Association between weekend catch-up sleep and hypertension in U.S. middle-aged and older adults: Insights from NHANES 2017–2023

Abstract

Weekend catch-up sleep (WCS) has been proposed as a compensatory mechanism to mitigate the adverse cardiovascular effects of weekday sleep deprivation. However, evidence regarding its association with hypertension remains limited and inconsistent, particularly among middle-aged and older adults. Data were obtained from National Health and Nutrition Examination Survey 2017–2023, including 11,934 U.S. adults aged ≥ 40 years. WCS was calculated as the difference between self-reported weekend and weekday sleep duration. Hypertension was defined based on self-reported diagnosis, medication use, or measured blood pressure. Multivariable logistic regression was used to assess the association between WCS and hypertension. Subgroup analyses were performed by sex and weekday sleep duration. Compared to participants with no WCS (<0 hours), those with 0 to 2 hours and ≥ 2 hours of WCS had significantly lower odds of hypertension (odds ratios [OR]: 0.82, 95% confidence intervals [CI]: 0.71–0.93; OR: 0.77, 95% CI: 0.65–0.90, respectively). A consistent inverse association was observed in males, especially among those with ≥ 2 hours of WCS (OR: 0.68, 95% CI: 0.54–0.86), while the association in females was not statistically significant after full adjustment. Among individuals with > 7 hours of weekday sleep, a clear dose–response relationship was identified between WCS duration and reduced hypertension risk (≥2 hours: OR: 0.64, 95% CI: 0.50–0.81), whereas no significant protective effect was found in those with ≤ 7 hours of weekday sleep. In U.S. middle-aged and older adults, moderate to extended WCS is associated with a reduced risk of hypertension, particularly among males and those with sufficient weekday sleep. These findings highlight the potential role of WCS in cardiovascular risk management and underscore the importance of considering baseline sleep patterns and sex differences in sleep-related interventions.

1. Introduction

Hypertension, characterized by persistently elevated blood pressure, is a significant global public health issue and a leading cause of cardiovascular and cerebrovascular diseases.^[1] Despite advances in medical treatments and public health initiatives, the global increase in hypertension cases emphasizes the need for a deeper understanding of its causes.^[2]

Sleep, an essential physiological function, is increasingly recognized as a modifiable risk factor for hypertension. Disrupted sleep patterns – such as insomnia, short sleep duration, and poor sleep quality – have been associated with changes in the autonomic nervous system and endocrine balance, potentially contributing to the development of hypertension.^[3–5] The relationship between sleep and hypertension remains a subject of ongoing research, with studies producing mixed results. While some research shows a positive association between short sleep duration and increased hypertension risk, others suggest a more complex relationship.^[6–8]

In today’s fast-paced world, many individuals accrue sleep debt during the workweek due to professional and social demands.^[9] As a result, weekend catch-up sleep (WCS), where individuals extend their sleep duration on weekends to compensate for weekday sleep loss, has become a common practice.^[10] This raises questions about its health implications, particularly concerning hypertension risk. Some may benefit from additional rest, while others might engage in behaviors that further compromise sleep quality, such as increased social activities or screen time, potentially increasing cardiovascular risk.^[11,12]

With growing interest in the role of sleep in hypertension, recent studies have increasingly focused on the role of WCS in cardiovascular health. Hwangbo et al^[13] identified a significant inverse relationship between WCS and hypertension in Korean adults, especially among individuals reporting subjective sleep insufficiency. Similarly, Gupta et al^[14] observed that children and adolescents with hypertension exhibited shorter WCS duration compared to normotensive peers. In more recent studies using U.S. National Health and Nutrition Examination Survey (NHANES) data, Zhu et al^[15] and Luo et al^[16] provided evidence that moderate WCS is associated with a lower prevalence of cardiovascular disease and hypertension, respectively. However, inconsistencies remain regarding the optimal duration and population-specific effects of WCS.

These findings underscore the need for further investigation into sex- and age-specific patterns, as well as baseline sleep behaviors. This study utilizes data from the NHANES from 2017 to 2023 to explore the relationship between WCS and the incidence of hypertension among middle-aged and older US population. The goal is to provide epidemiological evidence to clarify the role of WCS in hypertension development, potentially informing preventive strategies and public health policies.

2. Methods

2.1. Study population and survey design

The NHANES, conducted by the National Center for Health Statistics, is a nationwide program aimed at assessing the health and nutritional status of the U.S. population. The study drew on cross-sectional data from NHANES 2017–2023, using a probability sampling approach that involved multiple stages and methods. Initially, data from 27,493 participants were considered. After excluding those with missing data on sleep duration, hypertension status and younger than 40 years old, the final sample included 11,934 participants. All data are publicly accessible through the NHANES website, and informed consent was obtained from all participants. A flowchart detailing the sample selection process is presented in Figure 1.

Figure 1.

Flowchart of participant selection. NHANES = National Health and Nutrition Examination Survey.

2.2. Hypertension

In the NHANES study, hypertension was defined based on self-reported data and direct blood pressure measurements. Participants were classified as hypertensive if they met any of the following criteria: a self-reported previous diagnosis of hypertension by a healthcare professional, current use of antihypertensive medication, or an average systolic blood pressure of ≥ 140 mm Hg and/or an average diastolic blood pressure of ≥ 90 mm Hg, based on measurements taken during the physical examination.^[17] Blood pressure was measured by trained clinicians using a standardized procedure recommended by the American Heart Association. Participants were seated and rested for at least 5 minutes before 3 consecutive blood pressure readings were taken at 30-second intervals using a mercury sphygmomanometer. The average of these 3 readings was recorded as the participant’s blood pressure. While blood pressure measurements were taken directly, self-reported data may be subject to recall bias, potentially affecting data interpretation. Detailed measurement procedures are available in the NHANES Physician Examination Procedures Manual.

2.3. Assessment of WCS

WCS was assessed using responses from the NHANES 2017–2023 sleep disorder questionnaire. Participants reported their usual sleep durations on weekdays and weekends by answering the following questions: “What time do you usually fall asleep on weekdays or workdays?” “What time do you usually wake up on weekdays or workdays?” “What time do you usually fall asleep on weekends or non-workdays?” and “What time do you usually wake up on weekends or non-workdays?” For consistency in the analysis, sleep durations of <3 hours were recorded as 2.5 hours, and durations of 14 hours or more were capped at 14 hours. WCS duration was calculated as the difference between weekend and weekday sleep durations. In this study, WCS was defined as the sleep duration on weekends that surpassed the weekday sleep duration (WCS duration > 0).^[18,19]

2.4. Additional measures

Demographic characteristics, including sex (male, female), age (18–39, 40–59, and ≥ 60 years), race/ethnicity, educational level (less than high school, high school or equivalent, and college or above), smoking status (never smoked, former smoker, current smoker), and alcohol consumption status (never, mild, moderate, and heavy) were collected through standardized questionnaires. Diabetes status was determined based on a previous diagnosis by a healthcare professional or laboratory tests indicating fasting glucose ≥ 7.0 mmol/L or HbA1c ≥ 6.5%. Body mass index (BMI) was calculated by dividing weight in kilograms by height in meters squared (kg/m²), and waist circumference was measured to assess central obesity. Poverty status was assessed with the NHANES-derived ratio of family income to poverty (INDFMPIR), calculated as annual household income divided by the U.S. Federal Poverty Guideline; higher values reflect higher income relative to the poverty threshold. These assessments were conducted by experienced medical personnel at the Mobile Examination Center to ensure data collection consistency and accuracy.

2.5. Statistical analysis

Participants were classified into 4 groups according to their WCS duration: <0, 0 to 2, and ≥ 2 hours. Descriptive analyses of categorical variables were conducted using weighted percentages, with weights applied in accordance with NHANES sampling procedures. Continuous variables were reported as weighted means and standard deviations, while categorical variables were presented as frequencies and percentages. Differences in baseline characteristics among participants were assessed using t-tests or Kruskal–Wallis rank sum tests, as appropriate.

Multinomial logistic regression models were employed to estimate odds ratios (ORs) and 95% confidence intervals (CIs) to evaluate the association between WCS and hypertension incidence. The crude model did not adjust for any confounders; Adjusted Model I controlled for sex, age, and race; Adjusted Model II further accounted for additional confounders, including education level, BMI, smoking status, alcohol consumption, and diabetes status. Subgroup analyses were conducted based on sex and weekday sleep duration, with multivariable logistic regression employed to explore these associations.

Data analysis was performed using R (The R Foundation for Statistical Computing, Vienna, Austria, http://www.R-project.org) and Empower Statistics software (v. 4.2.1, X&Y Solutions, Inc., Boston). Statistical significance was defined as a two-tailed P-value < .05.

3. Results

Based on an analysis of 11,934 middle-aged and older adults, participants were stratified into 3 groups according to WCS duration: <0 hours (n = 1618), 0 to 2 hours (n = 8476), and ≥ 2 hours (n = 1840). Significant differences were observed across multiple demographic and health-related characteristics (P < .001). Notably, the ≥ 2 hours WCS group was significantly younger (55.39 ± 10.16 years), with shorter weekday sleep duration (6.63 ± 1.53 hours) and markedly prolonged weekend sleep time (9.62 ± 1.64 hours). Hemoglobin A1c (HbA1c) levels were slightly but significantly higher in the WCS ≥ 2 hours group (P = .001). Alanine aminotransferase differed modestly across WCS groups (P = .034), while aspartate aminotransferase, creatinine, uric acid, and triglyceride levels did not show significant variation (P > .05). However, total cholesterol levels were higher in individuals with longer WCS (P < .001). Non-Hispanic White individuals predominantly clustered in the 0 to 2 hours WCS group (54.24%), while Mexican Americans and other ethnicities showed higher proportions in the ≥ 2 hours WCS group. Although diabetes mellitus prevalence did not differ significantly (P = .542), substantial variations were evident in alcohol use, smoking status, and hypertension prevalence across different WCS groups (Table 1).

Table 1.

Baseline characteristics of middle-aged and older US population with different weekend catch-up sleep duration (WCS).

Variables	Total (n = 11,934)	Weekend catch-up sleep duration (WCS)			P value
		< 0 h (n = 1618)	0–2 h (n = 8476)	≥2 (n = 1840)
Age (yr)	61.25 ± 11.89	59.72 ± 11.80	62.82 ± 11.82	55.39 ± 10.16	<.001*
BMI (kg/m²)	30.13 ± 7.20	30.75 ± 7.53	29.85 ± 7.02	30.85 ± 7.60	<.001*
Poverty (PIR)	2.83 ± 1.64	2.61 ± 1.65	2.91 ± 1.64	2.70 ± 1.61	<.001*
Sleep hours (weekdays)	7.64 ± 1.66	8.38 ± 1.67	7.72 ± 1.59	6.63 ± 1.53	<.001*
Sleep hours (weekends)	8.15 ± 1.75	7.12 ± 1.64	8.03 ± 1.59	9.62 ± 1.64	<.001*
WCS	0.51 ± 1.39	−1.26 ± 0.96	0.31 ± 0.48	2.98 ± 1.22	<.001*
HbA1C	6.01 ± 1.18	6.06 ± 1.26	5.98 ± 1.13	6.08 ± 1.33	.001*
ALT (U/L)	21.85 ± 18.03	21.43 ± 15.91	21.59 ± 18.93	23.22 ± 15.96	.034*
AST (U/L)	22.06 ± 14.55	22.15 ± 13.62	21.96 ± 14.66	22.39 ± 14.85	.705
Creatinine (μmol/L)	83.77 ± 52.40	85.27 ± 53.97	83.88 ± 49.73	82.14 ± 60.46	.453
Uric acid(μmol/L)	328.69 ± 88.24	332.09 ± 90.12	328.81 ± 88.22	325.48 ± 86.74	.296
Triglyceride (mmol/L)	1.31 ± 1.14	1.27 ± 1.24	1.32 ± 1.15	1.31 ± 1.02	.754
Total cholesterol(mmol/L)	4.90 ± 1.11	4.83 ± 1.10	4.90 ± 1.13	4.98 ± 1.05	<.001*
Sex					.464
Male	5632 (47.19%)	774 (47.84%)	4013 (47.35%)	845 (45.92%)
Female	6302 (52.81%)	844 (52.16%)	4463 (52.65%)	995 (54.08%)
Ethnicity					<.001*
Non-Hispanic White	5859 (49.10%)	690 (42.65%)	4597 (54.24%)	572 (31.09%)
Non-Hispanic Black	2339 (19.60%)	454 (28.06%)	1369 (16.15%)	516 (28.04%)
Mexican American	974 (8.16%)	129 (7.97%)	590 (6.96%)	255 (13.86%)
Non-Hispanic Asian	1031 (8.64%)	107 (6.61%)	772 (9.11%)	152 (8.26%)
Other	1731 (14.50%)	238 (14.71%)	1148 (13.54%)	345 (18.75%)
Education level					<.001*
College	7002 (58.67%)	919 (56.80%)	5109 (60.28%)	974 (52.93%)
High school	2769 (23.20%)	397 (24.54%)	1912 (22.56%)	460 (25.00%)
Less than high school	2141 (17.94%)	299 (18.48%)	1438 (16.97%)	404 (21.96%)
Unknown	22 (0.18%)	3 (0.19%)	17 (0.20%)	2 (0.11%)
Sleep trouble					.011*
No	4311 (68.15%)	603 (66.78%)	2908 (67.46%)	800 (71.94%)
Yes	2015 (31.85%)	300 (33.22%)	1403 (32.54%)	312 (28.06%)
Diabetes mellitus					.542
No	8219 (68.87%)	1101 (68.05%)	5838 (68.88%)	1280 (69.57%)
Yes	2932 (24.57%)	408 (25.22%)	2069 (24.41%)	455 (24.73%)
Unknown	783 (6.56%)	109 (6.74%)	569 (6.71%)	105 (5.71%)
Alcohol use status					<.001*
Mild	3452 (28.93%)	393 (24.29%)	2574 (30.37%)	485 (26.36%)
Moderate	1492 (12.50%)	225 (13.91%)	1010 (11.92%)	257 (13.97%)
Never	873 (7.32%)	122 (7.54%)	609 (7.18%)	142 (7.72%)
Heavy	1372 (11.50%)	216 (13.35%)	872 (10.29%)	284 (15.43%)
Unknown	4745 (39.76%)	662 (40.91%)	3411 (40.24%)	672 (36.52%)
Smoking status					<.001*
Never	6510 (54.55%)	812 (50.19%)	4581 (54.05%)	1117 (60.71%)
Now	1912 (16.02%)	350 (21.63%)	1252 (14.77%)	310 (16.85%)
Former	3501 (29.34%)	453 (28.00%)	2638 (31.12%)	410 (22.28%)
Unknown	11 (0.09%)	3 (0.19%)	5 (0.06%)	3 (0.16%)
Hypertension					<.001*
No	5267 (44.13%)	646 (39.93%)	3704 (43.70%)	917 (49.84%)
Yes	6667 (55.87%)	972 (60.07%)	4772 (56.30%)	923 (50.16%)

Continuous variables are presented as mean ± standard deviation (SD), and categorical variables are presented as n (%).

ALT = alanine aminotransferase, AST = aspartate aminotransferase, BMI = body mass index, DM = diabetes mellitus; PIR = ratio of family income to poverty, WCS = weekend catch-up sleep.

^*P-values were calculated using ANOVA for continuous variables and chi-square tests for categorical variables.

Preliminary analysis in the total population (including all individuals aged ≥ 18 years) revealed a significant positive association between WCS duration and hypertension. However, further stratified analyses showed that this association was not significant in the <40 years age group (P > .05), but was consistently observed in participants aged 40 years and above (P < .05). These findings guided our decision to focus the main analysis on the middle-aged and older adult population (≥40 years), where the effect of WCS on hypertension risk was more robust and biologically plausible. The detailed results for the total population and the younger subgroup are provided in Table S1, Supplemental Digital Content, https://links.lww.com/MD/Q201.

Logistic regression analyses revealed significant associations between WCS duration and hypertension symptoms across multiple models (Table 2). In the crude model, continuous WCS duration showed an inverse association with hypertension risk (OR: 0.90, 95% CI: 0.88–0.93, P < .0001). After adjusting for demographic factors (Model I) and additional confounders including education, BMI, smoking status, alcohol use, and diabetes mellitus (Model II), the inverse relationship persisted. Compared to participants with no WCS (≤0 hours), those with WCS duration > 0 hours demonstrated a consistently lower odds of hypertension (Adjusted Model II: OR: 0.81, 95% CI: 0.71–0.92, P = .0011). When stratified into categories, participants with 0 to 2 hours and ≥ 2 hours of WCS exhibited significantly reduced hypertension risk (0–2 hours: OR: 0.82, 95% CI: 0.71–0.93, P = .0028; ≥2 hours: OR: 0.77, 95% CI: 0.65–0.90, P = .0013) compared to the reference group.

Table 2.

Regression analyses for associations between weekend catch-up sleep duration (WCS) and hypertension symptoms in U.S. middle-aged and older adults.

	Crude model	Adjusted Model I	Adjusted Model II
	OR (95% CI) P value
WCS duration (continuous)	0.90 (0.88–0.93) < .0001*	0.97 (0.94–1.00) .0437*	0.97 (0.93–1.00) .0315*
WCS (duration > 0 h)
No	Reference	Reference	Reference
Yes	0.82 (0.74–0.91) .0002*	0.79 (0.70–0.89) < .0001*	0.81 (0.71–0.92) .0011*
WCS duration (multicategory)
≤ 0 h	Reference	Reference	Reference
0–2 h	0.86 (0.77–0.95) .0050*	0.79 (0.70–0.88) < .0001*	0.82 (0.71–0.93) .0028*
≥ 2 h	0.67 (0.58–0.77) < .0001*	0.80 (0.69–0.93) .0029*	0.77 (0.65–0.90) .0013*

Crude model: unadjusted.

Adjusted Model I: adjusted by sex, age, and race.

Adjusted Model II: adjusted by Model I + education, BMI, smoke status, alcohol user status, and DM status.

BMI = body mass index, CI = confidence interval, DM = diabetes mellitus, OR = odds ratio, WCS = weekend catch-up sleep duration.

^*P < .05.

Gender-stratified regression analyses unveiled nuanced associations between WCS duration and hypertension risk (Table 3). In males, WCS demonstrated a consistent and statistically significant inverse relationship with hypertension across all analytical models. Continuous WCS duration showed a protective effect (Adjusted Model II: OR: 0.95, 95% CI: 0.90–0.99, P = .0187), with participants experiencing > 0 hours of WCS exhibiting substantially reduced hypertension risk (Adjusted Model II: OR: 0.76, 95% CI: 0.63–0.92, P = .0044). The most pronounced risk reduction was observed in males with ≥ 2 hours of WCS (OR: 0.68, 95% CI: 0.54–0.86, P = .0011). In stark contrast, females displayed markedly different results, with most protective associations becoming statistically nonsignificant after comprehensive adjustments. While the crude model suggested some potential correlations, adjusted models failed to confirm a robust relationship between WCS duration and hypertension risk.

Table 3.

Regression analyses for associations between weekend catch-up sleep duration (WCS) and hypertension symptoms by different genders in U.S. middle-aged and older adults.

Grouped by genders	Crude model	Adjusted Model I	Adjusted Model II
	OR (95% CI) P value
Male
WCS duration (continuous)	0.90 (0.87–0.94) < .0001*	0.95 (0.92–0.99) .0224*	0.95 (0.90–0.99) .0187*
WCS (duration > 0 h)
No	Reference	Reference	Reference
Yes	0.78 (0.67–0.92) .0021*	0.74 (0.63–0.88) .0004*	0.76 (0.63–0.92) .0044*
WCS duration (multicategory)
≤ 0 h	Reference	Reference	Reference
0–2 h	0.82 (0.70–0.96) .0148*	0.75 (0.64–0.89) .0008*	0.79 (0.65–0.95) .0132*
≥ 2 h	0.62 (0.51–0.76) < .0001*	0.71 (0.58–0.88) .0014*	0.68 (0.54–0.86) .0011*
Female
WCS duration (continuous)	0.90 (0.87–0.94) < .0001*	0.99 (0.95–1.03) .6424	0.99 (0.95–1.04) .7558
WCS (duration > 0 h)
No	Reference	Reference	Reference
Yes	0.85 (0.74–0.99) .0343*	0.84 (0.71–0.99) .0331*	0.86 (0.72–1.04) .1233
WCS duration (multicategory)
≤ 0 h	Reference	Reference	Reference
0–2 h	0.89 (0.77–1.03) .1212	0.82 (0.70–0.97) .0211*	0.86 (0.71–1.04) .1169
≥ 2 h	0.71 (0.59–0.86) .0003*	0.90 (0.73–1.10) .3088	0.88 (0.70–1.11) .2885

Crude model: unadjusted.

Adjusted Model I: adjusted by age and race.

Adjusted Model II: adjusted by Adjusted Model I + education, BMI, smoke status, alcohol user status, and DM status.

BMI = body mass index, CI = confidence interval, DM = diabetes mellitus, OR = odds ratio, WCS = weekend catch-up sleep duration.

^*P < .05.

Stratified analysis by weekday sleep duration revealed distinct patterns of association between WCS and hypertension risk (Table 4). For participants with weekday sleep ≤ 7 hours, the protective effects of WCS were largely attenuated, with no statistically significant associations observed in adjusted models. In contrast, individuals with weekday sleep > 7 hours demonstrated a consistent and significant inverse relationship between WCS and hypertension risk across all analytical models. In the > 7 hours weekday sleep group, continuous WCS duration showed a significant protective effect (Adjusted Model II: OR: 0.92, 95% CI: 0.87–0.97, P = .0009). Participants with > 0 hours of WCS exhibited substantially reduced hypertension risk (Adjusted Model II: OR: 0.77, 95% CI: 0.66–0.89, P = .0007). The most pronounced risk reduction was observed in those with ≥ 2 hours of WCS (OR: 0.64, 95% CI: 0.50–0.81, P = .0003), suggesting a potential dose–response relationship that is particularly pronounced among individuals with longer weekday sleep duration.

Table 4.

Logistic regression analyses for associations between weekend catch-up sleep duration (WCS) and hypertension symptoms by different sleep hours on weekdays in U.S. middle-aged and older adults.

Grouped by sleep hours on weekdays	Crude model	Adjusted Model I	Adjusted Model II
	OR (95% CI) P value
Sleep hours ≤ 7
WCS duration (continuous)	0.93 (0.89–0.96) < .0001*	1.00 (0.96–1.04) .8600	1.01 (0.96–1.05) .8071
WCS (duration > 0 h)
No	Reference	Reference	Reference
Yes	0.84 (0.68–1.04) .1141	0.83 (0.66–1.05) .1135	0.94 (0.73–1.22) .6438
WCS duration (multicategory)
≤ 0 h	Reference	Reference	Reference
0–2 h	0.90 (0.73–1.12) .3576	0.82 (0.65–1.03) .0937	0.94 (0.72–1.22) .6471
≥ 2 h	0.71 (0.56–0.90) .0042*	0.86 (0.67–1.10) .2266	0.94 (0.71–1.25) .6773
Sleep hours > 7
WCS duration (continuous)	0.87 (0.83–0.91) < .0001*	0.92 (0.88–0.97) .0005*	0.92 (0.87–0.97) .0009*
WCS (duration > 0 h)
No	Reference	Reference	Reference
Yes	0.82 (0.72–0.93) .0019*	0.77 (0.67–0.88) .0001*	0.77 (0.66–0.89) 0.0007*
WCS duration (multicategory)
≤ 0 h	Reference	Reference	Reference
0–2 h	0.84 (0.74–0.95) .0074*	0.78 (0.68–0.89) .0003*	0.79 (0.67–0.92) .0025*
≥ 2 h	0.64 (0.53–0.79) < .0001*	0.72 (0.58–0.90) .0030*	0.64 (0.50–0.81) .0003*

Crude model: unadjusted.

Adjusted Model I: adjusted by sex, age, and race.

Adjusted Model II: adjusted by Adjusted Model I + education, BMI, smoke status, alcohol user status, and DM status.

BMI = body mass index, CI = confidence interval, DM = diabetes mellitus, OR = odds ratio, WCS = weekend catch-up sleep duration.

^*P < .05.

4. Discussion

This study is the first to examine the relationship between WCS and hypertension incidence in a nationally representative sample of U.S. middle-aged and older adults. Our findings suggest that WCS may significantly reduce hypertension risk, with particularly pronounced effects observed in men and those who obtain more than 7 hours of sleep on weekdays. These results highlight the potential of WCS as a beneficial practice for cardiovascular health, particularly in populations where the impact on hypertension risk is most substantial.

WCS may influence hypertension risk through several mechanisms, such as realigning disrupted circadian rhythms, reducing sympathetic nervous system activity, and lowering systemic inflammation. Our findings are supported by previous studies, such as Hwangbo et al,^[13] who observed that WCS significantly reduced hypertension risk, particularly in those with subjective sleep insufficiency. Similarly, Han et al^[20] found that WCS was associated with lower levels of inflammatory markers linked to hypertension, further supporting the protective role of WCS. However, our study also reveals some divergences; for instance, Gupta et al^[14] found that children with primary hypertension had less WCS, indicating that the impact of WCS may vary with age and health status. The lack of significant findings in some of our adjusted models suggests that while WCS can reduce hypertension risk, the effects may depend on the duration and consistency of catch-up sleep. These results emphasize the potential for WCS to be used as a preventive strategy against high blood pressure, particularly in those unable to achieve adequate sleep during the week, but also suggest the need for personalized sleep strategies to maximize its benefits.

The findings of this study are generally consistent with prior research on the association between WCS and hypertension, while also offering new perspectives to deepen understanding in this area. Luo et al,^[16] using data from the U.S. population, applied generalized additive models and revealed a nonlinear association between WCS and hypertension risk. They reported that a moderate amount of WCS (<4 hours) was associated with a significant reduction in hypertension risk, whereas longer durations (>4 hours) did not confer additional benefit. In contrast, the present study categorized WCS duration into 0 to 2 hours and ≥ 2 hours, and observed that both groups experienced significantly reduced hypertension risk (OR = 0.82 and 0.77, respectively). These results suggest that, within the study population – predominantly middle-aged and older adults – longer durations of WCS may still maintain or even enhance its protective effect. This discrepancy may be attributable to differences in participant age, baseline cardiovascular risk, or sociocultural and behavioral sleep patterns. Whereas Luo et al study included a broader age range, the current study focused on adults aged 45 and older, potentially contributing to the observed heterogeneity in outcomes.^[16]

Similarly, Zhu et al^[15] identified that WCS (>2 hours) was protective against cardiovascular disease, especially in individuals with < 6 hours of weekday sleep, highlighting the compensatory role of WCS for short baseline sleep. Notably, Hwangbo et al^[13] demonstrated that each additional hour of WCS was associated with a 17% lower risk of hypertension among Korean adults, particularly among those experiencing subjective sleep insufficiency. This aligns with our findings in older adults with sufficient weekday sleep. Moreover, Gupta et al^[14] observed that hypertensive children obtained less WCS and reported lower subjective sleepiness, indicating a potentially diminished physiological or behavioral sleep compensation mechanism in pediatric hypertension. Taken together, these studies highlight the complex interplay between WCS, weekday sleep duration, age, and subjective sleep perception. Our study observed a consistent inverse association between WCS and hypertension even among individuals with sufficient weekday sleep (>7 hours), suggesting a broader benefit beyond compensation for sleep debt. While, we found this association to be more prominent in males, whereas previous studies either did not assess sex differences or reported mixed results, thereby providing a more nuanced understanding of WCS and hypertension risk.

Son et al^[21] investigated a postmenopausal female population and found that WCS was associated with a decreased risk of hypertension among individuals with insufficient weekday sleep. These findings echo earlier research conducted in younger cohorts and pediatric populations,^[22] supporting the potential of WCS as an intervention to mitigate the adverse cardiovascular effects of chronic sleep restriction. In addition, Li et al^[22] examined WCS in the context of all-cause mortality and hypertension, finding that WCS may exert indirect protective effects on cardiovascular outcomes, even when statistical interactions were more evident for mortality endpoints. Unlike previous studies that primarily focused on sleep duration or sleep disorders in relation to hypertension,^[23,24] the present study emphasized the role of WCS as a unique compensatory sleep behavior. The findings contribute to a nuanced understanding of how sleep timing and recovery patterns influence blood pressure regulation. Existing evidence has consistently identified insufficient sleep and sleep fragmentation as risk factors for hypertension. However, the current study highlights that appropriate amounts of compensatory sleep may mitigate physiological stress induced by chronic sleep deprivation, thereby supporting cardiovascular health.

Importantly, this study also conducted a gender-stratified analysis, revealing notable sex-specific differences in the association between WCS and hypertension risk. These differences may arise from both physiological and behavioral mechanisms. From a biological standpoint, sex differences in sleep regulation and cardiovascular responses are well-established. Evidence suggests that among men, WCS may help reduce sympathetic nervous system overactivity, attenuate inflammatory responses, and improve vascular function in the context of chronic sleep loss, thus lowering the risk of hypertension.^[25,26] In contrast, among women – particularly those undergoing menopause – hormonal fluctuations and endocrine dysregulation may increase cardiovascular vulnerability to sleep disturbances, thereby diminishing the protective effects of WCS.^[27] Additionally, women often report lower subjective sleep quality and more frequent sleep fragmentation, which may confound or attenuate the independent impact of WCS in multivariate models.

Behavioral and lifestyle factors may also contribute to these gender differences. Compared to men, women tend to experience greater external stressors related to family, career, and social responsibilities, which may disrupt sleep continuity and reduce the effectiveness of weekend sleep compensation.^[28,29] Furthermore, psychological factors such as anxiety, emotional stress, and depression – more prevalent among women – may further influence both sleep quality and cardiovascular outcomes.^[30] Consequently, the protective association between WCS and hypertension may be less robust or statistically nonsignificant in women after adjustment for these confounders. These observations underscore the importance of considering gender-specific mechanisms and behavioral patterns in sleep research.

In comparing with earlier studies, this work supports the protective role of moderate WCS, as reported by Luo et al,^[16] in reducing hypertension risk in U.S. adults. However, the identification of sex-related heterogeneity in this study offers valuable insights for future research. Longitudinal studies incorporating objective sleep measurements and biomarker analyses are needed to further elucidate the gender-specific pathways through which WCS influences cardiovascular health. Moreover, future investigations should incorporate psychosocial variables to better understand additional risk modifiers, particularly in women.

The stratified analysis based on weekday sleep duration revealed that the protective effect of WCS on hypertension varied depending on baseline sleep patterns. Among participants with ≤ 7 hours of sleep on weekdays, WCS was not significantly associated with reduced hypertension risk after adjustment. However, among those with > 7 hours of weekday sleep, a clear inverse association was observed, with longer durations of WCS associated with progressively lower odds of hypertension – especially for those with ≥ 2 hours of WCS. These findings suggest that baseline sleep sufficiency may modulate the effectiveness of WCS in promoting cardiovascular health. Individuals with adequate weekday sleep may possess more stable circadian rhythms and better sleep quality, thereby deriving greater physiological benefit from additional weekend sleep.^[31] Mechanistically, extended sleep in these individuals may reduce sympathetic activation, systemic inflammation, and endothelial dysfunction – all key contributors to hypertension.^[31] Conversely, in those with chronic sleep deprivation, short-term WCS may be insufficient to reverse accumulated physiological stress and its effects on blood pressure regulation.

This observation aligns with studies by Kohyama et al and Chaput et al,^[32,33] which reported that the benefits of WCS are more pronounced among individuals with sufficient baseline sleep. Device-based monitoring data from Chaput et al further revealed a dose–response reduction in hypertension and cardiovascular events with increasing WCS duration among those with longer weekday sleep. These results underscore the importance of considering both sleep quantity and structure when evaluating the health impacts of sleep patterns.

Several limitations exist in this study, which are outlined as follows. First, the cross-sectional nature of this study precludes definitive conclusions regarding causality, making it unclear whether sleep patterns influence hypertension or if existing hypertension alters sleep behaviors. Second, the NHANES database does not include data on subjective sleep insufficiency or perceived sleepiness, which previous research suggests significantly impact the effectiveness of WCS. This lack of data prevents us from assessing these variables in our analysis. Third, the specific sleep needs of adolescents, who may require more precise sleep recovery strategies, were not adequately addressed due to the lack of targeted data. These limitations suggest that while our findings offer valuable insights, they should be interpreted cautiously, and further research is needed to address these gaps.

5. Conclusion

This study shows a link between catching up on sleep on weekends and a lower risk of hypertension in American middle aged and old adults. We found this effect was especially strong for men and those who sleep more than 7 hours during the week. They suggest that sleep interventions aimed at cardiovascular risk reduction should take into account an individual’s baseline sleep patterns, distribution, and long-term behavioral rhythms. Future research should employ longitudinal designs integrating objective sleep tracking and physiological biomarkers to clarify the long-term cardiovascular implications of WCS under different baseline sleep conditions. Such efforts will provide a foundation for developing personalized sleep-based strategies for hypertension prevention and cardiovascular health promotion.

Acknowledgments

We are extremely grateful to all the families participating in this study.

Author contributions

Conceptualization: Lisheng Zheng.

Data curation: Yudong Ba.

Formal analysis: Yemeng Zhang, Haidong Wang, Yanan Li, Fangyuan Li.

Funding acquisition: Haidong Wang.

Investigation: Yemeng Zhang, Yanan Li.

Methodology: Yanliang Yin.

Software: Min Wang, Jinxiu Zhuo.

Writing – original draft: Haidong Wang, Yudong Ba.

Writing – review & editing: Lisheng Zheng.

Supplementary Material

medi-104-e44858-s001.docx ^{(17.4KB, docx)}