Health Information Seeking Behaviour Study in Bangladesh: Household Survey and Qualitative Study, 2012-2016

To explore the role of ICTs in health information seeking, the UK Institute of Development Studies (IDS) and the International Center for Diarrhoeal Disease Research, Bangladesh (ICDDR,B) jointly conducted a household survey during March to May 2014 in three locations of Bangladesh; Chakaria (a rural sub-district), Mirzapur (a peri-urban sub-district) and Dhaka (five largest slums of the capital city). In the absence of prior variance estimates of the outcome variables, a value of 0.5 (the maximum for dichotomous variables) was used to calculate the required sample size to obtain 95% confidence limits with a precision of ±10%, assuming a design effect of 2. This implied a sample size of 840 households for each location, allowing a 5% buffer for the probable non-response rate.
In the rural and peri-urban sites (Chakaria and Mirzapur), the sample was selected using systematic cluster sampling from pre-existing frames (Health and Demographic Surveillance Sites). For each location, the 840 households were selected from 28 villages (30 households/village). In Dhaka, three slums were randomly selected from the six largest in the city. Using household lists, 10 locations were selected in each slum and 28 households sampled from each location using systematic sampling.
In the majority of cases (over 81%) information was gathered from the head of the household or the spouse of the head. Where this person was not present or unwilling to respond, the respondent was usually an adult child of the head or the spouse of a child. The sampling at household level was designed to produce a sample which obtained data from twice as many women as men.
The survey of college students aimed to provide further information on health information seeking and possible behavioural change among college students in Chakaria and Mirzapur. In the assumed absence of clustering effects, the target sample size for each site was 420 and the survey was designed to select approximately equal numbers of men and women in each study site. It was undertaken in August 2015 and in practice provided data on 421 students attending Chakaria Degree College, the largest government institution, and the private Chakaria Abashik Mohila College, which is an important residential women’s college. In Mirzapur, 436 students were interviewed in two government institutions, Mirzapur Degree College and Mirzapur Government Saadat College, chosen as being the most prominent colleges, with high student numbers. All the colleges offer education from Higher Secondary School Certificate up to Masters-level degrees, and are easily accessible by road. Trained enumerators spent time at the colleges, recruiting survey respondents using a combination of non-probability sampling approaches based on convenience and snowballing.
Qualitative methods were used to probe more deeply into the attitudes of people to different sources of health information and explore the process of diffusion of new types of health-information seeking behaviour. The first stage of qualitative work provided insights that both informed the final shape of the quantitative survey and provided qualitative data. Thereafter special studies examined in greater depth special areas of interest, such as individuals and groups with specific conditions or exemplifying emergent behaviours.
All the research undertaken in this project received ethical approval from the icddr,b Research Review Committee (RRC). This included the provision of detailed information about the nature of the research and participants' voluntarily consent (either written or recorded verbally).
Qualitative research was undertaken in in three districts – Chakaria, Mirzapur and Dhaka – to feed into the quantitative survey and as a source of information in its right. Chakaria is a relatively remote district, where access to government health services is poor and households rely largely on informal providers; Mirzapur, is a mixed urban/rural setting with relatively easy road access to Dhaka; and a low income area in Dhaka, where households rely primarily on local markets for services, including health.
All qualitative research was conducted in Bangla by research officers with Master's degrees in anthropology and extensive experience in conducting. These were recorded, transcribed, and then translated into English. Peer debriefing was conducted within the study team to help understand the issues and consolidate the findings. The data obtained were coded for emerging themes, which were triangulated using information collected through the different qualitative exercises.

The Data Collection is available for download to users registered with the UK Data Service.