Abstract
Study Objectives:
The relationship between time-limited eating (TLE) and sleep quality is a topic of growing interest in the field of chronobiology. Data in adult cohorts shows that TLE may improve sleep quality, but this has not been evaluated in adolescents. The aim of this secondary analysis was to (1) examine the impact of 8-hour TLE on sleep parameters in youth with obesity and (2) explore if there was any association between sleep patterns and glycemic profiles.
Methods:
Adolescents with obesity were randomized into one of three groups: 8-hour TLE (participants self-selected their eating window) + real-time continuous glucose monitor, 8-hour TLE + blinded continuous glucose monitor, or a prolonged eating window. In the primary analysis, it was found that participants in the real-time continuous glucose monitor group + 8-hour TLE group did not access their continuous glucose monitor data and thus for this analysis the two TLE groups were combined and only completers who had available Pittsburgh Sleep Quality Index (PSQI) data at all three time points were included. Participants completed the PSQI at baseline, week 4, and week 12. Mixed-effects generalized linear regression models were utilized to examine the change in PSQI score and assess association between glycemic variability and PSQI total score overtime by intervention arm.
Results:
The median PSQI total score for the TLE groups (n = 27) was 6 at week 0 (interquartile range = 5 to 10) and 5 at week 12 (interquartile range = 2 to 7). There was no significant difference in the change in total PSQI score or sleep latency between TLE and control over the study period (P > .05). There was no association between PSQI score and change in weight or glycemic profile between groups (all P values > 0.05).
Conclusions:
These results suggest that in adolescents with obesity, an 8-hour TLE approach did not negatively impact sleep quality or efficiency when compared to a prolonged eating window. The potential effects of TLE on sleep should be further investigated in larger randomized trials.
Citation:
Jayakumr A, Gillett ES, Wee CP, Kim A, Vidmar AP. Impact of 8-hour time-limited eating on sleep in adolescents with obesity. J Clin Sleep Med. 2023;19(11):1941–1949.
Keywords: sleep, obesity, adolescent, endocrinology, nutrition, PSQI, TLE, quality, glucose
BRIEF SUMMARY
Current Knowledge/Study Rationale: There is a paucity of data investigating the use of time-limited eating in adolescents with obesity. Sleep-related issues are more common in adolescents with obesity and thus it is imperative to understand how adjusting the timing of eating affects sleep in this high-risk cohort.
Study Impact: As part of a secondary analysis to a 12-week pilot study of time-limited eating in youth with obesity, sleep quality was assessed over time to determine if adjusting the eating windows affected sleep. The results highlight that 8-hour time-limited eating does not negatively impact sleep in youth with obesity when compared to a prolonged eating window.
INTRODUCTION
Rising prevalence rates of childhood obesity have been accompanied by an increasing incidence of life-limiting obesity-related comorbidities.1 Pediatric guidelines have historically focused on nutrition quality and quantity and physical activity intensity and duration.1 There has been growing interest in simplifying interventional approaches that can be implemented across all communities without requiring access to additional resources. One promising approach is based on the timing of eating.2–4
Intermittent fasting regimens emerged from chronobiology and circadian science findings indicating that mistimed eating is associated with metabolic dysfunctions, while aligning circadian rhythms with peripheral neuroendocrine biology may improve health outcomes overtime. Time-restricted eating, or time-limited eating (TLE), refers to the practice of confining meals to a specific time window, usually 6 to 10 hours, while fasting for the remaining hours of the day.5,6 In adults, TLE has been shown to have various health benefits, including weight loss and improved glucose homeostasis.7 It has been hypothesized that the metabolic effects of TLE result from synchronization of the central circadian and peripheral molecular clocks, as well as a decrease in energy intake; thus there may be a relationship between TLE regimens and sleep patterns.8,9
The relationship between TLE and sleep quality is a topic of growing interest in the field of chronobiology.9,10 The results of TLE in adults appear to depend on when eating occurs, with evidence to support that early TLE may be more effective at improving cardiometabolic risk factors than a late eating window.11–14 The timing of the eating window is pertinent in relationship to sleep in that previous evidence has shown that eating later in the evening may change sleep architecture.7 There is growing evidence investigating the effects of TLE on sleep in adult cohorts. Two studies have shown that 10-hour TLE improves morning restlessness and sleep quality after 12 and 16 weeks, respectively.15,16 Several studies also showed that neither 4-hour, 6-hour, 8-hour, nor 9-hour TLE resulted in change in sleep quality or duration.5,6,17,18 In addition, neither 4- nor 6-hour TLE affected insomnia severity or risk of obstructive sleep apnea.5
Adolescence is a critical period of rapid physical and psychological development, and many adolescents struggle with poor sleep patterns and quality.19,20 There is growing evidence to suggest that insufficient sleep is associated with unhealthy nutrition status, processed food consumption, sedentary behaviors, and worse health outcomes across the lifespan. Data shows that for each hour of lost sleep in this population, the odds of obesity increased by 80%.21–24 Therefore, it is essential when investigating the use of TLE in youth to examine if adjusting the eating window affects sleep patterns. Given the limited data on intermittent fasting in youth with obesity, to date there is no data investigating how 8-hour TLE affects sleep in this age group.
A 12-week pilot study was completed to evaluate the safety and feasibility of 8-hour TLE vs a prolonged eating window in adolescents with obesity. The primary results from this study showed that the 8-hour TLE was safe, feasible, and acceptable in this age group.25 During this pilot study, sleep quality was monitored over the course of 12 weeks using the Pittsburg Sleep Quality Index (PSQI) questionnaire. The aim of this secondary analysis was to (1) examine the impact of an 8-hour eating window on sleep parameters in youth with obesity and (2) explore if there was any association between sleep patterns and glycemic profiles in this age group when compared to a prolonged eating window. Previous data suggest youth experience a shift in their chronobiology resulting in the consumption of the majority of their food later in the evening and night; therefore we hypothesize that 8-hour TLE will not negatively impact sleep quality in adolescents with obesity and that defining a specific eating window that concludes between 7 and 8 pm may actually improve sleep quality in this cohort.25,26
METHODS
Study design of primary study protocol
This is a secondary analysis of data collected as part of a 12-week randomized, parallel-arm, feasibility trial comparing the effects of 8-hour TLE with real-time continuous glucose monitoring (CGM) and 8-hour TLE with blinded CGM to a prolonged eating window (no meal timing restrictions) in adolescents with obesity.22,25,26, Inclusion criteria were as follows: age 14–18 years; obesity as defined as a body mass index (BMI) greater than the 95th percentile; participant and/or parent/guardian or family member had a personal smartphone and/or was willing to come to the study center for manual data upload monthly for the study duration; and participant was willing and able to adhere to the assessments, visit schedules, and eating/fasting periods. Exclusion criteria included documented diagnosis of Prader-Willi syndrome, type 2 diabetes, brain tumor, hypothalamic obesity, binge eating disorder, serious developmental or intellectual disability, or previously diagnosed eating disorder. Those who were unable or unwilling to complete study assessments, were enrolled in a weight loss intervention or previously underwent bariatric surgery, and/or were taking weight-altering medication were also excluded. The Children’s Hospital Los Angeles Institutional Review Board approved the study protocol, and all research participants and their parent or guardian gave their written informed consent to participate in the trial. In summary, the trial consisted of a 12-week intervention period, with three study visits total conducted at week 0 (baseline), week 4, and week 12. Adolescents self-selected their 8-hour eating window with the majority selecting an afternoon eating window, between 11 am and 7 pm ± 1 hour, and wore a continuous glucose monitor daily. They were instructed to eat ad libitum during the 8-hour eating window and then fast for 16 hours each day. Adolescents were not required to monitor caloric intake or macronutrients. During the fasting window, adolescents were permitted to consume energy-free beverages.
Sleep measures
For the secondary analysis, to investigate the effect of TLE on sleep, participants who had available PSQI data at all three time points were included. Youth completed the PSQI questionnaires at baseline and weeks 4 and 12 via Research Electronic Data Capture. The PSQI is widely used in research and in clinical settings as a measure of sleep quality.27,28 The PSQI is a self-rated questionnaire that assesses sleep quality over the course of a one-month period. The PSQI score is calculated from seven components of sleep quality (derived from 19 individual items) that are rated from 0 to 3, yielding a global score ranging from 0 to 21, where the higher score indicates worse sleep. The seven components of sleep quality include measures of self-reported sleep quality, sleep onset latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medications, and daytime dysfunction. A score of 7 or higher indicates poor sleep quality. Components of the PSQI were used to assess details of sleep monitoring including sleep efficiency, prolonged sleep latency, and daytime function. Sleep efficiency was calculated by comparing the self-reported hours spent sleeping compared to the overall hours participants spent in bed. Prolonged sleep onset latency was also extrapolated by responses to the PSQI question regarding whether they were able to fall asleep within 30 minutes. Difficulty with daytime function relative to sleep was evaluated by questions in the PSQI regarding enthusiasm during the day and difficulty staying awake during social activity.27,29
Body weight measures
Adolescents’ height and weight were collected by the adolescent and parent/guardian at home with the research coordinator monitoring the measurement collection via a virtual platform during a research visit. Adolescents wore minimal clothing during the height and weight measurements. Height was measured using a portable wall height indicator tape ruler, accurate to 0.5 cm (Posh Rulers, Quick Medical, Issaquah, WA), and weight was measured on a self-calibrating Bluetooth enabled Etekcity Digital Body Weight Scale, accurate to 0.2 kg (Etekcity, San Diego, CA). BMI in excess of the 95th percentile was determined utilizing the Centers for Disease Control and Prevention growth charts.
Glycemic profiles
Glycemic profiles were captured with CGM worn continuously for the duration of the study. CGM data was downloaded weekly by the research team. Upon completion of the study, several measures from the standard ambulatory glucose profile were computed including mean, maximum, and minimum glucose levels; standard deviation of glucose; mean amplitude of glycemic excursion; and overall percent of total time spent in euglycemic range (percent time in blood glucose range = 70–140 mg/dL).30,31
Statistical analysis
Data are described in mean and standard deviation or median and interquartile range for continuous variables, frequency, and percentage for categorical variables. Differences in distribution of outcome variables and other sleep key variables at baseline and demographics between study groups were evaluated using chi-squared or Fisher’s exact test for categorical variables, and Wilcoxon rank-sum test or two-sample t-test for continuous variables. A nonadditive effect was assessed using mixed-effects generalized linear model with gamma or ordinal distribution to evaluate whether the difference in TLE and control varied over the course of study period on PSQI score, sleep ratio (sleep efficiency), sleep latency, components of sleep, and daily function by including an interaction term into the models. Questions in the PSQI regarding self-reported bedtime and waketime were compared against the individual’s total sleep time to obtain general sleep efficiency. The results are described in percent change or odds ratio with their associated 95% confidence interval, unadjusted and adjusted P values for multiple testing correction using Simes’ false discovery rate. A two-tailed P-value of less than 0.05 was considered statistically significant. All statistical analysis were performed using Stata/SE 17.0 (StataCorp, College Station, TX, USA).
RESULTS
Adolescent baseline characteristics
As previously reported, 83 adolescents were assessed for eligibility. A total of 50 adolescents were randomized into the 8-hour TLE + real-time CGM (n = 16), 8-hour TLE + blinded CGM (n = 19), or the control group (n = 15). A total of five adolescents withdrew from the study (TLE = 3, control = 2) due to scheduling conflicts, new-onset type 2 diabetes, or pregnancy.30 In the primary analysis, it was found that participants in the real-time CGM + 8-hour TLE group did not access their data, and thus for the secondary analysis the two TLE groups were combined, and only completers were included that had available PSQI data at all three time points (TLE = 27 and control = 15). In total, 27 adolescents from the TLE groups and 15 adolescents from the control group had complete PSQI data for analysis (Table 1). At baseline, there were no significant differences between groups for any demographic features and no difference between completers and noncompleters.
Table 1.
Demographics, body weight, and sleep variables at baseline during the 12 weeks of TLE.
| Total (n = 42) | Control (n = 15) | TLE (n = 27) | P | |
|---|---|---|---|---|
| Age (in years), mean (SD) | 16.34 (1.17) | 16.09 (1.09) | 16.45 (1.20) | 0.08a |
| Sex, n (%) | 0.5b | |||
| Female | 30 (74) | 10 (67) | 19 (71) | |
| Male | 12 (38) | 5 (33) | 8 (29) | |
| Ethnicity, n (%) | 0.30b | |||
| Non-Hispanic | 11 (26) | 5 (33) | 6 (21) | |
| Hispanic | 30 (70) | 9 (60) | 20 (76) | |
| Missing | 1 (2) | 1 (6.7) | 1 (3) | |
| Average glucose, median (IQR) | 108.9 (104.7–121.4) | 110.4 (106.4–120.2) | 109.87 (103.5–121.4) | 0.79c |
| Average estimated hemoglobin A1C, mean (SD) | 5.8 (1.0) | 5.6 (0.4) | 5.8 (1.2) | 0.73a |
| PSQI Data | Total (n = 43) | Control (n = 14) | TLE (n = 29) | |
| Sleep duration (h) | 7 (1.8) | 7.54 (1.6) | 6.75 (1.8) | 0.2a |
| Effect of keeping enthusiasm in getting things done, n (%) | 0.8b | |||
| None during the past month | 18 (34.0) | 6 (37.5) | 12 (32.4) | |
| Less than once a week | 10 (18.9) | 4 (25.0) | 6 (16.2) | |
| Once or twice a week | 7 (13.2) | 1 (6.25) | 6 (16.2) | |
| Three or more times a week | 7 (13.2) | 2 (12.5) | 5 (13.5) | |
| Missing | 11 (20.8) | 3 (18.8) | 8 (21.6) | |
| Effect on staying awake while driving, eating meals, and social activities, n (%) | 0.5b | |||
| None during the past month | 33 (62.3) | 9 (56.3) | 24 (64.9) | |
| Less than once a week | 8 (15.1) | 4 (25.0) | 4 (10.8) | |
| Once or twice a week | 1 (1.9) | 0 (0) | 1 (2.7) | |
| Three or more times a week | ||||
| Missing | 11 (20.8) | 3 (18.8) | 8 (21.6) |
aTwo-sample t-test. bFisher’s exact test. cWilcoxon rank-sum test. IQR = interquartile range, PSQI = Pittsburgh Sleep Quality Index, SD = standard deviation, TLE = time-limited eating.
Sleep measures
An overall PSQI score greater than 7 indicates poor sleep quality.27,28 In these cohorts, the median PSQI total scores at baseline were 6 [interquartile range (IQR) = 5 to 10 for TLE and 5 (IQR = 4 to 10) for controls] (Table 2). Across the total cohort, with univariate analysis there was a significant linear trend that change in PSQI score was decreasing over time (Wald test = 6.04 on 1df, P = .01). The median PSQI total score for the TLE groups (n = 27) was 6 at week 0 (IQR = 5 to 10) and 5 at week 12 (IQR = 2 to 7). There was no significant difference in the change in total PSQI score or sleep latency between TLE and controls over the study period (both unadjusted P > .05 and multiple testing adjusted P > .05). There was an increase in the sleep efficiency scores over the course of the 12-week study period in the TLE group from 0.86 (IQR = 0.78 to 1) to 0.95 (IQR = 0.84 to 1), whereas sleep efficiency scores decreased from 1 (IQR = 0.89 to 1.6) to 0.86 (IQR = 0.75 to 1) in the control group at the end of study (unadjusted P = .04). However, after adjusting for multiple testing, there was no significant difference in these changes between groups (adjusted P = .07, Table 3). There was no significant association of TLE on responses to the questions with regards to staying awake for daytime activities or enthusiasm for completing these activities with and without multiple testing (adjusted interaction P = .22 and adjusted interaction P = .99, respectively).
Table 2.
Change in PSQI score, sleep efficiency, and sleep latency over study period between TLE and control.
| Control (n = 15) | TLE (n = 27) | Difference over Study Period between Control and TLE | ||
|---|---|---|---|---|
| Median (IQR) | Median (IQR) | Unadjusted P | Adjusted P* | |
| PSQI score | 0.7 | 0.7 | ||
| Pre | 5 (4–10) | 6 (5–10) | ||
| Week 4 | 4 (2–13) | 6 (3–9) | ||
| Week 12 | 7 (4–9) | 5 (2–7) | ||
| Sleep ratio (sleep efficiency) | 0.01 | 0.04 | ||
| Pre | 1 (0.9–1.6) | 0.9 (0.8–1) | ||
| Week 4 | 0.9 (0.7–1) | 0.9 (0.7–1) | ||
| Week 12 | 0.8 (0.8–0.9) | 1 (0.8–1) | ||
| Sleep latency | 0.08 | 0.1 | ||
| Pre | 1 (0–2) | 2 (0–3) | ||
| Week 4 | 1 (0–2) | 1 (0–2.5) | ||
| Week 12 | 1 (0.5–2.5) | 1 (0–2) | ||
Adjusted P for multiple testing using Simes’ false discovery rate. IQR = interquartile range, PSQI = Pittsburg Sleep Quality Index, TLE = time-limited eating.
Table 3.
Univariate analysis: mixed-effects generalized linear model based on gamma or ordinal distribution on total PSQI score, sleep efficiency, and sleep latency over the study period between groups.
| % Change | 95% CI | Unadjusted P | Adjusted P* | |
|---|---|---|---|---|
| PSQI score | ||||
| Time | ||||
| Pre | Ref | – | – | – |
| Week 4 | −13.3 | −261.5 | 0.08 | 0.2 |
| Week 12 | −19.5 | −31.8, −5.0 | 0.01 | 0.06 |
| Intervention group | ||||
| Control | Ref | – | – | – |
| TLE | 8.1 | −28.5, 63.4 | 0.7 | 0.9 |
| Sleep ratio (sleep efficiency) | ||||
| Time | ||||
| Pre | Ref | – | – | – |
| Week 4 | −15.0 | −27.6, –0.3 | 0.05 | 0.1 |
| Week 12 | −1.6 | −16.6, 16.1 | 0.9 | 0.9 |
| Intervention group | ||||
| Control | Ref | – | – | – |
| TLE | −3.6 | −26.8, 26.2 | 0.8 | 0.9 |
| Sleep latency | Odds Ratio | 95% CI | Unadjusted P | Adjusted P * |
| Time | ||||
| Pre | Ref | – | – | – |
| Week 4 | 0.9 | 0.3, 2.4 | 0.8 | 0.9 |
| Week 12 | 0.9 | 0.3, 2.6 | 0.9 | 0.9 |
| Intervention group | ||||
| Control | Ref | – | – | – |
| TLE | 1.2 | 0.1, 14.5 | 0.9 | 0.9 |
Adjusted P for multiple testing using Simes’ false discovery rate. CI = confidence interval, PSQI = Pittsburg Sleep Quality Index, TLE = time-limited eating.
Association of sleep measures and glycemic profiles
The standard 13 ambulatory glucose profile metrics were captured on the CGM data and evaluated over the study period across groups. On mixed-effects generalized linear model there was no significant association between total PSQI score and average glucose, estimated hemoglobin A1c, and standard deviation for either 24-hour values or day and nighttime values (Table 4).
Table 4.
Association between total PSQI score and glycemic profile as captured with the continuous glucose monitor as assessed by a mixed effects generalized linear model by intervention arm over the study period.
| Global PSQI | % Change | 95% CI of % Change | Unadjusted P | Adjusted P* |
|---|---|---|---|---|
| Average glucose | −3.1 | −7.9, 1.9 | 0.2 | 0.7 |
| Intervention group | ||||
| Control | Ref | – | – | – |
| TLE | −94.5 | −99.9, 5,405.0 | 0.4 | 0.7 |
| Average glucose × intervention group | 2.3 | −3.5, 8.7 | 0.5 | 0.7 |
| Standard deviation | 8.0 | −2.9, 19.9 | 0.2 | 0.7 |
| Intervention group | ||||
| Control | Ref | – | – | – |
| TLE | 50.6 | −85.4, 1,537.3 | 0.7 | 0.8 |
| Standard deviation × intervention group | −5.7 | −17.4, 6.1 | 0.3 | 0.7 |
| Estimated hemoglobin A1c | −59.3 | −90.4, 73.2 | 0.2 | 0.7 |
| Intervention group | ||||
| Control | Ref | – | – | – |
| TLE | −98.1 | −100.0, 2,9876.8 | 0.4 | 0.7 |
| Estimated hemoglobin A1c × intervention group | 90.8 | −66.4, 982.6 | 0.5 | 0.7 |
| Daytime average glucose | −2.7 | −7.6, 2.6 | 0.3 | 0.7 |
| Intervention group | ||||
| Control | Ref | – | – | – |
| TLE | −93.0 | −100.0, 7108.2 | 0.5 | 0.7 |
| Daytime average glucose × intervention group | 2.0 | −4.0, 8.5 | 0.5 | 0.7 |
| Daytime standard deviation | 5.0 | −3.1, 13.8 | 0.2 | 0.7 |
| Intervention group | ||||
| Control | Ref | – | – | – |
| TLE | −4.3 | −88.5, 694.9 | 1.0 | 1.0 |
| Daytime standard deviation × intervention group | −3.3 | −12.6, 6.9 | 0.5 | 0.7 |
| Nighttime average glucose | −3.3 | −7.3, 0.9 | 0.1 | 0.7 |
| Intervention group | ||||
| Control | Ref | – | – | – |
| TLE | −88.2 | −100.0, 7771.7 | 0.5 | 0.7 |
| Nighttime average glucose × intervention group | 1.6 | −4.1, 7.6 | 0.6 | 0.7 |
| Nighttime standard deviation | 7.4 | −16.9, 38.9 | 0.6 | 0.7 |
| Intervention group | ||||
| Control | Ref | – | – | – |
| TLE | 48.7 | −94.9, 4265.2 | 0.8 | 0.9 |
| Nighttime standard deviation × intervention group | −7.0 | −28.4, 20.7 | 0.6 | 0.7 |
Adjusted P for multiple testing using Simes’ false discovery rate. CI = confidence interval, PSQI = Pittsburg Sleep Quality Index, TLE = time-limited eating.
Summary of primary outcomes
See Vidmar et al.30
TLE feasibility and adherence
As reported previously, adherence to the 8-hour TLE window was excellent with youth assigned to the TLE group adhering 6.1 days/week (standard deviation = 1.2). In this cohort, youth self-selected the timing of their eating window and the majority selected an afternoon/evening eating window (52% eating between 11 am and 7 pm).
Dietary intake and physical activity
For all adolescents across both TLE groups and controls there was a decrease in daily caloric intake of about 375 calories/d with no significant difference in the quality of quantity of caloric reduction between TLE and control. Physical activity increased on average 1 day per week (range: −1 to 3 days/week) across all youth with no significant difference between groups.
Change in BMI in excess of the 95th percentile
After 12 weeks, there was a significant decrease in BMI in excess of the 95th percentile across all three groups: 8-hour TLE real-time CGM (−4.9 ± 5.1); 8-hour TLE blinded CGM (−3.8 ± 5.8); and control (−3.2 ± 3.34) with no significant difference in between groups (P = .7) (Figure 1).
Figure 1. The distribution of Pittsburg Sleep Quality Index score, sleep ratio (sleep efficiency), and body mass index in excess of the 95th percentile from baseline to week 12 by study groups (control: n = 15; time-limited eating: n = 27).
DISCUSSION
This study is the first to examine the effect of 8-hour TLE on sleep in adolescents with obesity. Given the novelty of a TLE approach in youth, it is imperative to assess how this nutrition approach affects other components of daily function and ensure no negative compensatory behaviors develop in response to adjusting the eating window. Consistent with our hypothesis, 8-hour TLE did not have a negative impact on sleep quality when compared to a prolonged eating window in adolescents with obesity, and the median PSQI score in the TLE group significantly decreased, which may suggest a trend toward improved sleep quality over the study period.
Studies in cohorts of adults with obesity have shown that weight loss improves sleep; however, it remains unclear the minimum amount of weight loss required to see these changes and how weight loss leads to sleep-related improvements independent of improvements in obesity-related comorbidities such as sleep apnea.32–34 There has been little consensus in studies of adult cohorts regarding the effects of TLE on sleep quality or duration. In some groups, TLE seems to have little effect on sleep, while others report that TLE can result in slight improvement in sleep quality. A recent study by Cienfuegos et al showed that neither 4-hour nor 6-hour TLE affected sleep quality, and there was no association between weight loss and sleep quality over the study period.5 Some postulate that the lack of association between TLE-induced weight loss and improvement in sleep quality may be due the low amount of weight loss reported with TLE interventions, which is also consistent with the low level of weight loss seen in this pediatric study.5,6,15,17,18
It also remains unclear how the exact timing of the eating window may affect sleep in TLE approaches. This is a particularly important question in the adolescent population because of the natural delay in circadian phase that occurs during adolescence. In adolescence, an increased resistance to sleep pressure contributes to a natural circadian phase delay that provides a drive to stay awake later in the evening and to sleep later in the morning.35 This shift in the circadian clock does not fit well with conventional school schedules and can contribute to insufficient sleep. In adolescents, delayed sleep phase syndrome and insufficient sleep associated with pubertal change-related circadian rhythm dysregulation are very common and often coexist.36 Some report that more than 45% of adolescents have insufficient sleep in the United States37 and the prevalence of delayed sleep phase syndrome in this population is 3.3% with 57% overlap between delayed sleep phase syndrome and insomnia.36
Interestingly, in our cohort, 89% of adolescents chose an afternoon/evening eating window from 11 am to 8 pm. This TLE window is relatively delayed compared to adult populations who underwent similar TLE studies.6 The increased propensity for insomnia, insufficient sleep, and delayed sleep phase syndrome provide some insight into the self-selection of an afternoon/evening eating window by this cohort. Nighttime eating may be associated with metabolic syndrome.33 Future studies evaluating this relationship are important to help us understand if simple interventions such as prescribing an end time for an eating window in these obese adolescents that extends the time between last meal and sleep onset may decrease nighttime eating and promote improved sleep quality and duration.38
While not statistically significant after multiple computation, there was a trend toward improvement in sleep efficiency over the course of the 12-week study period in the TLE group compared to the control group. The control group had 100% sleep efficiency at baseline and fell to the lower limit of normal at the end of the study, generally held to be >85%. However, in this age group, the definition of sleep efficiency is challenging. Traditionally, sleep efficiency is calculated by total sleep time divided by time spent in bed. With adolescents, we know that they spend more time participating in nonsleep-related activities in bed prior to falling asleep such as reading, texting, and watching television.39,40 It has been proposed that alternative denominators of the calculation of sleep efficiency can more accurately define sleep efficiency; these include specifics of sleep onset latency, wake after sleep onset, and time attempting to sleep after final awakening. Despite the likely inaccuracy without an objective polysomnogram, both TLE and control groups have sleep efficiencies within the normal range, suggesting that TLE did not significantly worsen sleep efficiency. High-carbohydrate meals may impair sleep quality and potentially increase the number of awakenings at night and reduce the amount of slow wave N3 sleep.38
Food choices were not regimented for adolescents in this study but could be an interesting aspect of nutrition to explore in future studies.
Finally, as pediatric obesity prevalence continues to rise, there is increased interest in better understanding how early metabolic changes can predict disease onset over the lifespan.40 This is pertinent for youth onset type 2 diabetes given the increasing prevalence, aggressive disease course, and life-limiting complications that can ensue with a youth-onset diagnosis.41 Continuous glucose monitors now provide scientists and clinicians an opportunity to track glycemic variability in youth at risk for diabetes prior to the development of the disease to better understand both how and when the onset occurs and how interventions may impact the disease course.30,31 Previous adult data has shown a relationship between obesity, sleep deprivation, insulin resistance, and reduced glucose tolerance and thus an increased risk of diabetes.42 Multiple mechanistic explanations have been postulated to account for these associations including increased sympathetic nervous system activity, increased cortisol and growth hormones levels leading to increased insulin resistance, and upregulation of orexin.43,44 While it remains unknown if these same mechanisms apply to youth-onset diabetes, the data in adults suggests that prolonged fasting reverses some of these findings and thus has the potential to not only improve sleep quality but possibly play a role in decreasing the risk of developing diabetes in this high-risk age group.43,44 We assessed the association of changes in glycemic profiles and sleep in our TLE and control cohorts. Overall, although we found no significant association between change in glycemic variables and sleep measures, it is an important association to consider for future investigation into how TLE approaches can impact short-term glycemic measures and long-term cardiometabolic health overtime in this age group.
Given the persistent and rising prevalence of pediatric obesity globally, there is an urgent need to identify alternative, effective, safe, and feasible nutrition approaches to utilize in this age group.1 Time-based approaches are simple, straightforward, and nonstigmatizing and can be implemented across all communities regardless of resources without requiring significant change to the home and school environments. As with any novel nutrition approach, TLE should be implemented by a care team with experience in the implementation of this intervention in a pediatric cohort to ensure appropriate monitoring and evaluation for any negative compensatory behaviors overtime. TLE may be an approach that can be used intermittently or continuously and alone or in combination with other therapies such as obesity pharmacotherapy or bariatric surgery. This flexibility, especially in pediatric cohorts that have variable academic and extracurricular schedules occurring simultaneously with development changes, require different treatment approaches at different times to promote efficacy and sustained engagement. In addition to its flexibility, another benefit of intermittent fasting is that the treatment is completely free.
Our study has several limitations. First, our sample size was small and designed as a feasibility trial and thus was not powered to identify significant differences in sleep measures between groups over the study period. Second, the PSQI was utilized to capture sleep metrics via self-report, and no actigraphy measures were included that would have provided a more objective measure of sleep and activity over the study course. Third, the PSQI captures sleep patterns over the previous 30 days and thus could only capture eight weeks of intervention data for this short study duration. Longer study durations are needed to assess the relationship between TLE and sleep patterns. Fourth, for this analysis we combined the two TLE groups and did not adjust for any possible confounding effects of participants having access to their real-time CGM data. Our primary outcomes did not show significant differences in primary outcomes between these two groups in that most adolescents in this cohort did not access their CGM data; however, we did not account for that factor in this secondary analysis. Finally, previous pediatric data has demonstrated that timing of physical activity can affect sleep in this age group, and we did not control for or prescribe a certain physical activity regimen, but tracking the effects of exercise on CGM and sleep measures would be fascinating to evaluate in the future.
CONCLUSIONS
This study is the first to examine the effect of 8-hour TLE on sleep in adolescents with obesity. The current results suggest that 8-hour TLE does not have a negative impact on sleep quality when compared to a prolonged eating window and may even improve sleep quality for some youth living with obesity. Further investigation into the relationship between TLE interventions and sleep in youth is required. A well-powered, longitudinal randomized controlled trial that includes objective measures of sleep, such as actigraphy or polysomnography, is needed to fully investigate the effect of TLE on sleep in this age group.
DISCLOSURE STATEMENT
All authors have seen and approved this manuscript. Research location: Children’s Hospital Los Angeles. The authors report no conflicts of interest.
ABBREVIATIONS
- BMI
body mass index
- CGM
continuous glucose monitoring
- PSQI
Pittsburgh Sleep Quality Index
- TLE
time-limited eating
REFERENCES
- 1. Hampl SE , Hassink SG , Skinner AC , et al . Clinical practice guideline for the evaluation and treatment of children and adolescents with obesity . Pediatrics. 2023. ; 151 ( 2 ): e2022060640 . [DOI] [PubMed] [Google Scholar]
- 2. Fanti M , Mishra A , Longo VD , Brandhorst S . Time-restricted eating, intermittent fasting, and fasting-mimicking diets in weight loss . Curr Obes Rep. 2021. ; 10 ( 2 ): 70 – 80 . [DOI] [PubMed] [Google Scholar]
- 3. Duregon E , Pomatto-Watson LCDD , Bernier M , Price NL , de Cabo R . Intermittent fasting: from calories to time restriction . Geroscience. 2021. ; 43 ( 3 ): 1083 – 1092 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Patterson RE , Sears DD . Metabolic effects of intermittent fasting . Annu Rev Nutr. 2017. ; 37 ( 1 ): 371 – 393 . [DOI] [PubMed] [Google Scholar]
- 5. Cienfuegos S , Gabel K , Kalam F , et al . The effect of 4-h versus 6-h time restricted feeding on sleep quality, duration, insomnia severity and obstructive sleep apnea in adults with obesity . Nutr Health. 2022. ; 28 ( 1 ): 5 – 11 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Gabel K , Hoddy KK , Haggerty N , et al . Effects of 8-hour time restricted feeding on body weight and metabolic disease risk factors in obese adults: a pilot study . Nutr Healthy Aging. 2018. ; 4 ( 4 ): 345 – 353 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Andriessen C , Fealy CE , Veelen A , et al . Three weeks of time-restricted eating improves glucose homeostasis in adults with type 2 diabetes but does not improve insulin sensitivity: a randomised crossover trial . Diabetologia. 2022. ; 65 ( 10 ): 1710 – 1720 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Manoogian ENC , Chow LS , Taub PR , Laferrère B , Panda S . Time-restricted eating for the prevention and management of metabolic diseases . Endocr Rev. 2022. ; 43 ( 2 ): 405 – 436 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Simon SL , Blankenship J , Manoogian ENC , Panda S , Mashek DG , Chow LS . The impact of a self-selected time restricted eating intervention on eating patterns, sleep, and late-night eating in individuals with obesity . Front Nutr. 2022. ; 9 : 1007824 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Karan M , Bai S , Almeida DM , Irwin MR , McCreath H , Fuligni AJ . Sleep-wake timings in adolescence: chronotype development and associations with adjustment . J Youth Adolesc. 2021. ; 50 ( 4 ): 628 – 640 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Charlot A , Hutt F , Sabatier E , Zoll J . Beneficial effects of early time-restricted feeding on metabolic diseases: importance of aligning food habits with the circadian clock . Nutrients. 2021. ; 13 ( 5 ): 1405 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Jamshed H , Beyl RA , Della Manna DL , Yang ES , Ravussin E , Peterson CM . Early time-restricted feeding improves 24-hour glucose levels and affects markers of the circadian clock, aging, and autophagy in humans . Nutrients. 2019. ; 11 ( 6 ): 1234 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Jamshed H , Steger FL , Bryan DR , et al . Effectiveness of early time-restricted eating for weight loss, fat loss, and cardiometabolic health in adults with obesity: a randomized clinical trial . JAMA Intern Med. 2022. ; 182 ( 9 ): 953 – 962 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Sutton EF , Beyl R , Early KS , Cefalu WT , Ravussin E , Peterson CM . Early time-restricted feeding improves insulin sensitivity, blood pressure, and oxidative stress even without weight loss in men with prediabetes . Cell Metab. 2018. ; 27 ( 6 ): 1212 – 1221.e3 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Wilkinson MJ , Manoogian ENC , Zadourian A , et al . Ten-hour time-restricted eating reduces weight, blood pressure, and atherogenic lipids in patients with metabolic syndrome . Cell Metab. 2020. ; 31 ( 1 ): 92 – 104.e5 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Gill S , Panda S ; S G . A smartphone app reveals erratic diurnal eating patterns in humans that can be modulated for health benefits . Cell Metab. 2015. ; 22 ( 5 ): 789 – 798 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Zhao L , Hutchison AT , Liu B , et al . Time-restricted eating improves glycemic control and dampens energy-consuming pathways in human adipose tissue . Nutrition. 2022. ; 96 : 111583 . [DOI] [PubMed] [Google Scholar]
- 18. Hutchison AT , Regmi P , Manoogian ENC , Fleischer JG , Wittert GA , Panda S , Heilbronn LK . Time-restricted feeding improves glucose tolerance in men at risk for type 2 diabetes: a randomized crossover trial . Obesity (Silver Spring). 2019. ; 27 ( 5 ): 724 – 732 . [DOI] [PubMed] [Google Scholar]
- 19. Kansagra S . Sleep disorders in adolescents . Pediatrics. 2020. ; 145 ( Suppl 2 ): S204 – S209 . [DOI] [PubMed] [Google Scholar]
- 20. Musshafen LA , Tyrone RS , Abdelaziz A , et al . Associations between sleep and academic performance in US adolescents: a systematic review and meta-analysis . Sleep Med. 2021. ; 83 : 71 – 82 . [DOI] [PubMed] [Google Scholar]
- 21. Zeb F , Wu X , Chen L , et al . Effect of time-restricted feeding on metabolic risk and circadian rhythm associated with gut microbiome in healthy males . Br J Nutr. 2020. ; 123 ( 11 ): 1216 – 1226 . [DOI] [PubMed] [Google Scholar]
- 22. Sluggett L , Wagner SL , Harris RL . Sleep duration and obesity in children and adolescents . Can J Diabetes. 2019. ; 43 ( 2 ): 146 – 152 . [DOI] [PubMed] [Google Scholar]
- 23. Lee JH , Cho J . Sleep and obesity . Sleep Med Clin. 2022. ; 17 ( 1 ): 111 – 116 . [DOI] [PubMed] [Google Scholar]
- 24. Tambalis KD , Panagiotakos DB , Psarra G , Sidossis LS . Insufficient sleep duration is associated with dietary habits, screen time, and obesity in children . J Clin Sleep Med. 2018. ; 14 ( 10 ): 1689 – 1696 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Vidmar AP , Naguib M , Raymond JK , Salvy SJ , Hegedus E , Wee CP , Goran MI . Time-limited eating and continuous glucose monitoring in adolescents with obesity: a pilot study . Nutrients. 2021. ; 13 ( 11 ): 3697 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Naguib MN , Hegedus E , Raymond JK , et al . Continuous glucose monitoring in adolescents with obesity: monitoring of glucose profiles, glycemic excursions, and adherence to time restricted eating programs . Front Endocrinol (Lausanne). 2022. ; 13 : 841838 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Buysse DJ , Reynolds CF 3rd , Monk TH , Berman SR , Kupfer DJ . The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research . Psychiatry Res. 1989. ; 28 ( 2 ): 193 – 213 . [DOI] [PubMed] [Google Scholar]
- 28. Mollayeva T , Thurairajah P , Burton K , Mollayeva S , Shapiro CM , Colantonio A . The Pittsburgh Sleep Quality Index as a screening tool for sleep dysfunction in clinical and non-clinical samples: a systematic review and meta-analysis . Sleep Med Rev. 2016. ; 25 : 52 – 73 . [DOI] [PubMed] [Google Scholar]
- 29. Fontes F , Gonçalves M , Maia S , Pereira S , Severo M , Lunet N . Reliability and validity of the Pittsburgh Sleep Quality Index in breast cancer patients . Support Care Cancer. 2017. ; 25 ( 10 ): 3059 – 3066 . [DOI] [PubMed] [Google Scholar]
- 30. Klonoff DC , Nguyen KT , Xu NY , Gutierrez A , Espinoza JC , Vidmar AP . Use of continuous glucose monitors by people without diabetes: an idea whose time has come? J Diabetes Sci Technol. 2022. : 19322968221110830 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Klonoff DC , Ahn D , Drincic A . Continuous glucose monitoring: a review of the technology and clinical use . Diabetes Res Clin Pract. 2017. ; 133 : 178 – 192 . [DOI] [PubMed] [Google Scholar]
- 32. Carneiro-Barrera A , Amaro-Gahete FJ , Guillén-Riquelme A , et al . Effect of an interdisciplinary weight loss and lifestyle intervention on obstructive sleep apnea severity: the INTERAPNEA randomized clinical trial . JAMA Netw Open. 2022. ; 5 ( 4 ): e228212 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Ryan DH , Yockey SR . Weight loss and improvement in comorbidity: differences at 5%, 10%, 15%, and over . Curr Obes Rep. 2017. ; 6 ( 2 ): 187 – 194 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Tahrani AA , Morton J . Benefits of weight loss of 10% or more in patients with overweight or obesity: a review . Obesity (Silver Spring). 2022. ; 30 ( 4 ): 802 – 840 . [DOI] [PubMed] [Google Scholar]
- 35. Carskadon MA , Acebo C , Jenni OG . Regulation of adolescent sleep: implications for behavior . Ann N Y Acad Sci. 2004. ; 1021 ( 1 ): 276 – 291 . [DOI] [PubMed] [Google Scholar]
- 36. Sivertsen B , Pallesen S , Stormark KM , Bøe T , Lundervold AJ , Hysing M . Delayed sleep phase syndrome in adolescents: prevalence and correlates in a large population based study . BMC Public Health. 2013. ; 13 ( 1 ): 1163 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Hagenauer MH , Perryman JI , Lee TM , Carskadon MA . Adolescent changes in the homeostatic and circadian regulation of sleep . Dev Neurosci. 2009. ; 31 ( 4 ): 276 – 284 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. St-Onge MP , Mikic A , Pietrolungo CE . Effects of diet on sleep quality . Adv Nutr. 2016. ; 7 ( 5 ): 938 – 949 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. St-Onge MP , Pizinger T , Kovtun K , RoyChoudhury A . Sleep and meal timing influence food intake and its hormonal regulation in healthy adults with overweight/obesity . Eur J Clin Nutr. 2019. ; 72 ( Suppl 1 ): 76 – 82 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Ogden CL , Martin CB , Freedman DS , Hales CM . Trends in obesity disparities during childhood . Pediatrics. 2022. ; 150 ( 2 ): e2022056547 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Yang W , Wang C , Zeitler PS . Long-term complications in youth-onset type 2 diabetes . N Engl J Med. 2021. ; 385 ( 21 ): 2014 – 2015 . [DOI] [PubMed] [Google Scholar]
- 42. Antza C , Kostopoulos G , Mostafa S , Nirantharakumar K , Tahrani A . The links between sleep duration, obesity and type 2 diabetes mellitus . J Endocrinol. 2021. ; 252 ( 2 ): 125 – 141 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Almeneessier AS , Alzoghaibi M , BaHammam AA , et al . The effects of diurnal intermittent fasting on the wake-promoting neurotransmitter orexin-A . Ann Thorac Med. 2018. ; 13 ( 1 ): 48 – 54 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Linehan V , Hirasawa M . Short-term fasting induces alternate activation of orexin and melanin-concentrating hormone neurons in rats . Neuroscience. 2022. ; 491 : 156 – 165 . [DOI] [PubMed] [Google Scholar]