Publication Abstract Display
Type: Poster
Title: Frailty and cognition: Cross-sectional comparison of the Fried Phenotype, Rockwood Frailty Index, and Veterans Aging Cohort Study (VACS) Index on HIV-associated neurocognitive disorders.
Authors: Sun-Suslow, N, Paolillo E, Saloner R, Letendre SL, Morgan EE, Moore DJ
Date: 02-03-2021
Abstract:Objective: Frailty has long been recognized as a geriatric syndrome, but to date there has yet to be an agreed upon standard operational definition, which poses challenges to both clinical and research settings. Despite this limitation, frailty is reliably associated with poorer cognitive functioning. Among people with HIV (PWH), rates of frailty are higher and are predictive of future cognitive decline. To determine which measure has the strongest association to cognition among PWH, this study compared common measures of frailty as they relate to HIV-Associated Neurocognitive Disorders (HAND). Participants and Methods: Participants were 289 PWH, older than 50 years (mean age=59.6, SD=7.3) enrolled in UC San Diego’s HIV Neurobehavioral Research Program from 2014 to 2020. Frailty measurements included the (1) Fried Phenotype criteria (0-5 symptoms: weight loss, exhaustion, low physical activity, slowness, weakness), the (2) Rockwood Frailty Index (proportion of general and HIV-specific health deficits ranging from 0 [no deficits] to 1 [all 34 deficits]), and the (3) Veterans Aging Cohort Study (VACS) Index 1.0. HAND was diagnosed according to the Frascati criteria using a seven-domain neuropsychological battery. Separate ANOVAs (i.e., one for each frailty measure) were used to examine differences in frailty severity between HAND subgroups, and ROC analyses evaluated sensitivity and specificity of each frailty measure to detect symptomatic HAND [mild neurocognitive disorder (MND) and HIV-associated dementia (HAD)] from cognitively normal cases. Results: Individuals diagnosed with HAD had higher rates of frailty than those without HAND among all three frailty measures (p's<.05, d's>.41). Significant differences in frailty severity were observed between no HAND and MND (p’s<.05, d's>.48), as well as ANI and HAD (p’s<.05, d's>.53) when using the Fried Phenotype or the Rockwood Frailty Index, but not with the VACS Index. To detect symptomatic cases from those who were cognitively normal, an optimal cutoff of ≥3 was identified for the Fried Frailty index (AUC = 0.71), with sensitivity of 37% and specificity of 92%; an optimal cutoff of ≥0.206 was identified for the Rockwood Frailty index (AUC = 0.66), with sensitivity of 85% and specificity of 43%; and an optimal cutoff of ≥29 was identified for the VACS index (AUC = 0.59), with sensitivity of 58% and specificity of 65%. Conclusions: Based on pairwise comparisons, the Fried Phenotype was the best at distinguishing among HAND subtypes; this is followed by the Rockwood Frailty Index and then the VACS Index. However, when considering an optimal balance between sensitivity and specificity, all three measures were poor at detecting HIV-related neurobehavioral dysfunction, and thus highlighting the heterogeneity in neurobehavioral status among PWH at varying frailty levels. Together, frailty may help clinicians stratify risk for HAND, but is an inadequate proxy for HAND and further supports the need for cognitive assessments in clinical settings.

return to publications listing