Socioeconomic Determinants of Cardiometabolic Diseases in Kenya: A Machine Learning Approach

Cardiometabolic diseases (CMDs) are major contributors to global morbidity and mortality, with significant disparities driven by socioeconomic factors. This study uses data collected in 2019 in Kenya by the AWI-Gen project, a collaborative initiative between the University of the Witwatersrand (Wits) and the INDEPTH Network. The work in Kenya was conducted within the Nairobi Urban Health and Demographic Surveillance System (NUHDSS), run by the African Population and Health Research Center (APHRC). The main objective is to apply machine learning methods to identify socioeconomic factors that influence rising CMD rates in Kenya, thereby guiding targeted interventions to mitigate CMDs. We performed translational research by leveraging the K-Prototypes clustering algorithm, a Random Forest classification model, and the SHAPley Additive Explanations (SHAP) technique to investigate and identify socioeconomic status factors influencing CMD risk. The K-Prototypes clustering model identified three groups: 31\% of participants in the high-vegetable intake group, 34\% in the overextended labour group, and 37\% in the economically disadvantaged group had CMDs. The economically disadvantaged group, which had the highest prevalence of CMDs, also exhibited lower average years of schooling and a higher proportion of individuals with no formal education. Unemployment was notably high in this group (62\%), with only 16\% being self-employed, and asset ownership was significantly lower. Consuming vegetables and fruits more than three times a week reduced CMD risk, while education level, income, employment status, overall socioeconomic status, and the number of hours worked per week contributed to CMD risk in varying degrees. This study highlights that targeted socioeconomic interventions, including improved education and job creation, can mitigate CMD risk, especially among disadvantaged groups. Access to affordable, healthy food should be increased through subsidies, community gardening, and public-private partnerships that expand urban food markets. Healthcare coverage must be enhanced and out-of-pocket costs minimized, with government-funded mobile clinics reaching underserved populations. Public health campaigns can boost awareness of early detection and regular check-ups, further reducing CMD prevalence.