Poor sleep quality can affect physical, mental and emotional function and has been frequently reported in people with HIV (PWH). We explored dimensions of sleep health, derived from objectively-measured sleep/wake activity, and their associations with health- and sleep-related quality of life (QoL) in PWH and lifestyle-matched controls.
A subset of PWH (18-49 and ≥50 yo) and HIV-negative controls (≥50 yo) participating in the POPPY study wore an actigraphy device for 7 days/nights. Physical and mental QoL, sleep-related impairment (perceptions of daytime functional impairment associated with sleep) and disturbance (perceptions of sleep quality) were derived from the SF-36 and PROMIS questionnaires. Exploratory factor analysis of 27 actigraphy variables was performed and 7 dimensions of sleep health were obtained. Linear regression was used to test associations of sleep dimensions with HIV-status and QoL measures (separately in PWH and controls) and whether they differed by HIV-status. All analyses accounted for age, gender and ethnicity.
The 343 PWH and 117 HIV-negative controls were predominantly male (87% and 68%) with a median (IQR) age of 57 (52-62) and 61 (57-66) years, respectively. The 7 actigraphy-derived dimensions of sleep health were fragmentation, irregularity in duration/timing, sleep duration, duration/variability of awake periods (after initial sleep), irregularity in fragmentation, onset latency and timing. None of these significantly differed between PWH and controls (all p’s>0.1). In PWH, longer duration and/or greater variability of awake periods was associated with poorer physical (p=0.01) and mental (p=0.04) health and greater sleep-related impairment (p<0.001). Greater irregularity in duration/timing and longer onset latency were both associated with greater sleep-related impairment (p<0.001 and p=0.004) and disturbance (p<0.001 and p=0.03). Irregularity in duration/timing was also associated with poorer mental health (p=0.03). Associations were generally similar to those seen among the HIV-negative controls; only the associations of onset latency and sleep-related QoL appeared to differ (p-interaction=0.01 and 0.003, Figure).
We found seven dimensions of sleep health based on objective actigraphy measures. Whilst these dimensions have differential impacts on self-reported health, the effects are generally similar between people with and without HIV. These findings could inform targeted strategies to improve sleep health and, in turn, QoL of PWH.