The relationships between cognitive function, literacy and HIV status knowledge among older adults in rural South Africa

INTRODUCTION: Although HIV prevalence is exceptionally high in South Africa, HIV testing rates remain below targeted guidelines. Older adults living with HIV are substantially more likely to remain undiagnosed than younger people. Cognitive function and literacy could play key roles in HIV status knowledge due to the decision‐making processes required around weighing the costs and benefits of testing, navigating testing logistics and processing results. We aimed to assess the independent relationships among each of cognitive function, literacy and education with HIV status knowledge in a population‐based sample of older adults living in a rural South African community with high HIV prevalence. METHODS: We analyzed data from a population‐based study of 5059 men and women aged 40 years and older in rural South Africa (Health and Aging in Africa: A Longitudinal Study of an INDEPTH community (HAALSI)). HAALSI surveys, conducted between 2014 and 2015, queried self‐reported literacy, educational attainment and HIV status knowledge. Laboratory tests were conducted to assess true HIV sero‐status. Cognitive function was assessed with a battery of cognitive tests measuring time orientation, immediate and delayed recall, and numeracy and coded using confirmatory factor analysis as a z‐standardized latent variable. We estimated the relationship between the outcome of HIV status knowledge and each of three exposures: (1) latent cognitive z‐score, (2) literacy and (3) education, using confounder‐adjusted modified Poisson regression models in the study population overall and stratified by HIV sero‐status. RESULTS: We found that HIV status knowledge was higher among those with higher cognitive z‐scores (adjusted Prevalence Ratio (aPR) (95% CI): 1.18 (1.14, 1.21) per standard deviation unit), and among literate participants (aPR (95% CI): 1.24 (1.16, 1.32) vs. non‐literate participants). Taken together, the associations with literacy and cognitive function completely attenuated the otherwise positive association between educational attainment and HIV status knowledge. The magnitudes of effect were generally similar among laboratory‐confirmed HIV‐negative and HIV‐positive participants. CONCLUSIONS: Campaigns that target older adults in rural South Africa with HIV testing messages should carefully consider the cognitive and literacy levels of the intended audience. Innovations to ease the cognitive load associated with HIV testing could prove fruitful to increase HIV status knowledge.