ASSOCIATION BETWEEN DAILY ENERGY EXPENDITURE AND SLEEP IN PHYSICALLY ACTIVE ADULTS

Abstract

Although epidemiologic surveys have consistently shown an association between exercise and better sleep (1), laboratory experimental studies have failed to provide compelling support for this association (2). This discrepancy might be partly explained by limitations in laboratory studies, including few nights of assessment and sleep disruptions associated with the novel sleep environment. The purpose of this investigation was to examine the association between daily energy expenditure and sleep in physically active adults. We examined 71 physically active, normal sleepers (n = 38 ages 18–30, n = 33 ages 60–75). During seven consecutive days of home recording, participants wore a wrist-mounted Actillume that recorded activity. Daily total energy expenditure (TEE) was estimated from diary-assessed physical activity ratios (PAR) and estimated basal metabolic rate (BMR) (TEE = PAR X BMR) (3). Sleep-wake was estimated with a validated algorithm associating wrist movement with electro-encepalographically measured sleep. The sleep variables assessed were sleep onset latency (SOL), total sleep time (TST), and wake after sleep onset (WASO). Volunteers recorded subjective estimates of SOL, TST, WASO, and insomnia with a daily sleep diary. Between-subjects associations between average TEE and average actillume and subjectively assessed sleep were assessed with partial correlations, controlling for age. Variables that were not normally distributed were log-transformed. To examine whether within-subjects daily variations in TEE moderated variations in sleep, repeated measures ANCOVAs were conducted with TEE as the covariate. The sleep variables were z-transformed before the ANCOVAs. In addition sleep was compared following the day of highest TEE vs. the day of lowest TEE with Wilcoxin nonparametric sign-ranks tests. A significant between-subjects association was found only for TEE and SOL (r = −0.26, p = 0.04). A significant (but modest) within-subjects moderating effect of TEE was found only for subjective WASO (p = 0.01, beta = −0.40). However, no significant Wilcoxin sign-ranks results were found. The results suggest little association between TEE and sleep in active adults. The two significant findings might be confounded by other factors we have not yet assessed (e.g., light exposure). Limitations of our analysis include a low daily variability in TEE, a restricted assessment of physical activity, a reliance on self-reported activity, and a sample of normal sleepers (ceiling effect). (1) Vuori et al., Acta Physiol Scand, 1988, 133, Supp 574, 3–7. (2) Youngstedt et al., Sleep, 1997, 20, 203–314. (3) Bouchard et al., Am J Clin Nutr, 1983, 37, 461–467. Supported by NIH HL61280, AG15763.