Sociodemographic and behavioral characteristics associated with self-reported diagnosed diabetes mellitus in adults aged 50+ years in Ghana and South Africa : results from the WHO-SAGE wave 1

Objective: The objective is to identify and describe thesociodemographic and behavioral characteristics of adults,aged 50 years and over, who self-reported having beendiagnosed and treated for diabetes mellitus (DM) in Ghanaand South Africa. Research design and methods: This is a cross-sectionalstudy based on the WHO Study on global AGEing and adulthealth (SAGE) wave 1. Information on sociodemographicfactors, health states, risk factors and chronic conditionsis captured from questionnaires administered in face-tofaceinterviews. Self-reported diagnosed and treated DM isconfirmed through a ‘yes’ response to questions regarding1having previously been diagnosed with DM, and2 havingtaken insulin or other blood sugar lowering medicines.Crude and adjusted logistic regressions test associationsbetween candidate variables and DM status. Analysesinclude survey sampling weights. The variance inflationfactor statistic tested for multicollinearity. Results: In this nationally representative sample ofadults aged 50 years and over in Ghana, after adjustingfor the effects of sex, residence, work status, bodymass index, waist-hip and waist-height ratios, smoking,alcohol, fruit and vegetable intake and householdwealth, WHO-SAGE survey respondents who were older,married, had higher education, very high-risk waistcircumference measurements and did not undertakehigh physical activity, were significantly more likelyto report diagnosed and treated DM. In South Africa,respondents who were older, lived in urban areas andhad high-risk waist circumference measurements weresignificantly more likely to report diagnosed andtreated DM. Conclusions: Countries in sub-Saharan Africa arechallenged by unprecedented ageing populations andtransition from communicable to non-communicablediseases such as DM. Information on those who arealready diagnosed and treated needs to be combinedwith estimates of those who are prediabetic or, as yet,undiagnosed. Multisectoral approaches that includesocioculturally appropriate strategies are needed toaddress diverse populations in SSA countries