Midlife cognitive testing in Africa: validity of the Harmonised Cognitive Assessment Protocol in the Kenya Life Panel Survey

Objectives Cohort studies of ageing and cognitive decline typically do not begin fielding comprehensive cognitive assessments until older adulthood. However, for identifying preventable dementia risk factors, there is strong value in beginning at earlier ages. The case is especially compelling in sub-Saharan Africa, where the number of older individuals is expected to triple in the next three decades, and where risk factors may operate more intensively at earlier ages. This study reports on the adaptation and validity of the Harmonised Cognitive Assessment Protocol (HCAP) approach in the Kenya Life Panel Survey (KLPS), collected among middle-aged respondents. Design To evaluate the validity of the HCAP approach in Kenya, this study assesses model fit statistics from confirmatory factor analyses (CFA) and tests measurement invariance by respondent characteristics. Setting Both rural and urban areas in Kenya. Participants A sample of n=5878 individuals from the KLPS, who have been surveyed regularly since they were schoolchildren in the 1990s. The HCAP assessment was administered in 2023 at an average age of 37 years (10-90 range 34 to 41). Primary and secondary outcome measures For each individual, the CFA generates a general cognitive performance score, and cognitive performance scores for five distinct domains, including memory, executive functioning, language, orientation to time and place, and visuospatial functioning. Results Fit of the models to the data was adequate for general cognitive performance (root mean squared error of approximation (RMSEA)=0.03; comparative fit index (CFI)=0.94; standardised root mean residual (SRMR)=0.05), language (RMSEA=0.02; CFI=0.95; SRMR=0.05) and good for memory (RMSEA=0.05; CFI=0.99; SRMR=0.02) and executive functioning (RMSEA=0.03; CFI=0.98; SRMR=0.03). The CFA indicate that the factor structure is consistent with findings from other countries and that reliability for the general cognitive performance score was high. Statistical models also suggest invariance at the scalar level for leading demographic (gender, age) and socioeconomic (education, occupational complexity) characteristics. Conclusions This study demonstrates that the cognitive functioning of mid-age Kenyans appears to be well captured by the adapted protocol. While there is a moderate decline in cognitive performance among older individuals, this relationship appears to be mediated by education, indicating that this KLPS HCAP provides a valuable baseline for studying future cognitive decline.