Weekend warrior and the risk of specific disease: a meta-epidemiology study

Abstract

Background

The "Weekend Warrior (WW)" physical activity (PA) pattern, involving intensive exercise on 1–2 days per week, has become increasingly popular. The WW PA pattern demonstrates protective effects against a broad spectrum of chronic diseases; however, regarding a comprehensive investigation into the disease-specific protective mechanisms and long-term health outcomes of the subject, it remains unclear. WW exhibits protective effects against various diseases; however, there is a conspicuous scarcity of literature investigating its protective mechanisms across different disease conditions. The objective of this meta-epidemiology study was to exam WW’s protective effects by synthesizing data from published observational studies.

Methods

A systematic search was conducted across databases including PubMed, Embase, Cochrane Library and Web of Science through February 19, 2025. The search focused on observational studies reporting the association between the WW PA pattern and various health outcomes, including cardiovascular diseases, mortality, metabolic syndrome, and mental health, compared to inactive individuals. Odds ratios (ORs) were pooled using random-effects models. Subgroup analyses were performed to investigate the association with ORs of factors, such as sex, study type, and PA assessment.

Results

Twenty-seven studies encompassing 1,204,486 participants were included. The pooled analysis indicated that the WW exercise pattern significantly reduced the risk of CVD mortality (OR = 0.742, 95% CI: 0.568–0.968), I² = 71.3%, P = 0.028). Additionally, WW showed lower risks of mental disorders and metabolic syndrome.

Conclusion

The WW PA pattern is associated with significant health benefits, including reduced risks of mortality, cardiovascular diseases, and metabolic syndrome. This pattern may be a viable alternative for individuals unable to engage in daily physical activity. Future research should further explore the long-term effects and refine exercise recommendations for various population subgroups.

PROSPERO registration number

CRD42024587216.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-025-22667-7.

Background

Sufficient literature indicates that physical activity (PA) is associated with a lower risk of non-communicable diseases [1–3]. WHO guidelines recommend that all adults should engage in 150–300 min of moderate-intensity PA, 75–150 min of vigorous-intensity PA, or an equivalent combination of moderate- and vigorous-intensity PA per week [4]. Although there are guideline recommendations and population-wide strategies to increase PA levels, the majority of the population does not adhere to PA recommendations. A primary reason for low adherence is a lack of time. A “Weekend Warrior (WW)” is an individual who engages in vigorous physical activities or sports primarily for 1–2 days per week, remaining relatively inactive during the rest of the time [5]. It is a kind of exercise mode that has become popular in recent years. Previous studies have shown that the WW PA pattern has a protective effect against a wide range of diseases compared to inactive individuals [2]. These benefits include reduced risks of cardiovascular diseases, all-cause mortality, metabolic syndrome, mental health disorders, and neurological conditions [6–10]. In recent years, interest in this group has increased. While existing studies have reviewed the association between WW and mortality [11], there are obviously few literatures that have conducted research on the relationship between WW and the risk of various diseases. Given the rising prevalence of WW in contemporary society, it is imperative to investigate the associated health benefits of this exercise pattern. Therefore, we conducted a systematic review of existing population-based longitudinal studies to elucidate the relationship between WW and the risk of specific diseases. Thereby exploring a new alternative to the traditional sports mode as a solution.

Methods

This meta-analysis was conducted in accordance with the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [12]. The protocol was pre-registered in the International Prospective Register of Systematic Reviews (PROSPERO) platform, and the approval number is CRD42024587216.

Retrieval strategy

A systematic review of the literature was conducted using PubMed, Embase, Cochrane Library and Web of Science from database inception to February 19, 2025, and the reference search was conducted for the included literature simultaneously. The search process and terms used are listed in Appendix 1– 4 in Supplement 1.

Eligibility criteria

All documents were selected based on the following Population, Exposure, Comparator, and Outcomes (PECO) model: (a) Participants: Adults who underwent physical examinations without any hidden diseases and exhibited various PA patterns; (b) Exposure: Exercise conditions, including the duration and intensity of exercise; (c) Comparator: WW and inactive physical activity patterns; (d) Outcomes: Presentation of quantitative point estimates [odds ratios (ORs)] and variance of the estimates of the association between exercise patterns and the risk of specific diseases in WW, along with adjustment for potential confounders; (e) Type of study: observational study.

Research papers were only taken into consideration for inclusion if they provided conclusive data once the exposure was finished. The exclusion criteria were as follows:

(1) Duplicate articles, (2) studies involving animals, (3) studies written in a language different from English, and (4) systemic diseases.

Data extraction

We developed a data extraction template using Microsoft Excel. Two authors (KQ Fu and JL Wang) independently extracted information from eligible studies, including details such as the first author, publication date, country, and confounders. Data extraction was cross-validated, and discrepancies were resolved through discussion with a third reviewer (XL Li).

Exposure and Outcome

Exposure was the WW PA pattern. The major outcome was the protective effect against various diseases.

Study quality assessment

The quality of included cohort studies was assessed using the Newcastle–Ottawa Quality Assessment Scale (NOS) (available from: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp), evaluating three domains: selection, comparability, and outcomes. Cohort studies received scores ranging from 0 to 9 stars, with higher scores indicating better quality. NOS ratings of ≥ 7, 4–6, and 0–3 were classified as high, moderate, and low quality, respectively. For cross-sectional studies, we employed the Agency for Healthcare Research and Quality (AHRQ) 11-item assessment, where a"yes"response scored 1 point and"no"or"unclear"scored 0 points. Scores of 0–3, 4–7, and 8–11 were categorized as low, medium, and high quality, respectively.

Statistical analysis

Meta-analysis data were analyzed using STATA statistical software version 14.0. A bilateral P-value with an α level of 0.05 was considered significant. We employed a random effects model (DerSimonian and Laird methods) to compute the combined OR with a 95% CI, assessing the association between WW and the risk of specific disease. Due to the relatively low incidence of WW and the risk of specific disease, OR was used as a proxy for RR. We prioritized risk estimates from multivariate models fully adjusted for confounding factors. Random effects models based on heterogeneity levels were utilized to synthesize specific risk estimates across all studies. Heterogeneity was quantified using I² statistics (0–25%: low, 25%− 50%: moderate, 50%− 75%: substantial, 75%− 100%: high). Subgroup analyses were conducted based on sex, study type, and PA assessment. Publication bias was evaluated through Egger’s test, or filled funnel plots.

Results

Study selection

Our initial search yielded 476 records, from which 243 duplicates were removed. Following title and abstract screening, 192 records were excluded due to irrelevance. Reading the full text, 14 records were excluded due to meeting abstracts or inability to extract data. The remaining 27 studies underwent further eligibility assessment, ultimately resulting in 27 observational studies included in the meta-analysis. The study selection process is illustrated in Fig. 1.

Fig. 1.

Flow chart of literature screening

Study characteristics

27 studies encompassing 1,204,486 individuals (192,734 WWs and 600,819 inactive controls) were included in this analysis. The majority of participants were from China, the United States, and Switzerland. Studies were published between 2004 and 2025, comprising 16 cohort studies and 11 cross-sectional studies. The analysis focused on five health outcomes: cardiovascular outcomes, specific mortalities, mental disorders, metabolic syndrome, and neurological diseases. Commonly adjusted confounders included age, sex, and education. Table 1 summarizes the study characteristics, while Table 2 details the exercise features of the WW programs.

Table 1.

Basic characteristics of included studies

Author	Study name	Year	Country	Population source	Study type	Sample size	Follow-up years	Baseline age (years)	Age range (years)	Males (%)	Outcomes assessed (no. of events)	Confounder adjusted for
Mahe et al. (2024) [13]	NHANES	2024	China	Population register	Prospective cohort	Total:6,067 WW:128 Regular:1,054 Inactive:4,885	6.1 averages	61.4 averages	≥ 18	52	All-cause mortality (25)	Age, gender, race and ethnicity, levels of education and income,marital status,cigarette smoking,alcohol intake,diet quality, and BMI
O'Donovan et al.(2024) [14]	Mexico City Prospective Study	2024	Colombia	Population register	Prospective cohort	Total:10,033 WW:726 Regular:1,362 Inactive:7,945	16 averages	51 averages	≥ 35	29.98	Mild dementia (2,400)	Age,sex, education, income,systolic blood pressure, smoking, body mass index, civil status, sleep duration, fruit and vegetable intake, and alcohol
Xue et al. (2024) [15]	NHANES	2024	China	Population register	Cross-sectional study	Total:19,223 WW:718 Regular:5,326 Inactive:10,266 Insufficient: 2,913	NA	50.02 averages	≥ 18	49.18	Obstructive sleep apnea symptoms (10,557)	Age,gender, racial background, educational level, the ratio of family income to poverty level, BMI, smoking status, and alcohol
Hui et al. (2024) [16]	Hong Kong Territory-Wide Physical Fitness Survey	2024	China	Population register	Cross-sectional study	Total:9,863 WW:626 Regular:2,893 Inactive:2,486 Insufficient: 3,858	NA	17 ~ 79	≥ 17	39.06	Obesity	Sex, age, educational level, household income, smoking status, hypertension, and sleep quality
Li et al. (2025) [17]	NHANES	2025	China	Population register	Cross-sectional study	Total:5,102	NA	62 averages	NA	NA	Diabetic retinopathy	Age, sex, ethnicity, education level, poverty income ratio,marital status, alcohol consumption, smoking, BMI, hypertension, and hyperlipidaemia
Chen et al. (2025) [18]	NHANES	2025	China	Population register	Cross-sectional study	Total:22,455 WW:943 Regular:6,534 Inactive:11,635 Insufficient: 3,343	NA	47.6 averages	≥ 20	49.2	Diabetes	Gender, age, race, education level, income level, and BMI
He et al. (2024) [19]	UK Biobank	2024	China	Population register	Prospective cohort	Total:1,303 WW:287 Regular:510 Inactive:506	7.8 averages	40 ~ 69	≥ 40	46.2	Incident bowel resection (3,132)	Age, sex, ethnicity, education attainment, Townsend deprivation index, employment, BMI, smoking status, drinking status, and diet
Wang et al. (2025) [20]	NHANES	2025	China	Population register	Retrospective cohort	Total:6,482 WW:326 Regular:2,098 Inactive:3,020 Insufficient: 1,038	13 averages	20 ~ 60	≥ 20	50.15	Bone mineral density	Age, gender, race, personal habits, custom, BMI, drinking habit, serum cotinine, serum phosporus, serum calcium, hypertension status, diabetes status, thyroid problem, and liver condition
Xiong et al. (2024) [21]	NHANES	2024	China	Population register	Cross-sectional study	Total:12,691 WW:387 Regular:6,143 Inactive:4,699 Insufficient: 1,732	NA	54.31 averages	NA	53.24	Albuminuria	Age, BMI, alanine aminotransferase (ALT), aspartate aminotransferase (AST), serum creatinine, uric acid, gender, race, education level, marital status, poverty income ratio (PIR), smoking, alcohol status, CVD, diabetes, hyperlipidemia, and medication use
Khurshid et al. (2023) [6]	UK Biobank cohort study	2023	USA	Population register	Prospective cohort	Total:89,573 WW:37,872 Regular: 21,473 Inactive: 30,228	6.3 averages	62 averages	≥ 40	44	Atrial fibrillation (2,642)/Myocardial infarction (1,229)/Heart failure (1,220)/Stroke(891)	Age, sex, racial and ethnic background, tobacco use, alcohol intake, Townsend Deprivation Index, employment status, self-reported health, and diet quality
Gubelmann et al.(2018) [22]	CoLaus study	2018	Switzerland	Population register	cross-sectional study	Total:2,605 WW:592 Regular: 1,162 Inactive:851	NA	45 ~ 86	≥ 45	45.6	Smoking/Obesity/Hypertension/Dyslipidemia/Diabetes	Age, gender, professional occupation, educational level, household income, and accelerometer diurnal wear-time
Yang et al. (2024) [7]	NHANES	2024	China	Population register	Prospective cohort	Total:1,702 WW:35 Regular:239 Inactive: 1,218 Insufficient: 210	3.25 ~ 8.67	64 averages	≥ 18	51.54	All-cause mortality(536)/CVD mortality(163)	Age, sex, racial background, education, family income-to-poverty level ratio, smoking, alcohol consumption, BMI, HbA1c, total cholesterol, MAP
Xie et al. (2024) [23]	UK Biobank	2024	China	Population register	Prospective cohort	Total:10,210	6.9 averages	58.1 averages	NA	56.5	Incident CVD (1,538)/CHD(392)/CVD-cause mortality(90)/All-cause mortality(358)	Age, sex, body mass index, ethnic background, economic status, education, and employment
O'Donovan et al.(2024) [14]	Mexico City Prospective Study	2024	Colombia	Population register	Prospective cohort	Total:154,882 WW:11,981 Regular: 22,576 Inactive: 120,325	14 ~ 22	39 ~ 65	≥ 35	32.76	All-cause mortality (26,006)/CVD mortality (8,750)/Cancer mortality (3,409)	Age, sex, education, income, smoking, diet, and alcohol
Dos Santos et al.(2022) [24]	NHIS	2022	Brazil	Population register	Prospective cohort	Total:350,978 WW:9,992 Regular: 150,906 Inactive: 190,080	10.4 averages	41.4 averages	≥ 18	49.2	All-cause mortality (21,898)/CVD mortality (4,130)/Cancer mortality (6,034)	Age, sex, race and ethnicity, marital status, income and educational attainment
Shiroma et al. (2019) [25]	NHANES	2019	USA	Population register	Prospective cohort	Total:3,438 WW:1,110 Regular:1,068 Inactive: 1,260	6.45 averages	57.1 averages	≥ 40	48.7	All-cause mortality (394)	Age, gender, race, education, smoking status, alcohol use, healthy eating index, body mass index, diabetes status, and physical function and limitations
O'Donovan et al.(2017) [26]	HSE and SHS	2017	UK	Population register	Prospective cohort	Total:63,591 WW:2,341 Regular:7,079 Inactive: 39,947 Insufficient: 14,224	8.8 averages	58.6 averages	≥ 40	45.9	All-cause mortality (8,802)/CVD mortality (2,780)/Cancer mortality (2,526)	Age, sex, smoking habit, long-standing illness, occupation, and ethnicity
Kohl et al. (2005) [27]	Harvard Alumni Health Study	2005	USA	Alumni register	Prospective cohort	Total:12,805	9 averages	66 averages	NA	100	All-cause mortality (0.99%)	Anthropomorphic data, lifestyle and medical history risk factors, and early parental mortality
Lee et al. (2004) [28]	Harvard Alumni Health Study	2004	USA	Alumni register	Prospective cohort	Total:8,421 WW:580 Regular:5,261 Sedentary: 1,453 Insufficient: 1,127	9 averages	66.3 averages	NA	100	All-cause mortality (1234)	Cigarette smoking, alcohol consumption, red meat intake, vegetable intake, vitamin/mineral supplements, and early parental mortality
Chen et al. (2023) [29]	NHANES	2023	China	Population register	Cross-sectional study	Total:21,125	NA	≥ 20	≥ 20	51.23	Depression risk (8.14%)	Gender, age, race, education level, income level, marital status, Body mass index (BMI), smoking status, alcoholism, CVD, diabetes and hypertension
Gubelmann et al.(2018) [22]	CoLaus study	2018	Switzerland	Population register	Cross-sectional study	Total:2,649 WW:617 Regular:1,150 Inactive:882	NA	61.6 averages	≥ 45	46.5	Short sleep/Poor sleep quality/Excessive daytime sleepiness/Increased risk of sleep apnea/Insomnia/Chronotype (2,649)	Age, gender, professional occupation, smoking, alcohol consumption, and BMI
Hamer et al. (2017) [30]	HSE and SHS	2017	UK	Population register	Cross-sectional study	Total:82,552 WW:8,921 Regular: 18,976 Inactive: 54,655	NA	47 averages	≥ 30	46.5	Psychological distress (82,552)	Age, sex, smoking habit, longstanding illness, social occupational class, and BMI
Ye et al. (2024) [31]	UK Biobank	2024	China	Population register	Prospective cohort	Total:77,977 WW:34,190 Regular: 18,003 Inactive: 25,784	6.8 averages	55.9 averages	≥ 37	44.2	CKD (1,324)/AKI(1,515)	Age, sex, ethnicity, Townsend Deprivation Index (TDI), income, education, employment status, smoking status, and alcohol consumption
Shin et al. (2024) [8]	Korea National Health and Nutrition Examination Survey (KNHANES)	2024	USA	Population register	Cross-sectional study	Total:26,197 WW:1,364 Regular:2,186 Inactive: 22,647	NA	44.1 averages	≥ 20	50.4	MS risk (26,197)	Sex, education, household income, drinking habits, and smoking status
Xiao et al. (2018) [32]	Nantong Metabolic Syndrome Study (NMSS)	2018	China	Population register	Cross-sectional study	Total:20,378 WW:1,540 Regular: 15,249 Inactive: 2,755 Insufficient: 834	NA	18 ~ 74	≥ 18	34.13	Metabolic syndrome(MS) (4,422)/Hypertension (7,363)/Diabetes (1,333)/Obesity(1,559)	Age, smoking status, alcohol consumption, foods intake, and education
Lin et al. (2024) [10]	UK Biobank	2024	China	Population register	Prospective cohort	Total:89,400 WW:37,241 Regular: 20,381 Inactive: 31,778	12.32 averages	56.05 averages	NA	43.7	Parkinson's disease (329)	Age, sex, drinking status, blood pressure status, diabetes, and family history of patients with PD
Ning et al. (2024) [33]	UK Biobank	2024	China	Population register	Prospective cohort	Total:92,784 WW:40,217 Regular: 21,053 Inactive: 31,514	8.85 averages	61.88 averages	≥ 40	43.62	Parkinson's disease (PD) (266)/Dementia (486)	Age, sex, ethnic type, educational attainment, employment status, lifestyle, clinical, medication and medical history

ICD International Classification of Diseases, MS Metabolic syndrome, CVD Cardiovascular disease, CKD Chronic kidney disease, AKI Acute kidney injury

Table 2.

WW characteristics of included studies

Author	Year	Country	Physical activity assessment	Physical activity patterns	Type of questionnaire	Type of physical activity exposure	Quantification of physical activity exposure	Definition of weekend warrior
Khurshid et al. (2023) [6]	2023	USA	Accelerometry	Inactive/Regular/WW	NA	MVPA	NA	≥ 150 min of MVPA/week (guideline based) with ≥ 50% over 1–2 day
Gubelmann et al (2018) [22]	2018	Switzerland	Accelerometry	Inactive/Regular/WW	NA	MVPA	> 338 g·min	MVPA ≥ 133 min/day/PA mainly on weekends
Mahe et al. (2024) [13]	2024	China	Self-reported	Inactive/Regular/WW	Global Physical Activity Questionnaire	MVPA	2*VPA + MPA	Accumulating at least 150 min/week of MVPA in 1 or 2 sessions
Yang et al. (2024) [7]	2024	China	Self-reported	Inactive/Regular/WW/Insufficient	Global Physical Activity Questionnaire	Total PA	2*vigorous PA + moderate PA	At least 150 min/week of total PA from 1–2 sessions
Xie et al. (2024) [23]	2024	China	Accelerometry	Inactive/Regular/WW	NA	MVPA	NA	> 150 min/week with > 50% of total MVPA achieved in 1–2 days
O'Donovan et al. (2024) [14]	2024	Colombia	Survey data	Inactive/Regular/WW	NA	Leisure-time physical activity	NA	Exercised or played sports up to once or twice per week
Dos Santos et al. (2022) [24]	2022	Brazil	Self-reported	Inactive/Regular/WW	NHIS adult physical activity questionnaire	MVPA	Summing the duration (in min) of moderate intensity plus vigorous intensity and multiplying by 2	≤ 2 MVPA sessions/wk
Shiroma et al. (2019) [25]	2019	USA	Accelerometry	Inactive/Regular/WW	NA	MVPA	Any minute where the accelerometer registered ≥ 1952 counts, where a count is a summary measure of the intensity of acceleration over time	Accrued ≥ 50% of their weekly MVPA on only 1 or 2 d
O'Donovan et al. (2017) [26]	2017	UK	Self-reported	Inactive/Regular/WW/Insufficient	An established questionnaire	MVPA	Moderate activities consisted of 3.0 to 5.9 metabolic equivalents (METs), and vigorous activities consisted of 6.0 METs or more, with 1 MET representing resting energy expenditure	At least 150 min/wk in moderate-intensity physical activity or at least 75 min/wk in vigorous-intensity physical activity from 1 or 2 sessions
Kohl et al. (2005) [27]	2005	USA	Self-reported	Sedentary/Reugular/WW/Insufficient	Havard Alumni Physical Activity Questionnaire	Baseline physical activity	Daily walking and stair climbing and frequency and duration of sports and recreational activities during the previous week	1000 kcal/wk expended in 1 or 2 activity sessions per week
Lee et al. (2004) [28]	2004	USA	Self-reported	Sedentary/Regular/WW/Insufficient	Havard Alumni Physical Activity Questionnaire	Sports and recreational activities undertaken in the past week and the frequency and duration of participation	Energy cost of each activity	Expending 1,000 kcal/week or more by participating in sports and recreational activities 1–2 times/week
O'Donovan et al. (2024) [14]	2024	Colombia	Self-reported	Inactive/Regular/WW	Physical activity questions	Exercise time	NA	Exercised or played sports up to once or twice per week
Xue et al. (2024) [15]	2024	China	Self-reported	Inactive/Regular/WW/Insufficient	Global Physical Activity Questionnaire	Total PA time	2*vigorous PA + moderate PA	PA 1–2 sessions per week totaling 150 min
Chen et al. (2023) [29]	2023	China	Self-reported	Inactive/Regular/WW/Insufficient	Global Physical Activity Questionnaire	Vigorous-intensity activity&Moderate-intensity activity	Exercises that cause large increases in breathing or heart rate&Exercises that cause relatively small increases in breathing or heart rate	At least 150 min of total PA in 1 or 2 sessions per week
Gubelmann et al. (2018) [22]	2018	Switzerland	Accelerometry	Inactive/Regular/WW	NA	PA and SB	The lowest and highest tertile of each behavior	High PA and PA mainly on weekends
Hamer et al (2017) [30]	2017	UK	Self-reported	Inactive/Regular/WW	An established questionnaire	MVPA	Moderate activities were of 3.0–5.9 metabolic equivalents (METs) and vigorous activities were of ≥ 6.0 METs, where one MET is considered to represent resting energy expenditure	≥ 150 min/wk
Ye et al. (2024) [31]	2024	China	Accelerometry	Inactive/Regular/WW	NA	MVPA	≥ 150 min per week	Meeting or exceeding the MVPA threshold and accumulating ≥ 50% of their total MVPA over 1–2 days
Hui et al. (2024) [16]	2024	China	Self-reported	Inactive/Regular/WW/Insufficient	An established questionnaire	MVPA	NA	Engage in at least 150 min/week of MVPA but only one or two weekly sessions
Li et al. (2025) [17]	2025	China	Self-reported	Inactive/Regular/WW/Insufficient	The physical activity questionnaire	Total PA	2*vigorous PA + moderate PA	Greater than 150 min of total PA per week, in 1–2 sessions
Chen et al. (2025) [18]	2025	China	Self-reported	Inactive/Regular/WW/Insufficient	A standardized questionnaire	Total PA	2*VPA + MPA	Achieved the recommended 150 min of MPA or 75 min of VPA in one or two sessions per week
Shin et al. (2024) [8]	2024	USA	Self-reported	Inactive/Regular/WW	Global Physical Activity Questionnaire (GPAQ)	MVPA	NA	Participated in MVPA only 1–2 days per week
Xiao et al. (2018) [32]	2018	China	Self-reported	Inactive/Regular/WW/Insufficient	International Physical Activity Questionnaire (IPAQ)	Moderate-intensity PA&Vigorous-intensity PA	Activity at a level of 3.0–5.9 metabolic equivalents (METs)&Activity at a level of 6.0 METs or more	At least 150 min/week of moderate intensity PA or at least 75 min/week of vigorous-intensity PA on 1 or 2 sessions during the weekend or equivalent combinations
Lin et al. (2024) [10]	2024	China	Accelerometery	Inactive/Regular/WW	NA	MVPA	NA	Complete the majority of their weekly exercise in 1–2 days
Ning et al. (2024) [33]	2024	China	Accelerometery	Inactive/Regular/WW	NA	MVPA	NA	Reached the MVPA threshold and concentrated more than 50% of their total weekly MVPA exercise minutes over 1–2 days
He et al. (2024)	2024	China	Accelerometery	Inactive/Regular/WW	NA	MVPA	PA performed at over 3 metabolic equivalents of task (METs), where 1 MET represents the energy expenditure of an individual at rest	≥ 150 min/week with ≥ 50% of MVPA achieved in 1–2 days
Wang et al. (2025) [20]	2025	China	Self-reported	Inactive/Regular/WW/Insufficient	A standardized questionnaire	Total PA	2*vigorous PA + moderate PA	Total PA ≥ 150 min in one or two sessions per week
Xiong et al. (2024) [21]	2024	China	Survey data	Inactive/Regular/WW/Insufficient	A series of survey questions	Total PA	2*vigorous PA + moderate PA	Total PA ≥ 150 min/week and frequency ≤ 2 sessions/week

Quality assessment

Among the cohort studies, twelve received scores of ≥ 7, indicating good quality, while the other half scored 5 or 6, representing moderate to good quality. All eleven cross-sectional studies scored ≥ 10, reflecting high quality. The specific evaluation scores are detailed in Tables 3– 4.

Table 3.

The quality assessment of cohort studies

Study	Year	Selection	Comparability	Outcome	Total
Cohort studies (n = 16)
Wang, Y	2025	***	***	**	8
Mahe, J	2024	**	***	**	7
O'Donovan, G	2024	**	***	***	8
He, Z	2024	**	***	**	7
Yang, Q	2024	**	**	**	6
Xie, Q	2024	**	***	**	7
O'Donovan, G	2024	**	***	***	8
Ye, Z	2024	**	***	**	7
Lin, F	2024	***	***	***	9
Ning, Y	2024	***	***	**	8
Khurshid, S	2023	***	***	*	7
Dos Santos, M	2022	**	***	***	8
Shiroma, E. J	2019	**	***	**	7
O'Donovan, G	2017	*	***	**	6
Kohl, H. W	2005	*	**	**	5
Lee, I. M	2004	*	**	**	5

The NOS scale was used to evaluate the quality of the cohort studies

Table 4.

The quality assessment of cross-sectional studies

Study	Year	1	2	3	4	5	6	7	8	9	10	11	Total
Cross-sectional studies (n = 11)
Chen, Z	2025	1	1	1	1	1	1	1	1	1	1	1	11
Li, B	2025	1	1	1	1	1	1	1	1	1	1	1	11
Xiong, B	2024	1	1	1	1	1	1	1	1	1	1	1	11
Xue, F	2024	1	1	1	1	1	1	1	1	1	1	1	11
Hui, S	2024	1	1	1	1	1	1	1	1	1	1	1	11
Shin, S. W	2024	1	1	1	1	1	1	1	1	1	0	1	10
Chen, R	2023	1	1	1	1	1	1	1	1	1	1	1	11
Gubelmann, C	2018	1	1	1	1	1	1	1	1	1	1	1	11
Gubelmann, C	2018	1	1	1	1	1	1	1	1	1	1	1	11
Xiao, J	2018	1	1	1	1	1	1	1	1	1	1	1	11
Hamer, M	2017	1	1	1	1	1	1	1	1	1	1	1	11

The AHRQ scale was used to evaluate the quality of the cross-sectional studies

Result of WW and the risk of specific mortality

Nine cohorts assessed the relationship between WW activity patterns and specific mortality [7, 13, 14, 23–27, 34]. Our pooled analysis revealed a pronounced decrease in CVD mortality [OR = 0.742, 95% CI (0.568, 0.968), I² = 71.3%, P = 0.028, Fig. 2] and a non-significant association trend for reduced cancer mortality [OR = 0.853, 95% CI (0.766, 0.949), I² = 0.0%, P = 0.004, Fig. 2], underscoring the heterogeneous impact of intermittent physical activity on mortality endpoints. Sensitive analysis showed that none of the individual studies reversed the pooled-effect size, which means that the results are robust (Appendix Fig. 7).

Fig. 2.

Forest plot for the risk of specific mortality in WW by mortality types

Subgroup of WW and the risk of specific mortality

We conducted a subgroup analysis of sex and PA assessment. In the subgroup analysis, the WW PA pattern shows a significant protective effect on all-cause mortality among females. However, for men, since the 95% confidence interval crossed the null line 1, there was no statistical significance (Table 5).

Table 5.

Subgroup analysis for the risk of specific mortality in WW

Subgroups	Included studies	OR] (95%Cl)	Heterogeneity
			I2(%)	P-values
Sex
Female	2	0.711(0.548–0.924)	0.0	0.011
Male	2	0.582(0.236–1.432)	55.9	0.239

Result of WW and the risk of mental disease

Five studies evaluated the risk of mental disease related to PA patterns. The pooled analysis indicated a reduced risk of mental disease: dementia [OR = 0.715, 95% CI (0.620, 0.824), I² = 0.0%, P < 0.001, Fig. 3] [33, 35] and sleep apnea [OR = 0.742, 95% CI (0.539, 1.021), I² = 70.2%, P = 0.067, Fig. 3] [15, 22]. Sensitive analysis showed that none of the individual studies reversed the pooled-effect size, which means that the results are robust (Appendix Fig. 8).

Fig. 3.

Forest plot for the risk of mental disease in WW by mental disease types

Subgroup of WW and the risk of mental disease

We conducted a subgroup analysis of sex, study type, and PA assessment. The results regarding the risk of mental disease in WW did not show any statistical significance in either male or female groups. Besides, the analysis stratified by study type further revealed consistent protective effects across both cohort studies [33, 35] and cross-sectional studies [15, 22, 29, 30], suggesting that the WW pattern confers significant mental health benefits regardless of methodological differences in observational designs. The results indicate that the population with self-reported WW exercise intensity has better health outcomes, which is inconsistent with the findings regarding specific mortality. However, the results obtained from the accelerometer measurements did not have statistical significance (Table 6) (Fig. 4).

Table 6.

Subgroup analysis for the risk of mental disease in WW

Subgroups	Included studies	OR] (95% CI)	Heterogeneity
			I2(%)	P-values
Sex
Female	2	0.677(0.266–1.719)	84.5	0.411
Male	2	0.774(0.341–1.759)	80.7	0.542
Study type
Cohort study	2	0.715(0.620–0.824)	0.0	0.000
Cross-sectional study	4	0.786(0.652–0.947)	75.3	0.011
PA assessment
Accelerometery	2	0.839(0.650–1.084)	69.4	0.179
Self-reported	4	0.720(0.613–0.846)	69.3	0.000

Fig. 4.

Forest plot for the risk of mental disease in WW by study types

Result of WW and the risk of metabolic syndrome

Five studies investigated the risk of metabolic syndrome associated with physical activity patterns. The pooled results demonstrated a reduced risk of metabolic syndrome: metabolic syndrome [OR = 0.703, 95% CI (0.574, 0.861), I² = 53.8%, P = 0.001, Fig. 5] [32, 36], hypertension [OR = 0.748, 95% CI (0.639, 0.877), I² = 0.0%, P < 0.001, Fig. 5] [32]. Among them, the protective effect against diabetes is extremely significant. Concurrently, the results indicate that WW has no definite protective significance for the risk of obesity onset [OR = 0.861, 95% CI (0.688, 1.077), I² = 52.9%, P = 0.190, Fig. 5] [16, 32]. Sensitive analysis indicated that no single study reversed the magnitude of the pooled effect, suggesting that the results were robust (Appendix Fig. 9).

Fig. 5.

Forest plot for the risk of metabolic syndrome in WW by metabolic syndrome types

Subgroup of WW and the risk of metabolic syndrome

We performed a subgroup analysis of sex. The analysis found that the effects of WW and metabolic syndrome were almost the same in male and female populations (Table 7).

Table 7.

Subgroup analysis for the risk of metabolic syndrome in WW

Subgroups	Included studies	OR] (95% CI)	Heterogeneity
			I2(%)	P-values
Sex
Female	2	0.725(0.587–0.896)	63.9	0.003
Male	2	0.683(0.572–0.815)	50.8	0.000

Result of WW and the risk of neurological disorder

Two studies examined the risk of neurological disease associated with physical activity patterns. The pooled analysis indicated a reduced risk of Parkinson’s disease among WWs [OR = 0.521, 95% CI (0.423, 0.643), I² = 0.0%, P < 0.001; Fig. 6] [10, 33]. Sensitive analysis shows that no study can reverse the degree of aggregation effect, which shows that the results are robust (Appendix Fig. 10).

Fig. 6.

Forest plot for the neurological disorder in WW

Result of WW and the risk of cardiovascular outcomes

Two studies evaluated the risk of cardiovascular disease (CVD) associated with physical activity patterns. No studies were included in the analysis. Qualitative analysis indicated that the WW pattern lowers the incidence of CVD compared to inactive populations [6, 37].

Result of WW and the risk of other health outcomes

There are four pieces of study that were not included in the above five categories for analysis and research. They respectively studied the relationship between WW and intestinal disease, bone density, and kidney diseases. Regardless of what disease or body index it is, WW has shown a positive aspect, once again demonstrating that the WW exercise mode can be advocated [19–21, 31].

Publication bias

In this study, we assessed the relationship between the WW PA pattern and the risk of specific disease. Since none of the specific disease categories included more than ten studies, a formal analysis of publication bias was not warranted. Egger’s regression test yielded the following values for each category: P = 0.001 < 0.05 (Mortality), P = 0.050 = 0.05 (Metabolic syndrome), P = 0.298 > 0.05 (Mental disease), P = 0.339 > 0.05 (Neurological disorder), P = 0.871 > 0.05 (Cardiovascular outcomes). There is a publication bias in the Mortality and Metabolic syndrome section, but not in the rest. Therefore, we carried out the correction of the above two results by using the method of interpolation and extrapolation (Appendix Fig. 11–12).

Discussion

Main findings

Our meta-analysis synthesizes evidence from 27 observational studies(1,204,486 participants) and demonstrates that the WW PA pattern is associated with a significantly reduced risk of specific diseases. Notably, we observed a substantial decrease in CVD mortality rates among individuals engaging in this exercise pattern, which aligns with findings from studies like the NHANES cohort analysis on adults living with type 2 diabetes [13], and extends evidence to understudied outcomes like neurological disorders. This suggests that infrequent but intense PA can lead to important longevity benefits, likely due to improved overall health and increased resilience against chronic diseases [13, 38].

Additionally, our analysis revealed a marked reduction in the incidence of CVD among WW. This reduction is likely linked to the positive effects of short, intense PA on cardiovascular health markers, including lower blood pressure, improved lipid profiles, and enhanced cardiac function [11, 13]. The WW lifestyle was also associated with a lower prevalence of metabolic syndrome, as concentrated bouts of exercise have been shown to enhance insulin sensitivity, reduce abdominal fat, and improve metabolic regulation, thereby mitigating key risk factors for diabetes and heart disease [30].

Beyond physical health benefits, our findings suggest that the WW pattern is associated with reduced risks of mental health disorders, such as anxiety and depression [30]. These benefits may stem from the psychological advantages of exercise, including the release of endorphins and improvements in mood, which contribute to enhanced emotional well-being. We also noted a reduction in the incidence of neurological diseases among WW, suggesting a potential protective effect stemming from improved blood flow and neuroplasticity associated with PA [11].

Overall, these results underscore the significant health benefits of engaging in intermittent PA, particularly for those who struggle to maintain regular exercise routines. Embracing a WW lifestyle can lead to meaningful improvements in both physical and mental health outcomes, making it a viable strategy for promoting overall well-being.

Interpretation of findings

The findings from our meta-analysis illuminate the considerable health benefits associated with the WW lifestyle, enhancing our understanding of how intermittent PA impacts the risk of specific diseases. The reduction in specific mortality among WW indicates that even those unable to commit to daily exercise can still achieve substantial health gains through concentrated bouts of PA. This suggests that intermittent, intense exercise sessions provide important longevity benefits, particularly in cardiovascular health, where reduced risks for heart disease, hypertension, and metabolic syndrome have been observed. For instance, a study from the UK Biobank showed that WW experienced a 23% reduction in hypertension risk, comparable to those engaging in more frequent exercise [38].

The observed decline in CVD risk emphasizes the importance of PA for heart health. The protective effects against CVD likely arise from improved cardiovascular fitness, enhanced endothelial function, and favorable changes in metabolic markers linked to PA. Consequently, healthcare providers should advocate for any form of PA, even if sporadic, as a means to promote cardiovascular health.

Moreover, the positive impact on metabolic syndrome underscores the potential of the WW pattern to mitigate key risk factors for chronic diseases. In light of the increasing prevalence of obesity and related metabolic disorders, promoting the WW lifestyle could serve as an accessible intervention for individuals facing challenges in weight management and metabolic health.

The associations with improved mental health outcomes and a lower risk of neurological diseases are particularly noteworthy, highlighting the multidimensional benefits of PA that extend beyond physical health to mental well-being. The mood-enhancing effects of exercise can serve as a compelling incentive for individuals to adopt an active lifestyle, even in a limited capacity. Improved mood regulation and emotional well-being may arise from the endorphin-releasing effects of exercise [38]. Additionally, the protective effects against neurological diseases, likely due to enhanced blood flow and neuroplasticity, make this exercise pattern beneficial beyond just physical health [38].

In summary, our findings support the idea that engaging in PA, even intermittently, can yield significant health benefits. This underscores the need to shift perspectives on exercise recommendations, promoting flexibility and inclusivity in active lifestyle initiatives. As we navigate the complexities of modern life, the WW pattern may offer a viable strategy for enhancing public health and well-being, particularly for those facing barriers to consistent exercise.

Strengths and limitations

Our study has several strengths. First, it is the most comprehensive meta-analysis to date examining the association between the WW PA pattern and multiple health outcomes. The inclusion of both cohort and cross-sectional studies, along with a large sample size, enhances the robustness of our findings. Additionally, we employed a random-effects model to account for potential heterogeneity among the studies, providing more conservative and reliable estimates of the associations between PA and the risk of specific diseases. Subgroup analyses were conducted to further explore potential sources of heterogeneity, such as sex, study type, and PA assessment, adding depth to our results.

However, our analysis also has limitations. Despite efforts to minimize bias, significant heterogeneity was observed in some pooled estimates, particularly in the CVD mortality risk analysis. This heterogeneity may reflect differences in population characteristics, variations in study design, or discrepancies in how PA was measured. While we focused on observational studies, which are valuable for assessing long-term outcomes, the inherent limitations of such designs—including potential residual confounding and reverse causality—should not be overlooked. Meanwhile, due to the varying definitions of WW and exercise intensity in the included studies, it is difficult to carry out the subgroup analysis based on different WW intensities. Furthermore, the reliance on self-reported PA in many studies may introduce recall bias and misclassification.

Conclusions

Our meta-analysis underscores the substantial health benefits associated with the WW lifestyle, demonstrating that engaging in intermittent PA can significantly reduce the risk of various health outcomes, including specific mortality, cardiovascular disease, metabolic syndrome, mental health disorders, and neurological disorders. These findings suggest that even sporadic bouts of exercise can lead to meaningful health improvements, providing an accessible alternative for individuals struggling to maintain regular exercise routines. In an increasingly sedentary society, promoting the WW pattern could serve as an effective public health strategy, encouraging more individuals to incorporate PA into their lives. Future research should further explore the long-term effects of this lifestyle, addressing potential risks and optimizing health benefits. Ultimately, the WW approach emphasizes that every bit of movement counts, reinforcing the idea that a flexible and attainable exercise regimen can significantly contribute to overall health and well-being.

Authors’ contributions

Dr Li had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Mr Fu and Wang contributed equally to this article. Concept and design: All authors. Acquisition, analysis, or interpretation of data: Fu, Wang, Pan, Li. Drafting of the manuscript: Fu, Wang, Pan. Critical review of the manuscript for important intellectual content: Huang, Li. Statistical analysis: All authors. Obtained funding: Huang. Administrative, technical, or material support: Huang, Li. Supervision: Huang, Li.

Funding

This study was funded by grant 361100E002 from the Academician Matching Program; and grant 2023 JKZKTS27 from the Research Project of Zhejiang Chinese Medical University.

Data availability

Data is provided within the manuscript or supplementary information files. Further inquiries can be directed to the corresponding authors.

Declarations

Role of the Funder/Sponsor

The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Competing Interests

The authors declare no competing interests.