Sub-Saharan Africa is experiencing accelerated growth in Alzheimer’s disease and related dementias (ADRDs), including among older people living with HIV (PLWH). Diagnostic options that are contextually appropriate for diverse cultures, languages, and variable education and literacy levels are limited in the region. The Tablet-based Cognitive Assessment Tool (TabCAT) is a digital platform providing brief cognitive tests with automated scoring designed to be administered by non-specialists and used across diverse settings. The Uganda Aging and Dementia Cohort Study (UADCS), a prospective cohort including older PLWH and age- and sex-similar adults not living with HIV in southwestern Uganda, provided an opportunity to translate, adapt, and evaluate a brief TabCAT battery for this context.
Four TabCAT tests were selected for translation and cultural adaptation and then administered alongside a previously validated reference-standard cognitive testing battery used in Uganda. The study assessed psychometric properties including floor and ceiling effects, associations between test scores and known demographic predictors (age, education), and concurrent and divergent validity against the reference-standard battery. Objective cognitive impairment was defined using Jak/Bondi criteria applied to the reference-standard assessment. A regression-based normative approach produced Z-scores for TabCAT tests adjusted for age, sex, education, urbanicity, and literacy. Receiver Operating Characteristic (ROC) curves were fit to evaluate the TabCAT Composite Score's ability to discriminate objective cognitive impairment in the total sample and stratified by HIV serostatus.
The four TabCAT tests underwent translation and cultural adaptation through expert review and focus groups composed of adults at the study sites. Local experts and focus group participants reported the adapted tests had acceptable face and content validity for the Ugandan context. The adaptation process aimed to ensure linguistic and cultural relevance and to support administration by non-specialists across varying education and literacy levels.
Psychometric validity was examined by checking for floor and ceiling effects and confirming expected associations between cognitive scores and demographic predictors such as age and education. Construct validity was assessed by comparing correlations between TabCAT tests and domain-matched reference-standard tests (for example, memory and executive function) and by examining weaker correlations with non-aligning domains (for example, motor measures). A regression-based normative model adjusted raw scores to Z-scores accounting for age, sex, education, urbanicity, and literacy. Diagnostic performance of the TabCAT Composite Score for identifying objective cognitive impairment (Jak/Bondi criteria) was quantified using ROC analysis to produce c-statistics with 95% confidence intervals.
A total of 563 participants completed both the reference-standard and TabCAT batteries. Mean age was 60 years (standard deviation 6.4); 50% were female. Educational attainment was limited for many participants: 51% had not completed primary school, and 16% were not literate. Approximately 49% of participants were people living with HIV.
Local reviews and focus groups supported the face and content validity of the adapted TabCAT tests. No notable floor or ceiling effects were observed for TabCAT measures. Test scores were associated with age and education in expected directions, supporting construct validity. Correlations between TabCAT tests and the reference-standard battery were stronger when cognitive domains aligned (memory and executive function) than when domains did not align (motor), supporting concurrent and divergent validity.
Using Jak/Bondi criteria applied to the reference-standard battery to define objective cognitive impairment, the TabCAT Composite Score discriminated impairment with good accuracy. The overall c-statistic was 0.77 (95% CI 0.72–0.81). This discriminatory performance was retained when analyses were restricted to PLWH, indicating potential applicability among older adults living with HIV.
The adapted TabCAT battery demonstrated multiple forms of validity — face, content, construct, concurrent, and criterion — for measuring cognition and detecting objective cognitive impairment among older adults in Uganda, including among PLWH. The tests showed no major floor or ceiling effects and behaved as expected with respect to demographic predictors such as age and education. Correlations with domain-aligned reference tests strengthen the case for construct validity, and the TabCAT Composite Score achieved good discrimination for objective cognitive impairment.
These results support the potential of brief, tablet-based cognitive assessment tools to measure cognitive performance and detect impairment in Uganda and in populations with lower levels of education and literacy. The automated scoring and design for administration by non-specialists may facilitate wider screening and research use in similar low-resource settings.
Ethical approval was obtained from the Massachusetts General Hospital Institutional Review Board, the Mbarara University of Science and Technology Research Ethics Committee (MUSTREC), and the Uganda National Council for Science and Technology (UNCST). The authors report that all necessary participant consent and oversight were obtained. Data produced in the study are available upon reasonable request to the corresponding author. Funders included the US National Institutes of Health National Institute on Aging, the Alzheimer’s Association, and The Michael J. Fox Foundation for Parkinson’s Research. One author disclosed receipt of unrelated financial honoraria; other authors reported no relevant conflicts of interest.