Publication Abstract Display
Type: Published Abstract
Title: NIH Toolbox Cognition Battery as a screen for HIV-associated neurocognitive disorders.
Authors: Blackstone K, Umlauf A, Franklin DR, Corkran S, Gershon R, Grant I, Heaton RK
Year: 2015
Publication: Society for Clinical Neuropsychology
Volume: Issue: Pages:
Abstract:Despite improved antiretroviral treatment, approximately a third to a half of HIV+ individuals evidence some level of HIV-associated neurocognitive disorder (HAND). As such, there is a growing demand for brief neurocognitive screening instruments in order to identify these individuals in the clinic, who may then be appropriate to undergo further testing. However, selecting assessment tools sensitive to the mild and spotty cognitive profile typically observed in HIV infection has proven challenging. The NIH Toolbox Cognition Battery (NIH-TB-CB) is a 30-minute, computerized assessment that covers six ability domains (i.e., executive functions, episodic memory, processing speed, working memory, attention, and language). Additionally, the NIH-TB-CB is available for testing ages 3-85 and is nonproprietary. Given its relative brevity, ease of administration (i.e., computerized), and nonproprietary nature, the NIH-TB-CB may be well poised as a potential screener for HAND. The current study therefore examined the utility of the NIH-TB-CB in detecting HAND, compared to a traditional comprehensive neuropsychological test battery. 107 HIV+ participants completed the NIH-TB-CB. For the purposes of this study, only the fluid measures of the NIH-TB-CB were examined (i.e., language was excluded) given that these are the areas most susceptible to impairment in HIV infection and that our gold standard neuropsychological battery (GS-NP) only measured fluid abilities; the fluid battery administration time is approximately 20 minutes. Participants additionally completed a comprehensive GS-NP covering seven ability domains, as well as psychiatric and neuromedical evaluations. For both batteries, demographically-corrected T-scores were converted to deficit scores (0=normal to 5=severely impaired). Following standard practice, impairment was classified as an average deficit score ≥0.5 across each battery. Overall, participants exhibited 37.4% and 32.7% global impairment on the Fluid NIH-TB-CB and GS-NP, respectively (Κ=0.84,p<0.001). HIV+ individuals demonstrated the largest rate of impairment on the Fluid NIH-TB-CB Picture Memory subtest (36%). Fluid NIH-TB-CB deficit scores were strongly correlated with GS-NP deficit scores (ρ=0.98, p<0.001), and within-subjects analyses indicated that participants did not differ across the two batteries (t=1.5,p=0.13). The Fluid NIH-TB-CB demonstrated 94.3% sensitivity and 90.3% specificity as compared to the GS-NP (PPV=82.5%; NPV=97.0%). HIV+ participants classified as impaired on the Fluid NIH-TB-CB demonstrated more depressed mood and cognitive symptoms in daily life, as well as more lifetime substance use disorders and unemployment, lower current CD4 counts, and higher HIV plasma viral loads than HIV+ classified as normal. Overall, the fluid measures of the NIH-TB-CB demonstrated excellent concordance with a traditional comprehensive neuropsychological battery in detecting HAND. Additionally, participants detected as impaired evidenced worse psychiatric, psychosocial, and HIV disease outcomes, providing construct validity to the fluid NIH-TB-CB. This 20-minute computerized neurocognitive screener may be especially amenable for use in clinics in which time and resources are often limited in order to identify HIV+ patients who are appropriate for further testing. Future studies are warranted to determine how the NIH-TB-CB may perform in detecting neurocognitive declines in other neurological populations.

return to publications listing