Background: Recent work by Marquine et al (2014) has shown an association between higher Veterans Aging Cohort Study (VACS) Index and increased risk of neuropsychological (NP) impairment; but there is limited evidence on longitudinal association between the VACS index and NP functioning and the potential clinical utility of the VACS index for HAND.
Methods: Study participants were from recruited from the city of Toronto, Canada. Baseline and one follow-up NP assessments were done (median follow-up time: 12 months) using a brief NP battery that included Hopkins Verbal Learning Test- Revised (HVLT-R), Grooved Pegboard, and WAIS-R Digit Symbol tests. Overall NP functioning was assessed using an unadjusted raw NP score (z-score), demographically adjusted NP score (Global T-score), and summary regression-based change scores. The VACS Index was computed by summing pre-assigned risk points for age, CD4 count, plasma HIV viral load, haemoglobin, fibrosis, renal glomerular filtration, and HCV infection following guidelines. Hierarchical linear regression method was used to determine the association between the VACS index score at baseline and change in NP functioning adjusting for NP performance, demographic factors, cognitive symptoms, depression, HIV disease markers, and medical comorbidities at baseline.
Results: Data from 523 participants (78% male, 85% on ART, and 71% with suppressed viral load) were included in the analyses. In an unadjusted analyses, higher VACS index score at baseline was associated with a decline in overall NP function measured by changes in z-score (B= -0.06, p<0.001), summary regression-based score (B= -0.03, p=0.009), or Global T-score (B=-0.27, p=0.001). This association persisted after adjustment for baseline NP performance, cognitive symptoms, depression, HIV disease markers, medical comorbidities, and demographic variables. In adjusted analyses, higher VACS index score at baseline predicted a decline in raw NP z-score (B= -0.06, p<0.001), summary regression-based change score (B= -0.02, p=0.004), and Global T-score (B=-0.25, p<0.001). The VACS index score also accounted for considerable proportion (9% -17%) of the total variances explained by the regression models.
Conclusions: Our results validate recent findings from the CHARTER study and suggest that the VACS index may be a useful clinical tool to identify and target those at higher risk for HIV–associated Neurocognitive Disorders.