Background HIV infection commonly results in neurological and neurocognitive deficits, including HIV-associated dementia. A standardized neuropsychological (NP) test battery is often administered to determine the presence of neurocognitive impairment. NP tests rely on normative data to produce standardized scores and may adjust for a patient's stated years of education. However, the stated years of education do not necessarily correlate with the quality of education obtained. Thus, a discrepancy in the quality of schooling may result in significantly different levels of functioning. This suggests that individuals with poor-quality education may be more likely to score as cognitively impaired on the NP tests, resulting in a false-positive diagnosis of impairment.
Objective To determine if standardized NP test scores based on self-reported years of education or the Wide Range Achievement Test-3 (WRAT-3) reading test grade-equivalent level are more accurate in predicting neurocognitive impairment.
Methods 120 HIV-positive participants were administered a NP test battery as part of a comprehensive evaluation. Subjects were classified by a neurologist as either (1) unimpaired/subsyndromic or (2) neurologically impaired (possible or probable HIV-associated motor/cognitive disorder). Raw neuropsychological test scores were converted to T-scores using published norms, with education determined in two ways: (1) self-reported education or (2) WRAT-3 reading grade level. T-scores across 7 measures were averaged, and individuals with a mean T-score of less than 41 were classified as neuropsychologically impaired. Cross-tabulation tables were used to determine sensitivity and specificity of the two methods in detecting neurological impairment.
Results T-scores derived from WRAT-3 reading scores were more sensitive than those derived from self-report (77.6% vs 13.1%) in detecting neurological impairment, although specificity was somewhat lower (81.4% vs 94.9%).
Conclusions Normative data based on WRAT-3 reading scores were more accurate than those based on self-report in classifying individuals as neurologically impaired vs unimpaired. The use of reading tests for determining true education level may be a useful alternative in deriving neuropsychological test norms.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.