Background: Hypertension is the most common modifiable risk factor for cardiovascular diseases in South Asia. Machine learning (ML) models have been shown to outperform clinical risk predictions compared to statistical methods, but studies using ML to predict hypertension at the population level are lacking. This study used ML approaches in a dataset of three South Asian countries to predict hypertension and its associated factors and compared the model's performances. Methods: We conducted a retrospective study using ML analyses to detect hypertension using population-based surveys. We created a single dataset by harmonizing individual-level data from the most recent nationally representative Demographic and Health Survey in Bangladesh, Nepal, and India. The variables included blood pressure (BP), sociodemographic and economic factors, height, weight, hemoglobin, and random blood glucose. Hypertension was defined based on JNC-7 criteria. We applied six common ML-based classifiers: decision tree (DT), random forest (RF), gradient boosting machine (GBM), extreme gradient boosting (XGBoost), logistic regression (LR), and linear discriminant analysis (LDA) to predict hypertension and its risk factors. Results: Of the 8,18,603 participants, 82,748 (10.11%) had hypertension. ML models showed that significant factors for hypertension were age and BMI. Ever measured BP, education, taking medicine to lower BP, and doctor's perception of high BP was also significant but comparatively lower than age and BMI. XGBoost, GBM, LR, and LDA showed the highest accuracy score of 90%, RF and DT achieved 89 and 83%, respectively, to predict hypertension. DT achieved the precision value of 91%, and the rest performed with 90%. XGBoost, GBM, LR, and LDA achieved a recall value of 100%, RF scored 99%, and DT scored 90%. In F1-score, XGBoost, GBM, LR, and LDA scored 95%, while RF scored 94%, and DT scored 90%. All the algorithms performed with good and small log loss values <6%. Conclusion: ML models performed well to predict hypertension and its associated factors in South Asians. When employed on an open-source platform, these models are scalable to millions of people and might help individuals self-screen for hypertension at an early stage. Future studies incorporating biochemical markers are needed to improve the ML algorithms and evaluate them in real life.

Child malnutrition remains one of the major public health problems in many parts of the world, especially in a developing country like Bangladesh. Several socioeconomic and demographic factors are responsible for this condition. The present study was conducted to uncover the risk factors associated with malnutrition among under-five children in Bangladesh by analyzing the data from a nationally representative Bangladesh Demographic and Health Survey (BDHS) in 2014. The ordinal dependent variable—child nutrition status (severely malnourished, moderately malnourished, and nourished)—was developed by calculating weight-for-age Z score (WAZ). Bivariate analysis was conducted by performing gamma measure and chi-square test of independence to explore the association between child nutrition status and selected independent variables. To know the adjusted effects of covariates, a popular ordinal model—namely, the proportional odds (PO) model—was considered. All the selected covariates were found highly significant (p < 0.01) in the bivariate setup. However, in the multivariate setup, father’s and mother’s education, wealth index, mother’s body mass index (BMI), and antenatal care service during pregnancy were found highly significant (p < 0.01) factors for child malnutrition. Among the divisions, only Dhaka had more control on child malnutrition, compared to the Sylhet division. Birth interval of children was also reported as a significant factor at a 5% level of significance. Finally, the results of this paper strongly highlighted the necessity of increasing parent’s education level, improving the mother’s nutritional status, and increasing facilities providing antenatal care service in order to achieve better nutrition status among under-five children in Bangladesh.

This study intends to explore the prevalence of diabetes mellitus (DM) and its associated factors in Bangladesh. The necessary information was extracted from Bangladesh Demographic and Health Survey (BDHS) 2011. In bivariate analysis, Chi-square test was performed to assess the association between selected covariates and diabetes status. A two-level logistic regression model with a random intercept at each of the individual and regional level was considered to identify the risk factors of DM. A total of 7,535 individuals were included in this study. From the univariate analysis, the prevalence of DM was found to be 33.3% in 50–54 age group for instance. In bivariate setup, all the selected covariates except sex of the participants were found significant for DM (p < 0.05). According to the two-level logistic regression model, the chance of occurring DM increases as age of the participants’ increases. It was observed that female participants were more likely to have DM. The occurrence of DM was 62% higher for higher educated participants, 42% higher for the individuals who came from rich family and 63% higher for the individuals having hypertension. The chance of developing diabetes among overweighed people was almost double. However, the individuals engaged in physical work had less chance to have DM. This study calls for greater attention of government and other concerned entities to come up with appropriate policy interventions to lower the risk of DM.

Contraceptive use among married women of reproductive age has showed a substantial progress over the last few decades in Bangladesh. This study explores the sociodemographic factors associated with contraceptive use among ever-married women of reproductive age in Bangladesh by utilizing the information extracted from three of the Bangladesh Demographic and Health Surveys (BDHSs) in 1993–1994, 2004 and 2014. Bivariate analysis was conducted by performing chi-squared test of independence to explore the relationship between selected sociodemographic factors and dependent variables. To know the adjusted effects of covariates, a popular binary logistic regression model was considered. Respondents’ current age, place residence, division religion, education, age at first marriage, family planning (FP) media exposure, ideal number of children and fertility preferences are the significant determinants according to the most recent survey, BDHS 2014. However, wealth index and a respondent’s current working status were found to be significant factors only in BDHS 2004. The results of the study strongly recommend efforts to increase the education level among poor people, particularly among women in Bangladesh. Program interventions, including health behavior education and family planning services and counselling, are especially needed for some categories of the population, including those living in rural areas, Sylhet, Chittagong and Dhaka divisions, as well as illiterate and Muslim ever-married women.