Examining the Prevalence of Hypertension by Urban–Rural Stratification

Abstract

Introduction: Nepal has one of the highest prevalences of hypertension in South Asia, which also causes other cardiovascular diseases. However, no studies investigated the prevalence and risk factors of hypertension by urban-rural stratification. Methods: We used a machine learning, Boruta algorithm to select risk factors and a tenfold random forest classifier to evaluate their performance. Finally, multivariate logistic regression estimated crude and adjusted odds ratios with 95% confidence intervals for knowledge generation. Results: The study included 7825 participants (urban: 4939; rural: 2886), where rural participants were slightly older (median: 37 years; interquartile range: 26–53) and females were more hypertensive (urban: n = 606, 34.5%; rural: n = 308, 31.2%). The prevalence of hypertension was 35.6% in urban and 34.1% in rural regions. The odds of hypertension increased in rural regions for advancing age, provinces (province 4 and 5), and ecological zones (hill and terai). Overweight and obese participants were more likely to have hypertension in both regions. Conclusion: The study recommends the rigorous improvement of public health programs in rural regions of province 4 and 5, concentrating on Dalit and Janajati older males from hill and terai ecological zones. Overweight and obese people from both regions also need special focus. Finally, policymakers and government officials have to tailor campaigns differently for robust implementation of the essential health-care package and multisectoral action plans to prevent and control hypertension.

Detecting Need-Attention Patients using Machine Learning

“…By using Boruta wrapper feature selection algorithms with CART decision tree, [19] have improved the classification accuracy of medical datasets effectively. Moreover, Boruta has been used to identify the significant features of medical datasets [10], [22], [23], [25]. The researchers recommended Boruta to select relevant variables in high-dimensional datasets [37], [38].…”

Uncovering Undiagnosed Hypertension Rates in Bangladeshi Adults: Insights from Advanced Statistical Models

“…Socioeconomic, demographic, and individual-level risk factors were selected based on literature from low-and middle-income countries (LMICs) [6,14,[19][20][21]. Variables included administrative division (Barishal, Chattogram, Dhaka, Khulna, Mymensingh, Rajshahi, Rangpur, Sylhet); place of residence (urban, rural); socioeconomic status (poorest, poorer, middle, richer, richest); age (in years); education level (no formal education, primary, secondary, higher); occupational status (employed; unemployed).…”