| Publication Abstract Display | | Type: Poster | | Title: Neuropsychological test combinations as a screening measure for detection of HIV-associated neurocognitive impairment in India. | | Authors: Sakamoto M, Ghate M, Deutsch R, Letendre S, Meyer R, Heaton RK, Grant I, Mehendale S, Marcotte TD, and the HNRP Group | | Date: 02-06-2013 | | Abstract:Objective: Existing screening measures, such as the International HIV Dementia Scale, lack good classification accuracy for detection of HIV-associated neuropsychological (NP) impairment. This study aimed to identify and validate a screening battery, by pairing up two NP tests, for HIV clinicians in settings where time and resources may limit the availability of full NP assessments.
Participants and Methods: 206 HIV-infected participants from Pune, India completed a comprehensive NP battery measuring 7 cognitive domains, translated into Marathi. 53% had an AIDS diagnosis, with the median CD4 cell count being 240 cells/mm3. 33% of the sample was classified as NP impaired based upon a large normative group from Pune. All 22 NP tests were compared using a Superiority Index (Deutsch et al., 2009), which quantifies and ranks relative performance of each diagnostic test based on sensitivity and specificity in comparison to the NP status. Cross-validated recursive partitioning was then executed on 2-test combinations of the highest ranking tests to identify which combinations demonstrated the best prediction of NP impairment.
Results: A test of visual (non-verbal) learning met the criteria for inclusion in each of the top-performing combinations. The best 2-test combinations were the Brief Visuospatial Memory Test – Revised (BVMT-R) learning and Color Trails 1 (sensitivity = 75%/specificity = 84%), the BVMT-R learning and Grooved Pegboard (71%/84%), and the BVMT-R learning and Digit Symbol (72%/83%).
Conclusions: The combination of 2 NP tests, primarily assessing non-verbal learning and processing speed, yielded very good sensitivity and specificity for the detection of NP impairment in this Indian cohort. Given that these NP test pairs have minimal language and literacy requirements, and take only 5 to 10 minutes to administer, HIV clinicians in India and other international settings may benefit from using the screening measures described here. |
return to publications listing
|