Neurocognitive impairment (NCI) is highly prevalent in older (≥50 years) people with HIV (PWH). Older PWH have
higher rates of NCI than the general population of the same age; prevalence rates are as high as 50% – even
among the virologically controlled. Having NCI is associated with increased mortality, decline in independence,
lower medication adherence, poor decision making, and possibly greater dementia risk. The pathogenesis of NCI
in HIV is likely multifactorial due to extensive diversity of PWH and factors that affect the brain (e.g., HIV,
demographic, socioeconomic, chronic inflammation, comorbidities, and psychosocial stress). This knowledge,
however, comes mostly from high-income countries (HICs). Yet, the burden of HIV is greatest in low- and middle-
income countries (LMICs) where our understanding of NCI (e.g., prevalence; risk; patient/provider needs) in aging
PWH is only just emerging and risk may differ than in HICs. Multimodal phenotypes of NCI risk can help elucidate
NCI’s mechanisms and assist in developing more targeted interventions for it. Critical aspects of HIV disease differ
between HICs and LMICs (e.g., immune responses, age of patients, duration of HIV infection, type of treatment,
and socio-economic factors), but most phenotyping studies have been done with PWH in HICs. We propose to
leverage multimodal data (e.g., demographic, medical, inflammation) from a cohort of diverse, aging, and treated
PWH in Malaysia from two time points across 4-6 years and add one more time point to develop longitudinal
phenotypes of NCI risk – a first for PWH in Malaysia. A pre-requisite, however, is high quality, accurate, unbiased,
and valid neurocognitive test data that can be easily collected in any setting and is suitable for cross-study/-country
comparisons and Big Data applications. Because few tests meeting these requirements exist in Malaysia, we
propose to adapt and preliminarily norm a battery of tests (NeuroScreen) that do. NeuroScreen is brief (~25
minutes), highly automated, easy-to-administer by all levels of staff, disseminated via the internet, designed for
adaptation across countries/languages and in harmonized cross-study data sets. It assesses six neurocognitive
domains most affected by HIV, and has a growing body of evidence demonstrating that it is unbiased, culturally
fair, and psychometrically valid in adolescent and adult populations with HIV and varying levels of computer
literacy in multiple countries and languages (US, South Africa, and neighboring Thailand). No testing apps have
been adapted and normed for ethnically diverse Malaysians, where the most commonly spoken languages are
Bahasa Malaysia, Mandarin, Tamil, and English. Using NeuroScreen’s data will enhance our phenotyping.
Moreover, having an easy-to-use and valid tool to measure neurocognition and screen for NCI can enhance
research and clinical care for PWH in Malaysia. We will build neuropsychological expertise in Malaysia (where
there is little); promote multidisciplinary research into the causes, risks, and detection of NCI in HIV; and expand
capacity to use state-of-the-art statistical analyses (i.e., machine learning).
