Abstract
Objective
We intend to explore whether sleep duration and sleep patterns were associated with obesity among the adult American population.
Methods
Our study analyzed data from 4540 adults (2007–2008, 2015–2018) in the National Health and Nutrition Examination Survey (NHANES) database. Self-reported questionnaires were used to collect data on sleep duration, excessive daytime sleepiness, and sleep problems. The overall sleep pattern score was calculated and categorized into three types: healthy, intermediate, and poor sleep patterns. Multivariable logistic regression models were used to calculate the associations between sleep duration, sleep patterns and obesity. We further conducted linearity tests using restricted cubic splines to explore the dose-response relationship between sleep duration and obesity. Additionally, we performed stratified and interaction analyses to understand if this relationship was stable in different subgroups.
Results
After adjusting for potential confounding factors, sleep duration (odds ratio [OR] = 0.91, 95% confidence interval [CI]:0.87–0.95, P < 0.001) and sleep pattern score (OR = 1.18, 95% CI: 1.1–1.27, P < 0.001) were independently associated with obesity. There was a non-linear relationship between sleep duration and obesity, with a threshold of approximately 9.73 h. The effects and CI below and above the threshold were 0.89 (0.849–0.936) and 2.023 (1.113–3.677), respectively.
Conclusions
Unhealthy sleep patterns and shorter sleep duration were positively correlated with obesity occurrence, and there was a non-linear relationship between sleep duration and the occurrence of obesity. Interventions aimed at promoting healthy sleep habits and appropriate sleep duration may be important in reducing the risk of obesity.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12889-025-22433-9.
Keywords: Sleep duration, Sleep patterns, Obesity, NHANES
Introduction
Obesity was considered a major public health issue and was listed as the fifth leading cause of death globally. According to the World Health Organization, in 2015, approximately 1.9 billion (39%) adults worldwide were overweight, and 609 million (13%) adults were obese. Obesity had reached epidemic levels in Western countries, with a 2017–2018 face-to-face survey of 12 European countries showing that nearly half of the participants (48.1%, 95% CI: 47.2–49.1) reported being overweight or obese (54.1% among men, 42.5% among women), with an obesity rate of 12.6% [1]. A large-scale study from 2019 showed that the prevalence of obesity and overweight in China is 34.8% and 14.1%, respectively [2]. However, the prevalence of obesity in the United States was much higher than the global obesity rate. In the United States, the national prevalence of adult obesity had risen from 30.5% in 1999–2000 to 42.4% in 2017–2018 [3]. It was projected that by 2030, the prevalence of overweight and obesity in the United States will reach 48.9% [4]. Additionally, obesity was becoming increasingly prevalent in developing countries. Over the past 20 years, the obesity rate in developing countries had been rapidly rising. In 2014, there were 641 million obese adults, compared to just 105 million in 1975 [5], showing a staggering increase.
Obesity poses significant risks to the human body, primarily by affecting metabolism and leading to a series of diseases such as insulin resistance, type 2 diabetes, non-alcoholic fatty liver disease, cardiovascular diseases, and cancer. Therefore, it is increasingly important to take proactive measures to prevent and control obesity.
Sleep was a normal biological behavior of the human body and was the foundation for maintaining good physical and mental health. It greatly influenced cardiovascular health, mental health, cognition, memory consolidation, immune function, reproductive health, and hormone regulation [6]. Despite this, many individuals did not achieve a sufficient amount or quality of sleep. As of 2014, 33.8% of the U.S. population did not get the recommended amount of sleep, and 5 to 7 million American adults had various sleep disorders [7].
More and more evidence showed that sleep duration and sleep quality were related to obesity [8]. A meta-analysis [9] included 11 relevant studies for analysis, a random effects model was used to calculate the combined OR and its 95% CI, which ultimately indicated that short sleep duration was significantly associated with incidence of obesity. Another study [10]conducted a random effects meta-analysis and regression analysis by including 30 studies on sleep and waist circumference, which also indicated that shorter sleep durations covary with central adiposity. It may be achieved through a series of pathways such as affecting metabolism, secretion of appetite hormones, and even the function of the nervous system. Our study aimed to establish a sleep pattern score, that was, the cumulative effect of sleep duration and sleep quality, to evaluate the relationship with obesity, in the hope of providing a more comprehensive explanation of the connection between sleep and obesity.
Materials and methods
Data sources
National Health and Nutrition Survey (NHANES) is a national representative survey conducted by the National Center for Health Statistics (NCHS), which aims to assess the health or nutritional status of the noninstitutionalized US population. Participants undergo random household interviews, physical examination, and laboratory tests every two years [11]. The survey’s design, methods, and data are available to the public. The website is https://www.cdc.gov/nchs/nhanes/irba98.htm).Ethical approval and consent were not required as this study was based on publicly available de-identified data.
This study complied with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Study design and population
This cross-sectional study used publicly available data from NHANES during 2007–2008, 2015–2016, and 2017–2018, without personally identifiable information, and in accordance with relevant regulations and guidelines. A total of 29,374 participants (aged 20 years or older) completed the questionnaire and examination. We excluded 182 women who reported being pregnant during the survey. In addition, we excluded participants with missing data on sleep, BMI, and covariates. After removing individuals with missing information from the dataset, a total of 4540 participants (2271 men and 2269 women) with complete interview and examination data were included in the final analysis. Figure 1 shows the flowchart of the exclusion criteria.
Fig. 1.
Flowchart of the study cohort
Sleep patterns
Sleep duration data was collected during the health interview process using a self-reported questionnaire (SLD010H). Each participant was asked the computer-assisted personal interview question, “How much sleep do you usually get at night on weekdays or workdays?“. Building on previous research [12], this study further categorized sleep duration as short (Less than 7 h ), normal (7–9 h per night), or long (9 h or more per night). In addition, sleep problems were measured by specific questions (SLQ050): “Have you ever told a doctor you had trouble sleeping?” and “How often do you feel overly sleepy during the day?” The response to SLQ120(How often feel overly sleepy during day? ) was categorized as usual (less than or equal to 4 times per month) or sleepy (more than 4 times per month). Refer to the previous relevant literature [13], we established a sleep pattern score using the above three sleep-related behaviors, that including the evaluation of sleep duration and sleep quality, to generate an overall sleep pattern score, normal and abnormal aspects of sleep behavior were classified as 0 and 1, respectively. These factors related to sleep behavior resulted in scores ranging from 0 to 3, if the score was 0, it indicated a healthy sleep pattern. If the score was 1, it meaned that one of the three abnormal sleep behaviors was present, indicating a moderate sleep pattern. If the score was between 2 and 3, it indicateed a poor sleep pattern(see supplementary materials).
Obesity
Height and weight data were obtained by physical examination, from which the body mass index (BMI )was calculated. When BMI ≥ 30 kg /m2, defined as obesity [14–16].
Covariate
We collected covariate data through questionnaires, physical examination, and laboratory tests in the NHANES. Mainly include age, sex, family income, marital, race and ethnicity, educational level, smoking status, alcohol drinking status, hypertension, diabetes, physical activity, Energy intake. Race and ethnicity was categorized as non-Hispanic White, non-Hispanic Black, Mexican American, or other races. Marital status was classified as married or living with a partner, and living alone. Education level were divided into three levels(<9(High school or less), 9–12(Some college), or > 12 years(College or above)). Smoking status was categorized as never smoked ((or smoked < 100 cigarettes), current smoker (smoked ≥ 100 cigarettes but already quit), or former smoker based on previously reported definitions. Participants who answered “yes” to the following question were classified as alcohol drinkers, “Had at least 12 alcohol drinks in one year or Ever had a drink of any kind of alcohol?” The 24-h total energy intake was obtained by the dietary review interview. The presence of hypertension was determined by the following questionnaire(Ever told you had high blood pressure and Told had high blood pressure more than 2 times), that could be considered with hypertension only if both answer yes, otherwise, there is no hypertension. Diabetes was based on the inquiry in the questionnaire of whether the doctor had been informed of the condition in the past. In accordance with the WHO analysis guide, PA was converted to metabolic equivalent (MET) minutes of moderate to vigorous PA per week [17]. In NHANES, the time spent on each type of physical activity was first converted into moderate-intensity exercise time. Then, based on the MET values for the type, frequency, and duration of activity per week, physical activity (PA) is calculated using the following formula: PA (MET-min/week) = MET × times per week × duration of each PA [18, 19]. Where PA = 0 indicates no physical activity for the week.
Statistical analyses
All normally distributed and skewed continuous variables are presented as mean (standard deviation) or median (interquartile range). Categorical variables are presented as frequencies (%). Continuous variables were compared between groups using Student’s t-test or Mann–Whitney U test, depending on the normality of the distribution. Fisher’s exact test or χ2 statistics were used to compare categorical variables between groups.
We calculated the odds ratio (OR) and 95% confidence interval (CI) using logistic regression to assess the relationship between obesity and sleep duration, sleep patterns. Participants were divided into two groups based on obesity status, with 1670 obese individuals and 2870 non-obese individuals, using the non-obese group as the reference. We conducted logistic regression analysis and assessed sleep duration and sleep pattern scores as both continuous and categorical variables. We used four models for covariate adjustment to estimate potential differences in confounding effects. Model 1 adjusted for sociodemographic variables (gender, age, race and ethnicity, education level, marital status). Model 2 further adjusted for smoking and alcohol consumption. Model 3 further adjusted for hypertension, diabetes. Model 4 further adjusted for physical activity and energy intake. Additionally, we used logistic regression models for interaction and subgroup analysis based on age, gender, race and ethnicity, education level, marital status, family income, physical activity, hypertension, diabetes, smoking status and alcohol consumption. And We also performed ROC analysis to compare the predictive power of using sleep duration alone and the sleep pattern score for obesity.
Restricted cubic spline analyses were conducted to investigate the relationship between sleep duration and Obesity. The piecewise linear regression model was employed to identify potential threshold effects. The likelihood ratio test and bootstrap resampling method were used to determine inflection points.
All analyses were performed using the statistical software packages R 4.2.2 and Free Statistics software version 1.9.2. A descriptive study was conducted on all participants. A P value of < 0.05 indicated significance by two-tailed testing.
Sensitivity analysis
To verify the stability of the research results, we used multiple imputation to analyze the missing covariates and then conducted a sensitivity analysis. We created and analyzed three datasets together.
Results
Participants and demographic characteristics
A total of 4540 participants were included in the study. The average age of the participants was 50.5 ± 17.6 years, with 2271 (50%) being male. The majority self-reported as non-Hispanic white (2232, 49.2%). Table 1 presents the clinical and biochemical characteristics of the study population based on obesity status. Individuals with obesity were more likely to be female and non-Hispanic white, engage in less physical activity, with hypertension, diabetes, and were more likely to have never smoked or consumed alcohol. Importantly, participants with obesity also exhibited poorer sleep conditions, with a higher prevalence of unhealthy sleep patterns and abnormal sleep duration.
Table 1.
Baseline characteristics of the study participants
| Variables | Total (n = 4540) | Non-Obesity (n = 2870) |
Obesity (n = 1670) |
p |
|---|---|---|---|---|
| Sex, n (%) | < 0.001 | |||
| Male | 2271 (50.0) | 1500 (52.3) | 771 (46.2) | |
| Female | 2269 (50.0) | 1370 (47.7) | 899 (53.8) | |
| Age, years | 50.5 ± 17.6 | 50.4 ± 18.3 | 50.8 ± 16.3 | 0.408 |
| Race and ethnicity, n (%) | < 0.001 | |||
| Non-Hispanic white | 2232 (49.2) | 1494 (52.1) | 738 (44.2) | |
| Non-Hispanic black | 923 (20.3) | 517 (18) | 406 (24.3) | |
| Mexican American | 755 (16.6) | 442 (15.4) | 313 (18.7) | |
| Other race | 630 (13.9) | 417 (14.5) | 213 (12.8) | |
| MET, min/wk | 3841.1 ± 6655.1 | 4097.2 ± 6740.3 | 3401.0 ± 6484.5 | < 0.001 |
| Educational level, n (%) | 0.192 | |||
| High school or less | 549 (12.1) | 338 (11.8) | 211 (12.6) | |
| Some college | 1911 (42.1) | 1188 (41.4) | 723 (43.3) | |
| College or above | 2080 (45.8) | 1344 (46.8) | 736 (44.1) | |
| marital, n (%) | 0.647 | |||
| married or living with partners | 2745 (60.5) | 1728 (60.2) | 1017 (60.9) | |
| living alone | 1795 (39.5) | 1142 (39.8) | 653 (39.1) | |
| hypertension, n (%) | < 0.001 | |||
| No | 3153 (69.4) | 2186 (76.2) | 967 (57.9) | |
| Yes | 1387 (30.6) | 684 (23.8) | 703 (42.1) | |
| diabetes, n (%) | < 0.001 | |||
| No | 3984 (87.8) | 2653 (92.4) | 1331 (79.7) | |
| Yes | 556 (12.2) | 217 (7.6) | 339 (20.3) | |
| smoking, n (%) | < 0.001 | |||
| Never | 2342 (51.6) | 1448 (50.5) | 894 (53.5) | |
| Former | 1165 (25.7) | 692 (24.1) | 473 (28.3) | |
| Current | 1033 (22.8) | 730 (25.4) | 303 (18.1) | |
| drinking, n (%) | < 0.001 | |||
| No | 1315 (29.0) | 769 (26.8) | 546 (32.7) | |
| Yes | 3225 (71.0) | 2101 (73.2) | 1124 (67.3) | |
| Sleep duration | 6.8 ± 1.5 | 6.9 ± 1.4 | 6.7 ± 1.5 | < 0.001 |
| Sleep duration, n (%) | < 0.001 | |||
| <7 | 1809 (39.8) | 1066 (37.1) | 743 (44.5) | |
| 7–9 | 2614 (57.6) | 1726 (60.1) | 888 (53.2) | |
| >9 | 117 ( 2.6) | 78 (2.7) | 39 (2.3) | |
| Family PIR, n (%) | 0.435 | |||
| low income | 1353 (29.8) | 844 (29.4) | 509 (30.5) | |
| medium income | 1762 (38.8) | 1106 (38.5) | 656 (39.3) | |
| high income | 1425 (31.4) | 920 (32.1) | 505 (30.2) | |
| Sleep pattern score | 0.8 ± 0.9 | 0.8 ± 0.9 | 0.9 ± 0.9 | < 0.001 |
| Sleep patterns, n (%) | < 0.001 | |||
| Healthy sleep | 1955 (43.1) | 1326 (46.2) | 629 (37.7) | |
| Intermediate sleep | 1603 (35.3) | 981 (34.2) | 622 (37.2) | |
| Poor sleep | 982 (21.6) | 563 (19.6) | 419 (25.1) | |
| Energy intake, kcal | 2072.2 ± 1033.0 | 2098.2 ± 1070.0 | 2027.4 ± 964.8 | 0.026 |
Associations between sleep pattern and obesity
The results of the multivariable logistic regression analysis on the relationship between sleep patterns and the incidence of obesity were presented in Table 2. When the sleep pattern score was analyzed as a continuous variable, it was found that in the unadjusted crude model (OR: 1.23, 95% CI: 1.15–1.32, P < 0.001), there was a significant independent positive correlation between the sleep pattern score and the risk of obesity, and further adjustments did not significantly affect the results. When the sleep pattern score was analyzed as a categorical variable (including healthy sleep, intermediate sleep, and unhealthy sleep), it was found that compared to those with healthy sleep patterns, those with unhealthy sleep patterns had a higher likelihood of obesity (OR = 1.57, 95% CI: 1.34–1.84, P < 0.01). In the fully adjusted model 4 (adjusted for age, sex, race and ethnicity, education level, marital status, household income, physical activity, hypertension, diabetes, smoking status, alcohol consumption, and energy intake), individuals with unhealthy sleep patterns had a 41% increased risk of obesity compared to those with healthy sleep patterns.
Table 2.
Associations between sleep pattern and obesity in the multiple regression model
| Variable | Sleep pattern Score | Sleep patterns | |||
|---|---|---|---|---|---|
| Healthy(1955) | Intermediate(1603) | Poor (982) |
|||
| OR(95% CI) | p value | OR(95%CI) | OR(95% CI) | OR(95% CI) | |
| Unadjusted | 1.23 (1.15 ~ 1.32) | < 0.001 | 1.0(ref) | 1.34 (1.16 ~ 1.53) | 1.57 (1.34 ~ 1.84) |
| Model 1 | 1.24 (1.15 ~ 1.32) | < 0.001 | 1.0(ref) | 1.33 (1.15 ~ 1.53) | 1.58 (1.35 ~ 1.86) |
| Model 2 | 1.26 (1.17 ~ 1.35) | < 0.001 | 1.0(ref) | 1.33 (1.16 ~ 1.54) | 1.63 (1.39 ~ 1.92) |
| Model 3 | 1.18 (1.1 ~ 1.27) | < 0.001 | 1.0(ref) | 1.29 (1.12 ~ 1.49) | 1.41 (1.19 ~ 1.67) |
| Model 4 | 1.18 (1.1 ~ 1.27) | < 0.001 | 1.0(ref) | 1.29 (1.12 ~ 1.5) | 1.41 (1.19 ~ 1.67) |
Model 1: Age, sex, family PIR, marital, race and ethnicity, educational level
Model 2: Model 1 + smoking status, alcohol drinking status
Model 3: Model 2 + hypertension, diabetes
Model 4: Model 3 + physical activity, energy intake
In addition, we used multivariable linear regression to analyze the relationship between sleep pattern scores and BMI. The results also indicated that higher sleep pattern scores, which reflect poorer sleep patterns, are associated with higher BMI (see supplementary materials ).
Moreover, by comparing the area under the ROC curves for both, we found that the sleep pattern score had a higher predictive efficacy for obesity than using sleep duration alone, and the difference between the two was statistically significant (P = 0.013)(see supplementary materials ). .
Associations between sleep duration and obesity
The results of the multivariable logistic regression analysis on the relationship between sleep duration and the incidence of obesity are presented in Table 3. In the unadjusted crude model (OR = 0.9, 95%CI: 0.87–0.94, P < 0.001), there was a significant independent negative correlation between sleep duration and the risk of obesity. Even after adjusting for all potential covariates (Table 3, fully adjusted model), this association remained significant, with sleep duration represented as a continuous variable (OR = 0.91, 95% CI, 0.87–0.95, P < 0.001). When sleep duration was entered as a categorical variable into the fully adjusted model, the obesity incidence was 24% lower in the normal sleep duration group compared to the short sleep duration group (OR = 0.76, 95% CI, 0.66–0.86 ).
Table 3.
Associations between sleep duration and obesity in the multiple regression model
| Variable | Sleep duration | Sleep duration | |||
|---|---|---|---|---|---|
| Short(1809) | Normal(2614) | Long(117) | |||
| OR(95% CI) | p value | OR(95%CI) | OR(95% CI) | OR(95% CI) | |
| Unadjusted | 0.9 (0.87 ~ 0.94) | < 0.001 | 1.0(ref) | 0.74 (0.65 ~ 0.84) | 0.72 (0.48 ~ 1.07) |
| Model 1 | 0.91(0.87 ~ 0.95) | < 0.001 | 1.0(ref) | 0.75 (0.66 ~ 0.85) | 0.72 (0.48 ~ 1.07) |
| Model 2 | 0.9(0.86 ~ 0.94) | < 0.001 | 1.0(ref) |
0.73 (0.65 ~ 0.83) |
0.72 (0.48 ~ 1.08) |
| Model 3 | 0.91(0.87 ~ 0.95) | < 0.001 | 1.0(ref) | 0.76 (0.67 ~ 0.87) | 0.723(0.48 ~ 1.1) |
| Model 4 | 0.91(0.87 ~ 0.95) | < 0.001 | 1.0(ref) | 0.76 (0.66 ~ 0.86) | 0.72 (0.47 ~ 1.09) |
Model 1: Age, sex, family PIR, marital, race and ethnicity, educational level
Model 2: Model 1 + smoking status, alcohol drinking status
Model 3: Model 2 + hypertension, diabetes
Model 4: Model 3 + physical activity, Energy intake
Nonlinear relationship between sleep duration and obesity
After adjusting for some covariates, we observed a non-linear dose-response relationship between sleep duration and the incidence of obesity (Fig. 2). Using a two-piecewise linear regression model, we found the threshold for sleep duration to be 9.73 h (Table 4). Below the threshold, for each additional unit increase in sleep duration, the risk of obesity decreased by 10.9% (OR = 0.891, 95% CI: 0.849–0.936, P < 0.001, Table 4; Fig. 2). Above the threshold, for each additional unit increase in sleep duration, the risk of obesity increased by 102.3% (OR = 2.023, 95% CI, 1.113–3.677, P = 0.0208, Table 4; Fig. 2).
Fig. 2.
Nonlinear dose-response relationship between sleep duration and obesity. Adjustment factors included age, gender, race and ethnicity, education level, marital status, family income, physical activity, hypertension, diabetes, smoking status, alcohol consumption, and energy intake. The red line and pink area represent the estimated values and their corresponding 95% confidence intervals, respectively
Table 4.
Threshold effect analysis of sleep duration on obesity
| Threshold of Sleep duration | OR | 95% CI | P value |
|---|---|---|---|
| <9.73 | 0.891 | (0.849 ~ 0.936) | < 0.001 |
| ≥ 9.73 | 2.023 | (1.113 ~ 3.677) | 0.0208 |
Adjustment factors included age, sex, marital, race and ethnicity, educational level, family PIR, smoking status, alcohol drinking status, hypertension, diabetes, physical activity, Energy intake
Subgroup analyses
We conducted stratified and interaction analyses to determine whether the relationship between sleep patterns and the incidence of obesity were consistent across several subgroups (Fig. 3). Stratified analyses were performed by age, gender, race, education level, marital status, family income, physical activity, smoking status, alcohol consumption, hypertension, and diabetes, and the results were consistent. Furthermore, in the study of effect modification, we did not find statistically significant interactions in the stratified analysis.
Fig. 3.
Associations between sleep pattern and obesity in different subgroups. Except for the stratification component itself, each stratification factor was adjusted for age, gender, race and ethnicity, education level, marital status, family income, physical activity, hypertension, diabetes, smoking status, alcohol consumption, and energy intake
Sensitivity analysis
After analyzing the missing data using multiple imputation, we found that even after adjusting for potential covariates (in the fully adjusted model), the relationship between sleep duration or sleep quality scores and obesity remained stable.(see supplementary materials).
Discussion
Obesity is a significant global public health issue caused by the interaction of multiple factors. The continuous increase in obesity rates has important health consequences, including an increased risk of a range of diseases such as degenerative joint disease, type 2 diabetes, cardiovascular disease, and obesity-related malignancies. In the past, excessive energy intake and insufficient physical activity were considered the main traditional risk factors, but this cannot fully explain the increase in obesity rates over the past few decades. In the last 20 years, an increasing amount of research has identified many new risk factors for obesity and related metabolic diseases, including short sleep and decreased sleep quality.
In this cross-sectional study, we used the NHANES databases from 2007 to 2008, 2015–2016, and 2017–2018 to explore the relationship between sleep and obesity, including not only sleep duration but also overall sleep patterns. Firstly, our study found that short sleep duration is associated with obesity. After adjusting for potential confounding factors, the obesity incidence rate decreased by 24% in individuals with normal sleep duration (7–9 h) compared to those with short sleep duration (< 7 h). Through curve fitting and inflection point analysis, our study also found a non-linear relationship between sleep duration and the incidence of obesity. The threshold we calculated was approximately10 hours, and the effects below and above the threshold were completely different. In the range of 7–10 h of sleep, the obesity incidence curve tended to stabilize. Therefore, it seems that maintaining an appropriate sleep duration (7–10 h) can better reduce the incidence of obesity. This result differs from previous research that classified 7–9 h as normal. The reasons for this discrepancy may be due to differences in the study population, research design, and other factors. Therefore, future studies should include a more diverse range of populations to better determine the classification of optimal sleep duration.
Multiple components of energy metabolism are influenced by circadian rhythms and the sleep-wake cycle, including energy expenditure, metabolism, and appetite hormones. Sleep has a regulatory effect on metabolic function, including immune and hormonal status [20, 21]. Specifically, sleep can regulate appetite, thus playing an important role in reducing obesity [15]. An early study [22] demonstrated the impact of sleep duration on appetite hormones. When sleep duration was restricted (4 h vs. 10 h), ghrelin levels in the body increased by 28%, while leptin decreased by 18%. Hunger and appetite increased by 24% and 23%, respectively, leading to higher energy intake and subsequent weight gain. The possible mechanism is that under conditions of sleep deprivation, the appetite assessment centers in the frontal and insular cortices related to the assessment of food choice were dulled, and the reactivity of the amygdala was amplified at the subcortical level. These brain mechanisms of change are important contributors to neuroendocrine and appetite hormone imbalances [23]. In summary, sleep affects changes in appetite hormone levels, and brain area activation related to changes in food choice. It is important to note that feeding is regulated at multiple levels and involves many hormones, such as leptin, PYY, in addition to just leptin and ghrelin [24].
However, hormones are not the only factor regulating human food intake. The research [25] has confirmed that short sleep increases the strength of connections between certain brain areas, such as the connectivity between the dorsolateral prefrontal cortex and bilateral putamen and the dorsolateral prefrontal cortex and bilateral insula. The increase in functional connectivity between these brain areas was positively correlated with fat intake and negatively correlated with laboratory-measured carbohydrate intake. Additionally, the research [26–28]had found that after sleep restriction, neural activity associated with reward and pleasure was enhanced by food stimuli, leading individuals to consume more high-fat, high-carbohydrate, and high-energy-dense foods, thus increasing the risk of obesity.
Currently, most studies have only confirmed the relationship between short sleep duration and obesity [9, 10, 29]. However, short sleep duration should not be the sole criterion for sleep deprivation. Instead, the cumulative effects of poor sleep quality and short sleep duration should be considered to provide a reliable assessment of the link between sleep and obesity. Research suggests that slow-wave sleep (SWS) can serve as a marker of sleep quality. Clinical studies have shown that sleep deprivation can have cardiovascular effects on young people, and selective suppression of SWS, mean reducing sleep quality without changing sleep duration, could lead to decreased insulin sensitivity without sufficient compensatory increase in insulin release. Therefore, it seems that not only sleep deprivation, but also sleep quality, play an important role in the biochemical pathways related to sugar metabolism [30]. In this study, we used a sleep pattern score, combining both sleep duration and sleep quality, to assess the relationship between sleep and obesity. From our research results, it could be seen that after adjusting for potential confounding factors, an unhealthy sleep pattern was independently associated with obesity (OR:1.18, 95% CI:1.1–1.27, P < 0.001), and the results remain stable in stratified analysis. Additionally, based on our research findings, classifying sleep duration as 7–10 h as normal and reconstructing the sleep patterns also confirmed that there is an independent positive correlation between sleep pattern scores and obesity.(see supplementary materials ).
Additionally, to better eliminate potential bias in the population, we also used propensity score matching data for analysis, and the results remained consistent. Therefore, we must consider the relationship between sleep and obesity comprehensively. In daily life, it is important not only to provide sufficient and appropriate sleep time but also to ensure sleep quality on this basis.
This study analyzed nationally representative data from the NHANES database. Consistent with previous research, it confirmed the correlation between short sleep duration and obesity [31, 32]. However, our study incorporated data from a larger population, making the results more reliable. Additionally, we identified a non-linear relationship between sleep duration and obesity, finding an appropriate duration of sleep. Furthermore, to better and comprehensively assess the impact of sleep on obesity, in this study, we used a sleep pattern score, And categorized them into three types based on the sleep pattern scores: healthy sleep pattern, moderate sleep pattern, and poor sleep pattern, a method that reflects the cumulative effects of sleep duration and quality, which could complement previous work, and was more suitable for assessing clinical risk populations. And more importantly, in our study, through ROC analysis comparison, we confirmed that the sleep pattern score had a higher predictive efficacy for obesity than using sleep duration alone, this meaned that although sleep duration was important, the establishment of sleep pattern scoring provided a more comprehensive predictive indicator for obesity.
However, this study also had some limitations. Since it was a cross-sectional study, we could not establish a causal relationship. Additionally, the sleep pattern score we established may not be comprehensive enough. In the future, it may be worth exploring the relationship between sleep and obesity from different dimensions by establishing a more comprehensive scoring system. Additionally, the lack of weighted calculations limits the generalizability of the study findings beyond the sample data, hindering further exploration of the population of US adults. Then The dietary energy intake in this article is based on data obtained from individual dietary recall interviews, however, a single 24-hour dietary recall cannot represent an individual’s long-term habitual diet, which is also a limitation of our article. Lastly, the association between sleep and obesity may also be influenced by additional potential confounding variables. Therefore, future multicenter cohort studies were needed to confirm these findings by including more potential confounding factors and using standardized and consistent measures. Additionally, in future research, we also need to conduct more fundamental and mechanistic studies to better elucidate the relationship between obesity and sleep.
Conclusion
Our findings consolidate the relationship between sleep duration and obesity. It was also found that there was a non-linear relationship between sleep duration and obesity. When the sleep duration was appropriate(7–9 h), the incidence of obesity was lowest. Both too short and too long sleep durations could increase the incidence of obesity. By using an accumulated score for the effects of sleep duration and quality, the relationship between sleep pattern score and obesity was investigated. By categorizing them into three types based on the sleep pattern scores: healthy sleep pattern, moderate sleep pattern, and poor sleep pattern, then it was found that an unhealthy sleep pattern was positively correlated with the occurrence of obesity, and the study indicated that the sleep pattern score had a higher predictive efficacy for obesity than using sleep duration alone, the establishment of sleep pattern scoring provided a more comprehensive predictive indicator for obesity. So, interventions aimed at promoting healthy sleep habits and appropriate sleep duration may be important in reducing the risk of obesity.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Acknowledgements
We thank the Free Statistics team for providing technical assistance and valuable tools for data analysis and visualization.
Abbreviations
- WHO
World health organization
- BMI
Body mass index
- NHANES
National health and nutrition examination survey
- MEC
Mobile examination center
- SLQ
Sleep disorders questionnaire
- OR
Odds ratio
- 95%CI
95% confidence intervals
- PIR
Poverty income ratio
- MET
Metabolic equivalent
- PSM
Propensity score matching
- SMD
Standardized mean difference
Author contributions
Q: participated in the study of concepts and manuscript preparation and drafting, contributed to data collection, data analysis and revising the manuscript for important intellectual content. Z: Supervision and guidance of the research. Y: Specialised knowledge, literature, revision, and comprehensive analysis of the research. M: Technical support and data collection. All authors reviewed the manuscript.
Funding
This research received no external funding.
Data availability
NHANES data used in this work are publicly available. All raw data are available on the NHANES website (https://www.cdc.gov/nchs/nhanes/). Further inquiries can be directed to the corresponding author.
Declarations
Ethics approval and consent to participate
Not applicable.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Stival C, Lugo A, Odone A, Van Den Brandt PA, Fernandez E, Tigova O, et al. Prevalence and correlates of overweight and obesity in 12 European countries in 2017–2018. Obes Facts. 2022;15:655–65. 10.1159/000525792. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Chen K, Shen Z, Gu W, Lyu Z, Qi X, Mu Y, et al. Prevalence of obesity and associated complications in China: A cross-sectional, real‐world study in 15.8 million adults. Diabetes Obes Metabolism. 2023;25:3390–9. 10.1111/dom.15238. [DOI] [PubMed] [Google Scholar]
- 3.Ogden CL, Fryar CD, Martin CB, Freedman DS, Carroll MD, Gu Q, et al. Trends in obesity prevalence by race and Hispanic Origin—1999–2000 to 2017–2018. JAMA. 2020;324:1208. 10.1001/jama.2020.14590. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ward ZJ, Bleich SN, Cradock AL, Barrett JL, Giles CM, Flax C, et al. Projected U.S. State-Level prevalence of adult obesity and severe obesity. N Engl J Med. 2019;381:2440–50. 10.1056/NEJMsa1909301. [DOI] [PubMed] [Google Scholar]
- 5.Zhang C, Zhang J, Liu Z, Zhou Z. More than an Anti-diabetic bariatric surgery, metabolic surgery alleviates systemic and local inflammation in obesity. OBES SURG. 2018;28:3658–68. 10.1007/s11695-018-3400-z. [DOI] [PubMed] [Google Scholar]
- 6.Baranwal N, Yu PK, Siegel NS. Sleep physiology, pathophysiology, and sleep hygiene. Prog Cardiovasc Dis. 2023;77:59–69. 10.1016/j.pcad.2023.02.005. [DOI] [PubMed] [Google Scholar]
- 7.Ogilvie RP, Patel SR. The epidemiology of sleep and obesity. Sleep Health. 2017;3:383–8. 10.1016/j.sleh.2017.07.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.St-Onge M, ‐P. Sleep–obesity relation: underlying mechanisms and consequences for treatment. Obes Rev. 2017;18:34–9. 10.1111/obr.12499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Capers PL, Fobian AD, Kaiser KA, Borah R, Allison DB. A systematic review and meta-analysis of randomized controlled trials of the impact of sleep duration on adiposity and components of energy balance. Obes Rev. 2015;16:771–82. 10.1111/obr.12296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sperry SD, Scully ID, Gramzow RH, Jorgensen RS. Sleep duration and waist circumference in adults: A Meta-Analysis. Sleep. 2015;38:1269–76. 10.5665/sleep.4906. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Rothwell CJ, Madans JH, Porter KS. National center for health statistics. Vital Health Stat. 2013;1:2013:1–37. [Google Scholar]
- 12.Sen A, Tai XY. Sleep duration and executive function in adults. Curr Neurol Neurosci Rep. 2023;23:801–13. 10.1007/s11910-023-01309-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Chen X, Wang C, Dong Z, Luo H, Ye C, Li L, et al. Interplay of sleep patterns and oxidative balance score on total cardiovascular disease risk: insights from the National health and nutrition examination survey 2005–2018. J Glob Health. 2023;13:04170. 10.7189/jogh.14.04170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Safaei M, Sundararajan EA, Driss M, Boulila W, Shapi’i A. A systematic literature review on obesity: Understanding the causes & consequences of obesity and reviewing various machine learning approaches used to predict obesity. Comput Biol Med. 2021;136:104754. 10.1016/j.compbiomed.2021.104754. [DOI] [PubMed] [Google Scholar]
- 15.Meurling IJ, Shea DO, Garvey JF. Obesity and sleep: a growing concern. Curr Opin Pulm Med. 2019;25:602–8. 10.1097/MCP.0000000000000627. [DOI] [PubMed] [Google Scholar]
- 16.Mercy UC, Elenwa F, Ogunsola AS, Eniola OA, Yunusa I, Karaye IM. Sex differences in the association between short sleep duration and obesity among US adults: findings from NHANES, 2015–2020. Sleep Med. 2022;92:59–63. 10.1016/j.sleep.2022.03.004. [DOI] [PubMed] [Google Scholar]
- 17.Bull FC, Maslin TS, Armstrong T. Global physical activity questionnaire (GPAQ): nine country reliability and validity study. J Phys Activity Health. 2009;6:790–804. 10.1123/jpah.6.6.790. [DOI] [PubMed] [Google Scholar]
- 18.Liang J, Huang S, Jiang N, Kakaer A, Chen Y, Liu M, et al. Association between joint physical activity and dietary quality and lower risk of depression symptoms in US adults: Cross-sectional NHANES study. JMIR Public Health Surveill. 2023;9:e45776. 10.2196/45776. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tang H, Zhang X, Luo N, Huang J, Zhu Y. Association of dietary live microbes and nondietary prebiotic/probiotic intake with cognitive function in older adults: evidence from NHANES. Journals Gerontology: Ser A. 2024;79:glad175. 10.1093/gerona/glad175. [DOI] [PubMed] [Google Scholar]
- 20.Dorsey A, De Lecea L, Jennings KJ. Neurobiological and hormonal mechanisms regulating women’s sleep. Front Neurosci. 2021;14:625397. 10.3389/fnins.2020.625397. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Irwin MR, Opp MR. Sleep health: reciprocal regulation of sleep and innate immunity. Neuropsychopharmacol. 2017;42:129–55. 10.1038/npp.2016.148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Spiegel K, Tasali E, Penev P, Cauter EV. Brief communication: sleep curtailment in healthy young men is associated with decreased leptin levels, Elevated Ghrelin Levels, and Increased Hunger and Appetite n.d. [DOI] [PubMed]
- 23.Brondel L, Romer MA, Nougues PM, Touyarou P, Davenne D. Acute partial sleep deprivation increases food intake in healthy men. Am J Clin Nutr. 2010;91:1550–9. 10.3945/ajcn.2009.28523. [DOI] [PubMed] [Google Scholar]
- 24.Chaput J-P, McHill AW, Cox RC, Broussard JL, Dutil C, Da Costa BGG, et al. The role of insufficient sleep and circadian misalignment in obesity. Nat Rev Endocrinol. 2023;19:82–97. 10.1038/s41574-022-00747-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Fang Z, Spaeth AM, Ma N, Zhu S, Hu S, Goel N, et al. Altered salience network connectivity predicts macronutrient intake after sleep deprivation. Sci Rep. 2015;5:8215. 10.1038/srep08215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Benedict C, Brooks SJ, O’Daly OG, Almèn MS, Morell A, Åberg K, et al. Acute sleep deprivation enhances the brain’s response to hedonic food stimuli: an fMRI study. J Clin Endocrinol Metabolism. 2012;97:E443–7. 10.1210/jc.2011-2759. [DOI] [PubMed] [Google Scholar]
- 27.St-Onge M-P, Wolfe S, Sy M, Shechter A, Hirsch J. Sleep restriction increases the neuronal response to unhealthy food in normal-weight individuals. Int J Obes. 2014;38:411–6. 10.1038/ijo.2013.114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.St-Onge M-P, McReynolds A, Trivedi ZB, Roberts AL, Sy M, Hirsch J. Sleep restriction leads to increased activation of brain regions sensitive to food stimuli. Am J Clin Nutr. 2012;95:818–24. 10.3945/ajcn.111.027383. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Wu Y, Zhai L, Zhang D. Sleep duration and obesity among adults: a meta-analysis of prospective studies. Sleep Med. 2014;15:1456–62. 10.1016/j.sleep.2014.07.018. [DOI] [PubMed] [Google Scholar]
- 30.Fatima Y, Doi SAR, Mamun AA. Sleep quality and obesity in young subjects: a meta-analysis. Obes Rev. 2016;17:1154–66. 10.1111/obr.12444. [DOI] [PubMed] [Google Scholar]
- 31.Giannos P, Prokopidis K, Candow DG, Forbes SC, Celoch K, Isanejad M, et al. Shorter sleep duration is associated with greater visceral fat mass in US adults: findings from NHANES, 2011–2014. Sleep Med. 2023;105:78–84. 10.1016/j.sleep.2023.03.013. [DOI] [PubMed] [Google Scholar]
- 32.Li Q. The association between sleep duration and excess body weight of the American adult population: a cross-sectional study of the National health and nutrition examination survey 2015–2016. BMC Public Health. 2021;21:335. 10.1186/s12889-021-10369-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
NHANES data used in this work are publicly available. All raw data are available on the NHANES website (https://www.cdc.gov/nchs/nhanes/). Further inquiries can be directed to the corresponding author.