2.8 1.4
Education 0.002
None 635 14.7 1.1 42.5 36.4 4.1 0.9
1–5 years 355 13.2 0.6 34.6 46.5 3.1 2.3
6+ years 227 19.4 0.4 27.8 48.4 3.1 1.3
Asset quintile <0.001
Lowest 246 15.0 0.4 48.4 31.3 2.9 2.0
2nd 242 10.3 0.0 43.0 37.2 8.3 1.2
Middle 244 11.1 1.6 43.0 41.4 2.5 0.8
4th 242 16.5 1.7 30.6 46.7 2.9 0.8
Highest 243 22.6 0.4 22.2 51.4 0.8 2.1
Das et al. BMC Public Health (2017) 17:203 Page 5 of 10

Table 3 Health care provider consultation in the last year by socio- demographic characteristics
Socio-demographic No. of respondents Consulted with drug MBBS doctor SACMO Health and family Others P-value
Characteristics seller/Village doctor planning worker
All 987 51.6 39.4 4.9 1.9 2.2
Sex 0.027
Male 404 53.5 39.6 5.2 0.5 1.2
Female 583 50.3 39.3 4.6 2.9 2.9
Age 0.003
20–29 77 51.9 32.5 9.1 5.2 1.3
30–39 237 59.9 29.5 7.2 2.1 1.3
40–49 221 50.7 43.4 3.6 1.4 0.9
50–59 203 47.3 42.4 4.4 1.5 4.4
60+ 249 47.8 45.0 2.8 1.6 2.8
Education <0.001
None 525 56.6 33.9 4.6 1.7 3.2
1–5 years 300 49.7 40.7 6.0 2.0 1.6
6+ years 162 38.9 54.9 3.7 2.5 0.0
Wealth index <0.001
Lowest 197 64.0 24.9 5.6 1.5 4.0
Second 202 59.9 29.2 6.4 4.0 0.5
Middle 197 53.3 38.1 5.1 2.5 1.0
Fourth 196 52.0 38.8 5.1 0.5 3.6
Highest 195 28.2 66.6 2.1 1.0 2.1

males (20.9%). The proportion of respondents reporting status compared to those with no education (16.1% versus
family members as their source of information about dia- 36.3%) and low socioeconomic status (10.0% versus 39.7%).
betes increased from 8.6% among illiterates to 13.2% The prevalence of family members and relatives as a source
among those who had more than 6 years of education. of information increased with the respondents’ socioeco-
Table 6 shows the likely sources of information about nomic status (9.1% versus 19.7%; 13.9% versus 16.7%) and
hypertension. Neighbours were the most important source education (11.2% versus 13.9%; 14.2% versus 19.3%).
of information about hypertension irrespective of age, sex, Interestingly, for both diabetes (85.7%) and hyperten-
education and household socioeconomic status. The pro- sion (78.0%), neighbours were the most frequently re-
portion of respondents seeking information from MBBS ported source of information followed by drug sellers/
doctors was higher among respondents with 6 or more village doctors (18.5% and 38.2%), relatives (27.9% and
years of education and those with high socioeconomic 15.0%), MBBS doctors (20.4% and 23.2%), and family
members (9.8% and 12.2%) (Tables 5 and 6).
Table 4 Distribution of sources of information regarding
childhood immunizationa Diabetes and hypertension among respondents
Sources of information % of responses % of respondents A total of 1185 respondents had heard about diabetes.
EPI worker 35.0 45.4 Among them, only 5.7% reported that they were dia-
Health worker 31.4 40.7 betic, while 79.8% claimed to be non-diabetic and 14.5%
Neighbours/villagers 15.0 19.5
did not know whether they were diabetic or not. Sixty-
nine percent of the identified diabetics learned of their
Miking (loud speaker) 10.0 13.0
status through blood and/or urine tests. Nearly one-
Family member 2.6 3.4 third of the diabetics were informed about their disease
Book 1.4 1.9 by another source (Table 7).
Others 4.6 5.9 Out of 1161 respondents who had heard about hyper-
N 698 538 tension, the majority (71.1%) reported not being hyper-
a
Multiple responses were reported by a given respondent, therefore the
tensive and only 11.2% reported being hypertensive,
number of responses are higher than number of respondents while the rest (17.7%) did not know whether they were
Das et al. BMC Public Health (2017) 17:203 Page 6 of 10

Table 5 Sources of information on diabetes by socio- demographic characteristicsa

Socio-demographic characteristics No. of respondents Family Relatives Neighbour Drug seller/Village Doctor MBBS doctor Others P-value
All 1185 9.8 27.9 85.7 18.5 20.4 5.6
Sex <0.001
Male 536 7.5 21.8 86.4 23.0 20.9 8.4
Female 649 11.7 32.8 85.2 14.8 20.0 3.2
Age (years) 0.143
20–29 102 10.8 34.3 85.3 19.6 15.7 3.9
30–39 301 7.0 29.2 87.7 19.9 17.3 7.3
40–49 268 8.2 28.0 86.6 18.7 18.3 3.4
50–59 242 7.4 26.3 87.2 21.5 26.4 6.6
60+ 272 16.2 25.4 81.6 13.6 22.4 5.5
Education <0.001
None 605 8.6 24.6 87.9 17.8 14.7 4.0
1–5 years 353 9.6 27.8 87.0 17.3 23.8 3.7
6+ years 227 13.2 36.6 78.0 22.0 30.4 12.9
Wealth index <0.001
Lowest 232 3.5 22.4 91.4 23.3 12.5 4.3
Second 233 4.7 22.7 94.0 10.2 11.6 5.6
Middle 239 6.3 27.2 85.4 15.5 20.1 5.4
Fourth 239 13.0 31.8 81.6 20.1 23.0 5.4
Highest 242 21.1 34.7 76.9 13.6 34.3 7.0
a
Multiple responses were recorded

Table 6 Sources of information on hypertension by socio demographic characteristicsa

Socio-demographic characteristics No. of respondents Family Relatives Neighbour Drug seller/Village Doctor MBBS doctor Others P-value
All 1161 12.2 15.0 78.0 38.2 23.2 3.6
Sex 0.002
Male 526 12.2 11.4 77.9 44.3 23.0 4.4
Female 635 12.3 18.0 78.1 33.1 23.3 2.7
Age (years) 0.249
20–29 100 13.0 19.0 79.0 34.0 20.0 3.0
30–39 292 9.3 16.8 82.9 40.1 17.5 1.7
40–49 262 11.5 14.5 76.3 40.5 25.2 3.1
50–59 239 13.0 15.1 73.2 42.3 27.2 6.7
60+ 268 15.3 11.9 78.4 31.7 25.0 3.7
Education <0.001
None 590 11.2 14.2 78.8 39.2 16.1 2.9
1–5 years 348 12.9 13.5 79.3 37.4 26.7 2.9
6+ years 223 13.9 19.3 74.0 36.8 36.3 6.7
Wealth index <0.001
Lowest 230 9.1 13.9 80.9 47.4 10.0 0.9
Second 229 10.0 12.7 83.4 39.7 16.2 3.9
Middle 235 10.6 11.9 75.7 38.3 23.4 3.8
Fourth 233 11.6 19.7 77.3 36.1 26.2 3.0
Highest 234 19.7 16.7 73.1 29.5 39.7 6.4
a
Multiple responses were recorded
Das et al. BMC Public Health (2017) 17:203 Page 7 of 10

Table 7 Distribution of modes of detecting diabetes among Table 9 Diabetes control measures by sex of the respondents
respondents Mentioned control measure Male Female Total P-value
Mode of detection Male Female Total P-value Low carbohydrate diet, Physical 72.7 73.4 73.1 0.781
Blood test and/or Urine test 68.4 69.6 69.3 0.973 activity or exercise, Drugs,
Reducing anxiety, Eating low fat
Symptoms, told by doctor 31.6 30.4 30.7 food, Avoiding sweetened food
No. of respondents 17 51 68 Eating more vegetables, stop 27.3 26.6 26.9
eating when stomach is partially
full, quitting smoking
hypertensive or not. Ninety-seven percent of the hyper- Number of respondents 198 300 498
tensive respondents learned of their diagnosis through a
blood pressure (BP) measurement by a health care pro-
vider and the rest learned of their diagnosis from sus- Health facility and care seeking
pected symptoms. The differences observed in the mode In Bangladesh, the archetypal public sector health ser-
of detecting hypertension between males and females vice in rural areas is distributed through a vast network
were not statistically significant (p = 1.00) (Table 8). of a multi-tier public health facilities and providers ar-
ranged according to administrative level (national, div-
isional, district, upazila/sub district, union and ward).
Control of diabetes and hypertension Primary health care is delivered through upazila and
In total, 44.4% (300 out of 676) of females and 36.6% lower down public health facilities, while district or
(198 out of 541) of males knew about control measures upper-level facilities provide secondary and tertiary care
for diabetes (p < 0.001). Among them, 73.1% of respon- [18]. Despite the presence of formal providers from pub-
dents mentioned a low-carbohydrate diet, physical activ- lic and private sectors, “informal providers” provide care
ity or exercise, drugs, lowering anxiety, low-fat food and to the majority of the Bangladeshi rural population due
the avoidance of sweetened foods as measures to control to the poor quality of the public health care and scarce
diabetes, with similar levels of knowledge among males resources and facilities [11]. “Informal providers” are not
and females (Table 9). registered with any government regulatory body and can
In total, 24.2% (131 out of 541) of males and 31.1% (210 be found in both sectors [19]. They include community
out of 676) of females knew about measures to control health workers (CHWs), village doctors, drug vendors,
hypertension (p < 0.001). Among them, 79.4% of respon- traditional healers, faith healers, traditional birth atten-
dents mentioned drugs, physical exercise, anxiety reduction, dants, and homeopaths, among others [19]. They prac-
reduced salt consumption, and nutritious food as methods tise allopathic, homeopathic, and traditional medicine
for controlling hypertension. A slightly higher percentage with little or no formal training in medicine. The infor-
of males (82.1%) compared to females (77.8%) mentioned mal providers are close to the community’s homes and
these measures for controlling hypertension (Table 10). are rooted in their belief system, despite the presence of
formal providers [19]. Our results also support this fact
(Table 1). Despite the presence of both state and pri-
Discussion vately run facilities, the chamber of the local village doc-
Health decision making is a dynamic interaction influenced tor, who is an informal provider, was the most widely
by an individual’s ability to access, appraise, understand and known health care facility among the respondents. Earlier
decide while interacting with information, systems, support, studies from Chakaria reported that as much as 96% of
resources and the environment. This interaction is driven the all health care providers were informal [20]. In
by the “contextual demand” of the specific disease, commu- terms of health care seeking, in actual practice, nearly
nication characteristics of the practised medical culture, the half of the respondents consulted village doctors des-
health care system and society’s value of its members. Thus, pite stating a preference for MBBS doctors (Table 3). A
HL is context- and content-specific [1]. study performed in Chakaria in 2007 reported that
Table 10 Hypertension control measures by sex of the
respondents
Table 8 Distribution of modes of detecting hypertension
Mentioned control measure Male Female Total P-value
among respondents
Low fat diet, Physical exercise, Decrease 82.1 77.8 79.4 0.249
Mode of detection Male Female Total P-value
food intake, Lower anxiety, Reduce salt
Measurement of BP by any provider 96.4 97.1 96.9 1.00 consumption, Eating nutritious food
Suspected because of symptom 3.6 2.9 3.1 Resting and seeking advice from doctor 17.9 22.2 20.6
No. of respondents 28 102 130 Number of respondents 131 210 341
Das et al. BMC Public Health (2017) 17:203 Page 8 of 10

consultations with village doctors (65%), irrespective of children when the full schedule is completed with the
disease types, were much more frequent compared to right doses administered at the right times. Health
consultations with MBBS doctors and other types of workers play a very important role in driving this mes-
health care providers (14%) [21]. In our study, we also sage home. Communities’ reliance on community health
found that a higher proportion of respondents received workers for vaccinating their children is thought to be
consultations from informal healthcare providers, i.e., more important than the physical existence of health
village doctors, but consultations from MBBS doctors care facilities [30].
had risen to 39%. With the rise in education level and
socio economic status, more people went to MBBS doc- Diabetes and hypertension
tors and fewer went to village doctors. Interestingly, Despite making progress in maternal and child health
males and females consulted MBBS doctors with the and in a few communicable diseases, Bangladesh is fa-
same frequency in this study. cing a new threat from NCDs. NCDs are responsible for
Informal providers comprise 51%-96% of providers in 59% of the proportional mortality in Bangladesh (% of
developing countries such as Bangladesh, India, and total deaths in both sexes and all age groups) [8]. Be-
Uganda due to their convenience, affordability and cul- tween 1986 and 2006, there has been a nearly nine-fold
tural acceptability [22]. Although they may provide ques- increase in the proportion of deaths due to NCDs in
tionable and even harmful care, village doctors are a Matlab, a rural HDSS in Bangladesh [11]. The same
trusted source of care and are intricately embedded in HDSS also reported a 3527% increase in mortality due
the social fabric in Bangladesh [23]. With the critical to cardio- and cerebro-vascular diseases during the same
shortage in the health workforce and the epidemiological period [31]. The death rate from CVDs (cardio vascular
transition at hand, non-physician health care provider diseases) in 2025 is projected to be 21 times the 2003
empowerment and task shifting are favoured for improving rate, and Bangladesh is projected to be among the top
access and coverage [24]. Bangladesh itself has successfully ten countries in terms of the number of diabetics [11].
used village doctors in its tuberculosis control programme The “urban-ness” of diseases such as hypertension,
to refer suspected cases and to distribute DOTS therapy popularly known as “blood pressure”, and diabetes not-
[25]. China’s “bare foot doctors” were systematically trained withstanding, the prevalence of these diseases is increas-
for a short duration to provide basic curative services and ing alarmingly among rural communities in Bangladesh.
health education to rural populations [26]. In a cross-sectional study conducted in a Bangladeshi
rural community, the prevalence of hypertension among
Immunization adult males and females was reported to be as high as
Bangladesh has admirably reduced child mortality. 31.5 and 29.3%, respectively [32]. In our study, the ma-
Childhood immunization through EPI is a major catalyst jority or participants had heard about diabetes and
in this success. According to recent national data, 78% hypertension. Interestingly, the most common sources of
of 12- to 23-month-old children were vaccinated with all information about diabetes and hypertension for the ma-
recommended vaccines before their first birthday [27]. jority of participants were neighbours and not health
In our study, we found that knowledge about childhood care providers. Knowledge about the control of diabetes
vaccination was universal among the respondents. Their and hypertension was patchy, and respondents reported
source of knowledge is mostly public sector grass root a variety of control measures. The majority of respon-
health workers, with much smaller contributions from dents who reported having diabetes and hypertension
family and community members. Another study con- were diagnosed through blood/urine tests and blood
ducted in a rural area of Bangladesh also reported that pressure measurements, respectively. HL is crucial for
the public sector health workers are the major source of diseases such as diabetes and hypertension as they re-
immunization information for the rural masses [28]. Un- quire considerable self-care and management, with com-
doubtedly, these grass roots-level health workers, such pliance to treatment being a major issue. Individuals
as health assistants, are the driving force of the success- with poorly controlled diabetes and low HL believed that
ful EPI campaign and the source of immunization infor- they were optimally controlling their blood glucose, al-
mation. This issue is an important point to consider though they did not take measures to improve their gly-
because a shortage of this crucial health force can im- caemic control [33]. Bangladeshis are among the South
pede the progress of the programme. In 2013, 16.7% of Asian populations most likely to have salient NCD be-
the sanctioned posts for domiciliary staff (health inspec- havioural risk factors such as physical inactivity, low in-
tors, assistant health inspectors, and health assistants) take of fruits and vegetables and tobacco consumption
were vacant [18]. Understanding only the benefit of [34]. Many of these risk factors can be modified by
immunization is not going to be useful for ensuring full adopting a healthy lifestyle. Diabetes and hypertension
immunization coverage [29]. Immunization can benefit management include patient involvement in optimizing
Das et al. BMC Public Health (2017) 17:203 Page 9 of 10

individual blood glucose and blood pressure levels to the findings do not reflect the nature and magnitude of
prevent complications. It is reported that patients with association between one independent variable and a
low HL have difficulty understanding their health care dependent variable when the effects of other independ-
providers’ health communications, are less keen to be- ent variables are controlled for.
come involved in medical decision making and are
dependent on their family members, friends and health Conclusions
care providers for final decision making [35]. HL is pos- In conclusion, we found that HL is improving in this
tulated to help diabetics achieve better glycaemic control community with low levels of general literacy. This re-
by influencing their self-efficacy and self-care behaviour sult was possible because of the large-scale dissemin-
[36]. A lack of understanding of the long-term effects of ation of health messages through health workers and
diabetes and hypertension can lead to poor adherence to various media sources promoting science-based health
medication among hypertensive and diabetic patients. In messages, which began a shift from traditional beliefs to
a study conducted in rural China, only 49% of hyperten- modern ones. In Bangladesh, the context-specific pro-
sive respondents knew that they had to take medicine motion of oral rehydration therapy (ORT) for the man-
daily. This study also elucidated that most of the rural agement of diarrhoea and the aggressive promotion of
population received their knowledge about hypertension immunization were of great importance. In Bangladesh,
from village clinics [37]. It is very important point for us ORT is now used in over 80% [38] of cases of diarrhoea,
to consider these findings from a health systems point of and immunization coverage by 12 months is now 78%
view. Low and middle income countries such as [27], which is a big increase from less than 2% in the
Bangladesh have undergone an epidemiologic transition mid-eighties [39]. Appropriate HL enhancement pro-
and face a huge surge in NCDs [7]. The health systems grammes need to be undertaken to achieve universal
of LMICs (low and middle income countries), which are health coverage by tackling the future challenges of hyper-
mostly acute care-oriented, are not prepared to tackle tension, diabetes and other emerging health threats.
the NCD challenge [7]. A very recent NCD score card
Abbreviations
ranked Bangladesh’s health system very poorly in terms BP: Blood pressure; EPI: Expanded Programme on Immunization; HL: Health
of NCD preparedness [10]. The Global Strategy for the Literacy; icddr,b: International Centre for Diarrhoeal Disease Research,
Prevention and Control of Non Communicable Diseases Bangladesh; LMIC: Low and middle income country; MBBS: Bachelor of
Medicine and Bachelor of Surgery; NCD: Non communicable disease;
advocates the combination of primary prevention inter- SACMO: Sub assistant community medical officer
ventions for whole populations, targeting high-risk indi-
viduals and improving access to care [7]. The chronic Acknowledgements
This study was based on icddrb’s Health and Demographic Surveillance
nature and the expensive, technology-intensive treat- System (HDSS), Chakaria which is supported by icddr,b’s core donors. icddr,b
ment modalities of advanced stage NCDs necessitates is grateful to the Governments of Bangladesh, Canada, Sweden and the UK
focus on primary and secondary prevention efforts [7]. for providing core/unrestricted support.
Countries such as Bangladesh will need to come up with Funding
a comprehensive health system response to face the This study was based on icddrb’s Health and Demographic Surveillance
enormous challenge posed by NCDs, which will involve System (HDSS), Chakaria which is supported by icddr,b’s core donors:
Governments of Bangladesh, Canada, Sweden and the UK.
finding solutions to many lingering health system short-
comings such as workforce issues, capacity building, re- Availability of data and materials
ferral systems, and patient empowerment, among others The authors confirm that all data underlying the findings are fully available
without restriction. All data underlying the findings in this paper are freely
[7]. The large number of informal providers practising in available upon request subject to adherence to icddr,b’s data sharing policy
rural areas can be explored for this purpose. They can (http://www.icddrb.org/policies). The request for the data can be sent to
take up health promotion with tailor made messages for [email protected], the co-principal investigator of the project.
the rural population of Bangladesh with low literacy
Authors’ contributions
levels. This area requires further in depth exploration in AB, SMAH, MNM, SH, SD conceived and designed the study. MNM, SMAH,
terms of existing disease literacy and the optimum con- SH performed the experiments. AB, SMAH, MNM analysed the data. AB, SD,
MNM wrote the paper. AB, SMAH, SD, MNM provided critical review, input
tent of messaging for health promotion.
and revision of the manuscript. All authors read and approved the final
manuscript. We acknowledge Sharif Al Hasan, Ashish Paul for field
Limitations implementation and data management.
Our study was conducted in a select number of villages
Competing interests
in a rural area of Bangladesh and was not based on a The authors declare that they have no competing interests.
nationally representative sample. Thus, the study find-
ings cannot be generalized for the whole country. The Consent for publication
The manuscript does not contain any individual data in any form (including
association between the dependent and independent individual details, images or videos); hence, consent for publication was not
variables was assessed through univariate analysis. Thus, applicable.
Das et al. BMC Public Health (2017) 17:203 Page 10 of 10

Ethics approval and consent to participate 19. Ahmed SM, Hossain MA, Chowdhury MR. Informal sector providers in
The Ethical Review Committee of the International Centre for Diarrhoeal Bangladesh: how equipped are they to provide rational health care? Health
Disease Research, Bangladesh (icddr,b) approved the project. Informed Policy Plan. 2009;24(6):467–78.
written consent was taken from all interviewees, and confidentiality and 20. Rasheed Sabrina, Iqbal Mohammad, Urni Farhana: Inventory of Facilities. In
anonymity were ensured. Health for the Rural Masses Insights from Chakaria Edited by Bhuiya Abbas.
Dhaka: ICDDR,B; 2009:25–37.
Author details 21. Mahmood Shehrin Shaila, Iqbal Mohammad, Hanifi SMA: Health -seeking
1
Health Systems and Population Studies Division, International Centre for Behaviour. In Health for the Rural Masses Insights from Chakaria. Edited by
Diarrhoeal Disease Research, Bangladesh (icddr,b), Dhaka, Bangladesh. Bhuiya Abbas. Dhaka: ICDDR,B; 2009. p. 67–94.
2
Partners in Population and Development, Dhaka, Bangladesh. 22. Sudhinaraset M, Ingram M, Lofthouse HK, Montagu D. What is the role of
informal healthcare providers in developing countries? a systematic review.
Received: 7 April 2016 Accepted: 1 February 2017 PLoS ONE. 2013;8(2):e54978.
23. Mahmood SS, Iqbal M, Hanifi SMA, Wahed T, Bhuiya A. Are 'Village Doctors'
in Bangladesh a curse or a blessing? BMC Int Health Human Rights. 2010;
10(1):1–10.
References 24. Sharma A, Ladd E, Unnikrishnan MK: Healthcare Inequity and Physician
1. Dodson S, Good S, Osborne RH. Health literacy toolkit for low- and middle- Scarcity: Empowering Non-Physician Healthcare. Economic & Political
income countries: a series of information sheets to empower communities Weekly (EPW). 2013;48(13):112–17.
and strengthen health systems. New Delhi: World Health Organization, 25. Hamid Salim MA, Uplekar M, Daru P, Aung M, Declercq E, Lönnroth K.
Regional Office for South-East Asia; 2015. Turning liabilities into resources: informal village doctors and tuberculosis
2. Ratzan SC. Health literacy: communication for the public good. Health control in Bangladesh. Bull World Health Organ. 2006;84(6):479–84.
Promot Int. 2001;16(2):207–14. 26. Weiyuan Cui China’s village doctors take great strides. Bulletin of the World
3. Sørensen K, Van den Broucke S, Fullam J, Doyle G, Pelikan J, Slonska Z, Health Organization, 2008, Volume 86 (Number 12):914–915.
Brand H. Health literacy and public health: A systematic review and 27. National Institute of Population Research and Training (NIPORT), Mitra and
integration of definitions and models. BMC Public Health. 2012;12(1):1–13. Associates, ICF International: Bangladesh Demographic And Health Survey
4. World Health Organization, Regional Office for Europe: Health literacy. The 2014 Key Indicators April 2015.
solid facts. 2013. 28. Adhikary M, Haque R, Tanira S: Determinants of child Immunization under
5. Parker R: Health literacy: a challenge for American patients and their health expanded programme on Immunization (EPI) in a rural setting of
care providers. Health Promot Int. 2000;15(4):277–83. Bangladesh. J Dhaka Med Coll. 2013;22(2):201–06.
6. Nutbeam D. Health literacy as a public health goal: a challenge for 29. Angadi MM, Jose AP, Udgiri R, Masali KA, Sorganvi V. A Study of Knowledge,
contemporary health education and communication strategies into the 21st Attitude and Practices on Immunization of Children in Urban Slums of
century. Health Promot Int. 2000;15(3):259–67. Bijapur City, Karnataka. India J Clin Diagn Res. 2013;7(12):2803–6.
7. World Health Organization: Global status report on noncommunicable 30. Vyas Priyanka, Kim Dohyeong: Bangladesh Immunization Divide: Going
diseases 2010. Geneva: WHO; 2011. beyond rural, urban, and regional differences?. In 13th International
8. World Health Organization: Noncommunicable diseases country profiles Conference on GeoComputation: 2015; Richardson, Texas, USA; 2015.
2014. 2014. 31. Ahsan Karar Z, Alam N, Kim Streatfield P: Epidemiological transition in rural
9. National Institute of Population Research and Training (NIPORT), Mitra and Bangladesh, 1986–2006. Glob Health Action 2009, 2:10.3402/gha.v3402i3400.
Associates, ICF International: Bangladesh Demographic And Health Survey 1904.
2011. 2011. 32. Rahim M, Rahman M, Rahman M, Ahmed F, Chowdhury J, Islam F: The
10. NCD Scorecard Country Profile Bangladesh 2014. http://www. Prevalence rate of Hypertension in Rural Population of Bangladesh. J Dhaka
ncdglobalscorecard.org/wp-content/uploads/2015/09/Bangladesh-NCD- National Med Coll Hos. 2012; 18(1):12–17.
Scorecard-executive-report.pdf. Accessed 15 Jan 2016. 33. Ferguson MO, Long JA, Zhu J, Small DS, Lawson B, Glick HA, Schapira MM.
11. Ahmed SM, Alam BB, Anwar I, Begum T, Huque R, Khan JA, Nababan H, Low Health Literacy Predicts Misperceptions of Diabetes Control in Patients
Osman FA: Health Systems in Transition Bangladesh Health System Review. With Persistently Elevated A1C. Diabetes Educ. 2015;41(3):309–19.
Volume 5 No 3. Edited by Naheed A, Hort Krishna: Asia Pacific Observatory 34. Saquib N, Saquib J, Ahmed T, Khanam M, Cullen M. Cardiovascular diseases
on Public Health Systems and Policies. Manila: WHO Regional Office for the and Type 2 Diabetes in Bangladesh: A systematic review and meta-analysis
Western Pacific; 2015. of studies between 1995 and 2010. BMC Public Health. 2012;12(1):434.
12. Hanifi MA, Mamun AA, Paul A, Hasan SA, Hoque S, Sharmin S, Urni F, Khan 35. Ishikawa H, Yano E. Patient health literacy and participation in the health-
IR, Mahmood SS, Rasheed S, et al. Profile: The Chakaria health and care process. Health Expect. 2008;11(2):113–22.
demographic surveillance system. Int J Epidemiol. 2012;41(3):667–75. 36. Lee Y-J, Shin S-J, Wang R-H, Lin K-D, Lee Y-L, Wang Y-H: Pathways of
13. Hanifi S.M.A, Sultana A, Mia MN, Hoque S, Bhuiya A: Chakaria health and empowerment perceptions, health literacy, self-efficacy, and self-care
demographic surveillance system: focusing on the poor and vulnerable. behaviors to glycemic control in patients with type 2 diabetes mellitus.
Demographic events and safe motherhood practices- 2012. Dhaka: icddr,b; Patient Educ Couns. 2016;99(2):287–94.
2014. 37. Li X, Ning N, Hao Y, Sun H, Gao L, Jiao M, Wu Q, Quan H. Health Literacy in
14. Hanifi SMA, Mahmood SS, Bhuiya A: Cause-specific mortality and Rural Areas of China: Hypertension Knowledge Survey. Int J Environ Res
socioeconomic status in Chakaria, Bangladesh. Glob Health Action 2014, 7. Public Health. 2013;10:1125–38.
http://dx.doi.org/10.3402/gha.v7.25473. 38. Chowdhury AMR, Bhuiya A, Chowdhury ME, Rasheed S, Hussain Z, Chen LC:
15. Bangladesh Bureau Of Statistics (BBS), Statistics And Informatics Division The Bangladesh paradox: exceptional health achievement despite economic
(SID), Ministry Of Planning, Government Of The People's Republic Of poverty. Lancet. 2013;382(9906):1734–45.
Bangladesh: District Statistics 2011 Cox's Bazar. Dhaka: Bangladesh Bureau 39. Chowdhury AMR, Bhuiya A, Mahmud S, Abdus Salam AK, Karim F. Immunization
Of Statistics (BBS) December, 2013. divide: who do get vaccinated in Bangladesh? J Health Popul Nutr. 2003;21(3):
16. Bhuiya A, Sharmin T, Hanifi SMA. Nature of domestic violence against 193–204.
women in a rural area of Bangladesh: implication for preventive
interventions. J Health Popul Nutr. 2003;21(1):48–54.
17. Filmer D, Pritchett LH. Estimating wealth effects without expenditure
data—or tears: an application to educational enrollments in states of India*.
Demography. 2001;38(1):115–32.
18. Management Information System, Directorate General of Health Services,
Ministry of Health and Family Welfare, Government of the People’s Republic
of Bangladesh: Health Bulletin 2014. 2nd edition. Dhaka: Government of the
People’s Republic of Bangladesh, Ministry of Health and Family Welfare;
December 2014.