Hostname: page-component-586b7cd67f-dsjbd Total loading time: 0 Render date: 2024-12-02T15:21:52.397Z Has data issue: false hasContentIssue false

Neuropsychological tests associated with symptomatic HIV-associated neurocognitive disorder (HAND) in a cohort of older adults in Tanzania

Published online by Cambridge University Press:  20 May 2024

Lachlan Fotheringham
Affiliation:
Translational and Clinical Research Institute, Newcastle University, Newcastle upon Tyne, UK Cumbria Northumberland Tyne and Wear NHS Foundation Trust, UK
Rachael A. Lawson
Affiliation:
Translational and Clinical Research Institute, Newcastle University, Newcastle upon Tyne, UK
Sarah Urasa
Affiliation:
Kilimanjaro Christian Medical University College, Moshi, Kilimanjaro, Tanzania
Judith Boshe
Affiliation:
Kilimanjaro Christian Medical University College, Moshi, Kilimanjaro, Tanzania
Elizabeta B. Mukaetova-Ladinska
Affiliation:
Department of Neuroscience, Behaviour and Psychology, University of Leicester, Leicester, UK
Jane Rogathi
Affiliation:
Kilimanjaro Christian Medical University College, Moshi, Kilimanjaro, Tanzania
William Howlett
Affiliation:
Kilimanjaro Christian Medical University College, Moshi, Kilimanjaro, Tanzania
Marieke C.J. Dekker
Affiliation:
Kilimanjaro Christian Medical University College, Moshi, Kilimanjaro, Tanzania
William K. Gray
Affiliation:
Northumbria Healthcare NHS Foundation Trust, North Tyneside General Hospital, North Shields, UK
Jonathan Evans
Affiliation:
School of Health and Wellbeing, Glasgow University, UK
Richard W. Walker
Affiliation:
Northumbria Healthcare NHS Foundation Trust, North Tyneside General Hospital, North Shields, UK Population Health Sciences Institute, Newcastle University, Newcastle upon Tyne, UK
Philip C. Makupa
Affiliation:
Kilimanjaro Christian Medical University College, Moshi, Kilimanjaro, Tanzania Mawenzi Regional Referral Hospital, Kilimanjaro, Tanzania
Stella-Maria Paddick*
Affiliation:
Translational and Clinical Research Institute, Newcastle University, Newcastle upon Tyne, UK Gateshead Health NHS Foundation Trust, Gateshead, UK
*
Corresponding author: Stella-Maria Paddick; Email: stella-maria.paddick@ncl.ac.uk
Rights & Permissions [Opens in a new window]

Abstract

Objective:

Human immunodeficiency virus (HIV)-associated neurocognitive disorder (HAND) prevalence is expected to increase in East Africa as treatment coverage increases, survival improves, and this population ages. This study aimed to better understand the current cognitive phenotype of this newly emergent population of older combination antiretroviral therapy (cART)-treated people living with HIV (PLWH), in which current screening measures lack accuracy. This will facilitate the refinement of HAND cognitive screening tools for this setting.

Method:

This is a secondary analysis of 253 PLWH aged ≥50 years receiving standard government HIV clinic follow-up in Kilimanjaro, Tanzania. They were evaluated with a detailed locally normed low-literacy neuropsychological battery annually on three occasions and a consensus panel diagnosis of HAND by Frascati criteria based on clinical evaluation and collateral history.

Results:

Tests of verbal learning and memory, categorical verbal fluency, visual memory, and visuoconstruction had an area under the receiver operating characteristic curve >0.7 for symptomatic HAND (s-HAND) (0.70–0.72; p < 0.001 for all tests). Tests of visual memory, verbal learning with delayed recall and recognition memory, psychomotor speed, language comprehension, and categorical verbal fluency were independently associated with s-HAND in a logistic mixed effects model (p < 0.01 for all). Neuropsychological impairments varied by educational background.

Conclusions:

A broad range of cognitive domains are affected in older, well-controlled, East African PLWH, including those not captured in widely used screening measures. It is possible that educational background affects the observed cognitive impairments in this setting. Future screening measures for similar populations should consider assessment of visual memory, verbal learning, language comprehension, and executive and motor function.

Type
Research Article
Copyright
© The Author(s), 2024. Published by Cambridge University Press on behalf of International Neuropsychological Society

Introduction

Globally, 38.4 million people are living with human immunodeficiency virus (HIV), the majority of whom live in sub-Saharan Africa (SSA) (UNAIDS, 2021). HIV-associated neurocognitive disorder (HAND) is a common long-term complication of treated HIV, well-recognised in high-income countries (HICs). Older people appear to be at higher risk of HAND, although the mechanism for this is unclear and likely multifaceted (Hardy & Vance, Reference Hardy and Vance2009; Nightingale et al., Reference Nightingale, Ances, Cinque, Dravid, Dreyer, Gisslén, Joska, Kwasa, Meyer, Mpongo, Nakasujja, Pebody, Pozniak, Price, Sandford, Saylor, Thomas, Underwood, Vera and Winston2023). This poses a new challenge in SSA, as recent substantial progress toward the UNAIDS targets for HIV diagnosis, treatment, and viral suppression (Estill et al., Reference Estill, Marsh, Autenrieth and Ford2018; UNAIDS, 2014) has led to a recent rapid aging of the HIV population.

HAND is currently conceptualized as a spectrum of impairments, hypothesized to result from the direct and indirect effects of the HIV virus, potential neurotoxic effects of combination antiretroviral treatment (cART), comorbid disease, environmental factors, and sociodemographic vulnerabilities (Deeks et al., Reference Deeks, Lewin and Havlir2013; Manji et al., Reference Manji, Jäger and Winston2013). Current operationalized criteria describe asymptomatic neurocognitive impairment (ANI), mild neurocognitive disorder (MND), and HIV-associated dementia (HAD) in order of severity (Antinori et al., Reference Antinori, Arendt, Becker, Brew, Byrd, Cherner, Clifford, Cinque, Epstein, Goodkin, Gisslen, Grant, Heaton, Joseph, Marder, Marra, McArthur, Nunn, Price, Pulliam, Robertson, Sacktor, Valcour and Wojna2007). We acknowledge the critique particularly of ANI as a diagnostic entity of uncertain clinical significance and the substantial potential for false positives and associated negative consequences, and that revision of the current HAND diagnostic concept is advocated (Nightingale et al., Reference Nightingale, Ances, Cinque, Dravid, Dreyer, Gisslén, Joska, Kwasa, Meyer, Mpongo, Nakasujja, Pebody, Pozniak, Price, Sandford, Saylor, Thomas, Underwood, Vera and Winston2023). In seeking to provide clinically relevant findings, we have placed an emphasis here on HAND stages with a functional impairment (MND and HAD). Together, these are termed symptomatic HAND (s-HAND), which require clear evidence of functional impairment in addition to measured cognitive impairment.

Access to effective HIV treatment has a substantial impact on the prevalence of HAND subtypes (Habib et al., Reference Habib, Yakasai, Owolabi, Ibrahim, Habib, Gudaji, Karaye, Ibrahim and Nashabaru2013). HIV-associated dementia was classically described as a subcortical dementia, characterized by prominent motor slowing and the presence of frontal reflexes (Robinson-Papp et al., Reference Robinson-Papp, Elliott and Simpson2009). Modern treatment with cART can reduce the more severe stages of HAND, but it does not entirely prevent it (Saylor et al., Reference Saylor, Dickens, Sacktor, Haughey, Slusher, Pletnikov, Mankowski, Brown, Volsky and McArthur2016). Instead a milder and broader spectrum of clinical impairment is observed in an increasingly well-treated population, with both cortical and subcortical domains affected (Sacktor, Reference Sacktor2018). A recent global estimate suggests that while 40% of adult PLWH meet HAND criteria, only a minority have symptomatic HAND (MND 13–26% and HAD 5–9%) (Wang et al., Reference Wang, Liu, Lu, Farrell, Lappin, Shi, Lu and Bao2020; Wei et al., Reference Wei, Hou, Su, Jiang, Guo, Wang, Zhang, Chang, Wu and Zhang2020), with substantial differences seen between the pre- and post-cART era (Sacktor, Reference Sacktor2018; Wang et al., Reference Wang, Liu, Lu, Farrell, Lappin, Shi, Lu and Bao2020). Regional differences in previous and current cART provision, and differing hypothesized neurotoxic effects of different HIV clades, may also create important differences between populations (Habib et al., Reference Habib, Yakasai, Owolabi, Ibrahim, Habib, Gudaji, Karaye, Ibrahim and Nashabaru2013; Hardy & Vance, Reference Hardy and Vance2009; Tyor et al., Reference Tyor, Fritz-French and Nath2013), meriting specific attention to the East African and even Tanzanian context. Pooled prevalence of HAND in SSA was estimated at 53% in a meta-analysis of 2009–2019 studies, but prevalence varied between 14% and 88% and marked heterogeneity was reported (Nweke et al., Reference Nweke, Okemuo, Uduonu, Ugwu, Nwachukwu and Mshunqane2021).

Age can also impact the prevalence and the profile of impairments in older PLWH (Hardy & Vance, Reference Hardy and Vance2009). This is hypothesized to result from accelerated amyloid deposition and/or cerebrovascular disease associated with aging (Canet et al., Reference Canet, Dias, Gabelle, Simonin, Gosselet, Marchi, Makinson, Tuaillon, Van de Perre, Givalois and Salinas2018; Deeks et al., Reference Deeks, Lewin and Havlir2013; Mackiewicz et al., Reference Mackiewicz, Overk, Achim and Masliah2019). Where aging impacts the presentation of HAND due to an altered disease process, there is merit in considering older people separately. Relevant data are limited however and originate almost exclusively from HICs. A 2013 meta-analysis of African HAND studies was unable to identify any studies of individuals aged ≥50 years (Habib et al., Reference Habib, Yakasai, Owolabi, Ibrahim, Habib, Gudaji, Karaye, Ibrahim and Nashabaru2013), and a 2021 analysis identified only five studies with an identifiable subgroup aged ≥50 years, most of which used a screening tool only (Mwangala et al., Reference Mwangala, Mabrouk, Wagner, Newton and Abubakar2021). In SSA, the number of PLWH >50 years is expected to triple by 2040 (Hontelez et al., Reference Hontelez, de Vlas, Baltussen, Newell, Bakker, Tanser, Lurie and Bärnighausen2012).

Finally, culture and educational background play an important role on individual performance on neuropsychological testing (Ardila, Reference Ardila, Ardila, Uzzell and Ponton2007; Rosselli et al., Reference Rosselli, Velez Uribe, Ahne and Shihadeh2022). It is therefore important to study the pattern of cognitive impairment across cultural contexts, potentially highlighting diverse testing needs and even differences in observed cognitive phenotype. The Global NeuroAIDS Roundtable advised that a lack of culturally appropriate study instruments was limiting HAND research in a range of international settings (Joseph et al., Reference Joseph, Achim, Boivin, Brew, Clifford, Colosi, Ellis, Heaton, Gallo-Diop, Grant, Kanmogne, Kumar, Letendre, Marcotte, Nath, Pardo, Paul, Pulliam, Robertson and Wood2013). There are similar challenges around literacy, where the impression of cognitive phenotype can be distorted by tasks relying on reading, writing, and arithmetic (Ardila, Reference Ardila, Ardila, Uzzell and Ponton2007). This could be particularly important in low- and middle-income countries (LMIC) where low literacy is more common (Roser & Ortiz-Ospina, Reference Roser and Ortiz-Ospina2018).

To our knowledge, this is the only longitudinal cohort of older PLWH with a comprehensive neuropsychological battery and a rigorous consensus diagnostic process in East Africa. The current limited data on older PLWH are focused on South Africa, where cross-sectional screening (Joska et al., Reference Joska, Dreyer, Nightingale, Combrinck and De Jager2019) and incidence using a neuropsychological battery have been examined (Asiimwe et al., Reference Asiimwe, Farrell, Kobayashi, Manne-Goehler, Kahn, Tollman, Kabudula, Gómez-Olivé, Wagner, Montana, Berkman, Glymour and Bärnighausen2020). These populations differ from East Africa demographically, socioculturally, and in HIV prevalence (Asiimwe et al., Reference Asiimwe, Farrell, Kobayashi, Manne-Goehler, Kahn, Tollman, Kabudula, Gómez-Olivé, Wagner, Montana, Berkman, Glymour and Bärnighausen2020; Gómez-Olivé et al., Reference Gómez-Olivé, Montana, Wagner, Kabudula, Rohr, Kahn, Bärnighausen, Collinson, Canning, Gaziano, Salomon, Payne, Wade, Tollman and Berkman2018). A better understanding of the cognitive phenotype in this unique and important cohort will greatly aid the development of HAND screening tools for older adults in East Africa and more widely in the region. While understanding HAND across diverse settings in an evolving context is challenging, it is crucial to address.

Eleven million adults are estimated to have HAND in SSA currently (Wang et al., Reference Wang, Liu, Lu, Farrell, Lappin, Shi, Lu and Bao2020), and as this population ages, HAND may become a leading cause of cognitive impairment in SSA. As treatment provision improves throughout SSA, the aging, stable, longstanding, and well-managed cohort described in this study is likely to be increasingly representative of PLWH throughout the region.

Our primary aim was therefore to identify the neuropsychological tests associated with s-HAND diagnosed according to Frascati criteria (Antinori et al., Reference Antinori, Arendt, Becker, Brew, Byrd, Cherner, Clifford, Cinque, Epstein, Goodkin, Gisslen, Grant, Heaton, Joseph, Marder, Marra, McArthur, Nunn, Price, Pulliam, Robertson, Sacktor, Valcour and Wojna2007) in older PLWH in Tanzania assessed using a locally normed low-literacy neuropsychological test battery including subcortical and cortical domains (Flatt et al., Reference Flatt, Gentry, Kellett-Wright, Eaton, Joseph, Urasa, Howlett, Dekker, Kisoli, Rogathe, Henderson, Lewis, Thornton, McCartney, Yarwood, Irwin, Mukaetova-Ladinska, Akinyemi, Gray, Walker, Dotchin, Quaker, Makupa and Paddick2023). This would inform the development of future screening tools for s-HAND in older people in similar populations.

Methods

Participants and setting

This secondary analysis used longitudinal data from an initial cohort of individuals aged ≥50 years (n = 253) systematically sampled from attendees of a government-funded HIV clinic in Northern Tanzania and offered detailed neuropsychological assessment and consensus HAND diagnosis by Frascati criteria at baseline and annually thereafter for 2 years. The recruitment methods and characterization of the baseline study cohort are previously published (Flatt et al., Reference Flatt, Gentry, Kellett-Wright, Eaton, Joseph, Urasa, Howlett, Dekker, Kisoli, Rogathe, Henderson, Lewis, Thornton, McCartney, Yarwood, Irwin, Mukaetova-Ladinska, Akinyemi, Gray, Walker, Dotchin, Quaker, Makupa and Paddick2023). Recruitment of this cohort took place between March and June 2016. Informed consent was requested following provision of verbal and written information, with assent sought from a close relative where participants lacked capacity to consent due to cognitive impairment. The secondary analysis was approved by Newcastle University’s Faculty of Medical Sciences Research Ethics Committee (ref: 2125/12519) in addition to Kilimanjaro Christian Medical University College Research Ethics and Review Committee and the National Institute for Medical Research, Tanzania, including necessary data transfer agreements. The human data included in this manuscript were obtained in compliance with the Helsinki Declaration (World Medical Association, 2001).

HIV disease severity and other clinical measures

HIV disease severity measures were taken from a standardized clinical data sheet maintained by the clinic for each patient from diagnosis. Available data included nadir CD4 (with ≤200 cells/mm3, categorized as “low”), current CD4, current and previous cART regimen, World Health Organization (WHO) HIV disease stage, current and previous tuberculosis (TB), and, from mid-2017, HIV viral load when this became accessible locally. HIV viral load measurements above 10,000 copies/mm3 were categorized as “high.” Non-HIV measures included visual acuity using a Landholt C illiterate logmar chart (categorized as per WHO as mild, moderate, and severe visual impairment) and hearing impairment by self-report and clinician subjective rating. Demographic data included self-reported year of birth (cross-checked with clinic records), biological sex, and highest level of education (in years) attained by self-report. This methodology and measures are described in more detail in previously published work (Flatt et al., Reference Flatt, Gentry, Kellett-Wright, Eaton, Joseph, Urasa, Howlett, Dekker, Kisoli, Rogathe, Henderson, Lewis, Thornton, McCartney, Yarwood, Irwin, Mukaetova-Ladinska, Akinyemi, Gray, Walker, Dotchin, Quaker, Makupa and Paddick2023), notably finding no association between HAND and HIV disease stage or CD4 nadir.

Neuropsychological testing

Participants underwent a series of neuropsychological tests aimed at assessing a range of cognitive domains at each time point (Table 1). Low-literacy measures were selected from those validated in the original cross-cultural WHO HAD studies (Maj et al., Reference Maj, D’Elia, Satz, Janssen, Zaudig, Uchiyama, Starace, Galderisi and Chervinsky1993; Maj et al., Reference Maj, Janssen, Satz, Zaudig, Starace, Boor, Sughondhabirom, Bing, Luabeya, Ndetei, Riedel, Schulte and Sartorius1991; Maj et al., Reference Maj, Satz, Janssen, Zaudig, Starace, D’Elia, Sughondhabirom, Mussa, Naber and Ndetei1994) and to cover domains required by Frascati criteria (Alkali et al., Reference Alkali, Bwala, Nyandaiti and Danesi2013). Neuropsychological tests included in the battery aimed to assess the following cognitive domains: working memory, visual memory, visuoconstruction, verbal learning, learning interference, delayed recall, recognition memory, psychomotor speed, executive function, language comprehension, orientation, fine motor/2D spatial awareness, and categorical verbal fluency. These are detailed in Table 1. Additional low-literacy versions of tests examining typically cortical domains were added to account for a potentially evolving picture of HAND in the post-combination-ART era (Cysique & Brew, Reference Cysique and Brew2009; Hardy & Vance, Reference Hardy and Vance2009). Where possible, these were locally or regionally developed or validated (Table 1). Normative data were derived from local controls who self-identified as HIV negative. These were recruited from another chronic disease clinic at the same government hospital, matched by age band and education. Timed tests where the time recorded was greater than the maximum allowed were assigned the maximum time. Other test scores out of range were voided as the score was felt to be unreliable.

Table 1. Descriptions of individual cognitive tests and associated areas of impairment.

Note: WAIS = Wechsler Adult Intelligence Scale, AVLT = auditory verbal learning test, HIV = human immunodeficiency virus, WHO = World Health Organization, UCLA = University of California, Los Angeles, HAND = HIV-associated neurocognitive disorder.

a A WHO/UCLA adapted version of the AVLT (Maj et al., Reference Maj, D’Elia, Satz, Janssen, Zaudig, Uchiyama, Starace, Galderisi and Chervinsky1993; Peaker & Stewart, Reference Peaker, Stewart, D.P. and rawford1989; Rey, Reference Rey1958) was used, consistent with initial descriptions of HAND. This draws on a standard lexicon of culturally neutral concepts (Snodgrass & Vanderwart).

b Color trails 1 and 2 are intended to be culture neutral versions of Trail Making Tests A and B (Army Individual Test Battery, 1944; Strauss et al., Reference Strauss, Sherman and Spreen2006), with good applicability to non-English speaking, low-literacy populations (Llorente et al., Reference Llorente, Williams, Satz and D’Elia2003).

c Orientation and commands tests were taken from a version of the Alzheimer’s Disease Assessment Scale – Cognitive (ADAS-Cog), validated in Tanzania and Nigeria for the diagnosis of Alzheimer’s disease in rural-dwelling older adults (Paddick et al., Reference Paddick, Kisoli, Mkenda, Mbowe, Gray, Dotchin, Ogunniyi, Kisima, Olakehinde, Mushi and Walker2017). “Season” was removed from the original ADAS-Cog as it was not well understood in the validation pilot.

d Presumed to be culture neutral due to lack of culturally contingent concepts (Maj et al., Reference Maj, Janssen, Satz, Zaudig, Starace, Boor, Sughondhabirom, Bing, Luabeya, Ndetei, Riedel, Schulte and Sartorius1991).

e A version of the Brazilian “supermarket items” categorical fluency test was used (Lopes et al., Reference Lopes, Brucki, Giampaoli and Mansur2009), with items substituted for locally relevant ones. This test has not been locally validated.

HAND diagnosis

HAND diagnoses were made by consensus panel review in line with the process outlined in Fig. 1. This included detailed case note review and additional confirmatory bedside neurological and mental state examination to consider other sources of cognitive impairment such as Alzheimer’s disease type changes and previous stroke based on clinical presentation and collateral history. Further details of this assessment are previously published (Kellett-Wright et al., Reference Kellett-Wright, Flatt, Eaton, Urasa, Howlett, Dekker, Kisoli, Duijinmaijer, Thornton, McCartney, Yarwood, Irwin, Mukaetova-ladinska, Akinyemi, Lwezuala, Gray, Walker, Dotchin, Makupa and Paddick2021). For the purposes of the analysis, a distinction was made between those with impaired function in everyday living attributable to HAND, that is, symptomatic HAND (s-HAND), and those without. Individuals meeting criteria for MND and HAD were classified s-HAND and compared to individuals with ANI and no cognitive impairment. Assessment of function to distinguish s-HAND was supported by a locally validated Instrumental Activities of Daily Living scale (Paddick et al., Reference Paddick, Gray, Ogunjimi, Lwezuala, Olakehinde, Kisoli, Kissima, Mbowe, Mkenda, Dotchin, Walker, Mushi, Collingwood and Ogunniyi2015), clinician-rated Karnofsky Performance Status (Karnofsky et al., Reference Karnofsky, Abelmann, Craver and Burchenal1948), observation of clear functional difficulties on assessment (i.e., persistent difficulties following instructions), self-reported impairments using a standardized questionnaire and collateral informant history, available for the majority of participants (Collingwood et al., Reference Collingwood, Paddick, Kisoli, Dotchin, Gray, Mbowe, Mkenda, Urasa, Mushi, Chaote and Walker2014). Functional impairment was judged by the consensus panel using all the information available.

Figure 1. Simplified diagnostic flowchart based on Frascati criteria (Antinori et al., Reference Antinori, Arendt, Becker, Brew, Byrd, Cherner, Clifford, Cinque, Epstein, Goodkin, Gisslen, Grant, Heaton, Joseph, Marder, Marra, McArthur, Nunn, Price, Pulliam, Robertson, Sacktor, Valcour and Wojna2007) and demonstrating the relationship between HAND stages and s-HAND. *After an assessment of at least the following domains: verbal/language, attention/working memory, abstraction/executive, memory (learning, recall), speed of information processing, sensory-perceptual, and motor skills. Scores compared to age–education appropriate norms. Standard deviation (SD) in relation to age- and education-matched comparison group; symptomatic HAND (s-HAND).

The Confusion Assessment Method (Inouye et al., Reference Inouye, van Dyck, Alessi, Balkin, Siegal and Horwitz1990; Paddick et al., Reference Paddick, Lewis, Duinmaijer, Banks, Urasa, Tucker, Kisoli, Cletus, Lissu, Kissima, Dotchin, Gray, Muaketova-Ladinska, Cosker and Walker2018), the 15-item Geriatric Depression Scale (GDS) (Yesavage & Sheikh, Reference Yesavage and Sheikh1986), and the Mini-International Neuropsychiatric Interview (Sheehan et al., Reference Sheehan, Lecrubier, Sheehan, Amorim, Janavs, Weiller, Hergueta, Baker and Dunbar1998) supported clinical assessment to help screen for delirium, depression, and other psychiatric disorders, respectively. Where any of these was the primary diagnosis, the patient was excluded from the analysis (Supplementary Materials, S1).

Statistical analysis

Analysis was carried out in R software (version 3.6.3; R Foundation for Statistical Computing, Austria (R Core Team, 2020)). Data were examined visually for normality of distribution. Groups were compared using Skillings–Mack and Kruskal–Wallis tests as appropriate. The statistical significance level was set at alpha = 0.05. Benjamini–Hochberg multiple comparison corrections were carried out with a 5% false discovery rate.

The outcome of the analysis was s-HAND. Educational subgroups were created using a median split of 4 years of school education: “low education” ≤4 years and “high education” >4 years. Locally, there is an important transition from elementary education at this age, and local work has highlighted the importance of little to no formal education on cognitive testing (Paddick et al., Reference Paddick, Longdon, Gray, Dotchin, Kisoli, Chaote and Walker2014). It was not feasible to dichotomize at no education due to the small numbers in this category (n = 30). Dichotomozing at 4 years was the most statistically significant on exploratory preliminary modeling and resulted in the highest Youden index on receiver operating characteristic (ROC) curve analysis (data not shown). This confounder was prioritized due to the wide range of education in the cohort, from university level to none at all.

Accuracy of neuropsychological test scores to identify s-HAND was determined using area under the ROC curves (AUC). Using cutpointr (Thiele & Hirschfeld, Reference Thiele and Hirschfeld2021), sensitivity and specificity were calculated, and 95% confidence intervals (95% CI) were produced by bootstrapping. Youden’s index was used to identify optimal cutoffs. Neuropsychological test scores were dichotomized as normal or impaired using these cutoffs.

Binomial mixed effects modeling to examine the relationship between neuropsychological test performance and s-HAND from baseline to year 2 was performed using lme4 (Bates et al., Reference Bates, Mächler, Bolker and Walker2015), using all available assessments throughout this period. A random intercept model was used, varying at the individual and time level. This was able to account for repeated testing of individuals at more than one time point and does not exclude subjects with missing data from the analysis. A basic model was established, including variables of age, sex, GDS score, CD4 nadir, TB status, education, viral load, hearing and visual impairment, as fixed effects. These were expected to influence cognitive impairment or performance on neuropsychological tests. A backward stepwise approach was taken to exclude coefficients with p > 0.05.

To examine each neuropsychological test individually, dichotomized neuropsychological test (impaired vs. not impaired) scores and an interaction with time were added to the basic model as fixed effects. A further model was constructed including all neuropsychological tests that were significantly associated with s-HAND individually including the basic model, all as fixed effects. Non-significant neuropsychological tests and interactions were excluded using a backward stepwise approach. Each analysis was repeated with low- and high-education subgroups, with education removed from the basic model.

Results

Of 830 patients ≥50 years of age registered at the clinic, 310 were systematically sampled in accordance with the study protocol. Expected clinic capacity determined whether every 2nd or 3rd patient could be approached each day. Based on acute illness, intoxication, or refusal, 21 patients were excluded; 36 patients could not complete a full assessment and were also excluded, leaving a baseline study sample of 253 participants (Table 2). Further details regarding the study protocol and those excluded are previously published (Flatt et al., Reference Flatt, Gentry, Kellett-Wright, Eaton, Joseph, Urasa, Howlett, Dekker, Kisoli, Rogathe, Henderson, Lewis, Thornton, McCartney, Yarwood, Irwin, Mukaetova-Ladinska, Akinyemi, Gray, Walker, Dotchin, Quaker, Makupa and Paddick2023). Of the 253 participants, n = 117 (46.2%) were assessed at all 3 time points (baseline, year 1 and year 2), 85 were assessed twice, and 51 were assessed on only one occasion. This provided 572 observations over all the time points. A collateral history was possible in 452 of these observations (79%). HIV control was good where measured (median latest CD4 count ≥497, median viral load = 0 where available although this was not done in 2016). At baseline, the mean time since diagnosis was 7.1 years (range = 0.7–23.94; SD = 3.3) (Flatt et al., Reference Flatt, Gentry, Kellett-Wright, Eaton, Joseph, Urasa, Howlett, Dekker, Kisoli, Rogathe, Henderson, Lewis, Thornton, McCartney, Yarwood, Irwin, Mukaetova-Ladinska, Akinyemi, Gray, Walker, Dotchin, Quaker, Makupa and Paddick2023). Depression scores were significantly higher at baseline (p < 0.001); there was a higher proportion of individuals with suppressed HIV viral load (i.e., <105) at year 2 compared to year 1 (84.2% vs. 61.5%, respectively, p = 0.004) and poorer visual acuity at year 1 compared to baseline (p = 0.003). There were no other significant differences between sessions for expected confounders included in the model.

Table 2. Participant characteristics and other relevant measures for each time point.

Note: TB = tuberculosis, GDS = Geriatric Depression Scale, ANI = asymptomatic neurocognitive impairment, MND = mild neurocognitive disorder, HAD = HIV-associated dementia, HAND = HIV-associated neurocognitive disorder.

Data presented are median (interquartile range, IQR), unless otherwise stated.

a Skillings–Mack test.

b Kruskal–Wallis test.

c Prior to dichotomizing at baseline, educational groups consisted of none, n = 30 (11.9%); 1–4 years, n = 58 (22.9%); 5–7, n = 107 (42.3%); completed primary (7+), n = 38 (15.0%); completed secondary, n = 13 (5.1%); higher education, n = 4 (1.6%); not known, n = 3 (1.2%).

d One patient was excluded from the analysis for the baseline assessment as they were diagnosed with pseudo-dementia. They were subsequently diagnosed with HAD at a follow-up assessment and so were included as part of the year 1 (2017) assessments. The total number of patients included is therefore 253, rather than 252 as would be expected from the number included at baseline.

Neuropsychological test performance in relation to symptomatic HAND at baseline

All cognitive tests could significantly discriminate between groups (with or without s-HAND) at baseline assessment (p < 0.05, Table 3, educational subgroups in Supplementary Materials, S2), with the exception of Auditory Verbal Learning Test (AVLT) trial IX,false (p > 0.05). Baseline market items (categorical verbal fluency); AVLT trials IV, V, VII and VIII and sum of I–V (verbal learning with delayed recall and recognition memory); and matchstick construction 1 and 2 (visual memory and visuoconstruction, respectively) had the highest area under the receiver operating characteristic (AUROC) (Mandrekar, Reference Mandrekar2010) of ≥ 0.70 (p < 0.001 for all; AUROC = 0.70–0.72; sensitivities = 0.51AVLT 0.73; specificities = 0.60–0.81).

Table 3. Accuracy and optimal cutoffs of baseline cognitive tests identifying s-HAND.

Note: s-HAND = symptomatic HIV-associated neurocognitive disorder, AUC = area under the curve, AVLT = auditory verbal learning test.

Significant results after applying the Benjamini–Hochberg correction are highlighted in bold (p ≤ 0.05).

Neuropsychological test performance in relation to symptomatic HAND at any time point

Controlling for confounders, impairment in all neuropsychological tests apart from AVLT trial IX – false and digit span – forwards was significantly associated with s-HAND at any time point but not with change over time (i.e., impairment at a given time point was associated with s-HAND regardless of time point, but not developing s-HAND over time, Table 4). The strongest associations were between s-HAND and impaired Matchstick construction 1 (visual memory, odds ratio (OR) = 7.8, 95% CI = 3.0–20.4, p < 0.001), AVLT trial V (verbal learning, OR = 7.3, 95% CI = 3.3–16.0, p < 0.001), AVLT trial VIII (delayed recall, OR = 5.8, 95% CI = 2.5–13.1, p < 0.001), and color trails 1 (psychomotor speed, OR = 5.4, CI = 2.5–12.1, p < 0.001) performance at any time point.

Table 4. Summary of the association between impairment on each neuropsychological test and s-HAND diagnosis using logistic mixed effects modeling.

Note: s-HAND = symptomatic HIV-associated neurocognitive disorder, OR = odds radio, CI = confidence interval, AVLT = auditory verbal learning test.

Significant results after Benjamini–Hochberg procedure are highlighted in bold p ≤ 0.013.

Covariates included in each model: the dichotomized neuropsychological test specified, neuropsychological test × time, current age, and Geriatric Depression Scale. Participant ID and time as random effects throughout to account for repeated measures at baseline, year 1 and year 2.

Interactions with time (test * time) were included for each model; however, none were significant.

Distinct results were seen for educational subgroups (Supplementary Materials S3). For the low-education group (≤ 4 years), after correcting for multiple comparisons, the largest ORs for s-HAND were for impaired AVLT trial V (verbal learning (OR = 11.7, CI = 2.7–49.9, p < 0.001), AVLT trial VII (learning interference, OR = 11.5, CI = 2.5–52.1, p = 0.001), and AVLT trial VIII (delayed recall, OR = 10.9, 95% CI = 2.4–50.9, p = 0.002) test performance at any time point.

For the high-education group (>4 years), the highest ORs for s-HAND were for Matchstick construction 1 (visual memory, OR = 20.2, CI = 3.3–122.7, p = 0.001), color trails 1 (psychomotor speed, OR = 10.4, CI = 2.9–37.8, p < 0.001), Matchstick construction 2 (visuoconstruction, OR = 7.9, CI = 2.5–24.6, p < 0.001), and Alzheimer’s Disease Assessment Scale – Cognitive (ADAS-Cog) Commands (language comprehension, OR = 7.6, CI = 2.1 = 27.9, p = 0.002) test performance at any time point.

To determine which tests might be useful as part of a more limited testing battery or screening tool identifying those at risk of s-HAND at any time point, all neuropsychological test performances were individually included in a backward mixed effects model, with interactions with time and covariates (Table 5). Impaired performance on Matchstick construction 1 (OR = 3.3, CI = 1.7–6.6, p < 0.001), AVLT trial IV (OR = 2.1, CI = 1.2–3.9, p = 0.014), AVLT trial VIII (OR = 2.8, CI = 1.4–5.5, p = 0.003), AVLT trial IX – difference (OR = 2.6, CI = 1.3–5.1, p = 0.007), color trails 1 (OR = 2.7, CI = 1.4–5.4, p = 0.004), commands (OR = 2.8, CI = 1.4–5.7, p = 0.004), and market items (OR = 3.2, CI = 1.7–6, p < 0.001) were significantly associated with s-HAND. Impaired Pegboard with the dominant hand was associated with s-HAND over time (OR = 2.8, CI = 1.3–6.3, p = 0.011). For the low-education group, the strongest associations with s-HAND were for impaired AVLT trial VII (learning interference) and matchstick construction 2 (visuoconstruction, Table 5, p < 0.01 for all), whereas in the high-education group, the strongest associations were for impaired AVLT Trial IX – difference (recognition memory), color trails 1 (psychomotor speed), and ADAS-Cog Commands (language comprehension) test performance (p < 0.01 for all).

Table 5. Summary of the association between impairment on neuropsychological test combinations and s-HAND diagnosis using logistic mixed effects modeling.

Note: s-HAND = symptomatic HIV-associated neurocognitive disorder, AVLT = auditory verbal learning test, GDS = Geriatric Depression Scale, TB = tuberculosis.

All participants, n = 498; ≤4 years of education only (low), n = 168; >4 years of education only (high), n = 327.

Significant results after Benjamini–Hochberg procedure are highlighted in bold (all participants, p ≤ 0.013; low education, p ≤ 0.002; high education, p ≤ 0.024). Covariates included in each model: the neuropsychological test specified, neuropsychological test × time, and the basic model. Basic model for each educational group: all participants, age + GDS; low education, age only; and high education, GDS + history of TB. Participant ID and time as random effects throughout to account for repeated measures at baseline, year 1, and year 2.

Discussion

This study describes the cross-sectional association of s-HAND with cognitive impairments as measured by a battery of neuropsychological tests in cART treated older adults in SSA. The analysis used data from all time points, accounting for this using mixed effects models; however, very few tests were associated with s-HAND over time. A broad range of cortical and subcortical impairments were seen. Broadly, tests with real-life applicability, such as recalling items from the market, tended to be most strongly associated with s-HAND. There were prominent differences in the pattern of association seen in different educational groups. These points will be discussed further in turn.

It is widely acknowledged that increasingly available cART has changed the clinical presentation of HAND (Clifford & Ances, Reference Clifford and Ances2013; Sacktor, Reference Sacktor2018). These results provide further evidence that impairments typically associated with cortical pathology play an important role in the pathophysiology of HAND in older adults, as well as traditionally subcortical impairments. In fact, almost every test was significantly associated with s-HAND when considered individually, and when combined into a single model with all participants, matchstick construction 1 (visual memory), market items (categorical verbal fluency), commands (language comprehension), color trails 1 (psychomotor speed), and AVLT trials IV, VIII, and IX – difference (verbal learning, delayed recall, and recognition memory respectively) were independently associated with s-HAND. Categorical verbal fluency had the highest AUC at 0.72 (sensitivity 55%, specificity 81%) supporting the established importance of executive function in HAND, although the low sensitivity here suggests that this may be insufficient to provide clinical utility as a stand-alone test.

The cognitive impairments assessed in this study are broadly similar to those used in other longitudinal cohort studies of HIV in working age adults and children in the USA (Elicer et al., Reference Elicer, Byrd, Clark, Morgello and Robinson-Papp2018; Heaton et al., Reference Heaton, Franklin, Deutsch, Letendre, Ellis, Casaletto, Marquine, Woods, Vaida, Atkinson, Marcotte, McCutchan, Collier, Marra, Clifford, Gelman, Sacktor, Morgello, Simpson, Abramson, Gamst, Fennema-Notestine, Smith, Grant, McCutchan, Ellis, Marcotte, Franklin, Ellis, McCutchan, Alexander, Letendre, Capparelli, Heaton, Atkinson, Woods, Dawson, Smith, Fennema-Notestine, Taylor, Theilmann, Gamst, Cushman, Abramson, Vaida, Marcotte, Marquie-Beck, McArthur, Rogalski, Morgello, Simpson, Mintz, McCutchan, Toperoff, Collier, Marra, Jones, Gelman, Head, Clifford, Al-Lozi and Teshome2015; Sacktor et al., Reference Sacktor, Skolasky, Seaberg, Munro, Becker, Martin, Ragin, Levine and Miller2016), France (Vassallo et al., Reference Vassallo, Fabre, Durant, Lebrun-Frenay, Joly, Ticchioni, DeSalvador, Harvey-Langton, Dunais, Laffon, Cottalorda, Dellamonica and Pradier2017), Uganda (Nakasujja et al., Reference Nakasujja, L Skolasky, Musisi, Allebeck, Robertson, Ronald, Katabira, Clifford and Sacktor2010; Sacktor et al., Reference Sacktor, Nakasujja, Skolasky, Rezapour, Robertson, Musisi and Quinn2009; Sacktor et al., Reference Sacktor, Nakasujja, Skolasky, Robertson, Wong, Musisi, Ronald and Katabira2006), and Zambia (Adams et al., Reference Adams, Mwanza-Kabaghe, Mbewe, Kabundula, Potchen, Maggirwar, Johnson, Schifitto, Gelbard, Birbeck and Bearden2019), however, with the addition of language comprehension and orientation. These data indicate that such cortical processes might be important to assess as part of a comprehensive battery in future research. Without supporting biomarker and neuroimaging data, it is difficult to relate this impression to the underlying pathology with any more certainty.

These findings are relevant to the current difficulties in finding culturally relevant tools to screen for HAND. Existing tools are insufficiently sensitive or specific to provide clinical utility, including the widely used International HIV Dementia Scale (IHDS) (Haddow et al., Reference Haddow, Floyd, Copas and Gilson2013; Kellett-Wright et al., Reference Kellett-Wright, Flatt, Eaton, Urasa, Howlett, Dekker, Kisoli, Duijinmaijer, Thornton, McCartney, Yarwood, Irwin, Mukaetova-ladinska, Akinyemi, Lwezuala, Gray, Walker, Dotchin, Makupa and Paddick2021; Milanini et al., Reference Milanini, Paul, Bahemana, Adamu, Kiweewa, Langat, Owuoth, Allen, Polyak, Ake and Valcour2018). This may be because the IHDS relies heavily on motor speed, whereas in this and many other settings, a broad range of cognitive processes are impacted. Furthermore, tools developed in HICs for working age adults may have limited utility for older adults in SSA. These data suggest some tests which might have broad applicability to the future development of screening tools. Categorical verbal fluency, assessed by asking participants to name market items, not only had the highest AUC in all participants but also performed well in both high-education (AUC 0.77, sensitivity 0.70, specificity 0.66) and low-education subgroups (AUC 0.70, sensitivity 0.63, specificity 0.74. It is well established that executive dysfunction is often a feature of HAND (Sacktor, Reference Sacktor2018) though language ability will of course potentially impact on a task such as category fluency. Less expected was the finding that matchstick construction 1 was the single test most strongly associated with s-HAND in all participants (OR = 7.8, 95% CI = 3.0–20.4, p < 0.001). This was thought to assess visual memory primarily, alongside visuoconstruction. This is not generally thought of as a priority cognitive domain to assess in HAND (Antinori et al., Reference Antinori, Arendt, Becker, Brew, Byrd, Cherner, Clifford, Cinque, Epstein, Goodkin, Gisslen, Grant, Heaton, Joseph, Marder, Marra, McArthur, Nunn, Price, Pulliam, Robertson, Sacktor, Valcour and Wojna2007). While these tests were intended to reflect specific cognitive domains, they may simply reflect abilities which are more relevant to people’s day-to-day lives, more closely resembling tasks that participants have to perform day to day to maintain their function. This may explain the stronger association between these tests and s-HAND, which necessarily involves a degree of functional impairment, rather than specifically representing impaired executive function/language or visuoconstruction/visual memory. A further consideration is the limitations of trying to directly relate these neuropsychological tests with specific cognitive domains, when in reality they do not each map directly onto isolated cognitive processes (Howieson, Reference Howieson2019). The neuropsychological testing protocol used here does not allow us to make any conclusions as to whether executive function or visual memory, for example, was particularly impaired or whether, instead, the strong association in these data simply reflects a global impairment seen most prominently in the most culturally relevant tests.

These results support the importance of education in the diagnosis of s-HAND. Previous work on this and other cohorts has suggested that education is an important risk factor for developing HAND (Cross et al., Reference Cross, Önen, Gase, Overton and Ances2013; Eaton et al., Reference Eaton, Lewis, Kellett‐Wright, Flatt, Urasa, Howlett, Dekker, Kisoli, Rogathe, Thornton, McCartney, Yarwood, Irwin, Mukaetova‐Ladinska, Akinyemi, Gray, Walker, Dotchin, Makupa, Quaker and Paddick2020). These results add to this and suggest that the experience of HAND, or at least the clinical presentation, may be influenced by education too. In the group with ≤4 years of education, the strongest associations were seen with verbal learning, learning interference, delayed recall, and categorical verbal fluency. In the group with >4 years education, visual memory, psychomotor speed, visuoconstruction, language comprehension, and delayed recall were most strongly associated. As a possible explanation, the experience of even a few years of school education and the problem-solving abilities developed may be affecting neuropsychological test performance (Stern, Reference Stern2009). Alternatively, those with less education, and therefore lower cognitive reserve, may simply be vulnerable to small insults to their brain in a different way, leading to the cognitive and functional impairments seen. While the neuropsychological tests chosen for this battery were designed for low-literacy settings, the heterogeneity within the population poses challenges, with 30 participants having never been to school and some with a university education. Taking account of such difference is likely to be an important factor in any successful HAND screening program, which may require separate screening tests for those with little or no formal education.

Strengths and limitations

These data are derived from a routine clinic population with relatively good HIV control, and thus, they reflect well the types of patients likely to characterize the East African HIV pandemic of the future (Deeks et al., Reference Deeks, Lewin and Havlir2013; Estill et al., Reference Estill, Marsh, Autenrieth and Ford2018; Ortblad et al., Reference Ortblad, Baeten, Cherutich, Wamicwe and Wasserheit2019; Sacktor, Reference Sacktor2018; UNAIDS, 2014). The neuropsychological tests were designed to minimize floor and ceiling effects and to be culturally appropriate.

Viral load only became available locally in 2017, limiting analysis of this variable due to missing data. This may have been an important correlate to take account of given the association that has been described between viral suppression and motor speed (Sacktor et al., Reference Sacktor, Skolasky, Tarwater, McArthur, Selnes, Becker, Cohen, Visscher and Miller2003). Similarly, the data on visual and hearing impairment were insufficiently complete to control for sensory impairments. Many participants did not attend follow-up visits as seen in Table 2, leading to possible spectrum bias.

The consensus diagnostic process, while rigorous, lacked neuroimaging data which might have identified relevant changes such as hippocampal atrophy, previous stroke, and substantial small vessel disease. This limited the ability of this protocol to more confidently exclude other sources of cognitive impairment such as Alzheimer’s and vascular dementia. On the other hand, neuroimaging is not routinely available at this outpatient clinic primarily due to affordability. When unavailable, presumed central nervous system (CNS) infections are treated empirically. A diagnostic process more faithful to routine practice might be more generalizable to other clinics in LMICs. While the study protocol was able to exclude a wide range of potential comorbidities, there was no screening for hepatitis B or post-traumatic stress disorder (PTSD), which may have been present in a subset. The consensus panel also lacked a collateral history in 79% of encounters included in this analysis. This may have underestimated the proportion of those with s-HAND, in whom a history of functional impairment might only have been clear from speaking to an informant.

The mixed effects modeling analysis did not identify interactions with time, which may be partly due to the relatively short follow-up period of 3 years. Additionally, a fifth of participants had s-HAND at baseline and at each time point, and this proportion did not significantly change over time, therefore limiting the ability of the model to detect changes in those who subsequently developed s-HAND. A future study could include an analysis of those without HAND at baseline to better determine tests predictive of this. A longer period of follow-up might have been more able to detect any changes over time.

Conclusions

Functional impairment secondary to HAND experienced by older PLWH in this cohort was associated with a broad range of cortical and subcortical cognitive impairments. These pilot findings suggest that screening measures for HAND in similar populations should include measures of verbal learning and motor function, as the IHDS does presently, but that also assessing executive function, visual memory, and language comprehension may offer increased accuracy. Cultural adaptation or use of tests designed specifically for a population may offer further benefits. For older adults in SSA, a combination of a matchsticks construction task, verbal fluency with locally relevant items, a measure of language comprehension, and verbal learning with delayed recall may be an effective combination. Assessing the impact or practicalities of this in a potential screening tool was beyond the scope of this study.

The specifics of the clinical setting are likely to determine the balance between sensitivity and specificity that is considered optimal in a screening tool; however, education, culture, and age are likely to influence accuracy and should be considered in design, testing, and implementation. Future research should aim to establish which tests are most associated with HAND-attributable functional impairment over time, in order to identify and target interventions for individuals at greatest risk of future cognitive and functional decline.

Supplementary material

The supplementary material for this article can be found at https://doi.org/10.1017/S1355617724000201.

Acknowledgments

This study was part-funded by Grand Challenges Canada (Grant Number: 0086-04) and Newcastle University MRes Programme. We wish to acknowledge the help of the nursing and medical staff and hospital management team of Mawenzi Regional Referral Hospital care and treatment center (MRRH-CTC) as well as all the patients and family members who assisted in this study.

Competing interests

No competing interests.

References

Adams, H. R., Mwanza-Kabaghe, S., Mbewe, E. G., Kabundula, P. P., Potchen, M. J., Maggirwar, S., Johnson, B. A., Schifitto, G., Gelbard, H. A., Birbeck, G. L., & Bearden, D. R.The HIV-Associated Neurocognitive Disorders in Zambia (HANDZ) Study: Protocol of a research program in pediatric HIV in sub-Saharan Africa, 2019. medRxiv, 19003590.CrossRefGoogle Scholar
Alkali, N. H., Bwala, S. A., Nyandaiti, Y. W., & Danesi, M. A. (2013). NeuroAIDS in sub-Saharan Africa: A clinical review. Annals of African Medicine, 12(1), 110.CrossRefGoogle Scholar
Antinori, A., Arendt, G., Becker, J. T., Brew, B. J., Byrd, D. A., Cherner, M., Clifford, D. B., Cinque, P., Epstein, L. G., Goodkin, K., Gisslen, M., Grant, I., Heaton, R. K., Joseph, J., Marder, K., Marra, C. M., McArthur, J. C., Nunn, M., Price, R. W., Pulliam, L., Robertson, K. R., Sacktor, N., Valcour, V., Wojna, V. E. (2007). Updated research nosology for HIV-associated neurocognitive disorders. Neurology, 69(18), 17891799.CrossRefGoogle Scholar
Ardila, A. (2007). The impact of culture on neuropsychological test performance. In Ardila, A., Uzzell, B. P., & Ponton, M. (Eds.), International handbook of cross-cultural neuropsychology (pp. 2344). Taylor & Francis.Google Scholar
Army Individual Test Battery (1944). Manual of directions and scoring, War Department, Adjutant General’s Office.Google Scholar
Asiimwe, S. B., Farrell, M., Kobayashi, L. C., Manne-Goehler, J., Kahn, K., Tollman, S. M., Kabudula, C. W., Gómez-Olivé, F. X., Wagner, R. G., Montana, L., Berkman, L. F., Glymour, M. M., & Bärnighausen, T. (2020). Cognitive differences associated with HIV serostatus and antiretroviral therapy use in a population-based sample of older adults in South Africa. Scientific Reports, 10(1), 16625.CrossRefGoogle Scholar
Baiyewu, O., Unverzagt, F. W., Lane, K. A., Gureje, O., Ogunniyi, A., Musick, B., Gao, S., Hall, K. S., & Hendrie, H. C. (2005). The Stick Design test: A new measure of visuoconstructional ability. Journal of the International Neuropsychological Society, 11(5), 598605.CrossRefGoogle Scholar
Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67(1), 48, 2015.CrossRefGoogle Scholar
Canet, G., Dias, C., Gabelle, A., Simonin, Y., Gosselet, F., Marchi, N., Makinson, A., Tuaillon, E., Van de Perre, P., Givalois, L., & Salinas, S. (2018). HIV neuroinfection and Alzheimer’s disease: similarities and potential links? Frontiers in Cellular Neuroscience, 12(307), 113.CrossRefGoogle Scholar
Carey, C., Woods, S., Rippeth, J., Gonzalez, R., Moore, D., Marcotte, T., Grant, I., Heaton, R., HNRC Group (2004). Initial validation of a screening battery for the detection of HIV-associated cognitive impairment. The Clinical Neuropsychologist, 18(2), 234248.CrossRefGoogle Scholar
Choi, H. J., Lee, D. Y., Seo, E. H., Jo, M. K., Sohn, B. K., Choe, Y. M., Byun, M. S., Kim, J. W., Kim, S. G., Yoon, J. C., Jhoo, J. H., Kim, K. W., & Woo, J. I. (2014). A normative study of the digit span in an educationally diverse elderly population. Psychiatry Investigation, 11(1), 3943.CrossRefGoogle Scholar
Clifford, D. B., & Ances, B. M. (2013). HIV-associated neurocognitive disorder. Lancet Infectious Diseases, 13(11), 976986.CrossRefGoogle Scholar
Collingwood, C., Paddick, S. M., Kisoli, A., Dotchin, C. L., Gray, W. K., Mbowe, G., Mkenda, S., Urasa, S., Mushi, D., Chaote, P., & Walker, R. W. (2014). Development and community-based validation of the IDEA study Instrumental Activities of Daily Living (IDEA-IADL) questionnaire. Global Health Action, 7(1), 25988.CrossRefGoogle Scholar
Cross, S., Önen, N., Gase, A., Overton, E. T., & Ances, B. M. (2013). Identifying risk factors for HIV-associated neurocognitive disorders using the international HIV dementia scale. Journal of Neuroimmune Pharmacology, 8(5), 11141122.CrossRefGoogle Scholar
Cysique, L. A., & Brew, B. J. (2009). Neuropsychological functioning and antiretroviral treatment in HIV/AIDS:A review. Neuropsychology Review, 19(2), 169185.CrossRefGoogle Scholar
D’Elia, L., Satz, P., Uchiyama, C., & White, T. (1996). Color trails test. Psychological Assessment Resources Inc.Google Scholar
de Almeida, S. M., Kamat, R., Cherner, M., Umlauf, A., Ribeiro, C. E., de Pereira, A. P., Franklin, D., Heaton, R. K., & Ellis, R. J. (2017). Improving detection of HIV-associated cognitive impairment: Comparison of the International HIV Dementia Scale and a Brief Screening Battery. JAIDS Journal of Acquired Immune Deficiency Syndromes, 74(3), 332338.CrossRefGoogle Scholar
de Paula, J. J., Costa, M. V., Bocardi, M. B., Cortezzi, M., De Moraes, E. N., & Malloy-Diniz, L. F. (2013). The Stick Design Test on the assessment of older adults with low formal education: Evidences of construct, criterion-related and ecological validity. International Psychogeriatrics, 25(12), 20572065.CrossRefGoogle Scholar
Deeks, S. G., Lewin, S. R., & Havlir, D. V. (2013). The end of AIDS: HIV infection as a chronic disease. The Lancet, 382(9903), 15251533.CrossRefGoogle Scholar
Eaton, P., Lewis, T., Kellett‐Wright, J., Flatt, A., Urasa, S., Howlett, W., Dekker, M., Kisoli, A., Rogathe, J., Thornton, J., McCartney, J., Yarwood, V., Irwin, C., Mukaetova‐Ladinska, E. B., Akinyemi, R., Gray, W. K., Walker, R. W., Dotchin, C. L., Makupa, P. C., Quaker, A‐Leon S., … Paddick, S‐Maria (2020). Risk factors for symptomatic HIV-associated neurocognitive disorder in adults aged 50 and over attending a HIV clinic in Tanzania. International Journal of Geriatric Psychiatry, 35(10), 11981208.CrossRefGoogle Scholar
Elicer, I., Byrd, D., Clark, U. S., Morgello, S., & Robinson-Papp, J. (2018). Motor function declines over time in human immunodeficiency virus and is associated with cerebrovascular disease, while HIV-associated neurocognitive disorder remains stable. Journal of Neurovirology, 24(4), 514522.CrossRefGoogle Scholar
Estill, J., Marsh, K., Autenrieth, C., & Ford, N. (2018). How to achieve the global 90-90-90 target by 2020 in sub-Saharan Africa? A mathematical modelling study. Tropical Medicine & International Health, 23(11), 12231230.CrossRefGoogle Scholar
Flatt, A., Gentry, T., Kellett-Wright, J., Eaton, P., Joseph, M., Urasa, S., Howlett, W., Dekker, M., Kisoli, A., Rogathe, J., Henderson, L., Lewis, T., Thornton, J., McCartney, J., Yarwood, V., Irwin, C., Mukaetova-Ladinska, E. B., Akinyemi, R., Gray, W. K., Walker, R. W., Dotchin, C. L., Quaker, A.-L. S., Makupa, P. C., … Paddick, S.-M. (2023). Prevalence and 1-year incidence of HIV-associated neurocognitive disorder (HAND) in adults aged ≥50 years attending standard HIV clinical care in Kilimanjaro, Tanzania. International Psychogeriatrics, 35(7), 339350.CrossRefGoogle Scholar
Gómez-Olivé, F. X., Montana, L., Wagner, R. G., Kabudula, C. W., Rohr, J. K., Kahn, K., Bärnighausen, T., Collinson, M., Canning, D., Gaziano, T., Salomon, J. A., Payne, C. F., Wade, A., Tollman, S. M., & Berkman, L. (2018). Cohort profile: Health and ageing in Africa: A longitudinal study of an INDEPTH Community in South Africa (HAALSI). International Journal of Epidemiology, 47(3), 689690j.CrossRefGoogle Scholar
Habib, A. G., Yakasai, A. M., Owolabi, L. F., Ibrahim, A., Habib, Z. G., Gudaji, M., Karaye, K. M., Ibrahim, D. A., & Nashabaru, I. (2013). Neurocognitive impairment in HIV-1-infected adults in Sub-Saharan Africa: A systematic review and meta-analysis. International Journal of Infectious Diseases, 17(10), e820831.CrossRefGoogle Scholar
Haddow, L. J., Floyd, S., Copas, A., & Gilson, R. J. (2013). A systematic review of the screening accuracy of the HIV Dementia Scale and International HIV Dementia Scale. PLoS One, 8(4), e61826.CrossRefGoogle Scholar
Hardy, D. J., & Vance, D. E. (2009). The neuropsychology of HIV/AIDS in older adults. Neuropsychology Review, 19(2), 263272.CrossRefGoogle Scholar
Heaton, R. K., Franklin, D. R., Deutsch, R., Letendre, S., Ellis, R. J., Casaletto, K., Marquine, M. J., Woods, S. P., Vaida, F., Atkinson, J. H., Marcotte, T. D., McCutchan, J. A., Collier, A. C., Marra, C. M., Clifford, D. B., Gelman, B. B., Sacktor, N., Morgello, S., Simpson, D. M., Abramson, I., Gamst, A. C., Fennema-Notestine, C., Smith, D. M., Grant, I., McCutchan, J. A., Ellis, R. J., Marcotte, T. D., Franklin, D., Ellis, R. J., McCutchan, J. A., Alexander, T., Letendre, S., Capparelli, E., Heaton, R. K., Atkinson, J. H., Woods, S. P., Dawson, M., Smith, D. M., Fennema-Notestine, C., Taylor, M. J., Theilmann, R., Gamst, A. C., Cushman, C., Abramson, I., Vaida, F., Marcotte, T. D., Marquie-Beck, J., McArthur, J., Rogalski, V., Morgello, S., Simpson, D., Mintz, L., McCutchan, J. A., Toperoff, W., Collier, A., Marra, C., Jones, T., Gelman, B., Head, E., Clifford, D., Al-Lozi, M., … Teshome, M. (2015). Neurocognitive change in the era of HIV combination antiretroviral therapy: The longitudinal CHARTER study. Clinical Infectious Diseases, 60(3), 473480.CrossRefGoogle Scholar
Hontelez, J. A., de Vlas, S. J., Baltussen, R., Newell, M. L., Bakker, R., Tanser, F., Lurie, M., & Bärnighausen, T. (2012). The impact of antiretroviral treatment on the age composition of the HIV epidemic in sub-Saharan Africa. Aids, 26(0 1), S1930.CrossRefGoogle Scholar
Howieson, D. (2019). Current limitations of neuropsychological tests and assessment procedures. The Clinical Neuropsychologist, 33(2), 200208.CrossRefGoogle Scholar
Inouye, S. K., van Dyck, C. H., Alessi, C. A., Balkin, S., Siegal, A. P., & Horwitz, R. I. (1990). Clarifying confusion: The confusion assessment method. A new method for detection of delirium. Annals of Internal Medicine, 113(12), 941948.CrossRefGoogle Scholar
Joseph, J., Achim, C. L., Boivin, M. J., Brew, B. J., Clifford, D. B., Colosi, D. A., Ellis, R. J., Heaton, R. K., Gallo-Diop, A., Grant, I., Kanmogne, G. D., Kumar, M., Letendre, S., Marcotte, T. D., Nath, A., Pardo, C. A., Paul, R. H., Pulliam, L., Robertson, K.Wood, C. (2013). Global neuroAIDS roundtable. Journal of Neurovirology, 19(1), 19.CrossRefGoogle Scholar
Joska, J. A., Dreyer, A. J., Nightingale, S., Combrinck, M. I., De Jager, C. A. (2019). Prevalence of HIV-1 infection in an elderly rural population and associations with neurocognitive impairment. Aids, 33(11), 17651771.CrossRefGoogle Scholar
Karnofsky, D., Abelmann, W., Craver, L., & Burchenal, J. (1948). The use of the nitrogen mustards in the palliative treatment of carcinoma – with particular reference to bronchogenic carcinoma. Cancer, 1(4), 634656.3.0.CO;2-L>CrossRefGoogle Scholar
Kellett-Wright, J., Flatt, A., Eaton, P., Urasa, S., Howlett, W., Dekker, M., Kisoli, A., Duijinmaijer, A., Thornton, J., McCartney, J., Yarwood, V., Irwin, C., Mukaetova-ladinska, E., Akinyemi, R., Lwezuala, B., Gray, W. K., Walker, R. W., Dotchin, C. L., Makupa, P., & Paddick, S.-M. (2021). Screening for HIV-Associated Neurocognitive Disorder (HAND) in adults Aged 50 and over attending a government HIV clinic in Kilimanjaro, Tanzania. Comparison of the International HIV Dementia Scale (IHDS) and IDEA Six Item Dementia Screen. AIDS and Behavior, 25(2), 542553.CrossRefGoogle Scholar
Llorente, A. M., Williams, J., Satz, P., & D’Elia, L. F. (2003). Children’s color trails test: Professional manual. Psychological Assessment Resources.Google Scholar
Lopes, M., Brucki, S. M. D., Giampaoli, V., & Mansur, L. L. (2009). Semantic Verbal Fluency test in dementia: Preliminary retrospective analysis. Dementia & Neuropsychologia, 3(4), 315320.CrossRefGoogle Scholar
Mackiewicz, M. M., Overk, C., Achim, C. L., & Masliah, E. (2019). Pathogenesis of age-related HIV neurodegeneration. Journal of Neurovirology, 25(5), 622633.CrossRefGoogle Scholar
Maj, M., D’Elia, L., Satz, P., Janssen, R., Zaudig, M., Uchiyama, C., Starace, F., Galderisi, S., & Chervinsky, A. (1993). Evaluation of two new neuropsychological tests designed to minimize cultural bias in the assessment of HIV-1 seropositive persons: A WHO study. Archives of Clinical Neuropsychology, 8(2), 123135.CrossRefGoogle Scholar
Maj, M., Janssen, R., Satz, P., Zaudig, M., Starace, F., Boor, D., Sughondhabirom, B., Bing, E. G., Luabeya, M’ A. K., Ndetei, D., Riedel, R., Schulte, G., Sartorius, N. (1991). The World Health Organization’s cross-cultural study on neuropsychiatric aspects of infection with the human immunodeficiency virus 1 (HIV-1). Preparation and pilot phase. British Journal of Psychiatry, 159(3), 351356.CrossRefGoogle Scholar
Maj, M., Satz, P., Janssen, R., Zaudig, M., Starace, F., D’Elia, L., Sughondhabirom, B., Mussa, M., Naber, D., Ndetei, D., & et al. (1994). WHO neuropsychiatric AIDS study, cross-sectional phase II. Neuropsychological and neurological findings. Archives of General Psychiatry, 51(1), 5161.CrossRefGoogle Scholar
Mandrekar, J. N. (2010). Receiver operating characteristic curve in diagnostic test assessment. Journal of Thoracic Oncology, 5(9), 13151316.CrossRefGoogle Scholar
Manji, H., Jäger, H. R., & Winston, A. (2013). HIV, dementia and antiretroviral drugs: 30 years of an epidemic. Journal of Neurology, Neurosurgery & Psychiatry, 84(10), 11261137.CrossRefGoogle Scholar
Matthews, C., & Klove, K. (1964). Instruction manual for the adult neuropsychology test battery. University of Wisconsin Medical School.Google Scholar
Milanini, B., Paul, R., Bahemana, E., Adamu, Y., Kiweewa, F., Langat, R., Owuoth, J., Allen, E., Polyak, C., Ake, J., Valcour, V. (2018). Limitations of the International HIV Dementia Scale in the current era. Aids, 32(17), 24772483.CrossRefGoogle Scholar
Mwangala, P. N., Mabrouk, A., Wagner, R., Newton, C., & Abubakar, A. A. (2021). Mental health and well-being of older adults living with HIV in sub-Saharan Africa: A systematic review. BMJ Open, 11(9), e052810.CrossRefGoogle Scholar
Nakasujja, N., L Skolasky, R., Musisi, S., Allebeck, P., Robertson, K., Ronald, A., Katabira, E., Clifford, D. B., Sacktor, N. (2010). Depression symptoms and cognitive function among individuals with advanced HIV infection initiating HAART in Uganda. BMC Psychiatry, 10(1), 44.CrossRefGoogle Scholar
Nightingale, S., Ances, B., Cinque, P., Dravid, A., Dreyer, A. J., Gisslén, M., Joska, J. A., Kwasa, J., Meyer, A.-C., Mpongo, N., Nakasujja, N., Pebody, R., Pozniak, A., Price, R. W., Sandford, C., Saylor, D., Thomas, K. G. F., Underwood, J., Vera, J. H., & Winston, A. (2023). Cognitive impairment in people living with HIV: Consensus recommendations for a new approach. Nature Reviews Neurology, 19(7), 424433.CrossRefGoogle Scholar
Nweke, M. C., Okemuo, A. J., Uduonu, E. M., Ugwu, P. I., Nwachukwu, C., & Mshunqane, N. (2021). Meta-analysis of factors affecting prevalence estimates of HIV-associated neurocognitive disorder in sub-Saharan Africa. South African Journal of Science, 117(9/10), 110.CrossRefGoogle Scholar
Ortblad, K. F., Baeten, J. M., Cherutich, P., Wamicwe, J. N., & Wasserheit, J. N. (2019). The arc of HIV epidemics in sub-Saharan Africa: New challenges with concentrating epidemics in the era of 90-90-90. Current Opinion in Hiv and Aids, 14(5), 354365.CrossRefGoogle Scholar
Paddick, S.-M., Longdon, A., Gray, W. K., Dotchin, C., Kisoli, A., Chaote, P., & Walker, R. (2014). The association between educational level and dementia in rural Tanzania. Dementia & Neuropsychologia, 8(2), 117125.CrossRefGoogle Scholar
Paddick, S. M., Gray, W. K., Ogunjimi, L., Lwezuala, B., Olakehinde, O., Kisoli, A., Kissima, J., Mbowe, G., Mkenda, S., Dotchin, C. L., Walker, R. W., Mushi, D., Collingwood, C., & Ogunniyi, A. (2015). Validation of the Identification and Intervention for Dementia in Elderly Africans (IDEA) cognitive screen in Nigeria and Tanzania. BMC Geriatrics, 15, 53.CrossRefGoogle Scholar
Paddick, S. M., Kisoli, A., Mkenda, S., Mbowe, G., Gray, W. K., Dotchin, C., Ogunniyi, A., Kisima, J., Olakehinde, O., Mushi, D., & Walker, R. W. (2017). Adaptation and validation of the Alzheimer’s Disease Assessment Scale - Cognitive (ADAS-Cog) in a low-literacy setting in sub-Saharan Africa. Acta Neuropsychiatrica, 29(4), 244251.CrossRefGoogle Scholar
Paddick, S. M., Lewis, E. G., Duinmaijer, A., Banks, J., Urasa, S., Tucker, L., Kisoli, A., Cletus, J., Lissu, C., Kissima, J., Dotchin, C., Gray, W. K., Muaketova-Ladinska, E., Cosker, G., & Walker, R. W. (2018). Identification of delirium and dementia in older medical inpatients in Tanzania: A comparison of screening and diagnostic methods. Journal of The Neurological Sciences, 385, 156163.CrossRefGoogle Scholar
Peaker, A., & Stewart, L. E. (1989). Rey’s auditory verbal learning test — A review. In D.P., J. R., & rawford, C. (Eds.), Developments in clinical and experimental neuropsychology. Springer.Google Scholar
R Core Team 2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing.Google Scholar
Rey, A. (1958). L’examen clinique en psychologie. [The clinical examination in psychology]. Presses Universitaries De France.Google Scholar
Robinson-Papp, J., Elliott, K. J., & Simpson, D. M. (2009). HIV-related neurocognitive impairment in the HAART era. Current HIV/AIDS Reports, 6(3), 146152.CrossRefGoogle Scholar
Roser, M., & Ortiz-Ospina, E. Literacy, 2018. https://ourworldindata.org/literacy.Google Scholar
Rosselli, M., Velez Uribe, I., Ahne, E., & Shihadeh, L. (2022). Culture, ethnicity, and level of education in Alzheimer’s disease. Neurotherapeutics, 19(1), 129.CrossRefGoogle Scholar
Sacktor, N. (2018). Changing clinical phenotypes of HIV-associated neurocognitive disorders. Journal of Neurovirology, 24(2), 141145.CrossRefGoogle Scholar
Sacktor, N., Nakasujja, N., Skolasky, R., Rezapour, M., Robertson, K., Musisi, S., & Quinn, T. (2009). HIV subtype D is associated with dementia, compared with subtype A, in immunosuppressed individuals at risk of cognitive impairment in Kampala, Uganda. Clinical Infectious Diseases, 49(5), 780786.CrossRefGoogle Scholar
Sacktor, N., Nakasujja, N., Skolasky, R., Robertson, K., Wong, M., Musisi, S., Ronald, A., & Katabira, E. (2006). Antiretroviral therapy improves cognitive impairment in HIV+ individuals in sub-Saharan Africa. Neurology, 67(2), 311314.CrossRefGoogle Scholar
Sacktor, N., Skolasky, R. L., Seaberg, E., Munro, C., Becker, J. T., Martin, E., Ragin, A., Levine, A., & Miller, E. (2016). Prevalence of HIV-associated neurocognitive disorders in the Multicenter AIDS Cohort Study. Neurology, 86(4), 334340.CrossRefGoogle Scholar
Sacktor, N., Skolasky, R. L., Tarwater, P. M., McArthur, J. C., Selnes, O. A., Becker, J., Cohen, B., Visscher, B., & Miller, E. N. (2003). Response to systemic HIV viral load suppression correlates with psychomotor speed performance. Neurology, 61(4), 567569.CrossRefGoogle Scholar
Sacktor, N., Wong, M., Nakasujja, N., Skolasky, R. L., Selnes, O. A., Musisi, S., Robertson, K., McArthur, J. C., Ronald, A., & Katabira, E. (2005). The International HIV Dementia Scale: A new rapid screening test for HIV dementia. Aids, 19(13), 13671374.Google Scholar
Saylor, D., Dickens, A. M., Sacktor, N., Haughey, N., Slusher, B., Pletnikov, M., Mankowski, J. L., Brown, A., Volsky, D. J., & McArthur, J. C. (2016). HIV-associated neurocognitive disorder--pathogenesis and prospects for treatment. Nature Reviews Neurology, 12(4), 234248.CrossRefGoogle Scholar
Sheehan, D., Lecrubier, Y., Sheehan, K., Amorim, P., Janavs, J., Weiller, E., Hergueta, T., Baker, R., & Dunbar, G. (1998). The Mini-International Neuropsychiatric Interview (M.I.N.I.): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. The Journal of Clinical Psychiatry, 59, 2233.Google Scholar
Silverstein, S. M., Berten, S., Olson, P., Paul, R., Willams, L. M., Cooper, N., & Gordon, E. (2007). Development and validation of a World-Wide-Web-based neurocognitive assessment battery: WebNeuro. Behavior Research Methods, 39(4), 940949.CrossRefGoogle Scholar
Snodgrass, J. G., & Vanderwart, M. A standardized set of 260 pictures: Norms for name agreement, image agreement, familiarity, and visual complexity. (0096-1515 (Print)).Google Scholar
Stern, Y. (2009). Cognitive reserve. Neuropsychologia, 47(10), 20152028.CrossRefGoogle Scholar
Strauss, E., Sherman, E. M. S., & Spreen, O. (2006). A compendium of neuropsychological tests: Administration, norms, and commentary. Oxford University Press.Google Scholar
Thiele, C., & Hirschfeld, G. (2021). cutpointr: Improved estimation and validation of optimal cutpoints in R. Journal of Statistical Software, 98(11), 127.CrossRefGoogle Scholar
Tyor, W., Fritz-French, C., & Nath, A. (2013). Effect of HIV clade differences on the onset and severity of HIV-associated neurocognitive disorders. Journal of Neurovirology, 19(6), 515522.CrossRefGoogle Scholar
UNAIDS. 90-90-90 - An ambitious treatment target to help end the AIDS epidemic, 2014. https://www.unaids.org/sites/default/files/media_asset/90-90-90_en.pdf.Google Scholar
Vassallo, M., Fabre, R., Durant, J., Lebrun-Frenay, C., Joly, H., Ticchioni, M., DeSalvador, F., Harvey-Langton, A., Dunais, B., Laffon, M., Cottalorda, J., Dellamonica, P., & Pradier, C. (2017). A decreasing CD4/CD8 ratio over time and lower CSF-penetrating antiretroviral regimens are associated with a higher risk of neurocognitive deterioration, independently of viral replication. Journal of Neurovirology, 23(2), 216225.CrossRefGoogle Scholar
Wang, Y., Liu, M., Lu, Q., Farrell, M., Lappin, J., Shi, J., Lu, L., & Bao, Y. (2020). Global prevalence and burden of HIV-associated neurocognitive disorder: A meta-analysis. Neurology, 95(19), e2610e2621.CrossRefGoogle Scholar
Wechsler, D. (2008). Wechsler adult intelligence scale--Fourth edition (WAIS-IV). APA PsycTests.Google Scholar
Wei, J., Hou, J., Su, B., Jiang, T., Guo, C., Wang, W., Zhang, Y., Chang, B., Wu, H., & Zhang, T. (2020). The prevalence of frascati-criteria-based HIV-Associated Neurocognitive Disorder (HAND) in HIV-infected adults: A systematic review and meta-analysis. Frontiers in Neurology, 11(581346), 1–9.CrossRefGoogle Scholar
World Medical Association (2001). World Medical Association declaration of Helsinki - Ethical principles for medical research involving human subjects. Bulletin of the World Health Organization, 79, 373374.Google Scholar
Yesavage, J. A., & Sheikh, J. I. (1986). Geriatric depression scale (GDS). Clinical Gerontologist, 5(1-2), 165173.CrossRefGoogle Scholar
Figure 0

Table 1. Descriptions of individual cognitive tests and associated areas of impairment.

Figure 1

Figure 1. Simplified diagnostic flowchart based on Frascati criteria (Antinori et al., 2007) and demonstrating the relationship between HAND stages and s-HAND. *After an assessment of at least the following domains: verbal/language, attention/working memory, abstraction/executive, memory (learning, recall), speed of information processing, sensory-perceptual, and motor skills. Scores compared to age–education appropriate norms. Standard deviation (SD) in relation to age- and education-matched comparison group; symptomatic HAND (s-HAND).

Figure 2

Table 2. Participant characteristics and other relevant measures for each time point.

Figure 3

Table 3. Accuracy and optimal cutoffs of baseline cognitive tests identifying s-HAND.

Figure 4

Table 4. Summary of the association between impairment on each neuropsychological test and s-HAND diagnosis using logistic mixed effects modeling.

Figure 5

Table 5. Summary of the association between impairment on neuropsychological test combinations and s-HAND diagnosis using logistic mixed effects modeling.

Supplementary material: File

Fotheringham et al. supplementary material 1

Fotheringham et al. supplementary material
Download Fotheringham et al. supplementary material 1(File)
File 16.3 KB
Supplementary material: File

Fotheringham et al. supplementary material 2

Fotheringham et al. supplementary material
Download Fotheringham et al. supplementary material 2(File)
File 64.5 KB
Supplementary material: File

Fotheringham et al. supplementary material 3

Fotheringham et al. supplementary material
Download Fotheringham et al. supplementary material 3(File)
File 62 KB