Does Exercise Truly Enhance Sleep?

Shawn D. Youngerstedt, PhD

THE PHYSICIAN AND SPORTSMEDICINE - VOL 25 - NO. 10 - OCTOBER 1997

In Brief: Two recent studies support the commonsense view that exercise improves sleep in individuals with insomnia, but questions still abound. Do patients have to be fit to reap the benefits of exercise? Does exposure to bright light during exercise enhance sleep? Is shorter, more intense exercise more helpful than longer, more moderate activities? Does exercising too close to bedtime inhibit sleep? In considering recent research on sleep and exercise, this article addresses such questions and also looks at the relationship of exercise and sleep to anxiety, depression, circadian rhythms, and age-associated problems, such as sleep apnea.

Insomnia afflicts about a third of the adult population and is associated with increased mortality, psychiatric disturbances, and decreased work productivity (1,2). Because sleeping pills—the most common treatment for insomnia—are associated with increased mortality (2), medication tolerance and dependence, and a host of negative side effects (3), they are seldom recommended for long-term use. Hence, clinicians and patients have shown increased interest in cognitive and behavioral strategies for improving sleep.

One behavioral strategy is exercise. Epidemiologic survey studies (4,5) indicate that daytime exercise is the behavior most closely associated with improved sleep in the general populace. Since exercise elicits physical fatigue and because its physiologic and psychological calming effects are well established, the notion that exercise promotes sleep may seem like common sense. This commonsense view has been reinforced for years in the lay and scientific literature, but the experimental evidence that exercise promotes sleep is not compelling. In addition, the changes in sleep that follow exercise have not been shown to influence daytime cognitive functioning or alertness, and little is known about the effects of exercise on people who have sleep disturbances.

Exercise and Insomnia

Two recent studies (6,7), however, suggest that exercise training may improve sleep among insomniacs. Guilleminault et al (6) randomly assigned 30 individuals (average age 44) who had psychophysiologic insomnia to three different 4-week treatments. One treatment consisted of sleep hygiene education, such as encouraging the subjects to maintain a fixed sleep-wake schedule and to avoid daytime napping. Another treatment included sleep hygiene education and light therapy that involved patients sitting in front of a bright light (3,000 lux) for 45 minutes beginning 5 minutes after awakening. (Bright light exposure can improve sleep, particularly for people with abnormally phased circadian rhythms (8)). A third treatment involved sleep hygiene education and 45 minutes of brisk daily walking in the early evening.

Wrist actigraphy measurements taken the week before and after each 4-week protocol (figure 1: not shown) indicated that the regimens that included exercise and light elicited greater improvements in sleep (an increase of 17 and 54 min/night respectively) compared with sleep hygiene education alone (a decrease of 3 min/night). Similar findings were observed from self-reported sleep assessments via sleep diaries.

Though the data suggest that exercise can relieve insomnia, other issues may confound the study's findings. The exercise-related improvements in sleep could have been linked to the subjects' exposure to outdoor light during exercise and at other times of the day. Unfortunately, the study did not provide information regarding such exposure, nor about patient adherence to the treatments. Another confounder may have been the subjects' different expectations for improved sleep associated with each protocol.

In another study, King et al (7) randomly assigned 43 older individuals (ages 50 to 76) with moderate sleep complaints to a 16-week exercise program or to a waiting list control condition. The exercise group performed 30 to 40 minutes of moderate aerobic exercise four times per week during the day or early evening. Based on sleep diaries and the Pittsburgh Sleep Quality Index (9)—a standardized, subjective measure of sleep quality—the exercise group showed significantly greater improvements in sleep (figure 2: not shown).

Self-reported measures of sleep, however, do not always correspond with objective measures of sleep, especially in insomniacs. Self-reported measures can be skewed because individuals commonly expect exercise to promote sleep. The data may also have been confounded by environmental influences, such as exposure to bright light, or by the adoption of sleep-promoting habits that were not assessed, such as reduced caffeine consumption. Moreover, the data failed to delineate whether the reported sleep improvements were the result of chronic or repeated acute exercise.

Despite the limitations of these pioneering studies, they provide tantalizing evidence that exercise may promote sleep in individuals with insomnia. Whether exercise promotes sleep among all individuals or whether improvements in sleep can be attributed to exercise per se will require further research.

Acute Exercise and Sleep

Most studies on the relationship between exercise and sleep have focused on the influence of acute exercise. A recent meta-analysis by Youngstedt et al (10) of 38 studies examined the average effect of acute exercise on sleep and the variables that may moderate this effect. In all studies, sleep was assessed by polysomnography, and the stages of sleep were determined by standard criteria.

The analysis concluded that a single bout of exercise:

• had no effect on the time it took to fall asleep;
• elicited statistically significant but small increases in total sleep time (average, 10 minutes) and in the amount of slow-wave sleep (average, 4 minutes); and
• elicited statistically significant but small decreases in rapid eye movement (REM) sleep (average, 7 minutes) and increases in the time for REM sleep to occur following sleep onset (average, 13 minutes).

These studies focused exclusively on subjects who are good sleepers. Although these findings indicate that the effects of acute exercise are small, these effects may still be noteworthy. Since even sleeping pills can have little influence on good sleepers (11,12), the small size of the exercise-related effects doesn't necessarily undermine their importance. On the other hand, because the studies focus on good sleepers, the findings indicate nothing about the effects of acute exercise on individuals suffering from sleep disturbances. In fact, I am unaware of any studies that have examined the effects of acute exercise in these individuals.

Factors Affecting Exercise and Sleep

Several factors may potentially moderate the effects of sleep on exercise, including the subjects' level of fitness, the heat load and duration of exercise, when exercise is performed, and the subjects' exposure to bright light. Analyzing these factors may help us understand the mechanisms by which exercise influences sleep and the parameters that most influence sleep.

Fitness. Some researchers (13) have speculated that only fit subjects are capable of performing the intense, exhaustive exercise assumed necessary for sleep enhancement. Since unfit subjects cannot achieve the level of exercise that may be required for improved sleep, the argument goes, exercise may benefit them less than it benefits fit individuals. An alternative explanation might be that exercise benefits the unfit less because it elicits more prolonged physiologic arousal in the unfit, thus inhibiting their sleep. However, experimental evidence (10) indicates that fitness does not influence the effects of acute exercise on sleep, and these data are consistent with surveys that show that exercise promotes sleep in the general population.

Body temperature. A popular explanation for the effects of exercise on sleep has been the thermogenic hypothesis, which posits that exercise promotes sleep by heating the body or brain. This hypothesis is consistent with evidence that passive body heating—via hot tub or sauna—increases slow-wave sleep (14). Neurophysiologic evidence (15) also suggests an interaction between heat loss and sleep mechanisms in the anterior hypothalamus that are activated when temperature is elevated. Hence, exercise may act as a thermogenic stimulus to enhance sleep.

The primary impetus behind the thermogenic hypothesis has been a study by Horne and Moore (16) showing that increases in slow-wave sleep following exercise were reversed by cooling the body during exercise. The results of this study, though, are questionable because body temperature during sleep was not assessed, and exercise was performed approximately 6 hours before bedtime, apparently enough time for the body to return to a normal temperature before sleep. Furthermore, our meta-analysis (10) revealed no modulating thermogenic effect of exercise on sleep. These inconsistencies suggest the need for further testing of the thermogenic hypothesis.

Another issue for such research is the level of exercise that would benefit the most people. Since body temperature increases during exercise as a function of exercise intensity, the thermogenic hypothesis would suggest that the more intense the exercise, the greater the sleep enhancement. Since the general public tends to exercise at lower intensities, the thermogenic hypothesis would also suggest that exercise would not enhance most people's sleep. However, if less intense exercise is just as efficacious for sleep, as evidence suggests (10), then more people may be able to improve their sleep with exercise.

Exercise duration. Our recent meta-analysis indicated that exercise duration was one of the most important factors modulating the influence of exercise on sleep. For example, as exercise duration increased beyond 1 hour per day, total sleep time became progressively greater. Further research is needed to confirm whether very prolonged exercise may increase total sleep time. Even if research does so, the practical usefulness of such evidence is questionable since most people are unlikely to exercise for the 1 hour per day that the studies found necessary to reliably improve sleep.

Exercise timing. A common assumption is that vigorous exercise right before bedtime disrupts sleep. However, this issue has not been adequately addressed in the literature because most studies' exercise protocols were completed at least 4 hours before sleep.

Recent evidence (17) challenges the assumption that late-night exercise disturbs sleep. In aerobically fit subjects, sleep was not adversely affected by a 1-hour bout of exercise at 60% VO2 max or by 3 hours of exercise at 70% VO2 max completed 30 minutes before bedtime. These results could be explained by quicker postexercise physiologic recovery among physically fit individuals. On the other hand, a population survey (5) indicated that exercise within 2 hours of bedtime improved or had no effect on sleep for most individuals. Since the evening is often convenient for exercise, individuals should include evening as they experiment to find the most suitable exercise time.

Exposure to bright light. Inadequate exposure to bright light has been associated with disturbed sleep. Conversely, bright light exposure can elicit quick and dramatic improvements in sleep. While the average adult receives only about 20 minutes of daily exposure to bright light—more than 2,500 lux (18)—it is reasonable to suppose that individuals who exercise regularly outdoors receive at least three times this much.

Though our meta-analysis (10) generally failed to reveal whether exercise performed indoors or outdoors affected subjects' sleep differently, our analysis was limited by the studies' lack of information about their subjects' illumination during exercise and throughout the day. In light of the recent study (6) suggesting that bright light exposure may enhance sleep more than exercise, this issue needs to be examined more carefully. There may be interesting synergistic effects of exercise and bright light on sleep.

Anxiety and Depression

Anxiety is perhaps the primary cause of insomnia, particularly transitory, situational insomnia. Since exercise has been shown to reduce psychophysiologic and subjective indices of anxiety, and anxiolytic behavioral treatments have been effective for insomnia, it is plausible that exercise may promote sleep by reducing anxiety. The one study (19) that explored this mechanism was inconclusive: Although subjects who exercised had significantly reduced anxiety 20 minutes after exercise when compared with sedentary controls, there was no difference between the two groups' anxiety levels at bedtime (4 to 6 hours after exercise) and no association between anxiety level at bedtime and sleep.

Chronic exercise may also promote sleep by its antidepressant effects. Depression is associated with disturbed sleep and REM sleep abnormalities, which are frequently reversed with remission of depression. Acute exercise delays REM onset and reduces the total amount of REM sleep (10). Vogel et al (20) have argued that these REM effects are the mechanism by which all antidepressant treatments work.

The Biological Clock

Poor sleep is one of the primary symptoms resulting from working different shifts (shift work) or jetting across multiple time zones. It can largely be attributed to a person's circadian pacemaker—which controls sleep and wakefulness—being out of synchrony with his or her changed sleep-wake schedule. (Poor sleep is also associated with aging [see "Exercise and Sleep Problems of Older Patients," below]). Because about 48 million passengers return from international flights annually and 50 million people in the United States are engaged in some form of shift work (21), the potential health consequences of air travel and shift work are significant.

Evidence now suggests that exercise can elicit substantial shifts in circadian rhythm phase in humans (22) and rodents (23), comparable to shifts that occur in response to bright light. Moreover, research on rodents suggests that appropriately timed exercise and exposure to light can have synergistic phase-shifting effects (24). Therefore, it is plausible that appropriately timed exercise can improve sleep for shift workers and air travelers.

Circadian phase delays have been consistently observed following late-night or early-morning exercise (10 pm to 4 am). Recently, Eastman et al (25) found that exercise during a simulated graveyard shift (midnight to 8 am) allowed subjects to adjust more quickly to a daytime-sleep and nighttime-work schedule.

Exercise before bedtime (10 pm to midnight local time) following westward travel across two to five time zones might similarly enhance circadian adjustment and, therefore, improve sleep. It is unknown whether exercise in humans might have analogous circadian phase-advancing effects appropriate for eastward travel across multiple time zones.

The Exercise-Sleep Riddle

While most people have assumed that exercise promotes sleep, the potential adverse effects of exercise on sleep have received little attention. Anecdotal evidence suggests that overtraining can disturb sleep. Overtraining occurs in competitive athletes and fitness enthusiasts and is associated with other physiologic and psychological markers of distress such as elevated cortisol and depression. Experimental (26) and anecdotal (4) evidence demonstrates that discontinuing regular exercise will cause disrupted sleep, a sign of possible dependency. Such evidence may mean that exercise, like other sleep medicine, could have limitations and potential dangers.

Clearly, further research is needed to explore the effects of exercise on sleep and to establish whether they can be attributed to exercise or associated factors. A formal exercise prescription for sleep problems is not justified until such research results are available.

Nonetheless, clinicians and scientists still can suggest exercise as a means of improving sleep (see "Steps to Promote Sleep," below). Based on epidemiologic evidence and recent experimental evidence with insomniacs, the suggestions seem reasonable, especially if exercise is part of an overall treatment program and suggested with the patient's understanding that it may produce little or no improvement in sleep. Even if exercise does not improve sleep, it should still be recommended because exercise can reduce overall mortality and morbidity.

References

1. National Commission on Sleep Disorders Research: Wake Up America: A National Sleep Alert, Executive Summary and Executive Report. 1993;1:1-76
2. Kripke DF, Simons RN, Garfinkel L, et al: Short and long sleep and sleeping pills: is increased mortality associated? Arch Gen Psychiatry 1979;36(1):103-116
3. American Psychiatric Association. Benzodiazepine Dependence, Toxicity, and Abuse: A Task Force Report of the American Psychiatric Association. Washington DC, American Psychiatric Association, 1990
4. Vuori I, Urponen H, Hasan J, et al: Epidemiology of exercise effects on sleep. Acta Physiol Scand 1988;574:3-7
5. Hasan J, Urponen H, Vuori I, et al: Exercise habits and sleep in a middle-aged Finnish population. Acta Physiol Scand 1988;574:33-35
6. Guilleminault C, Clerk A, Black J, et al: Nondrug treatment trials in psychophysiologic insomnia. Ann Intern Med 1995;155(8):838-844
7. King AC, Oman RF, Brassington GS, et al: Moderate-intensity exercise and self-rated quality of sleep in older adults: a randomized controlled trial. JAMA 1997;277(1):32-37
8. Campbell SS, Dawson D, Anderson MW: Alleviation of sleep maintenance insomnia with timed exposure to bright light. J Am Geriatr Soc 1993;41:829-836
9. Buysse DJ, Reynolds CF III , Monk TH, et al: The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res 1989;28(2):193-213
10. Youngstedt SD, O'Connor PJ, Dishman RK: The effects of acute exercise on sleep: a quantitative synthesis. Sleep 1997;20(3):203-214
11. Mendelson WB: Effects of flurazepam and zolpidem on the perception of sleep in normal volunteers. Sleep 1995;18(2):88-91
12. Terzano MG, Parrino L: Evaluation of EEG cyclic alternating pattern during sleep in insomniacs and controls under placebo and acute treatment with zolpidem. Sleep 1992;15(1):64-70
13. Horne JA: The effects of exercise upon sleep: a critical review. Biol Psychol 1981;12(4):241-290
14. Bunnell DE, Agnew JA, Horvath SM, et al: Passive body heating and sleep: influence of proximity to sleep. Sleep 1988;11(2):210-219
15. McGinty D, Szymusiak R: Keeping cool: a hypothesis about the mechanisms and functions of slow wave sleep. Trends Neurosci 1990;13(12):480-487
16. Horne JA, Moore VJ: Sleep EEG effects of exercise with and without additional body cooling. Electroencephal Clin Neurophysiol 1985;60(1):33-38
17. Youngstedt SD, Kripke DF: Late night exercise does not disrupt sleep in physically active individuals. Sleep Res 1997;26:222
18. Espiritu RC, Kripke DF, Ancoli-Israel S, et al: Low illumination by San Diego adults: association with atypical depressive symptoms. Biol Psychiatry 1994;35(16):403-407
19. Youngstedt SD, O'Connor PJ, Dishman RK, et al: Influence of exercise on caffeine-induced insomnia. Sleep Res 1995;24:144
20. Vogel GW, Buffenstein A, Minter K, et al: Drug effects on REM sleep and on endogenous depression. Neurosci Biobehav Rev 1990;14(1):49-63
21. Monk TH: Shift work, in Kryger MH, Roth T, Dement WC (eds): Principles and Practices of Sleep Medicine, Philadelphia, WB Saunders, 1994, pp 471-476
22. Van Reeth O, Sturis J, Byrne MM, et al: Nocturnal exercise phase delays circadian rhythms of melatonin and thyrotropin secretion in normal men. Am J Physiol 1994;266(6 pt 1):E964-E974
23. Reebs SG, Mrosovsky N: Large phase-shifts of circadian rhythms caused by induced running in a re-entrainment paradigm: the role of pulse duration and light. J Comp Physiol (A) 1989;165(6):819-825
24. Mrosovsky N: Double-pulse experiments with nonphotic and photic phase-shifting stimuli. J Biol Rhythms 1991;6(2):167-179
25. Eastman CI, Hoese EK, Youngstedt SD, et al: Phase-shifting human circadian rhythms with exercise during the night shift. Physiol Behav 1995;58(6):1287-1291
26. Baekeland F: Exercise deprivation: sleep and psychological reactions. Arch Gen Psychiatry 1970;22(4):365-369

Exercise and Sleep Problems of Older Patients

Many of the sleep problems associated with aging, including sleep fragmentation, early-morning awakening, daytime sleepiness, and frequent napping, may be attributed to abnormal circadian phase or a dampened circadian rhythm amplitude, thought to indicate reduced strength of the circadian oscillator. In addition to shifting circadian phase, regular exercise may also strengthen the circadian oscillator, thus helping the older person to sleep at night and remain awake in the day.

Support for this hypothesis is provided by studies that show that complete activity restriction results in dramatic fragmentation of sleep throughout the day, even in young, healthy individuals (1) Further, a recent study by Van Someren et al (2) found that 3 months of exercise training significantly reduced fragmentation of the rest-activity rhythm in older men (average age 73), a change that may enhance daytime wakefulness and nighttime sleep.

Sleep apnea. Sleep apnea and periodic leg movements during sleep (PLMS) increase with age and have been associated with disturbed sleep. Obstructive sleep apnea can often be attributed to excessive fat deposits in the upper airway. This condition is treated by surgery or with the use of a nasal mask that provides continuous positive pressure to the upper airway.

Many patients, however, cannot tolerate the masks, and surgery is not always effective. Exercise may be an effective alternative or adjuvant treatment because it may reduce the upper-airway fat deposits.

Leg movements. Exercise may also help patients whose sleep is disturbed by leg movements. In our lab, we have observed that PLMS can occur from 100 to 1,000 times throughout the night. PLMS is related to restless legs syndrome (RLS), characterized by "creepy-crawly" sensations in the lower legs and an irresistible urge to move the legs that occur just after lying down for sleep. Both conditions may be attributed to inadequate dopamine stores or neurotransmission. Dopaminergic drugs have been used to treat PLMS and RLS but have adverse side effects.

Leg exercise such as walking is the only known behavioral method for temporary relief of RLS. This exercise-induced attenuation of symptoms suggests that these conditions are related to dopamine stores, since body movement is integrally related to dopamine systems, and chronic exercise increases dopamine synthesis and metabolism in many brain regions (3). Regular exercise may effectively treat both conditions, but, to my knowledge, this hypothesis has not been tested.

References

1. Campbell SS: Duration and placement of sleep in a 'disentrained' environment. Psychophysiol 1984;21(1):106-11
2. Van Someren EJ, Lijzenga C, Mirmiran G, et al: Long-term fitness training improves the circadian rest-activity rhythm in healthy elderly males. J Biol Rhythms 1997;12(2):146-156
3. Chaouloff F: Physical exercise and brain monoamines: a review. Acta Physiol Scand 1989;137(1):1-13

Steps to Promote Sleep

Though current research does not offer compelling evidence that exercise promotes sleep, several practical recommendations seem prudent in light of current knowledge on exercise and sleep:

• exercise in bright outdoor light if possible.
• experiment with exercise at different times of the day in order to find a convenient time; late-night exercise does not necessarily disrupt sleep.
• whether you are fit or unfit, try exercise to enhance sleep, since level of fitness does not determine the effects of exercise on sleep.
• exercise duration is more important than intensity. Thirty minutes of moderate exercise most days is the standard recommendation for general health, and sleep time seems to increase when exercise continues an hour or longer.

Dr Youngstedt is a postdoctoral fellow in the Department of Psychiatry at the Sam and Rose Stein Institute for Research on Aging at the University of California, San Diego. Address correspondence to Shawn D. Youngstedt, PhD, Dept of Psychiatry, 0667, University of California, San Diego, 9500 Gilman Dr, La Jolla, CA 92093-0667.

RETURN TO OCTOBER 1997 TABLE OF CONTENTS