Source and study population
Bangladesh is a tropical country located in the north-eastern part of South Asia. According to the Population and Housing Census 2011 (PHC), Bangladesh is the most densely populated country in the world (among countries larger than 10,000 square kilometres), and the population of the country is estimated to be about 157.9 million (9). We used data from the BDHS, a nationally representative cross-sectional survey conducted during June – November 2014. The survey was conducted using a two-stage stratified sampling approach. In the first stage, 600 enumeration areas (207 urban vs 393 rural) were selected using probabilities proportional to their size. In the second stage, an average of 30 households was selected from each enumeration area using a systematic random sampling technique. Thus 17,989 residential households were chosen from all divisions in the country. Then, a total of 17,300 households were interviewed, and 18,245 ever-married women aged 15-49 years were identified as eligible participants. Finally, a total of 17,863 ever-married women were interviewed at a response rate of 98% (10). Figure.1 shows the study design and data screening of BDHS 2014.
Study variables and definitions
The outcome variable was the occurrence or non-occurrence of diarrhea among children under five years old, as reported by the child's mother. Diarrhoea was defined as the passage of three or more loose stools in a day in the preceding two weeks during the conduct of the survey. No data on clinical examination were collected.
To identify the risk of diarrhea, we used data at the individual and community levels according to the hierarchical nature of the BDHS data. Individual variables were related to information of the child, their mother (respondent), and the household, while the community variables were related to divisions and place of residence of the respondents. Variables were selected for analysis based on existing literature and contemporary evidence from the study area (6,8,9,11–21,28,31,34–37).
There are almost 700 non-government microcredit organizations in Bangladesh. Microcredit organizations provide small loans to marginalized women in Bangladesh to support entrepreneurship and alleviate poverty (38). In this study, we considered five major microcredit organizations (Grameen Bank, ASA, PROSHIKA, BRAC and BRDB) to assess associations between the memberships of microcredit organizations and the incidence of childhood diarrhoea. Mothers were grouped as those who were and were not a member of any micro-credit organization. Through developing awareness, mothers’ access to mass media such as newspapers, radio, and television may help to reduce their children’s illness (31). The original categorization for three variables are set as not at all (0), less than once a week (1), at least once a week (2) and almost every day (3) in DHS study. Therefore mothers were grouped into two levels not having access to any media and having access to at least any one of this media. The DHS data also contains a wealth index score as a proxy for household wealth status, which was generated using principal component analysis, combining variables related to household assets (televisions and bicycles), dwelling characteristics (source of drinking water and sanitation facilities) and construction materials (roof, ceiling, and floor). The details of the calculation of the wealth score are available in the BDHS 2014 (10). Children were classified according to their household wealth levels as: lower (40%), middle (40%), and higher (20%).
The Joint Monitoring Programme (JMP) of WHO/UNICEF categorized the source of drinking water as improved (piped water, tube well or borehole, protected well, protected spring, rainwater and bottled water) and unimproved (unprotected well, unprotected spring, river/irrigation channel, tanker truck, cart with small tank) (39). Similarly, the source of sanitation facilities was classified as improved (flush toilet—flush to the piped sewer system, flush to a septic tank, flush to pit latrine, flush to somewhere else—pit toilet latrine, ventilated improved pit latrine, pit latrine with slab) and unimproved (no facility: bush/field, composting toilet, bucket toilet, pit latrine without a slab, hanging toilet/latrine, and other).The number of adults per living room was used to define household crowding. Households were classified as those with up to two adults (non-crowded or standard) or more than two adults per living room (over-crowded). Household floor material was categorized as “made of mud”, or other. Place of residence was defined as urban and rural areas. Bangladesh is demarcated by the seven administrative divisions or regions where a child is from. The seven divisions were recorded as Barisal, Khulna, Rangpur, Sylhet, Rajshahi, Dhaka, Chittagong.
Respondents and their children’s socio-economic and demographic characteristics were descriptively analyzed. Pearson’s chi-squared tests were used to find the association between diarrheal disease among under-five children and their parents’ socio-economic, demographic and community variables. We then perfromed a multilevel logistic regression analysis to identify the associations of these predictors with diarrheal disease according to the hierarchical nature of the BDHS data (40). Three multilevel models were developed. In model-1, division was considered as the second level; in Model-2, place of residence was considered as the second level; and in Model-3, both division and the place of residence was considered as the third level source of variation.We used Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) to select the best fitting model. All statistical analyses were performed using SPSS 25.0, and the lme4 package of thatistical software, R. In all these analyses, p-value<0.05 was considered statistically significant.
Ethical approval was obtained from the Human Research Ethics Committee (HREC) of the Australian National University (Approval No: 2019/893). The study was also approved by the National Research Ethics Committee (NREC) of the Bangladesh Ministry of Health and Family Welfare.