Sleep health differences and night-to-night variability in people with and without HIV in Mwanza, Tanzania: Insights from actigraphy and sleep diaries
Godfrey A Kisigo, Martha Oshosen, Vanessa Nyamkomola, Ezekiel Mgema, Megan Willkens, Shelley Lees, Robert N Peck, Ana C Krieger, Saidi Kapiga, Kathy BaisleyAbstract
Study Objectives
This study had two objectives: (1) to compare objectively and subjectively measured sleep health between people living with HIV (PLWH) and people without HIV (PWoH) in Mwanza, Tanzania, and (2) to determine the number of nights of actigraphy and sleep diary monitoring required to obtain reliable estimates of habitual sleep.
Methods
We conducted a cross-sectional study among participants enrolled in the Mwanza HIV & CVD Cohort. Participants completed seven consecutive nights of wrist actigraphy and daily sleep diaries. Sleep health parameters were compared between PLWH and PWoH using linear mixed-effects models for repeated measures and linear regression for participant-level variability measures. Intraclass correlation coefficients and the Spearman–Brown formula were used to estimate the number of nights of monitoring required for reliable measurement.
Results
Of the 100 participants (50 PLWH and 50 PWoH), 70% were female, and the median age was 49 years (IQR 44–54). Based on actigraphy data, PLWH had shorter mean sleep duration than PWoH (adjusted mean difference (aMD) −32.1 minutes, 95% confidence interval (CI) −52.6, −11.7). Bedtimes were slightly later (aMD 17.4 minutes, 95% CI −2.4, 37.3) and wake-up times slightly earlier (aMD −2.1 minutes, 95% CI −21.6, 17.3) among PLWH compared with PWoH. PLWH also exhibited greater night-to-night variability in sleep duration (aMD in standard deviation 13.1 minutes, 95% CI 1.2, 25.0). Actigraphy-derived sleep measures required at least 7 nights for reliable estimates, whereas self-reported sleep quality required at least 3 nights.
Conclusions
These findings highlight the importance of multi-day actigraphy for accurately characterizing sleep in sub-Saharan Africa and suggest that HIV is associated with shorter, less consistent sleep without altering circadian timing.