Abstract
The frequency and correlates of dynapenic abdominal obesity (DAO) in ageing adults have not been assessed in a longitudinal study in sub-Saharan Africa. This study aimed to identify these characteristics and their prevalence using three-wave longitudinal data from South Africa. Data from three waves of the “Health and Ageing in Africa: A Longitudinal Study of an INDEPTH Community in South Africa (HAALSI)” (2014/2015–2022) were utilized. Community residents (≥ 40 years) with complete measurements of waist circumference and handgrip strength were included in the sample (analytic sample: n = 11073 observations). Dynapenia (= D) was defined as a handgrip strength of less than 26 kg for men and less than 16 kg for women, and abdominal obesity (= AO) a measured waist circumference (> 88 cm for women and > 102 cm for males). DAO was characterized by both AO and D using valid measures. The longitudinal relationships between independent factors and DAO, D, and AO were estimated using multinomial logistic fixed effects (FE) regressions (no D and no AO served as reference). The prevalence of DAO was 7.7% (95% CI: 7.3%-8.3%). FE logistic regressions showed that among men increases in age and among women decreases in age were associated with DAO. An increase of fruit and vegetable intake was associated with a decrease in DAO in the entire sample. In the total sample, the transition to having hypertension was positively associated with DAO in the total sample and among women. Furthermore, among women, transitioning to non-current tobacco use and increases in pain interference were associated with increases in DAO. Largely similar results were found with Cox regression and incident DAO, but an increase in age was in both men and women associated with higher hazard of incident DAO and among women depressive symptoms were associated with a higher hazard of incident DAO. The study found no significant association between transitioning to non-heavy alcohol use, diabetes, functional limitations, cardiovascular disease, living with HIV, kidney disease, impaired cognition, hospital admission, and DAO. Our knowledge of the factors that are associated with DAO is enhanced by this long-term investigation in persons 40 years of age and beyond. The work is an observational study produces some findings which are interesting but some which appear to be weak and/or contradictory. Therefore, the results are presented for information but are insufficient to draw strong conclusions at this time until future work is undertaken.