No association of sleep with total daily physical activity in normal sleepers Shawn D. Youngstedt a, * , Michael L. Perlis b , Patrick M. O’Brien a , Christopher R. Palmer a , Michael T. Smith c , Henry J. Orff b , Daniel F. Kripke a a Department of Psychiatry, Sam and Rose Stein Institute for Research on Aging, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0667, USA b Department of Psychiatry, University of Rochester, Rochester, NY, USA c Department of Psychiatry and Behavioral Sciences, Johns Hopkins University, Baltimore, MD, USA Received 17 July 2002; received in revised form 31 October 2002; accepted 5 December 2002 Abstract The aim of two studies was to examine both between-subjects and within-subjects associations between daily amounts of physical activity and sleep in the home environment. Study 1 examined self-reported exercise durations and sleep diaries for 105 consecutive days in 31 college students who were normal sleepers. Between-subjects associations of mean exercise with mean sleep were assessed with Spearman rank-order correlations. Within-subjects correlations were determined across 105 days, and by comparing sleep on the 11 most active vs. the 11 least active days. Study 2 examined 71 physically active adults (n = 38 ages 18 – 30 years, and n = 33 ages 60 – 75 years), the majority of whom were normal sleepers. Over seven consecutive days, physical activity was assessed via actigraphy and a diary-derived estimate of energy expenditure, and sleep was assessed via actigraphy and sleep diaries. Between-subjects associations of mean physical activity with mean sleep were assessed with partial correlations, controlling for age. Within-subjects associations were assessed with ANCOVAs, with daily physical activity serving as the covariate, and by comparing sleep on the most active vs. the least active day. No significant within-subjects associations between physical activity and sleep were found in the main analyses of either study. Two small, but significant, between-subjects correlations between different physical activity measures and subjective sleep were found in Study 2. These results fail to support epidemiologic data on the value of exercise for sleep, but are consistent with experimental evidence showing only modest effects of exercise on sleep. D 2003 Elsevier Science Inc. All rights reserved. Keywords: Actigraphy; Energy expenditure; Exercise; Sleep diary; Normal sleepers 1. Introduction The notion that a physically demanding day will lead to a good night’s sleep has been widely accepted for centuries [1]. In large population surveys, people report that exercise is perhaps the most important sleep-promoting behavior [2– 4]. Epidemiologic studies have consistently shown an asso- ciation between self-reports of exercise and better sleep [5– 8]. Moreover, exercise is recommended as an important sleep aid by many sleep experts [9]. However, experimental studies have failed to provide compelling evidence that acute or chronic exercise promotes sleep to a substantial degree [10,11]. The lack of more compelling evidence might be ex- plained by several limitations in experimental studies. For example, exercise and sleep recording in the laboratory cannot duplicate the exercise and sleep conditions which have been associated with better sleep in surveys. Moreover, most experimental studies have examined exercise and sleep on only 1 or 2 days. Since sleep is influenced by multiple factors, several days may be necessary to delineate bene- ficial effects of exercise. Finally, experimental studies have focused largely on acute exercise of short duration ( 60 min). However, meta-analysis indicates more reliable sleep improvements following exercise of longer durations, and that exercise duration is more predictive of better sleep than other exercise parameters such as intensity or time-of-day [11]. Although experimental studies have tended to focus on short, acute bouts of exercise, it is plausible that the most important sleep-promoting parameter associated with phys- ical activity might be total energy expenditure (TEE). It is reasonable to expect, for example, that sleep might be better following a day of continuous light-to-moderate intensity 0031-9384/03/$ – see front matter D 2003 Elsevier Science Inc. All rights reserved. doi:10.1016/S0031-9384(03)00004-0 * Corresponding author. Tel.: +1-858-534-7347; fax: +1-858-534- 7405. E-mail address: syoungstedt@ucsd.edu (S.D. Youngstedt). Physiology & Behavior 78 (2003) 395 – 401