Mistry et al.

BMC Public Health (2022) 22:2063 Page 2 of 12

population in LMICs [3–6]. Like other LMICs, the preva- from rural areas, and 30 EAs from urban slums were
lence of hypertension is high in Bangladesh. According randomly selected to provide statistically reliable esti-
to the most recent data [7], the prevalence of hyperten- mates of the key health indicators for the rural areas
sion was 25.2% and 19.8% respectively in urban and rural and urban slums separately, where an EA is typically a
areas of Bangladesh among the adults aged 18–69 years. ward in slum areas, which is the lowest administrative
Previous studies conducted in LMICs including Bang- unit of urban areas in Bangladesh. A complete list of
ladesh [4, 5, 8–11] identified several risk factors asso- unions (rural) and wards (urban slum) from the most
ciated with hypertension. Some of these risk factors recent Population and Housing Census of Bangladesh
included higher age, overweight and/or obesity, seden- were collected and used as a sampling frame for the
tary lifestyle, and pre-existing co-morbidities. first stage of sampling. In the second stage of sampling,
UN-Habitat defined slum as “a heavily populated urban the EAs per administrative zone were selected follow-
area characterised by substandard housing and squalor” ing the probability proportional to size technique with
[12]. Globally, around one billion people live in slum set- a systematic random sample of five households. Start-
tlements [13], of which around 130 million lives in South ing from the north-west corner of an EA with a system-
Asian countries [13]. Like many other South Asian coun- atic random sample of five households, we selected on
tries, most of the urban growth has been taken place in an average of 54 households per EA [24]. In the present
urban slums of Bangladesh [14]. It has been estimated study, we considered 1197 adults aged 35 years and
that around 29 million people, which is around 55% of above residing in the slum households. In these slums
the total urban population in Bangladesh is currently BRAC operates its health and nutrition intervention
residing in urban slums [15]. where BRAC’s frontline community health workers
Most of the urban slum dwellers in Bangladesh exhibit deliver behavioral change communication messages,
poor living conditions and are deprived of basic health particularly on maternal and child health care, essential
care services, making them vulnerable to a range of ill- health services and nutrition among the slum dwellers
nesses including maternal and child health related through door-to-door visits [25]. Inclusion criteria for
problems as well as chronic conditions such as hyper- the present study were age being 35 years or above and
tension [15–19]. Poverty contributes to lack of access to residing in slum settlements for more than six months.
basic health care services among these population with A total of 42 participants were also excluded due to
concomitant increases in the risk of hypertension and missing and/or inconsistent information; 1155 com-
uncontrolled blood pressure. This might be due to lack of pleted the questionnaire accurately and included in the
blood pressure screening, poor management of hyperten- final analysis (Fig. 1).
sion, and absence of health education [7]. Previous stud-
ies carried out in the slum settlements of LMICs [20–23]
reported a high prevalence of hypertension among slum Data collection tools and techniques
dwellers. For example, Banerjee et al. (2016) found that Household level socio-demographic information were
the prevalence of hypertension was 42% among the adult collected through face-to-face interviews with each
population in slums of Kolkata, India [20]. Likewise, the respondent using a structured questionnaire. The ques-
prevalence of hypertension was 21% among urban slum tionnaire was pre-tested in Gazipur, a suburb of Dhaka,
dwellers in Brazil [22]. and the feedback was incorporated into the final version
However, there is scarcity of evidence on the preva- of the questionnaire. A total of 110 skilled interviewers,
lence and risk factors for hypertension among the adult having prior experience of conducting large-scale health-
population from slum areas of Bangladesh. Therefore, care surveys were recruited and trained for data collec-
the current study aimed to investigate the prevalence and tion. Data collection was carried out electronically in
associated factors of hypertension among the adult slum ODK (Open Data Kit), an Android-based open-source
dwellers (aged 35 years and above) in Bangladesh. mobile platform software [26]. The ODK can be used to
collect data both online and offline and can be used by
Subjects and methods the people with limited educational qualifications. Its
Study setting and participants unique features such as GPS tracker and short message
Data used in this study were collected as part of a large- service (SMS) enable the real-time data collection moni-
scale cross-sectional survey conducted between Octo- toring [27]. A multi-layered monitoring system compris-
ber 2015 and January 2016 in Bangladesh. The main ing of field supervisors as the first layer, followed by the
survey was conducted by Building Resources Across researchers, and finally by statisticians positioned at the
Communities (BRAC) in rural areas and urban slums head office, was employed to validate, standardize and
of Bangladesh. Briefly, 180 enumerations areas (EAs) maintain data quality.