This study evaluated whether access to household cash transfers (HCT) is associated with the cumulative incidence of prediabetes and diabetes in a rural South African population aged 40 years and older. The authors framed HCT as a potential social intervention that could reduce diabetes risk through pathways such as lower chronic stress, improved food security, greater physical activity, and increased preventive care. The analysis addresses an evidence gap on whether household cash transfer access predicts new-onset glycemic impairment in this setting.
Analyses used the Health and Ageing in Africa: Longitudinal Studies in South Africa (HAALSA) Indepth cohort. The analytic sample comprised Black South African participants aged 40+ with N = 5059. The evaluation covered cumulative incidence between baseline in 2014/15 and follow-up in 2021/22. Public use HAALSA Waves 1–3 data and Agincourt Health and socio-Demographic Surveillance System data are cited as sources and available through the repositories noted in the original report.
The exposure of interest was household cash transfer eligibility, operationalized as units of HCT available to the household. The source reports analyses of the relationship between incremental HCT exposure and later glycemic outcomes. Specific operational details of HCT units and eligibility criteria were described in the original study but are not further detailed in the abstract.
Primary outcomes were cumulative incidence of diabetes and prediabetes assessed over the follow-up period. In addition to categorical outcomes, the authors examined continuous glucose values and changes across the glucose distribution to identify whether associations differed by baseline glycemic level.
To estimate associations between HCT and categorical outcomes, the investigators fitted log binomial models to derive cumulative incidence ratios. For continuous glucose outcomes, quantile regression was applied to estimate how additional HCT related to glucose changes at different points of the glucose distribution. The combination of models allowed evaluation of average relative risk shifts and heterogeneous effects across glucose levels.
No association between HCT and incident diabetes was observed in the reported models.
For prediabetes, each additional unit of household cash transfer eligibility was associated with a reduced risk of incident prediabetes. The adjusted cumulative incidence ratio reported was aCIR 0.94 (95% CI 0.90–0.98) with p = 0.007, indicating a statistically significant association in favor of lower prediabetes incidence with greater HCT access.
Quantile regression results indicated that the largest reductions in continuous glucose values associated with additional HCT occurred at the highest end of the glucose distribution. This pattern suggests that participants with the most severe glucose impairment experienced the greatest absolute glucose decreases related to HCT exposure.
The reported findings suggest that household cash transfer access in this rural South African cohort was associated with lower risk of developing prediabetes but did not show a measurable effect on progression to diabetes within the study period. The observation that glucose reductions were largest among those with the highest baseline glucose implies HCT could have particular benefit for individuals with more severe glycemic impairment, potentially reducing risk of hyperglycemic complications.
These results are consistent with hypothesized mechanisms by which economic supports can influence health determinants—reducing stress and improving food security and access to preventive services—but the abstract does not provide mediation analyses or detailed mechanistic data. The absence of an observed diabetes association could reflect multiple possibilities, including duration of follow-up, measurement or power considerations, or differential effects across disease stages; the source does not provide additional detail in the abstract about these factors.
The HAALSA Indepth study received ethical approvals from the Mpumalanga Provincial Research and Ethics Committee, the University of the Witwatersrand Human Research Ethics Committee (ref. M141159), and the Harvard T.H. Chan School of Public Health Office of Human Research Administration (ref. C13-1608-02). The Indiana University Institutional Review Board deemed the present analysis not human subjects research (ref. 28496). The authors declared no competing interests and confirm adherence to reporting and consent procedures. Public use data for Waves 1–3 are available through the repositories referenced in the source. Funding included support from the National Institute on Aging (R01AG069128) and the National Center for Advancing Translational Sciences (TR004389).