Sleep. 2015 Dec 1;38(12):1893-903. doi: 10.5665/sleep.5236.

Diet/Energy Balance Affect Sleep and Wakefulness Independent of Body Weight.

Perron IJ1,2, Pack AI1, Veasey S1.
  • 1Center for Sleep and Circadian Neurobiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA.
  • 2Neuroscience Graduate Group, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA.



STUDY OBJECTIVES: Excessive daytime sleepiness commonly affects obese people, even in those without sleep apnea, yet its causes remain uncertain. We sought to determine whether acute dietary changes could induce or rescue wake impairments independent of body weight.

DESIGN: We implemented a novel feeding paradigm that generates two groups of mice with equal body weight but opposing energetic balance. Two subsets of mice consuming either regular chow (RC) or high-fat diet (HFD) for 8 w were switched to the opposite diet for 1 w. Sleep recordings were conducted at Week 0 (baseline), Week 8 (pre-diet switch), and Week 9 (post-diet switch) for all groups. Sleep homeostasis was measured at Week 8 and Week 9.

PARTICIPANTS: Young adult, male C57BL/6J mice.

MEASUREMENTS AND RESULTS: Differences in total wake, nonrapid eye movement (NREM), and rapid eye movement (REM) time were quantified, in addition to changes in bout fragmentation/consolidation. At Week 9, the two diet switch groups had similar body weight. However, animals switched to HFD (and thus gaining weight) had decreased wake time, increased NREM sleep time, and worsened sleep/wake fragmentation compared to mice switched to RC (which were in weight loss). These effects were driven by significant sleep/wake changes induced by acute dietary manipulations (Week 8 → Week 9). Sleep homeostasis, as measured by delta power increase following sleep deprivation, was unaffected by our feeding paradigm.

CONCLUSIONS: Acute dietary manipulations are sufficient to alter sleep and wakefulness independent of body weight and without effects on sleep homeostasis.

KEYWORDS: diet switch; fragmentation; high fat diet; obesity; sleep

PMID: 26158893



Why are we so sleepy, especially during the day? For some, sufficient sleep (≥7 hours)1 is difficult to get because of busy schedules, restless newborns, or sleep disorders such as sleep apnea or insomnia. However, studies from the Centers for Disease Control and Prevention (CDC) have found that more than 30% of people report unintentionally falling asleep during the day, regardless of how much sleep they get on average2. Excessive daytime sleepiness is not only a major public health concern (e.g., ‘drowsy driving’ contributes to >80,000 car accidents annually)3 but can reduce productivity at work, potentially costing the U.S. economy an estimated $63 billion4. Understanding the root causes of daytime sleepiness may allow for new therapies or strategies to improve daytime attention and vigilance.

Concomitant with these wake impairments, nearly 70% of Americans are overweight and half of those people are considered obese5. The links between obesity and sleep are well established, specifically insufficient sleep (<7 hours) as a risk factor for developing obesity6. On the other hand, obese individuals report being much sleepier during the day compared to their lean counterparts7. This observation was confirmed in a sleep laboratory setting, even when controlling for multiple confounding variables (diabetes, sleep apnea, narcolepsy, etc.)8. Therefore, we wanted to know: why are obese people so sleepy?

The house mouse (mus musculus) is a common animal model to experimentally study which variables affect behavior, including sleep. Obese mice exhibit profound changes in sleep including hypersomnolence (i.e., sleeping an additional 1-2 hours per day) and increased fragmentation (i.e., frequent flip-flopping between sleep and wakefulness)9. The increase in total sleep and fragmentation is most pronounced during the nighttime, when the nocturnal mouse is usually most active. This increased sleepiness during periods of normal wakefulness is reminiscent of excessively sleepy humans who have trouble staying awake during their day-job or behind the wheel.

In our lab, we assess the effect of obesity on sleep by feeding mice a high-fat diet (Fig 1). In this model, all genetically-identical mice begin the study lean. Then some of the mice are switched to a high-fat diet (HFD, 45% of calories from fat), while the other mice continue to consume their low-calorie regular chow (RC, 13.5% of calories from fat; Fig 1A). Over time, the HFD-fed mice develop obesity (Fig 1B) and exhibit all the classic sleep/wake impairments mentioned in the previous paragraph (Fig 1C). But the question still remains: why are obese mice so sleepy?



IP fig1Figure 1: The effects of different diets on body weight and sleep/wake time. (A) The two diets used in the study. (B) HFD-fed mice gain considerably more weight than RC-fed mice after eight weeks. (C) HFD-fed mice spend much less time awake compared to RC-fed mice, especially during the beginning of the dark phase (when they are usually most active). HFD-fed mice also have 75% more fragmentation during the dark phase (not shown here).


We thought there were two likely explanations. The first hypothesis was obvious: obese mice were sleepy because they were obese. In other words, a heavier mouse would be much sleepier than a comparatively lean mouse. If this were true, body weight would be the greatest predictor of how disrupted (or normal) their sleep and wake would be. However, we were already skeptical of this hypothesis because of follow-up studies with obese patients who had recently undergone bariatric surgery for weight loss. Our lab and others found that these obese individuals report dramatic improvements in attention and vigilance after only moderate (~10-12 pounds) weight loss7. Thus, our alternate hypothesis was that weight changes, due to an imbalance of calories consumed vs calories burned, was the primary determinant of how sleepy mice become. We decided to use the diet-induced obesity mouse model, whose weight can be increased by consuming HFD or decreased by consuming RC, to test these hypotheses.

But the obvious problem was staring us in the face: how can we tease apart ‘diet/weight changes’ from ‘body weight’ when changing the diet will invariably affect the animals’ body weight? We came upon the solution when we developed the ‘Diet Switch’ protocol (Fig 2)10.



IP fig2Figure 2: Diet Switch Protocol. After consuming either RC or HFD for 8 weeks, mice were randomly assigned to either undergo a ‘diet switch’ or continue consuming their respective diet. Body weight (Fig 3) and sleep/wake behavior (Fig 4) was measured one week later (Week 9).


In this protocol, we fed half the mice RC and other half of mice HFD. After 8 weeks, the HFD-fed mice were much heavier than RC-fed controls, and displayed the hallmark increases in total sleep and sleep/wake fragmentation (shown in Fig 1). Then subsets of mice from each dietary group had their diets switched. This caused newly-fed HFD mice (RC → HFD group) to rapidly gain weight, while newly-fed RC mice (HFD → RC group) began to quickly lose weight (Fig 3AB). Exactly one week later (Week 9), the two groups of diet switch mice had nearly identical weight (Fig 3C), especially when normalized to their initial weight at start of the study (Fig 3D). We kept some mice on their respective diets (non-diet switch, RC → RC or HFD → HFD groups) as controls. We then measured sleep/wake patterns of all mice at this time point, allowing us to determine the relative influence ‘body weight’ had on sleepiness. If sleep/wake patterns were similar between diet-switch groups, this would support our first hypothesis (that body weight is the primary factor that influences sleepiness) because these groups’ body weights were also similar.



IP fig3Figure 3: Body weights before and after the diet switch protocol. (A) Actual body weight or (B) the percent increase in body weight from Week 0 throughout the study. Notice how before the diet switch (up to and including Week 8), HFD-fed mice weigh considerably more than RC-fed mice, but after the diet switch, these two groups of mice have converging (similar) body weight. (C) Actual body weight or (B) percent increase in body weight from Week 0 at the Week 9 time point (one-week after the diet switch). The two diet switch groups weigh slightly more than RC-fed mice and slightly less than HFD-fed mice, but most importantly, are of similar body weight to each other. +: p<0.1, a: p<0.05, a’’: p<0.001, a’’’: p<0.0001 compared to RC → RC. b: p<0.05, b’: p<0.01 compared to HFD → HFD.


Instead, we found that the two diet switch groups had completely different sleep/wake patterns (Fig 4AB). Mice that were losing weight due to recent consumption of RC (HFD → RC group) had significantly increased total wake time (Fig 4A) and decreased sleep/wake fragmentation (Fig 4B) compared to mice gaining weight on HFD (RC → HFD group), despite being similar in body weight at this time.  In fact, just one week of the diet switch was sufficient to completely induce or rescue the sleep/wake abnormalities. As seen in Fig 4CD, mice that consumed RC for only one week after eight weeks of eating HFD (HFD → RC group) had indistinguishable sleep/wake patterns as mice that consumed RC for the entire study (RC → RC group). Conversely, mice that only ate HFD for the final week of the study (RC → HFD group) had similar total wake times and fragmentation levels observed in mice consuming HFD for nine straight weeks (HFD → HFD group). We also did some advanced statistical analysis (linear regression modeling) to confirm that ‘body weight’ has separate and independent effects on sleep/wake parameters from ‘diet/weight changes’ (not shown here). Therefore, body weight changes (caused by swapping healthy RC diets for unhealthy HFD, or vice versa) determine how sleepy or awake animals are, regardless of the animals’ actual body weight.



IP fig4

Figure 4: Sleep and waking behavior following the diet switch. (A) Total time spent awake and (B) wake fragmentation comparisons between the two diet switch groups at Week 9. Mice losing weight (HFD → RC) spend significantly more time awake and have less fragmentation than mice gaining weight (RC → HFD). (C) Total time spent awake and (D) wake fragmentation for all four groups of mice at Week 9. Notice how both metrics of wakefulness are very similar between groups that consumed the same diet for the final week of the study (Week 8 → Week 9), regardless of the diet consumed the previous eight weeks. * p<0.05.


A common question we receive is whether it’s the diet (RC vs HFD) or the weight changes (weight loss vs weight gain) that are really driving the sleep/wake effects. Unfortunately, our study alone is unable to differentiate between these two variables; our study can only claim that body weight (lean vs obese) is not the crucial factor. However, a study using obesity-resistant and obesity-susceptible rodents suggests that weight changes, not the diet per se, may be the most important factor11. Just like humans, some animals respond to a HFD by gaining weight rapidly (obesity-susceptible), while others do not gain weight as quickly or severely (obesity-resistant). This study found that obesity-susceptible rats exhibited the hallmark sleep/wake impairments (hypersomnolence and increased fragmentation) compared to obesity-resistant rats, even though all animals were consuming the exact same diet11. Thus, taken together, these results suggest that weight changes may play the strongest role in determining our daytime sleepiness. If true, weight loss via exercising should have similar beneficial effects to eating healthy and/or consuming fewer calories, although this has yet to be confirmed.

So what does this mean for the general public? First and foremost, it should be stressed that this study was performed in mice, so follow-up studies in humans are required in order to validate these findings. But if the results hold, it would indicate that dieting and/or eating healthy for only a short period of time (one week) could improve daytime sleepiness, even before your ‘target weight’ is achieved. In fact, a recent study found that obese people who only lost 5% of their body weight had many aspects of their peripheral health restored, such as pancreatic beta-cell function and insulin sensitivity in liver, adipose, and skeletal tissues12. Although the link between these studies has yet to be established, there is accumulating evidence that for overweight or obese individuals, initiating weight loss may be sufficient to improve overall health.

As stated in the introduction, there are many causes of daytime sleepiness including insomnia, sleep apnea, and, of course, restless newborns. While our work indicates that weight loss should improve daytime sleepiness, it is important to note that this may not work for everyone. For example, complete food restriction in lean mice greatly increases wakefulness and causes insomnia-like symptoms (unpublished observations). Therefore, weight loss below a target weight may promote wakefulness in a detrimental way. This current study barely scratches the surface towards understanding how diet/weight changes affect sleep and wakefulness. As research on this topic continues, we hope to better inform the public on dietary strategies, including which foods to eat and when to eat them, to best promote quality sleep.

All in all, many Americans are overweight/obese and suffer from excessive daytime sleepiness, two factors that can heavily impact one another. We used regular mice and two diets—a ‘healthy’ RC and an ‘unhealthy’ HFD—to modify their weight while measuring their sleep/wake profile. By using our ‘Diet Switch’ protocol, we found that only one-week consumption of HFD or RC was sufficient to completely cause or reverse sleep/wake impairments, respectively, regardless of whether the animals were lean or obese the week prior. Therefore, cutting calories and/or eating healthier can stimulate weight loss and improve daytime wake impairments, even before all of the excess weight has been lost. And if someone feels less tired during the day, it may encourage a healthier, more active lifestyle, which could perpetuate weight loss even further.



  1. Hirshkowitz M, Whiton K, Albert SM, et al. National Sleep Foundation’s sleep time duration recommendations: methodology and results summary. Sleep Health 2015;1:40-43.
  2. Centers for Disease Control and Prevention (CDC). (Mar 2011). Unhealthy Sleep-Related Behaviors—12 States, 2009. MMWR Morb Mortal Wkly Rep. Retrieved from:
  3. U.S. Department of Transportation (Mar 2011). Traffic Safety Facts: Drowsy Driving (DOT HS 811 449). NHTSA Nat Highway Traff Safe Admin. Retrieved from:
  4. Kessler RC, Berglund PA, Coulouvrat C, et al. Insomnia and the Performance of US Workers: Results from the America Insomnia Survey. Sleep 2011;34(9):1161-1171.
  5. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011-2012. JAMA 2014;311:806–14.
  6. Taheri S. The link between short sleep duration and obesity: we should recommend more sleep to prevent obesity. Arch Dis Child 2006;91(11):881-4.
  7. Panossian LA, Veasey SC. Daytime sleepiness in obesity: mechanisms beyond obstructive sleep apnea–a review. Sleep 2012;35:605–15.
  8. Vgontzas AN, Bixler EO, Tan W-L, et al. Obesity without sleep apnea is associated with daytime sleepiness. JAMA Inter Med 1998;158(12):1333-7.
  9. Mavanji V, Billington CJ, Kotz CM, Teske JA. Sleep and obesity: a focus on animal models. Neurosci Biobehav Rev 2012;36:1015–29.
  10. Perron IJ, Pack AI, Veasey SC. Diet/Energy Balance Affect Sleep and Wakefulness Independent of Body Weight. Sleep 2015;38(12):1893-903.
  11. Mavanji V, Teske JA, Billington CJ, Kotz CM. Elevated sleep quality and orexin receptor mRNA in obesity-resistant rats. Int J Obes (Lond) 2010;34:1576–88.
  12. Magkos F, Fraterrigo G, Yoshino J, et al. Effects of Moderate and Subsequent Progressive Weight Loss on Metabolic Function and Adipose Tissue Biology in Humans with Obesity. Cell Metab 2016; [epub ahead of print].



Multiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier SchönmannMultiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier Schönmann