Identifying Factors Associated with Hypertension in Slum-dwelling Women Using Structural Equation Modeling (SEM)

Abstract

Globally, an estimated 1.56 billion people will have hypertension by 2025, with about two thirds of cases in low- and middle-income countries. India accounts for 17.6% of the worldwide burden of hypertension with 30.7% of Indians having a systolic blood pressure ≥140/90 mmHg. Between October 2017 and May 2018, 607 women in urban slums in Mysore, India, were recruited into a cross-sectional assessment of coronary heart disease (CHD) and associated risk factors. This is a secondary data analysis of the same data. To test for mediation and moderation of these risk factors, SEM and cascade modelling were used. Covariates of interest included demographic and behavioral risk factors. The overall prevalence of hypertension (based on cut-offs of systolic blood pressure over 140 mm Hg or diastolic blood pressure over 90 mm Hg) was 57.5% (95% CI: 0.535, 0.614). The mean SBP and DBP were 137.7 mm Hg (± SD = 21.4) and 91.0 mm Hg (± SD = 10.8), respectively. Demographic factors (p-value < 0.001) and exercise frequency (p = 0.003) were significantly associated with the modifiable risk factors. The coefficient for the path between the modifiable risk factors and SBP was -3.88, while that for path between demographic factors and DBP was 0.03. Education and the demographic risk factors index were positively associated with DBP, and the modifiable risk factors index was negatively associated with SBP. Variables whose effects reversed in direction after inclusion of potential mediators represent an interesting finding that deserve further study as potential effect modifiers.