Prevalence, patterns, and determinants of multimorbidity in South Africa: Insights from a nationally representative survey.

Background: Multimorbidity in Sub-Saharan Africa is under researched and includes distinct disease combinations to those seen in high income countries. The aim of this study was to determine the prevalence and distribution of multimorbidity in South Africa, as well as the associated individual, area-level and contextual factors.

Methodology: Multilevel logistic regression analyses were conducted on nationally representative 2016 South Africa Demographic Health Survey Data. The sample included 5,592 individuals (level 1) living in 691 neighbourhoods (level 2).

Principal Findings: Multimorbidity was present in 45.3% of the study population, ranging from 35.6% in Limpopo to 52.1% in Eastern Cape. Hypertension was the most prevalent condition (46.4%) followed by diabetes (22.6%). Individuals aged 65-95 had 11.57 times higher odds (95% CI 8.50-15.74) of multimorbidity compared to those aged 15-24. Women had nearly twice the odds of men (OR 1.95, 95% CI 1.68-1.27). Formerly married individuals had 1.63 times higher odds (95% CI 1.32-2.02) than never married. Compared to Black Africans, White individuals had 44% lower odds (OR 0.56, 95% CI 0.39-0.82) and those of mixed ethnicity had 31% lower odds (OR 0.69, 95% CI 0.51-0.92). Obesity increased the odds by 38% (OR 1.38, 95% CI 1.17-1.64) and occupational smoke exposure by 26% (OR 1.26, 95% CI 1.07-1.49). There was variation in multimorbidity at the neighbourhood level, with 2.9% of the variation attributable to contextual factors in the empty model. The median odds ratio was 1.35, indicating substantially higher odds of multimorbidity if an individual moved to a higher risk neighbourhood.

Conclusions: This study found a high burden of multimorbidity in South Africa patterned by demographic, socioeconomic, lifestyle and contextual factors. The results highlight the need for multilevel strategies to reduce multimorbidity and its inequities by addressing individual risk factors as well as neighbourhood-level determinants of health.