Skip to main content
Download PDF
Download ePub

Diurnal patterns of accelerometer-measured physical activity and sleep and risk of all-cause mortality: a follow-up of the National Health and Nutrition Examination Surveys (NHANES)

International Journal of Behavioral Nutrition and Physical Activity volume 21, Article number: 120 (2024) Cite this article

A Correction to this article was published on 11 November 2024

This article has been updated

Abstract

Background

Physical activity and sleep are established modifiable lifestyle factors, but the optimal time of the day of these behaviours for health is unknown. This study examined the independent and joint associations of diurnal patterns of physical activity and sleep with all-cause mortality.

Methods

This prospective cohort study included 6,673 participants who have attended the accelerometer assessment in the 2011–2014 National Health and Nutrition Examination Surveys (NHANES). Diurnal patterns of accelerometer-measured physical activity and sleep were identified using K-means clustering analysis. All-cause mortality was ascertained from the accelerometer measurement to December 31, 2019 (median follow-up 6.8 years). Survey-weighted Cox proportional hazard models were performed to estimate the independent and joint associations of diurnal patterns of physical activity and sleep with all-cause mortality.

Results

Diurnal patterns identified were: early-morning (32.4%), midday (42.5%), and late-afternoon (25.1%) for physical activity; and irregular sleep (37.4%), morning lark (33.6%), and night owl (29.0%) for sleep. After adjusting for volume of physical activity, sleep duration and other potential covariates, the early-morning physical activity pattern (hazard ratio 1.36, 95% confidence interval 1.13–1.64) and irregular sleep pattern (1.42, 1.01–1.99) were independently associated with higher risk of all-cause mortality, compared with midday physical activity and morning lark sleep patterns, respectively. In addition, participants with the combined pattern of early-morning physical activity and irregular sleep had higher risk of all-cause mortality compared to those with midday physical activity combined with a morning lark sleep pattern (1.92, 1.33–2.78). Several sociodemographic differences were observed in the strength of these associations.

Conclusions

Wearable activity-rest monitoring data showed that peak physical activity in the early morning and irregular sleep diurnal patterns are associated with increased mortality risk, and the combination of these patterns further exaggerated the risk. Public health program should acknowledge that the diurnal patterns of physical activity and sleep, in addition to their duration and frequency, may play a crucial role in lifestyle-based health promotion and management strategies.

Background

Adhering to a healthy lifestyle, such as engaging in habitual physical activity (PA) and maintaining optimal sleep duration, is the cornerstone and cost-effective strategy for the primordial, primary, and secondary prevention of non-communicable diseases throughout the life span [1, 2]. There is strong evidence supporting the protective effects of regular PA and adequate sleep duration against a multitude of adverse health consequences (e.g., cardiovascular diseases [CVD], diabetes, dementia) and premature mortality [3, 4]. Accordingly, the American Heart Association has proposed Life’s Essential 8 which included participation in PA and healthy sleep as key targets for reducing the risk of major health problems [5]. While the benefits of higher volume and intensity of PA and appropriate sleep duration are well documented, less is known about whether the time of day we do physical activity and sleep affects health outcomes.

Given the consensus on the significant role of circadian rhythms in physiology and diseases [6], examining the impact of diurnal patterns of lifestyle behaviours on health is warranted. To date, the results on the association of time patterns of PA and sleep with health outcomes are inconsistent [7,8,9,10,11,12,13,14,15,16,17,18]. Using data from UK Biobank, some studies found that early-morning exercise may be associated with higher risks of all-cause and CVD mortality [8, 9], while contrasting findings suggested morning PA could aid in reducing the risks of CVD and diabetes [11, 13]. Conflicting results also exist in several randomized trail studies. One study based on 11 men with type 2 diabetes (T2D) showed that exercise in the morning was less efficacious than exercise in the afternoon in lowing blood glucose [7]; another randomized crossover trail study found morning exercise may benefit the management of glycaemia in patients with T2D [12]. For sleep patterns, some studies reported that individuals who are night owls (i.e., being active late into night and sleeping until relatively late in the day) may have higher risks of metabolic dysfunction and CVD [14,15,16], while others have suggested sleep regularity may be a stronger predictor of mortality risk [17, 18]. The measurement methods of sleep varied across these studies, with sleep chronotype being measured by self-reported questionnaires [15, 16], and sleep regularity being assessed by accelerometers [17, 18]. Considering these mixed results, further large-scale studies based on different populations with accelerometer-measured data are necessary to achieve more conclusive results.

Beyond their independent health effects, PA and sleep duration have been hypothesized to be interrelated. It is suggested, for instance, that more PA could help maintain a better sleep quality [19], while prolonged sleep duration might reduce the time for PA [20] among middle- or older-aged adults. The observed joint associations of PA and sleep with cognitive decline and all-cause, CVD and cancer mortality [21, 22] support this hypothesis. However, these studies focused only on the volume and intensity of PA and the duration of sleep. To date, we found no studies have examined the diurnal patterns of accelerometer-measured PA and sleep simultaneously. It is important to investigate not only the independent associations of time patterns of these lifestyle factors, but also how they interact to inform health outcomes.

Using data from the National Health and Nutrition Examination Survey (NHANES), this study aimed to identify diurnal patterns of PA and sleep in a US population, and to examine the independent and joint associations of these lifestyle patterns with all-cause mortality.

Methods

Study design and population

This study was conducted using individual-level data from NHANES, a program conducted by the National Centers for Health Statistics that recruited a representative sample of the civilian noninstitutionalized population in the US using a complex, multistage probability design. NHANES includes a series of cross-sectional surveys every two years since 1999. Data are collected through interviews and physical examinations, and mortality records have been linked from the date of survey participation to December 31, 2019. Further details of the protocol, study design, and data collection of NHANES are available elsewhere [23]. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline.

This study used data from the 2011–2012 and 2013–2014 cycles of NHANES because wearable accelerometer activity-rest monitoring was conducted during these rounds. There were 19,931 participants in the 2011–2014 cycles initially, and we included 9,342 non-pregnant participants aged ≥ 20 years who attended the accelerometer assessment. Further exclusion criteria included: (1) participants without at least three days valid accelerometer data (wear time ≥ 10 h for each day) (n = 475) [24]; (2) participants with no information on mortality (n = 18); and (3) participants with missing covariates (n = 2176). A total of 6,673 participants were enrolled in the final analyses (Supplementary Fig. 1).

Accelerometery data management

The measurement of rest-activity status with accelerometers (ActiGraph model GT3X+) was implemented in NHANES 2011–2014. Participants were asked to wear devices on their non-dominant hand and told not to do anything with the device except to wear it. The devices measured acceleration (i.e., on the x-, y-, and z-axes) 80 times a second (80 Hz) and ambient light (lux) once a second (1 Hz). In this study, we used the summary datasets produced and released by NHANES, which included individual-level data summarized in monitor-independent movement summary units (MIMS-units) by minute, hour, and day [25, 26]. Although there are no established cut-off points for MIMS-units to distinguish the intensity of PA to date, MIMS units could be used to measure peak values which represent minute-level epochs with higher rates of movements across the period of monitoring [27]. An open-source algorithm was used to predict time periods of sleep wear, wake wear, or non-wear, and this algorithm has identified a category for each minute which was used to identify the diurnal patterns of sleep. This algorithm could reliably estimate movements and was sensitive in detecting subtle wrist movements during sedentary behaviours [25]. More details of the algorithm generation can be found in NHANES website [26].

The diurnal patterns of PA and sleep were identified using K-means clustering algorithm, which is an unsupervised machine learning algorithm that assigns participants into different clusters in which participants share similar characteristics. This algorithm has been used to identify diurnal patterns of accelerometer-measured lifestyle factors in previous studies [8, 10]. We computed the Euclidean distance as the similarity criterion [10]. For the diurnal pattern of PA, we used the proportions of total MIMS-units accumulated in each hour, with higher MIMS-units indicating higher intensity of PA. For sleep pattern, we calculated the proportions of accumulated sleep minutes in each hour as the independent variable. To ensure discriminability, interpretability and sufficient statistical power, the number of clusters was set at 3 (k-value = 3) for both PA and sleep (for 4 to 9 clusters, see illustrations in Supplementary Figs. 2 and 3).

Mortality outcomes

The main outcome of this study was all-cause mortality, which was extracted from the National Death Index (NDI) dataset updated to December 31, 2019. In addition, we examined cause-specific mortality according to the following International Classification of Diseases 10th Revision (ICD-10): diabetes (E10-E14), CVD (I00-I09, I11, I13, I20-I51, I60-I69), cancer (C00-C97), and chronic lower respiratory diseases (J40-J47). Cardiometabolic mortality was defined as death due to diabetes or CVD.

Covariates

Baseline covariates included age, sex (male/female), ethnicity (Mexican American/non-Hispanic White/non-Hispanic Black/other Hispanic/other), educational level (less than high school/high school or equivalent/college or above), marital status (married or living with partner/widowed/divorced or separated/never married), annual family income (0–19,999/20,000–44,999/45,000–74,999/75,000–99,999/100,000 and over), body mass index (BMI, underweight/normal weight/overweight/obese), alcohol intake (yes/no), current smoking status (yes/no), alternative healthy eating index (AHEI [28]), and history of diabetes, CVD, cancer, hypertension, and hyperlipidemia. Socioeconomic status (SES) was assessed using the summed score (0, 1, or 2) of educational level (0: less than college; 1: college or above) and annual family income (0: <45,000; 1: 45,000 and over), with a higher score indicating higher SES. We dichotomised the SES score into low or middle (0–1) versus high SES (2). Further covariates included total volume of PA, calculated as the sum of accelerometer measured MIMS-units per day, and total sleep duration, calculated as the sum hour of predicted sleep periods per day. We also derived self-reported moderate-to-vigorous physical activity (MVPA) and sedentary behaviour based on the Global Physical Activity Questionnaire [29]. Total amount of MVPA was calculated as the hours of metabolic equivalent (MET) per week following the NHANES guidance: PA (MET-h/week) = MET × weekly frequency × duration of each PA [3, 30]. Participants were divided into three groups of no PA (< 1 MET-h/week), low intensity PA (1–48 MET-h/week), and high intensity PA (> 48 MET-h/week) [31]. Time of participants’ sedentary behaviours per day were divided into quartiles: 0–4 h/day, 4–6 h/day, 6–9 h/day, and > 9 h/day. Self-reported sleep duration at night was categorized into < 6 h, 6–8 h, and > 8 h per day.

Statistical analysis

All statistical analyses followed the complex sampling method with sample weight, stratification, and clustering in accordance with the NHANES analytic guidelines [32]. Characteristics of participant by PA and sleep diurnal patterns were presented as mean (standard deviation, SD) for continuous variables and numbers (percentage) for categorical variables, and were compared using analysis of variance adjusted for sampling weights and Rao-Scott χ2 tests.

We used the survey-weighted multinomial logistics regression models to assess factors associated with diurnal patterns of PA and sleep with all variables mutually adjusted in the same model. We used survey-weighted Cox proportional hazard models to examine the associations of diurnal patterns of PA, sleep, and their joint patterns with all-cause mortality. The proportional hazards assumption was tested by including product terms of follow-up time and PA and sleep patterns in the Cox model, and no significant deviation was found. Months from baseline to the date of death or end of follow-up (December 31, 2019), whichever occurred first, were calculated as the time scale. The fully adjusted model included age, sex, ethnicity, education level, marital status, family income, BMI, alcohol intake, smoking status, AHEI, disease history of diabetes, CVD, cancer, hypertension, hyperlipidemia, and accelerometer-measured PA volume and sleep duration per day. We adjusted for the diurnal sleep pattern in the analysis of PA, and diurnal pattern of PA in the analysis of sleep patterns. Stratified analyses were performed in subgroups of the following sociodemographic factors: sex (male and female), age (< 50 and ≥ 50 years), educational level (less than college and college or over), annual family income (< 45,000 and ≥ 45,000), and SES (low or middle and high).

A series of additional analyses were performed to test the robustness of the main results. First, we examined the associations of diurnal patterns of PA and sleep and their joint patterns with cause-specific mortality, including diabetes, CVD, cancer, cardiometabolic, chronic lower respiratory diseases, and other cause-specific mortality. Second, several sensitivity analyses were conducted after (1) including self-reported amount of MVPA, sedentary time, and sleep duration at night as covariates; (2) excluding participants with diabetes or CVD at baseline; (3) excluding participants with diabetes, CVD, cancer, hypertension, or hyperlipidemia at baseline; (4) excluding participants who reported no MVPA; (5) excluding participants with abnormal (< 1% or > 99%) accelerometer-measured PA volume or sleep duration; (6) excluding participants with less than two years follow-up; and (7) imputing missing covariates using multiple imputation for five times.

Odds ratios (ORs), hazard ratios (HRs) and 95% confidence intervals (95% CIs) were reported in this study. Significance tests were assessed at the 0.05 level using two-sided tests. All statistical analyses were performed using SAS (version 9.4, SAS Institute Inc., NC, USA) and R software (version 4.1.2).

Results

Population characteristics and diurnal patterns of physical activity and sleep

Among the 6,673 participants included in this study, the mean age was 49.7 (SD = 17.4) years and 51.3% of participants were females. Figure 1 shows the three diurnal patterns of PA among the participants, including early-morning (n = 2,159, 32.4%), midday (n = 2,837, 42.5%), and late-afternoon (n = 1,677, 25.1%). The early-morning pattern had a remarkable spike of PA between 8:00 and 9:00, and the intensity of PA decreased gradually during the rest of the day. For the midday group, there was a long plateau period of moderate-to-high intensity PA ranging from 11:00 to 17:00, with the first peak intensity occurring at 10:00–12:00 and a second peak occurring at 16:00–17:00. The peak intensity of PA for participants with the late-afternoon pattern occurred at 17:00–19:00. The three diurnal patterns identified for sleep were: irregular sleep (n = 2,493, 37.4%), morning lark (n = 2,245, 33.6%), and night owl (n = 1,935, 29.0%). Participants with the morning lark pattern slept early at night and rose early in the morning, while those with the night owl pattern stayed up later at night and rose later in the morning. Participants with the irregular sleep pattern demonstrated varying times of sleep onset and awakening throughout the day, resulting in heightened nighttime sleep fragmentation and increased daytime sleepiness.

Fig. 1
figure 1

Diurnal patterns of physical activity (A) and sleep (B)

A. Diurnal pattern of physical activity. Hourly monitor-independent movement summary (MIMS) units were calculated as the MIMS accumulated in each hour of the day. The average MIMS accumulated per hour across all subjects in each cluster was shown in different colours

B. Diurnal pattern of sleep. Hourly sleep duration was calculated as the sleep duration accumulated in each hour of the day. The average sleep duration accumulated per hour across all subjects in each cluster was shown in different colours

Full size image

The characteristics of participants by the diurnal patterns of PA and sleep were shown in Table 1. Participants with the early-morning PA pattern were more likely to be older, males, non-Hispanic, less educated, married or living with partner, with higher family income, obese, non-smokers, and had higher rates of prior diabetes, CVD, hypertension, and hyperlipidemia, compared to those with the late-afternoon or midday pattern. Participants classified in the irregular sleep group were more likely to be males, non-Hispanic, divorced or separated, with low family income, obese, current smokers, and had higher rates of diabetes, CVD, and hypertension. They also had lower AHEI score, lower PA volume, and higher sleep duration per day, compared to those classified in the morning lark or night owl groups.

Table 1 Basic characteristics of participants by diurnal patterns of physical activity and sleep
Full size table

Factors associated with diurnal patterns of physical activity and sleep

After mutually adjusted, several sociodemographic and lifestyle factors were associated with diurnal patterns of PA and sleep (Table 2). Compared to those with midday pattern, participants who were older; males; widowed, divorced or separated; or had higher family income were more likely to be physically active in early-morning. Compared to those with morning lark pattern, being younger; males; non-Hispanic Black; highly educated; single; with lower family income; obese; current smokers; and had lower AHEI score were associated with higher odds of experiencing irregular sleep. In addition, participants with diabetes (OR = 1.62, 95% CI = 1.23–2.13) or CVD (OR = 1.74, 95% CI = 1.28–2.38) were also at higher odds of experiencing irregular sleep.

Table 2 Factors associated with different diurnal patterns of physical activity and sleep
Full size table

Diurnal patterns of physical activity, sleep, and their joint associations with all-cause mortality

During the median follow-up of 6.8 years (interquartile range 5.8–7.8), 589 participants died. The independent and joint associations of diurnal patterns of PA and sleep with all-cause mortality are presented in Table 3. After adjusting for sociodemographic and lifestyle factors (Model 2), the risk of all-cause mortality was 33% higher for participants with the early-morning PA pattern (HR = 1.33, 95% CI = 1.11–1.60) and 44% higher for participants with the irregular sleep pattern (HR = 1.44, 95% CI = 1.09–1.90), compared to those with a midday PA and morning lark pattern, respectively. These associations were robust to additional adjustment for the accelerometer-measured PA volume and sleep duration per day (early-morning: HR = 1.36, 95% CI = 1.13–1.64; irregular sleep: HR = 1.42, 95% CI = 1.01–1.99). For the joint diurnal patterns of PA and sleep, participants with an early-morning and irregular sleep pattern had higher risk of all-cause mortality compared to those with a midday and morning lark pattern (HR = 1.92, 95% CI = 1.33–2.78).

Table 3 Independent and joint associations of diurnal patterns of physical activity and sleep with all-cause mortality
Full size table

Differences by sociodemographic factors in the associations with all-cause mortality

In all stratified analyses by sex, age, educational level, annual family income, and SES, the HRs for early-morning PA and irregular sleep, independently and jointly, exceeded one (range 1.23 to 5.16) (Fig. 2, Supplementary Figs. 4 and 5). The association between early-morning PA pattern and all-cause mortality was statistically significant among participants who were males, older, non-Hispanic white, less educated, with lower family income, and low or middle SES (Supplementary Fig. 4). The association between irregular sleep and all-cause mortality was statistically significant among those who were non-Hispanic white, higher educated, with higher family income and higher SES (Supplementary Fig. 5). For the joint pattern of PA and sleep, the association of the combination of early-morning PA and irregular sleep with all-cause mortality reached statistical significance among participants who were males, older, with higher education, or with higher family income and higher SES (Fig. 2).

Fig. 2
figure 2

Subgroup analysis of the joint associations of diurnal patterns of physical activity and sleep with all-cause mortality

Full size image

Cause-specific mortality and sensitivity analyses

The associations of early-morning PA pattern, irregular sleep pattern and their joint patterns with mortality caused by diabetes, CVD, cancer, cardiometabolic diseases, chronic lower respiratory and other causes were consistent with those observed for all-cause mortality (Supplementary Tables 1–2). The sensitivity analyses also showed similar results to our main findings after exclusion of specific participant groups (Supplementary Tables 3–8). Besides, when we included participants with missing covariates in the final analysis, diurnal patterns of PA and sleep identified were consistent with the main analysis (Supplementary Fig. 6). After imputing the missing data, the associations of diurnal patterns of PA and sleep and their combinations with all-cause mortality were similar to the main results (Supplementary Table 9).

Discussion

In this prospective study of a US population, we identified three diurnal patterns for both of accelerometer-measured PA and sleep: peak PA in early-morning, midday, and late-afternoon and irregular sleep, morning lark, and night owl sleep patterns. Participants who were older; males; widowed, divorced or separated or had higher family income were more likely to be physically active in early-morning, while those who were younger; males; non-Hispanic Black; single; with higher educational level; lower family income; obese; current smokers; low scorers in AHEI; or had diabetes or CVD were more likely to experience irregular sleep. The early-morning PA pattern, irregular sleep pattern, and their joint patterns were all associated with higher risk of all-cause mortality, independent of the volume of PA, sleep duration, and sociodemographic and health-related covariates. The lowest mortality risk was observed for midday physical activity combined with the morning lark sleep pattern, midday physical activity combined with the night owl pattern, late-afternoon physical activity combined with the night owl pattern, and early-morning physical activity combined with the morning lark pattern. Moreover, while sociodemographic differences existed among the associations between early-morning PA, irregular sleep and all-cause mortality, the effect estimates favoured increased risk for these specific diurnal patterns in all subgroups.

Comparison with other studies

The three diurnal patterns of PA we identified in a US population (2011–2014) are in agreement with those reported in a previous study based on MVPA in NHANES (2003–2006) [10]. To minimise potential bias arising from the influence of intensity levels of PA and sedentary behaviours [33], we conducted two sensitivity analyses by including self-reported amount of MVPA and sedentary time as covariates and by excluding participants who reported no MVPA. The results remained similar, suggesting that different intensities of PA and sedentary time are not driving our findings. Our findings were also robust to exclusions of participants with abnormal (< 1% or > 99%) accelerometer-measured physical activity volume or sleep duration, those with less than 2 years of follow-up and individuals with diabetes, CVD, cancer, hypertension or hyperlipidemia at baseline. This suggests that outliers and reverse causation due to pre-existing disease are unlikely explanations for our findings.

Our results on early-morning PA and mortality from CVD are in line with previous population-based observational findings [8, 9, 33, 34]. A cross-sectional study, for example, reported that being most active in the morning was associated with higher risk of CVD [33]. A prospective cohort study based on a UK population found a 1.56-fold increased risk of CVD mortality for participants in the early-morning PA group (peak intensity occurred at 7:00–9:00) compared to those in the midday group (peak intensity occurred at 10:00–13:00) [8]. In a large-scale prospective study based on 92,139 UK Biobank participants, the risk of all-cause and CVD mortality was lower in the midday-afternoon group than those in the morning group [9]. Consistent results were also reported in several randomized controlled trials [7, 35].

The unfavourable health effects of early-morning PA could be partially explained by the rapid increase of blood pressure accompanied with PA and the increased blood pressure reactivity to physical exertion in the early morning [36]. However, some conflicting results exist [11, 13]. A prospective cohort study based on 86,657 populations from UK Biobank suggests that morning physical activity was associated with lower risks of CVD [11]. Similarly, another study of UK Biobank populations (n = 93,095) found a protective effect of physical activity in the morning on diabetes risk [13]. These inconsistent results might partly result from (1) different study outcomes; (2) varied study populations; (3) different measurement methods of physical activity (ActiGraph model GT3X + in NHANES and Axivity AX3 in UK Biobank) and physical activity patterns. It is important to extend the concept of chronoactivity in further studies to verify the results.

This study also found three patterns of sleep using accelerometer-measured data. Unlike previous studies reporting that evening chronotype (i.e., get up and go to bed late) was associated with higher risk of morbidity and mortality compared with morning chronotype (i.e., get up and go to bed early) [37,38,39], our results showed no statistically significant association between the night owl pattern and mortality. The inconsistent results might partly attribute to the usage of self-reported question-defining sleep chronotypes in previous studies and different study populations. Our study showed that participants with the irregular sleep pattern were at higher risk of all-cause mortality, suggesting the sleep regularity might be a more important predictor of health than a consistently early or late onset of sleep. In line with our results, previous studies using self-reported sleep data have linked irregular sleep patterns with adverse health outcomes such as metabolic syndrome [40] and all-cause mortality [41], and such associations were independent of sleep durations. Other studies using accelerometer-measured data also provided similar evidence that sleep irregularity is an important predictor of reduced quality of life [18] and higher risk of mortality [17].

Participants in the irregular sleep pattern are characterised by increased nighttime sleep fragmentation and daytime sleepiness. Consequently, they are more likely to be exposed to irregular environmental stimulations, potentially resulting in the disruption of circadian rhythms and downstream adverse health outcomes [42, 43]. This interpretation is supported by animal studies showing that circadian disruption induced by irregular sleep is a causal risk factor of premature mortality [44].

Moreover, our results showed that participants with prior CVD and diabetes were at particularly high risk of experiencing irregular sleep, which could also contribute to the higher risk of mortality for participants in this group [45]. However, due to the difficulty of distinguishing the onset time of diseases and sleep irregularity, the temporal relationship could not be reliably inferred in the present study. This issue warrants future studies to validate such mechanisms.

We conducted joint analyses of the diurnal patterns of PA and sleep to address potential complex interactions between two behaviours. Our results supported a significant joint association of the diurnal patterns of PA and sleep with all-cause mortality which was stronger than the independent associations. Previous studies have suggested that higher intensity PA and optimal sleep duration in combination were associated with delayed cognitive decline, better mental health, improved cardiometabolic health, and reduced risk of all-cause, CVD, and cancer mortality across different age groups [21, 22, 46,47,48]. Our study extends previous evidence by capturing 9 different combinations of diurnal patterns and by demonstrating that the independent effects of diurnal PA and sleep patterns would be exaggerated among participants with both early-morning PA and irregular sleep. To the best of our knowledge, this is the first study examining the joint association of their diurnal patterns with mortality risk. Our findings, taken together with existing evidence, suggest that future lifestyle-based intervention strategies should consider the diurnal patterns of both PA and sleep, in addition to their volume, frequency and duration, to maximise the health benefits.

We observed several sociodemographic differences in the diurnal patterns of PA and sleep and their associations with mortality. The crucial role of sociodemographic factors in the associations of PA and sleep with health outcomes have been reported in previous studies [33, 49,50,51,52,53,54,55,56]. For example, it was suggested that the male morning PA groups had higher risk of developing CVD than the corresponding female groups, which could be partly explained by the difference of VO2max in response to the endurance training between men and women [33, 57]. For the role of age on circadian rhythm, studies have reported that older adults were more likely to be active in the morning and to have earlier circadian phase [50, 58]. This finding is consistent with our results and the suggestion that these people are more active during the night and less active during the day, resulting increased sleep fragmentation and daytime sleepiness with age [59, 60].

Interestingly, we found that the role of SES in the association between PA with mortality differed from that in the association between sleep with mortality. Thus, the association between early-morning PA pattern and mortality was stronger and that between irregular sleep and mortality weaker among participants with low or middle SES compared to those with high SES. According to our results on factors associated with PA and sleep patterns, participants with lower income were less likely to be physically active in the morning, but were more likely to experience irregular sleep. Several factors could explain these differences, including work stress, work schedules, and work-life imbalance. As a novel observation, it remains unknown why this preference difference existed. These sociodemographic variations imply that the policies targeting diurnal patterns of PA and sleep might have different impact in population sub-groups.

Strength and limitations

The main strengths of this study include its prospective study design, the accelerometer-measured PA and sleep, the usage of a machine learning algorithm to classify diurnal patterns, and the opportunity to adjust the effect estimates for multiple covariates. In addition, the sensitivity analyses confirmed the robustness of our results. Several limitations should also be acknowledged. First, there is currently a lack of established cut-off points for MIMS-units to determine levels of PA intensity. Therefore, we could not identify MVPA. Also, the recreational PA and occupational PA could not be differentiated. An algorithm to identify different intensity levels of PA is warranted for future analysis which would allow researchers to consider more detailed characteristics of PA. Second, although accelerometer-based measurement is widely used to determine sleep status [61, 62], the measure might overestimate sleep duration [63]. Moreover, the correlation of accelerometer-measured and self-reported sleep duration was low in NHANES [64]. To address this issue, we included self-reported amount of MVPA and sleep duration as covariates in the sensitivity analysis; the results did not materially change. Nonetheless, there remains a need to test the accuracy of the accelerometer-derived sleep and wake categories in NHANES. Third, as our study is based on observational data, we could not eliminate residual confounding even after adjusting for known covariates. Fourth, the sample size of this study is relatively small, limiting the generalizability of our results. The sample sizes in the subgroup analyses of the joint association of diurnal patterns of PA and sleep were particularly small, resulting in a large 95% CI for some subcategories. Fifth, the observational study design cannot demonstrate the causality relationship. Considering these limitations, further studies are warranted to validate the observed associations with larger sample sizes and, if feasible, using randomized controlled trial design.

Implications

This study has critical public health implications. PA and sleep have already been identified as two important targets for the prevention and intervention of major health problems by the World Health Organization guidelines [1]. Beyond previous suggestions that focused on PA volume and intensity as well as sleep duration, our study highlights the importance of the timing and diurnal patterns of lifestyle behaviours in disease prevention and health management. As the evidence on the effects of timing of these behaviours on different health outcomes remain relatively limited, future research is needed to validate the present evidence for health policies. In these interventional studies, it is particularly important to examine the effectiveness of simultaneously targeting the timing of PA and sleep to maximize the health benefits [21, 22]. Furthermore, considering the sociodemographic differences identified on the diurnal patterns of PA and sleep and their association with all-cause mortality, lifestyle-based health promotion strategies may need to be tailored for population sub-groups.

Conclusions

In this population-based prospective cohort study, we found that early-morning PA pattern and irregular sleep pattern were associated with higher risk of all-cause mortality. These independent associations were strengthened in individuals with a combined diurnal pattern of early-morning PA and irregular sleep. Our study suggests that diurnal patterns of physical activity and sleep, in addition to their duration and frequency, may play an important role in lifestyle-based health promotion strategies.

Data availability

The original data for the study are available on the NHANES website: https://wwwn.cdc.gov/nchs/nhanes/Default.aspx.

Change history

Abbreviations

AHEI:

alternative healthy eating index

CI:

confidence interval

CVD:

cardiovascular diseases

HR:

hazard ratio

ICD-10:

International Classification of Diseases 10th Revision

MIMS:

monitor-independent movement summary

MVPA:

moderate-to-vigorous physical activity

NDI:

National Death Index

NHANES:

National Health and Nutrition Examination Survey

OR:

odds ratio

PA:

physical activity

SD:

standard deviation

SES:

socioeconomic status

T2D:

type 2 diabetes

References

  1. Organization WH. Advancing the global agenda on prevention and control of noncommunicable diseases 2000 to 2020: looking forwards to 2030 2023 [ https://www.who.int/publications/i/item/9789240072695

  2. Yasin YJ, Banoub JAM, Husseini A. GBD 2017 causes of death collaborators. Global, regional, and national age- sex- specific mortality for 282 causes of death in 195 countries and territories, 1980–2017: a systematic analysis for the global burden of Disease Study 2017. Lancet. 2019;393(10190):E44–E.

    Google Scholar 

  3. U.S. Department of Health and Human Services. 2018 Physical activity guidelines advisory committee scientific report [ https://health.gov/sites/default/files/2019-09/PAG_Advisory_Committee_Report.pdf

  4. Kline CE, Hillman CH, Sheppard BB, Tennant B, Conroy DE, Macko RF et al. Physical activity and sleep: an updated umbrella review of the 2018 Physical Activity Guidelines Advisory Committee report. Sleep Med Rev. 2021;58.

  5. Lloyd-Jones DM, Allen NB, Anderson CAM, Black T, Brewer LC, Foraker RE, et al. Life’s essential 8: updating and enhancing the American Heart Association’s construct of Cardiovascular Health: a Presidential Advisory from the American Heart Association. Circulation. 2022;146(5):e18–43.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Crnko S, Du Pré BC, Sluijter JPG, Van Laake LW. Circadian rhythms and the molecular clock in cardiovascular biology and disease. Nat Rev Cardiol. 2019;16(7):437–47.

    Article  PubMed  Google Scholar 

  7. Savikj M, Gabriel BM, Alm PS, Smith J, Caidahl K, Björnholm M, et al. Afternoon exercise is more efficacious than morning exercise at improving blood glucose levels in individuals with type 2 diabetes: a randomised crossover trial. Diabetologia. 2019;62(2):233–7.

    Article  CAS  PubMed  Google Scholar 

  8. Ma T, Jennings L, Sirard JR, Xie YJ, Lee CD. Association of the time of day of peak physical activity with cardiovascular mortality: findings from the UK Biobank study. Chronobiol Int. 2023;40(3):324–34.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Feng HL, Yang LL, Liang YY, Ai SZ, Liu YP, Liu Y et al. Associations of timing of physical activity with all-cause and cause-specific mortality in a prospective cohort study. Nat Commun. 2023;14(1).

  10. Ma T, Bennett T, Lee CD, Wicklow M. The diurnal pattern of moderate-to-vigorous physical activity and obesity: a cross-sectional analysis. Obes (Silver Spring). 2023;31(10):2638–47.

    Article  Google Scholar 

  11. Albalak G, Stijntjes M, van Bodegom D, Jukema JW, Atsma DE, van Heemst D, et al. Setting your clock: associations between timing of objective physical activity and cardiovascular disease risk in the general population. Eur J Prev Cardiol. 2023;30(3):232–40.

    Article  PubMed  Google Scholar 

  12. Carrillo BJP, Cope E, Gurel S, Traslosheros A, Kenny A, Michot-Duval O et al. Morning exercise and pre-breakfast metformin interact to reduce glycaemia in people with type 2 diabetes: a randomized crossover trial. J Physiol-London. 2024.

  13. Tian CW, Buerki C, Westerman KE, Patel CJ. Association between timing and consistency of physical activity and type 2 diabetes: a cohort study on participants of the UK Biobank. Diabetologia. 2023;66(12):2275–82.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Reutrakul S, Knutson KL. Consequences of circadian disruption on Cardiometabolic Health. Sleep Med Clin. 2015;10(4):455–68.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Merikanto I, Lahti T, Puolijoki H, Vanhala M, Peltonen M, Laatikainen T, et al. Associations of Chronotype and Sleep with Cardiovascular diseases and Type 2 diabetes. Chronobiol Int. 2013;30(4):470–7.

    Article  PubMed  Google Scholar 

  16. Yu JH, Yun CH, Ahn JH, Suh S, Cho HJ, Lee SK, et al. Evening Chronotype is Associated with Metabolic disorders and Body Composition in Middle-aged adults. J Clin Endocr Metab. 2015;100(4):1494–502.

    Article  CAS  PubMed  Google Scholar 

  17. Windred DP, Burns AC, Lane JM, Saxena R, Rutter MK, Cain SW et al. Sleep regularity is a stronger predictor of mortality risk than sleep duration: a prospective cohort study. Sleep. 2024;47(1).

  18. Trivedi R, Man H, Madut A, Mather M, Elder E, Dhillon HM et al. Irregular Sleep/Wake patterns are Associated with reduced quality of life in Post-treatment Cancer patients: a study across three Cancer cohorts. Front Neurosci-Switz. 2021;15.

  19. Best JR, Falck RS, Landry GJ, Liu-Ambrose T. Analysis of dynamic, bidirectional associations in older adult physical activity and sleep quality. J Sleep Res. 2019;28(4).

  20. Bellavia A, Åkerstedt T, Bottai M, Wolk A, Orsini N. Sleep duration and survival percentiles across categories of physical activity. Am J Epidemiol. 2014;179(4):484–91.

    Article  PubMed  Google Scholar 

  21. Bloomberg M, Brocklebank L, Hamer M, Steptoe A. Joint associations of physical activity and sleep duration with cognitive ageing: longitudinal analysis of an English cohort study. Lancet Health Longev. 2023;4(7):E345–53.

    Article  Google Scholar 

  22. Huang BH, Duncan MJ, Cistulli PA, Nassar N, Hamer M, Stamatakis E. Sleep and physical activity in relation to all-cause, cardiovascular disease and cancer mortality risk. Br J Sports Med. 2022;56(13):718–24.

    Article  PubMed  Google Scholar 

  23. National Center for Health Statistics. NHANES survey methods and analytic guidelines 2024 [ https://wwwn.cdc.gov/nchs/nhanes/analyticguidelines.aspx#sample-design

  24. Doherty A, Jackson D, Hammerla N, Plötz T, Olivier P, Granat MH et al. Large Scale Population Assessment of Physical Activity using wrist worn accelerometers: the UK Biobank Study. PLoS ONE. 2017;12(2).

  25. John D, Tang Q, Albinali F, Intille S. An Open-Source Monitor-Independent Movement Summary for Accelerometer Data Processing. J Meas Phys Behav. 2019;2(4):268–81.

    Article  PubMed  PubMed Central  Google Scholar 

  26. ActiGraph. GT3X + and wGT3X + Device Manual, 2013 [ http://s3.amazonaws.com/actigraphcorp.com/wp-content/uploads/2018/02/22094126/GT3X-wGT3X-Device-Manual-110315

  27. Zheng PX, Pleuss JD, Turner DS, Ducharme SW, Aguiar EJ. Dose-response Association between Physical Activity (Daily MIMS, Peak 30-Minute MIMS) and cognitive function among older adults: NHANES 2011–2014. J Gerontol a-Biol. 2023;78(2):286–91.

    Article  Google Scholar 

  28. Wang DD, Leung CW, Li Y, Ding EL, Chiuve SE, Hu FB, et al. Trends in dietary quality among adults in the United States, 1999 through 2010. JAMA Intern Med. 2014;174(10):1587–95.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Schuna JM Jr., Johnson WD, Tudor-Locke C. Adult self-reported and objectively monitored physical activity and sedentary behavior: NHANES 2005–2006. Int J Behav Nutr Phys Act. 2013;10:126.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Chen L, Cai M, Li H, Wang X, Tian F, Wu Y, et al. Risk/benefit tradeoff of habitual physical activity and air pollution on chronic pulmonary obstructive disease: findings from a large prospective cohort study. BMC Med. 2022;20(1):70.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Tian X, Xue B, Wang B, Lei R, Shan X, Niu J et al. Physical activity reduces the role of blood cadmium on depression: A cross-sectional analysis with NHANES data. Environmental pollution (Barking, Essex: 1987). 2022;304:119211.

  32. National Center for Health Statistics. Weighting 2022 [ https://wwwn.cdc.gov/nchs/nhanes/tutorials/weighting.aspx

  33. Qian JY, Walkup MP, Chen SH, Brubaker PH, Bond DS, Richey PA, et al. Association of objectively measured timing of physical activity bouts with Cardiovascular Health in Type 2 diabetes. Diabetes Care. 2021;44(4):1046–54.

    Article  PubMed  PubMed Central  Google Scholar 

  34. van der Velde JHPM, Boone SC, Winters-van Eekelen E, Hesselink MKC, Schrauwen-Hinderling VB, Schrauwen P, et al. Timing of physical activity in relation to liver fat content and insulin resistance. Diabetologia. 2023;66(3):461–71.

    Article  PubMed  Google Scholar 

  35. Qian JY, Xiao Q, Walkup MP, Coday M, Erickson ML, Unick J, et al. Association of timing of moderate-to-vigorous physical activity with changes in Glycemic Control over 4 years in adults with type 2 diabetes from the look AHEAD Trial. Diabetes Care. 2023;46(7):1417–24.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Bilo G, Grillo A, Guida V, Parati G. Morning blood pressure surge: pathophysiology, clinical relevance and therapeutic aspects. Integr Blood Press Control. 2018;11:47–56.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Arora N, Richmond RC, Brumpton BM, Åsvold BO, Dalen H, Skarpsno ES, et al. Self-reported insomnia symptoms, sleep duration, chronotype and the risk of acute myocardial infarction (AMI): a prospective study in the UK Biobank and the HUNT study. Eur J Epidemiol. 2023;38(6):643–56.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Fan YJ, Wu YH, Peng Y, Zhao BB, Yang J, Bai L et al. Sleeping late increases the risk of myocardial infarction in the Middle-aged and older populations. Front Cardiovasc Med. 2021;8.

  39. Knutson KL, von Schantz M. Associations between chronotype, morbidity and mortality in the UK Biobank cohort. Chronobiol Int. 2018;35(8):1045–53.

    PubMed  PubMed Central  Google Scholar 

  40. Ogura Y, Koyama T, Ozaki E, Omichi C, Uehara R. Subjective irregular sleep is associated with metabolic syndrome: a cross-sectional study. Prev Med Rep. 2022;28:101844.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Omichi C, Koyama T, Kadotani H, Ozaki E, Tomida S, Yoshida T, et al. Irregular sleep and all-cause mortality: a large prospective cohort study. Sleep Health. 2022;8(6):678–83.

    Article  PubMed  Google Scholar 

  42. Vetter C. Circadian disruption: what do we actually mean? Eur J Neurosci. 2020;51(1):531–50.

    Article  PubMed  Google Scholar 

  43. Gao JJ, Huang XM, Park Y, Hollenbeck A, Blair A, Schatzkin A, et al. Daytime Napping, Nighttime sleeping, and Parkinson Disease. Am J Epidemiol. 2011;173(9):1032–8.

    Article  PubMed  PubMed Central  Google Scholar 

  44. Hurd MW, Ralph MR. The significance of circadian organization for longevity in the golden hamster. J Biol Rhythm. 1998;13(5):430–6.

    Article  CAS  Google Scholar 

  45. Fagiani F, Di Marino D, Romagnoli A, Travelli C, Voltan D, Di Cesare Mannelli L, et al. Molecular regulations of circadian rhythm and implications for physiology and diseases. Signal Transduct Target Therapy. 2022;7(1):41.

    Article  CAS  Google Scholar 

  46. Sampasa-Kanyinga H, Colman I, Goldfield GS, Janssen I, Wang JL, Podinic I et al. Combinations of physical activity, sedentary time, and sleep duration and their associations with depressive symptoms and other mental health problems in children and adolescents: a systematic review. Int J Behav Nutr Phy. 2020;17(1).

  47. Blodgett JM, Ahmadi MN, Atkin AJ, Chastin S, Chan HW, Suorsa K et al. Device-measured physical activity and cardiometabolic health: the prospective physical activity, sitting, and Sleep (ProPASS) consortium. Eur Heart J. 2023.

  48. Wennman H, Kronholm E, Heinonen OJ, Kujala UM, Kaprio J, Partonen T et al. Leisure time physical activity and sleep predict mortality in men irrespective of background in competitive sports. Progress Prev Med. 2017;2(6).

  49. Paudel S, Ahmadi M, Phongsavan P, Hamer M, Stamatakis E. Do associations of physical activity and sedentary behaviour with cardiovascular disease and mortality differ across socioeconomic groups? A prospective analysis of device-measured and self-reported UK Biobank data. Br J Sports Med. 2023;57(14):921–9.

    Article  PubMed  Google Scholar 

  50. Sartini C, Wannamethee SG, Iliffe S, Morris RW, Ash S, Lennon L et al. Diurnal patterns of objectively measured physical activity and sedentary behaviour in older men. BMC Public Health. 2015;15.

  51. Ramires VV, Wehrmeister FC, Böhm AW, Galliano L, Ekelund U, Brage S et al. Physical activity levels objectively measured among older adults: a population-based study in a southern city of Brazil. Int J Behav Nutr Phy. 2017;14.

  52. Salvo D, Torres C, Villa U, Rivera JA, Sarmiento OL, Reis RS et al. Accelerometer-based physical activity levels among Mexican adults and their relation with sociodemographic characteristics and BMI: a cross-sectional study. Int J Behav Nutr Phy. 2015;12.

  53. Copeland JL, Esliger DW. Accelerometer Assessment of physical activity in active, healthy older adults. J Aging Phys Activ. 2009;17(1):17–30.

    Article  Google Scholar 

  54. Lee GB, Kim HC, Jeon YJ, Jung SJ. Association between socioeconomic status and longitudinal sleep quality patterns mediated by depressive symptoms. Sleep. 2021;44(8).

  55. Whinnery J, Jackson N, Rattanaumpawan P, Grandner MA. Short and Long Sleep Duration Associated with Race/Ethnicity, Sociodemographics, and socioeconomic position. Sleep. 2014;37(3):601–.

    Article  PubMed  PubMed Central  Google Scholar 

  56. Grandner MA, Petrov MER, Rattanaumpawan P, Jackson N, Platt A, Patel NP. Sleep symptoms, Race/Ethnicity, and socioeconomic position. J Clin Sleep Med. 2013;9(9):897–905.

    Article  PubMed  PubMed Central  Google Scholar 

  57. Diaz-Canestro C, Montero D. Sex dimorphism of VO(2max) trainability: a systematic review and Meta-analysis. Sports Med. 2019;49(12):1949–56.

    Article  PubMed  Google Scholar 

  58. Monk TH, Buysse DJ, Reynolds CF, Kupfer DJ, Houck PR. Circadian temperature rhythms of older-people. Exp Gerontol. 1995;30(5):455–74.

    Article  CAS  PubMed  Google Scholar 

  59. Duffy JF, Zitting KM, Chinoy ED. Aging and circadian rhythms. Sleep Med Clin. 2015;10(4):423–34.

    Article  PubMed  PubMed Central  Google Scholar 

  60. Leng Y, Musiek ES, Hu K, Cappuccio FP, Yaffe K. Association between circadian rhythms and neurodegenerative diseases. Lancet Neurol. 2019;18(3):307–18.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Carazo-Vargas P, Salazar-Obando J, Vargas-Montero A, Alvarado-Barrantes R, Siles-Canales F, Moncada-Jiménez J. Lack of agreement between Polysomnography and Accelerometry Devices for Measuring Sleep Efficiency and Movement in University students. Meas Phys Educ Exerc. 2020;24(4):291–302.

    Article  Google Scholar 

  62. Full KM, Kerr J, Grandner MA, Malhotra A, Moran K, Godoble S, et al. Validation of a physical activity accelerometer device worn on the hip and wrist against polysomnography. Sleep Health. 2018;4(2):209–16.

    Article  PubMed  PubMed Central  Google Scholar 

  63. Conley S, Knies A, Batten J, Ash G, Miner B, Hwang Y, et al. Agreement between actigraphic and polysomnographic measures of sleep in adults with and without chronic conditions: a systematic review and meta-analysis. Sleep Med Rev. 2019;46:151–60.

    Article  PubMed  PubMed Central  Google Scholar 

  64. Lee PH. Validation of the National Health and Nutritional Survey (NHANES) single-item self-reported sleep duration against wrist-worn accelerometer. Sleep Breath. 2022;26(4):2069–75.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

We thank all team members and staff who have contributed to the NHANES study and all participants attending the NHANES study.

Funding

X. Xu was supported by Zhejiang University, Fundamental Research Funds for the Central Universities, the Key Laboratory of Intelligent Preventive Medicine of Zhejiang Province, and the Natural Science Foundation of China (72474197). M. Kivimaki was supported by the Wellcome Trust (221854/Z/20/Z), the MRC (R024227, Y014154), NIA (R01AG056477), the Academy of Finland (350426), and the Finnish Foundation for Cardiovascular Research (a86898) during the conduct of the study. The funding bodies had no roles in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.

Author information

Authors and Affiliations

  1. School of Public Health, The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, 310058, Zhejiang, China

    Yue Zhang, Yangyang Cheng, Yaguan Zhou, Hui Wang, Changzheng Yuan & Xiaolin Xu

  2. The Key Laboratory of Intelligent Preventive Medicine of Zhejiang Province, Hangzhou, China

    Yue Zhang, Yangyang Cheng, Yaguan Zhou, Hui Wang, Changzheng Yuan & Xiaolin Xu

  3. UCL Brain Sciences, University College London, London, UK

    Mika Kivimäki

  4. Emory Global Diabetes Research Center, Emory University, Atlanta, GA, USA

    Rodrigo M. Carrillo-Larco

  5. Hubert Department of Global Health, Rollins School of Public Health, Emory University, Atlanta, GA, USA

    Rodrigo M. Carrillo-Larco

  6. Department of Nutrition, Harvard T.H. Chan School of Public Health, Boston, MA, USA

    Changzheng Yuan

  7. School of Public Health, Faculty of Medicine, The University of Queensland, Brisbane, Australia

    Xiaolin Xu

Authors
  1. Yue Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Mika Kivimäki
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Rodrigo M. Carrillo-Larco
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Yangyang Cheng
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Yaguan Zhou
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Hui Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Changzheng Yuan
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Xiaolin Xu
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

X.X. contributed to the study conceptualization and supervised the whole project. Y.Z. made the analysis plan, conducted the statistical analyses, and drafted the initial manuscript. X.X., Y.C., YG.Z., and H.W. verified the underlying data and statistical analyses. M.K., R.C., and C.Y. provided support on the statistical methods and manuscript revision. All authors contributed to and approved the final manuscript. X.X. is the corresponding author, has full access to all the data and has final responsibility for the decision to submit for publication.

Corresponding author

Correspondence to Xiaolin Xu.

Ethics declarations

Ethical approval and consent to participate

The study protocol for the US NHANES was approved by the US NHANES institutional review board and National Center for Health Statistics Research ethics review board (Protocol #2011-17). All participants provided written informed consent. Institutional review board approval was exempted for this study because of the publicly available and deidentified data.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

The original version of this article has been revised: the legends of early-morning physical activity and midday physical activity were wrongly swapped in Figure 1 (A) – the blue and green line.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, Y., Kivimäki, M., Carrillo-Larco, R.M. et al. Diurnal patterns of accelerometer-measured physical activity and sleep and risk of all-cause mortality: a follow-up of the National Health and Nutrition Examination Surveys (NHANES). Int J Behav Nutr Phys Act 21, 120 (2024). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12966-024-01673-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12966-024-01673-9

Keywords

International Journal of Behavioral Nutrition and Physical Activity

ISSN: 1479-5868

Contact us