Assessing neurocognitive functioning among adults ageing with and without HIV at the Kenyan Coast: measurement issues and correlates

Background: Cognitive impairment is one of the most prevalent complications of HIV infection, with significant medical and functional impacts. However, valid, and reliable assessment tools are lacking for the newly emergent ageing population of people living with HIV (PLWH) in many parts of sub-Saharan Africa (SSA), including Kenya. Without these tools, critical intervention opportunities, e.g., psychoeducation, treatment and additional support, are missed. To bridge this gap in Kenya, we adapted the Oxford Cognitive Screen Plus (OCSPlus), a tablet-based cognitive assessment tool designed for low-literacy settings, with adults ageing with HIV ≥ 50 years and their uninfected peers. This study examines the acceptability, reliability, and validity of the OCSPlus tool among older Kenyan adults and provides an initial understanding of the cognitive performance of these adults (by HIV status) and the biopsychosocial factors associated with their cognitive performance. Methods: In a cross-sectional sample of 440 older adults (257 living with HIV), we administered the OCSPlus tool alongside the Raven’s Standard Progressive Matrices (RSPM), the International HIV Dementia Scale (IHDS), and health and sociodemographic assessments. Results: There was a high level of acceptability of OCSPlus by participants and test administrators. OCSPlus demonstrated good test–retest reliability. Acceptable correlations between individual OCSPlus sub-tasks and conventional tests (RSPM and IHDS) were also observed for convergent validity. Regarding cognitive performance, older adults living with HIV (OALWH) presented with significantly lower scores on language (picture naming task), executive function, and the IHDS overall score compared to their uninfected peers. However, OALWH performed significantly better on memory domain (orientation, word encoding and word recall tasks), non-verbal intelligence and processing speed. There were no differences in attention domain. Cognitive performance as assessed by OCSPlus was significantly associated with behavioural and lifestyle factors (physical activity, sleeping difficulties, obesity, and sexual activity), sociodemographic factors (age, sex, educational status, household income, household size and asset index), medical or treatment factors (self-reported urinary incontinence, hearing problems, history of TB, seeking services of traditional healers, antiretroviral therapy (ART) regimen change, being on 3rd line ART treatment) and psychosocial factors including ageism and food insecurity. Conclusion: We demonstrated the feasibility of OCSPlus administration by trained lay persons, its acceptability, and preliminary reliability and validity among low-literacy older adults on the Kenyan coast. Mean cognitive scores were mixed across the two groups. Cognitive performance was associated with several biopsychosocial factors spanning behavioural/lifestyle, sociodemographic, psychosocial, medical and treatment factors. Further validation studies and epidemiological research are needed to understand better the utility of OCSPlus and the cognitive function of these adults.