Publication Abstract Display
Type: Poster
Title: Intra-individual variability in objectively-measured sleep quality is associated with worse cognition in middle-aged and older adults with and without HIV.
Authors: Fisher A, Campbell LM, Sun-Suslow N, Kohli M, Tang B, Lee EE, Heaton A, Moore RC
Date: 02-03-2021
Abstract:Objective: Poor sleep quality is related to worse cognition. In older adults, greater intra-individual variability (IIV) in objectively-measured sleep quality has also been associated with worse cognition; however, this relationship has been less-studied in persons with HIV (PWH), a group with increased prevalence of sleep problems and cognitive impairment compared to older adults without HIV. Therefore, this study examined the association between cognition and IIV in sleep quality among adults aging with and without HIV. Participants and Methods: Sixty-one PWH (Mage=59.1, SDage=6.6; 82% Male) and 32 HIV- adults (Mage=59.2, SDage=6.2; 58% Male) completed comprehensive neuropsychological evaluations. Global and domain-specific demographically-adjusted T-scores were generated. Three objective sleep quality parameters (total sleep time [minutes], efficiency, and sleep fragmentation) were obtained via wrist actigraphy over 5–14 nights. Standard deviations of the objective sleep parameters were calculated to represent IIV for each participant. Subjective sleep quality was assessed using the Pittsburgh Sleep Quality Index (PSQI). Independent samples t-tests were used to compare sleep variables by HIV status. Spearman's rho correlations were used to examine the relationship between subjective sleep quality, average objective sleep, and IIV in objective sleep. Linear regressions, which included an HIV by IIV in sleep interaction and adjusted for use of sleep medication (yes/no), number of nights of sleep data, and hypertension, were used to examine the association between IIV in sleep and global cognition and examine if this relationship differed by HIV status. For significant findings, follow-up analyses examined which cognitive domains were driving this association. Results: Total sleep time, efficiency, and sleep fragmentation did not differ significantly by HIV status (p's>0.05). PWH had greater IIV in total sleep time than the HIV- group (PWH: M=69.4, HIV-: M=55.3, p=0.03); IIV in efficiency and IIV in sleep fragmentation did not differ by HIV status. Average total sleep time, efficiency, and sleep fragmentation were not associated with subjective sleep quality. Likewise, IIV in efficiency and IIV in sleep fragmentation were not associated with subjective sleep quality; however, greater IIV in total sleep time was associated with worse subjective sleep quality (ρ= 0.33, p<0.01). All HIV by IIV interactions were not significant (p’s>.50) and were thus removed from the linear regressions. Adjusting for covariates found greater IIV in total sleep time was associated with worse global cognition (β=-0.23, p=0.039); this effect was driven by executive function (β=-0.36, p<0.001). IIV in efficiency (β=-0.20, p=0.058) and IIV in sleep fragmentation (β=-0.07, p=0.48) were not associated with global or domain-specific cognition. Conclusions: We found that IIV, but not mean objective sleep measures, differed by HIV status; therefore, IIV is an important measure to include with overall sleep measures when assessing sleep’s association with cognition. While we found IIV in total sleep time was related to worse global cognition and executive functioning in middle-aged and older adults with and without HIV, longitudinal studies are needed to determine the directionality of this relationship. These data also highlight the usefulness of actigraphy in assessing hard-to-measure sleep variables that are related to cognition such as IIV in sleep quality.

return to publications listing