Validity of two scales in identifying HIV-associated dementia

Objectives: To examine and compare the validity of the HIV Dementia Scale (HDS) and Executive Interview (EXIT) in detecting HIV/AIDS dementia. Methods: The sample included hospitalized or skilled nursing/assisted-living facility (SNF) HIV/AIDS patients (N = 103). Participants completed the HDS, EXIT, and the Structured Clinical Interview from the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV). A chart review also was performed. Results: Sample was predominantly male (93%) and white (79%), with a mean age of 38 years (standard deviation [SD] 8 years). Twelve patients had a clinical diagnosis of HIV dementia. Cutoff scores of 10 or less for the HDS and 11 or more for the EXIT optimized sensitivity and specificity. Education level was related to performance for both tests. Using separate logistic regression analyses, the HDS and the EXIT were significant individual predictors of dementia. When entered together, the EXIT was the only significant predictor of dementia. Selected items of the HDS and EXIT also performed well in identifying patients with dementia. Conclusions: The HDS and the EXIT show promise as brief, well-tolerated screening tools for HIV dementia in ill patients. The HDS was more sensitive, but the EXIT added additional predictive power over the HDS in identifying dementia.