Screening for HIV-associated neurocognitive disorders in perinatally infected youth: Validation of the CAT-rapid version 2 screening tool

Mgweba-Bewana, Lihle

Screening for HIV-associated neurocognitive disorders in perinatally infected youth: Validation of the CAT-rapid version 2 screening tool

Thesis / Dissertation

2023

Abstract

Objectives: There is a lack of locally validated cognitive screening tools for youth living with HIV. To test the validity of the Cognitive Assessment Tool Rapid v2 tool in screening for all forms of neurocognitive impairment in a sample of youth living with HIV in Cape Town, South Africa. Design: We analysed the existing dataset of the neuro sub-study which was nested within the larger CTAAC study (Cape Town Adolescent Antiretroviral Cohort). The main study examined the long-term effects and outcomes of ARV treatment in 500 HIV-infected children and adolescents. HIV negative, matched controls were also included in the study. Component scores of the CAT-rapid version 1, the psychomotor speed and Luria hand sequence of the youth International HIV Dementia Scale were combined to form the CAT-rapid v2. SPSS version 28.0 was used for statistical analysis. Methods: For the CAT-rapid v2 validation, the sample of YLWH was dichotomised according to the impaired and not impaired groups, based on their neuropsychological test scores. Impairment in this sample, relates to all levels of neurocognitive impairment. The combined tool has a total score of 20. Cross tabulations were then conducted in SPSS. The results were then entered into the medcalc.org online calculator to determine the sensitivity and specificity, positive and negative predictive values, positive and negative likelihood ratios, disease prevalence and the overall accuracy of the screening tool. A Receiver Operator Characteristic (ROC) curve was constructed to investigate the sensitivity, specificity, positive and negative Running head: YLWH HAND SCREENING: CAT-RAPIDV2 VALIDATION 5 predictive values of the CAT-rapid v2 in detecting all forms of neurocognitive disorder in YLWH. Results: Statistical analyses did not detect significant between-group differences with regard to age, sex, language, or years of education. Significantly more participants in the YLWH however, had repeated at least one grade at school. Analyses detected significant betweengroup differences for CAT-rapid v2 total score, as well as for score in every cognitive domain except for attention. In each case, the healthy control group performed better than the YLWH group. A CAT-rapid v2 cut-off score of ≤17 yielded the best balance between sensitivity and specificity. At this cut off, the sensitivity of the CAT-rapid v2 screening tool was 69.31%, specificity was 50.35%, PPV of 49.65% and NPV of 69.90%. Finally, we conducted a ROC analysis for the CAT-rapid v2 total score, plotted against the gold standard battery neuropsychological tests in the YLWH group only. The ROC analysis showed that at the cut off score of ≤ 17, the CAT-rapid v2 had a significantly better chance of screening for any cognitive impairment in YLWH from those with no impairment. The area under the curve was acceptable at 0.651. Conclusion: Our study showed that the CAT-rapid v2 - at a cut off score of 17 or less – had reasonable sensitivity, however, it did not meet the generally acceptable statistical threshold in use for clinical settings. This screening tool may be useful for consideration in research settings and requires further development and validation for use in clinical settings.

Keywords

Neuropsychiatry

Reference:

Collections

Masters

Full item page