Objective measurements of free-living sleep and activity patterns in older Icelandic men and women

Purpose: To identify cross-sectional and seasonal sleeping and physical activity (PA) patterns in a community-dwelling Icelandic healthy aging population using accelerometers.

Methods: Participants (n=244, male/female 110/134, age 79.7±4.9 years, BMI 26.9±4.8 kg/m²) wore an Actiwatch Spectrum on the non-dominant wrist to assess sleep and an Actigraph GT3X on the right hip to assess PA for at least 4 days over a week of free living. A subpopulation (n=72) repeated the 7-day measurement during seasonal periods with greater (13.4±1.4 hours) and lesser (7.7±1.8 hours) daylight.

Results: Compared to women, men had a shorter sleep duration (462±80 vs. 487±68 minutes, p=0.008) mainly due to an earlier rise time, and a higher number of awakenings per night (46.5±18.3 vs. 40.2±15.7, p=0.007). No gender differences were observed in sleep efficiency (80.9±10.1% vs. 82.2±9.8%) or latency (34.4±28.7 vs. 39.3±33.6 minutes). Men and women had similar average daily PA (134±70 and 127±62 total counts/wear-time minute, respectively) but women were less sedentary (613±83 vs. 641±88 wear-time minutes, p=0.01) and had more light activity (181±56 vs. 158±55 minutes, p=0.001). In the subpopulation with two visits (n=72), the 5.7±2.6 hour difference in daylight had little impact on the within-individual differences in sleep patterns and PA for both men and women and differences in daylight hours were uncorrelated to differences in wear-time counts, sleep duration, and sleep efficiency.

Conclusions: The sleep and PA characteristics of this unique population revealed some gender differences, but the lack of variation in response to significant daylight changes may be due to long-term adaptation.