Buffering the effects of work stress on anxiety and depression: a moderated mediation model of physical activity enjoyment and sleep disturbance in China.

Buffering the effects of work stress on anxiety and depression: a moderated mediation model of physical activity enjoyment and sleep disturbance in China - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice BMC Psychol . 2026 Mar 9;14:539. doi: 10.1186/s40359-026-04320-z Search in PMC Search in PubMed View in NLM Catalog Add to search Buffering the effects of work stress on anxiety and depression: a moderated mediation model of physical activity enjoyment and sleep disturbance in China Kunpeng Li Kunpeng Li 1 Faculty for Physical Education, Zhejiang Yuexiu University, 2801 Qunxian Middle Road, Yuecheng District, Shaoxing, Zhejiang Province 312000 China Find articles by Kunpeng Li 1, ✉ , Chih-Hung Pai Chih-Hung Pai 2 School of Hospitality Management, Zhejiang Yuexiu University, 2801 Qunxian Middle Road, Yuecheng District, Shaoxing, Zhejiang Province 312000 China Find articles by Chih-Hung Pai 2 Author information Article notes Copyright and License information 1 Faculty for Physical Education, Zhejiang Yuexiu University, 2801 Qunxian Middle Road, Yuecheng District, Shaoxing, Zhejiang Province 312000 China 2 School of Hospitality Management, Zhejiang Yuexiu University, 2801 Qunxian Middle Road, Yuecheng District, Shaoxing, Zhejiang Province 312000 China ✉ Corresponding author. Received 2025 Nov 19; Accepted 2026 Mar 4; Collection date 2026. © The Author(s) 2026 Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, 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 you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. 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-nc-nd/4.0/ . PMC Copyright notice PMCID: PMC13085425  PMID: 41803930 Abstract Background This study aimed to investigate the pathways linking work stress to anxiety and depression among a sample of full-time white-collar professionals (spanning various industries and organizational hierarchies) in China’s first-tier cities and to test the buffering role of physical activity enjoyment in this adverse process. Methods A cross-sectional survey was conducted with 1,546 professionals in Beijing, Shanghai, and Guangzhou. A moderated mediation model was tested using the PROCESS macro (Models 4 and 59) to examine the mediating role of sleep disturbance and the moderating role of physical activity enjoyment. Results The results indicated that work stress was significantly and positively associated with both anxiety and depression (β = 0.184, p < 0.001) and sleep disturbance (β = 0.431, p < 0.001). Sleep disturbance, in turn, was positively associated with anxiety and depression (β = 0.126, p < 0.001). Sleep disturbance partially mediated the relationship between work stress and anxiety and depression (indirect effect β = 0.071, 95% CI [0.049, 0.094]). Crucially, physical activity enjoyment significantly moderated the associations between work stress and anxiety and depression (β = -0.166, p < 0.001), between work stress and sleep disturbance (β = -0.081, p < 0.001), and between sleep disturbance and anxiety and depression (β = -0.031, p < 0.05). This moderation ultimately attenuated the strength of the overall indirect effect. Conclusions The findings demonstrate that physical activity enjoyment is a key protective resource for mitigating the adverse mental health associations of work stress. This study extends the Effort–Reward Imbalance (ERI) and Stress-Buffering Models (SBM) and provides critical empirical evidence for workplace interventions, suggesting a strategic shift from emphasizing quantity of exercise to enhancing the quality of the exercise experience. Supplementary Information The online version contains supplementary material available at 10.1186/s40359-026-04320-z. Keywords: Work stress, Physical activity enjoyment, Sleep disturbance, Anxiety and depression, China Introduction Data from the World Health Organization (WHO) underscore the global scale of mental health challenges stemming from work stress. Workplace adversity is a primary contributor to 70% of cases involving anxiety and depression, and work-related stress is linked to 120 million cases of depression annually [ 1 ]. Furthermore, a significant association has been established between long working hours and the prevalence of depressive symptoms, with this relationship being more pronounced in Asia than in Europe [ 2 , 3 ]. Against this global backdrop, the issue of work-related stress is particularly acute in China. Recent socioeconomic transformations have cultivated a hypercompetitive professional environment, positioning the country as a compelling case for in-depth investigation. According to 2022 national monitoring data on occupational health literacy from the National Health Commission of China, 15% of the workforce reported experiencing negative emotions such as depression, tension, or anxiety during the previous year. This figure ranks as the third most prevalent health issue monitored, highlighting the problem’s severity and pervasiveness [ 4 ]. In an era of intensifying involution (a cycle of relentless internal competition), the demands of work and social comparison with peers have emerged as the foremost stressors for professionals [ 5 , 6 ]. These unique psychosocial pressures collectively contribute to the growing prominence of mental health issues such as anxiety and burnout among China’s working population, underscoring the urgent need for both rigorous scholarly inquiry and effective intervention strategies. Evidently, the modern professional landscape has become a significant psychosocial stressor. Work-related stress is evolving into a formidable global challenge and is a primary driver of mental health issues among employees, including anxiety and depression [ 7 ]. A body of research confirms that psychosocial factors—such as effort–reward imbalance, a negative workplace ethical climate, and high job demands—are significantly associated with adverse outcomes, including depression, anxiety, and psychological distress [ 8 – 11 ]. While the connection between work stress and anxiety and depression is well established, the intricate mechanisms underlying this relationship, particularly the mediating role of insomnia, remain underexplored [ 11 , 12 ]. Long working hours frequently come at the expense of sleep, and the deprivation of this critical resource for biopsychosocial recovery has been shown to directly exacerbate the vicious cycle between work-related stressors and insomnia [ 13 ]. In this context, physical activity has emerged as a valuable intervention, with its role in alleviating the burden of mental health issues becoming increasingly recognized [ 14 ]. Evidence indicates that different types of exercise offer distinct psychological benefits; aerobic exercise has been shown to significantly mitigate anxiety symptoms, whereas resistance training appears to be more effective at reducing depressive symptoms [ 15 ]. Furthermore, combined training programs that incorporate both aerobic and resistance elements have also proven effective at lowering depression levels, presenting a comprehensive intervention strategy for improving employee well-being [ 16 ]. Despite the well-documented benefits of physical activity, a paradoxical reality persists: many individuals who exercise regularly still suffer from work-related mental health problems. This suggests that the quantitative dimensions of exercise, such as frequency and duration, may not fully account for its protective effects [ 14 ]. Emerging research has begun to emphasize that the psychological attributes of exercise may be more critical than its physical parameters. Specifically, exercise motivation stemming from intrinsic interest has been found to be a more powerful predictor of positive mental health outcomes than exercise duration itself is [ 17 ]. This enjoyable exercise experience can serve as a vital psychological resource. It may not only replenish the positive affect and energy depleted by work stress but also proactively reshape an individual’s cognitive appraisal of that stress [ 18 , 19 ], thereby more effectively interrupting the negative cascade from stress to psychological distress. Accordingly, this study posits that the qualitative experience of exercise—specifically, the enjoyment derived from it, or physical activity enjoyment—represents a critical missing piece in the current literature. The present research aims to address this paradox through empirical investigation. Currently, few studies have incorporated physical activity enjoyment as a moderating variable within a comprehensive model examining the interplay of work stress, sleep, and mental health, with a particular dearth of validation using longitudinal data [ 20 ]. Therefore, this study aims to construct and test a moderated mediation model to elucidate the underlying mechanisms through which work stress is associated with anxiety and depression. The analysis specifically investigates the mediating role of sleep disturbance in this relationship and the moderating effect of physical activity enjoyment on the pathways involved. This research not only promises to deepen the theoretical understanding of the stress‒health nexus but also has significant practical implications. By clarifying the protective role of physical activity enjoyment, the findings provide more precise, evidence-based guidance for mental health interventions targeting urban populations. This could help foster a strategic shift from merely prescribing exercise to actively promoting enjoyable physical activity. Accordingly, this study seeks to address the following research questions (RQs): RQ1: What are the current profiles of work stress, anxiety and depression, sleep disturbance, and physical activity enjoyment in the Chinese working population? RQ2: What is the nature of the association between work stress and both anxiety and depression and sleep disturbance? RQ3: Does physical activity enjoyment moderate the associations of work stress with sleep disturbance and with anxiety and depression? If so, how does this moderation manifest? Literature review Work stress and anxiety and depression In the field of occupational health psychology, the Effort–Reward Imbalance (ERI) model provides a core theoretical framework for understanding how work stress affects mental health [ 21 ]. This model posits that psychological distress arises when an individual’s high level of effort at work is not met with adequate rewards. A consistent body of research indicates that adverse organizational environments, negative workplace ethical climates, and specific work-related stressors are significant factors associated with anxiety and depression among employees [ 22 – 24 ]. This relationship has been widely validated across diverse occupational groups [ 25 , 26 ], with numerous empirical studies providing robust evidence for the negative association between work stress and mental well-being [ 10 , 27 ]. The central mechanism of the ERI model is that a sustained state of imbalance between effort and reward can trigger negative emotions and feelings of frustration in individuals, in turn degrading their psychological health [ 28 – 30 ]. Grounded in ERI theory and supported by extensive empirical findings, work stress has been identified as a key factor related to anxiety and depression. Accordingly, this study proposes the following hypothesis: H1: Work stress is significantly and positively associated with anxiety and depression. Sleep disturbance mediates the relationship between work stress and anxiety and depression Prolonged work stress is known to significantly increase the risk of insomnia [ 31 – 33 ]. Stress not only depletes an individual’s psychological capital, making it more difficult to cope with sleep disruptions [ 33 ], but it can also lead to excessive work engagement, a state in which individuals are unable to mentally detach from work-related thoughts during nonwork hours, thereby impeding the sleep onset process [ 34 ]. Conversely, sleep deprivation or poor sleep quality is not merely a passive consequence of stress but also an active risk factor that can significantly amplify symptoms of anxiety and depression. Across various occupational populations, sleep disturbance has been identified as one of the most potent factors related to anxiety and depression [ 35 , 36 ]. Evidence from intervention studies further demonstrates that effectively alleviating insomnia symptoms can lead to significant improvements in depression, stress, and burnout, confirming that enhancing sleep is a viable pathway for promoting mental health [ 37 , 38 ]. Therefore, this study proposes the following hypotheses: H2: Work stress is significantly and positively associated with sleep disturbance. H3: Sleep disturbance is significantly and positively associated with anxiety and depression. Sleep disturbance is a partial mediator of the relationship between occupational stress and depression, anxiety and emotional exhaustion. Research has also suggested that insomnia is an important mediator between job burnout and depression, accounting for approximately 69.2% of the total effect [ 39 ]. Research on interventions that target sleep to improve mental health and that work through sleep also lend some indirect support to this pathway [ 37 ]. Collectively, these results indicate that sleep disturbance is an important linking mechanism in the negative cascade between work stress and anxiety and depression [ 40 , 41 ]. On the basis of this body of evidence, the following hypothesis is proposed: H4: Sleep disturbance plays a significant mediating role in the relationship between work stress and anxiety and depression. Physical activity enjoyment as a moderator of the relationship between work stress and psychological distress Within the theoretical framework of this study, physical activity enjoyment may function as a critical psychological resource. The Stress-Buffering Model (SBM) offers a theoretical explanation for the affronted protective mechanism [ 42 ]. This model suggests that psychosocial resources can effectively protect individuals against the harmful effects of stressors on physical and mental health. On the basis of this model, it is suggested that the subjective experience of physical activity enjoyment, which goes beyond the instrumental properties of the exercise, may exert its effects through different psychological routes [ 43 ]. First, physical activity enjoyment may buffer the positive association between work stress and anxiety and depression. When people experience work stress, it is likely that their positive psychological resources become depleted and that negative cognitive activity worsens their negative emotions. Physical activity enjoyment refers to an experience consisting of pleasure and intrinsic satisfaction, which is opposed to any work-related experience. Exercise can therefore be an effective tool for renewing these psychological resources [ 44 , 45 ]. This positive experience may facilitate a cognitive reappraisal of stress and enhance self-compassion, potentially mitigating the cognitive pathways through which stress translates into negative emotional outcomes [ 44 , 45 ]. Accordingly, the following hypothesis is proposed: H5a: Physical activity enjoyment significantly moderates the association between work stress and anxiety and depression. Specifically, the positive association between work stress and anxiety and depression is weaker for individuals with higher levels of physical activity enjoyment. Second, physical activity enjoyment may similarly attenuate the association between work stress and sleep disturbance. One of the primary psychological mechanisms through which work stress leads to sleep disturbance is believed to be hyperarousal. When physical activity is experienced as enjoyable, individuals may find it easier to become fully engaged in the activity itself. This positive engagement is thought to enhance emotional regulation skills, enabling individuals to better manage stress-related emotions. This, in turn, may lower their arousal levels, creating more favorable conditions for sleep onset [ 46 , 47 ]. Indirect support for this inference comes from studies on mind-body exercises; these forms of activity, characterized by their intrinsic enjoyment and focus, have been found to be more effective than conventional exercise in alleviating stress and improving sleep [ 48 , 49 ]. Therefore, the following hypothesis is proposed: H5b: Physical activity enjoyment significantly moderates the association between work stress and sleep disturbance. Specifically, the positive association between work stress and sleep disturbance is weaker for individuals with higher levels of physical activity enjoyment. Finally, we further postulate that, even when sleep has already been disrupted, physical activity enjoyment may still weaken the association between sleep disturbance and anxiety and depression. Sleep disturbance typically exacerbates states of anxiety and depression by depleting an individual’s cognitive and emotional resources, making them more vulnerable to negative moods the following day. However, physical activity itself is regarded as an independent antianxiety and antidepressant intervention [ 45 , 50 , 51 ]. It is therefore plausible that individuals who find enjoyment in exercise are more likely to employ it as a regular and proactive coping strategy. This type of enjoyable activity is theorized to improve mood through various psychological mechanisms. This mood enhancement could, in turn, alleviate the sleep difficulties that stem from emotional distress, thereby indirectly promoting sleep recovery [ 52 – 54 ]. On the basis of this reasoning, the following hypothesis is proposed: H5c: Physical activity enjoyment significantly moderates the association between sleep disturbance and anxiety and depression. Specifically, the positive association between sleep disturbance and anxiety and depression is weaker for individuals with higher levels of physical activity enjoyment. In summary, drawing upon the theoretical logic of SBM and insights from related empirical research, it is reasonable to conclude that physical activity enjoyment may play a critical moderating and mitigating role at multiple points along the pathway from work stress to anxiety and depression. It may not only weaken the initial impact of the stressor but also slow the deterioration of intermediate links and help counteract the final adverse health outcomes. Accordingly, this study proposes a moderated mediation hypothesis: H6: The indirect association between work stress and anxiety and depression via sleep disturbance is moderated by physical activity enjoyment. Specifically, the mediating effect of sleep disturbance in the relationship between work stress and anxiety and depression is weaker for individuals with higher levels of physical activity enjoyment. Methodology Data collection and participants This study employed a cross-sectional survey design with purposive sampling to obtain a representative sample, rather than a specific restricted cohort. This approach was selected to ensure that the findings could be cautiously generalized to the broader population of urban professionals sharing similar sociodemographic and environmental characteristics. Data were collected through the professional services of Wenjuanxing (WJX.cn), China’s leading online survey platform, which has been recognized as a preferred tool for researchers because of its reliable responses and participant impartiality [ 55 , 56 ]. The survey was administered over a 36-day period, from June 15 to July 20, 2025. The sampling frame was carefully constructed with specific inclusion criteria to obtain the required population. The participants must have been (1) full-time white-collar professionals (defined in this study as individuals engaged in mental or administrative labor across various sectors, such as technology, finance, education, healthcare, and enterprise management, encompassing organizational hierarchies ranging from entry-level staff to middle and senior management), (2) currently residing in one of China’s three first-tier cities (Beijing, Shanghai, Guangzhou), or (3) residents of their respective cities for at least six months to be familiar with the working landscape. The WJX.cn platform was responsible for managing the recruitment process; this platform automatically sends out survey invitations to sample pool members who met our requirements. The quality of the data and validity of the responses were controlled using several measures. First, the system of the platform did not allow a person to submit a form twice using the same IP address. Additionally, the platform flagged responses that were submitted for a very short completion time. The beginning of the survey contained a screening question to reconfirm eligibility (e.g., city of residence). Finally, an attention check item was included in the questionnaire to identify and exclude inattentive respondents. This multifaceted approach was designed to maximize the relevance and reliability of the collected data. The selection of China’s three first-tier cities was based on several key considerations. Geographically, Beijing, Shanghai, and Guangzhou serve as the respective cores of northern, eastern coastal, and southern coastal China, representing three major economic circles and encompassing regional and cultural diversity. Economically, their status as premier metropolitan centers reflects a high level of modernization, and their professional environments are indicative of leading-edge workplace trends. The survey was conducted in two phases. The first phase consisted of a pilot survey in which 314 questionnaires were distributed across the three cities. After invalid responses were excluded, 296 valid questionnaires remained, yielding a validity rate of 94.3%. The reliability and validity of the measurement instruments were assessed, with all the Cronbach’s alpha coefficients exceeding 0.80. On the basis of these promising results, we proceeded to the second phase: the formal survey. To ensure the statistical validity of the study and the robustness of the moderated mediation model, an a priori sample size calculation was conducted using two complementary approaches. First, utilizing G*Power software (version 3.1.9.7) [ 57 ], a power analysis for a multiple linear regression model was performed. To detect a conservative small effect size ( = 0.02) with an alpha level of 0.05, a statistical power of 0.90 [ 58 ], and accounting for up to 6 predictors in the full theoretical model, a minimum sample size of 855 was required. Second, based on Cochran’s formula for cross-sectional survey studies with an unknown target population size [ 59 ], a 95% confidence level, and a strict 3% margin of error (assuming maximum variability, = 0.5), the target sample size was calculated to be 1,067. The formal survey yielded a total of 1,700 questionnaires distributed, and after removing invalid entries, 1,546 valid responses were retained, for a final validity rate of 91.0%. This final sample size of 1,546 far exceeded the strict mathematically required thresholds from both calculations, thereby providing robust statistical power for hypothesis testing and ensuring high reliability. People with active engagement were the focus of this study. Given that more professionals in first-tier cities in China (Beijing, Shanghai, and Guangzhou) participate in different types of sports, we categorized the types of major exercise of the participants into two categories: aerobic exercise and resistance exercise. The American College of Sports and Medicine (ACSM) [ 60 ] and relevant recent literature [ 15 , 61 , 62 ] were used for classification to accurately capture the potential effects of different modalities of exercise. This strategy improved the relevance of the survey to the participants and generated higher interest levels from the target audience in the survey. It was reasoned that the subjects’ exercises involved would be likely to share their thoughts, therefore increasing the value of the data collected. The survey instrument consisted of three distinct sections. Survey respondents were asked for their demographic information, including their city, sex and age. The second section aimed to gather more data related to their exercise attributes, including the exercise they undertook and the exercise intensity they felt. In the last section, the researcher has the measurement items of the study core constructs. Scales The study’s measurement instruments assessed four core constructs: work stress, sleep disturbance, anxiety and depression, and physical activity enjoyment. All the scales underwent a rigorous, standardized “translation, back-translation, and expert panel review” procedure to ensure both linguistic and cultural validity. On the basis of this process, all the items were administered in Chinese. The Perceived Stress Scale (PSS-10) [ 63 ] was used to measure work stress. This scale moves beyond a simple inventory of objective stressors to assess an individual’s subjective perception of their life as unpredictable, uncontrollable, and overloaded over the previous month. The psychological core of the stress experience is captured exactly; it serves as a link between external demands and internal feelings. The PSS-10 is a suitable instrument for assessing the primary independent variable of the current study, given its excellent reliability and validity and extensive use and validation in occupational health [ 64 ]. We tailored the scale’s items to specifically ask about stress perceived in the work setting. An example item is “How often in the past month have you been upset because of something that happened unexpectedly at work?” Responses were given on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree); this scale was found to have good internal consistency in the current sample (Cronbach’s α = 0.93). Sleep disturbance was assessed using the Insomnia Severity Index (ISI) [ 65 ]. The ISI is a brief, 7-item self-report questionnaire designed to evaluate the nature, severity, and impact of insomnia symptoms. It is widely used in both clinical practice and research to screen for and quantify the severity of insomnia [ 66 ]. A sample item is “To what extent do you consider your sleep problem to interfere with your daily functioning (e.g., daytime fatigue, ability to function at work/daily chores, concentration, memory, mood)?” Items were rated on a 5-point Likert-type scale ranging from 1 (not at all severe) to 5 (very severe). The internal consistency of the scale was excellent in this study (Cronbach’s α = 0.92). Anxiety and depression were measured using the Hospital Anxiety and Depression Scale (HADS) [ 67 ]. The HADS is a widely validated instrument for assessing the severity of anxiety and depressive symptoms in various populations, including working adults [ 9 , 68 ]. The questionnaire comprises 14 items divided into two subscales: a 7-item scale for anxiety and a 7-item scale for depression. A sample item from the anxiety subscale is “I feel tense or ‘wound up’.” and the scores for each subscale range from 0 to 21. The established clinical cutoffs categorize scores of 8–10 as mild, 11–14 as moderate, and 15–21 as severe for both anxiety and depression. The total HADS scale demonstrated outstanding internal consistency in the present study (Cronbach’s α = 0.96). Physical activity enjoyment was measured using the Physical Activity Enjoyment Scale (PACES) [ 69 ]. While originally developed for assessing enjoyment in younger populations, the scale has been widely adapted and validated for use with adults [ 70 , 71 ]. For this study, a validated 4-item short version of the scale was utilized. A sample item is “I find physical activity to be pleasurable”. Items were rated on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). The internal consistency of the scale was excellent in the current sample (Cronbach’s α = 0.94). Statistical analysis The data were analyzed using IBM SPSS Statistics for Windows, version 27.0. To test the hypothesized mediation and moderated mediation effects, the PROCESS macro for SPSS (version 4.2) was employed. Following the recommendations of Hayes [ 72 ], PROCESS Model 4 was used to test the simple mediation model, and Model 59 was used to test the moderated mediation model. All the inferential tests were based on 5,000 samples. The 95% bias corrected confidence interval (CI), which is a measure of statistical significance, must not contain zero for an effect to be considered statistically significant [ 73 , 74 ]. To probe conditional effects, the moderator was tested at three levels: one standard deviation below the mean (M – 1 SD), the mean, and one standard deviation above the mean (M + 1 SD). To summarize the data, descriptive statistics, including frequencies, percentages and means, were calculated. To analyze the effects of sociodemographic factors and exercise modalities, mean comparison tests were performed. Confirmatory factor analysis was adopted to assess the goodness of fit of the proposed models, using AMOS version 29,0 [ 75 ]. Spearman’s rank correlation analysis was used to examine the bivariate relationships between the scale dimensions. Finally, the internal consistency reliability of the scales was assessed using Cronbach’s alpha. Furthermore, to ensure the robustness of our findings and explicitly account for the conceptual distinctions between anxiety and depression, sensitivity analyses were conducted by testing the moderated mediation models separately for the anxiety and depression subscales. Results Sample characteristics and descriptive statistics The sample consisted of 809 male (52.3%) and 737 female (47.7%) participants. The age distribution was concentrated among respondents aged 31–40 years (41.0%), followed by those aged 41–55 years (37.6%). A majority of the participants had completed a college or undergraduate degree (52.1%). In terms of monthly income, 50.6% reported earning between 5,000 and 10,000 RMB, while 27.5% reported earning between 10,000 and 20,000 RMB. Notably, there was considerable diversity in both the choice and intensity of physical activity. Among those engaging in aerobic exercise, spinning was the most popular activity in Beijing, Shanghai, and Guangzhou, particularly among younger participants (63.0%). This was followed by walking (34.1%) and yoga (33.8%). Cycling was also a preference of a significant portion (23.4%), reflecting the growing trend of nighttime riding. With respect to resistance training, bodyweight exercises were the most common (85.3%), followed by machine-based training (63.0%). Free weights (57.9%) and elastic resistance bands (46.9%) were also popular choices. In terms of exercise patterns, a weekly frequency of 2–3 sessions was the most reported (40.0%), and the majority of participants exercised for a duration of 30 to 60 min per session (49.2%). These diverse and realistic engagement patterns suggest that this sample was well suited for this investigation and possesses a high degree of ecological validity (see Table 1 ). Table 1. Sample characteristics (Beijing = 517, Shanghai = 515, Guangzhou = 514, Total = 1546) Work Stress Sleep Disturbance Anxiety and Depression Physical Activity Enjoyment N (%) Mean ± SD Mean ± SD Mean ± SD Mean ± SD Gender Male 809 52.3 3.39 ± 0.98 3.30 ± 1.06 2.85 ± 0.87 4.92 ± 1.71 Female 737 47.7 3.45 ± 0.96 3.36 ± 1.03 2.93 ± 0.84 4.95 ± 1.66 Age 18–30 128 8.3 3.39 ± 1.00 3.35 ± 1.06 2.84 ± 0.87 4.95 ± 1.71 31–40 634 41.0 3.41 ± 0.98 3.28 ± 1.05 2.87 ± 0.87 4.97 ± 1.69 41–55 582 37.6 3.43 ± 0.97 3.38 ± 1.03 2.95 ± 0.84 4.93 ± 1.68 Over 56 202 13.1 3.43 ± 0.97 3.33 ± 1.03 2.83 ± 0.88 4.84 ± 1.69 Education Level High school and below 318 20.8 3.36 ± 1.00 3.35 ± 1.05 2.88 ± 0.86 4.76 ± 1.77 College/Bachelor’s degree 805 52.1 3.45 ± 0.96 3.35 ± 1.03 2.91 ± 0.85 4.99 ± 1.65 Master’s degree or above 423 27.4 3.39 ± 0.98 3.29 ± 1.05 2.87 ± 0.87 4.96 ± 1.69 Monthly Income Less than ¥5,000 219 14.2 3.47 ± 0.98 3.39 ± 1.03 2.88 ± 0.86 4.85 ± 1.73 ¥5,000–10,000 782 50.6 3.42 ± 0.96 3.30 ± 1.05 2.89 ± 0.85 4.97 ± 1.61 ¥10,000–20,000 425 27.5 3.40 ± 0.99 3.32 ± 1.03 2.89 ± 0.87 4.98 ± 1.66 More than ¥20,000 120 7.8 3.34 ± 1.00 3.44 ± 1.03 2.89 ± 0.87 4.71 ± 1.75 Aerobic Exercise Walking 527 34.1 3.39 ± 0.99 3.36 ± 1.04 2.90 ± 0.85 4.94 ± 1.67 Running 293 19.0 3.43 ± 0.96 3.26 ± 1.04 2.82 ± 0.87 4.80 ± 1.79 Yoga 523 33.8 3.41 ± 0.97 3.30 ± 1.06 2.89 ± 0.86 4.98 ± 1.68 Dance 298 19.3 3.41 ± 0.97 3.40 ± 1.01 2.92 ± 0.85 4.96 ± 1.64 Cycling 362 23.4 3.41 ± 0.99 3.37 ± 1.04 2.91 ± 0.85 4.98 ± 1.64 Swimming 289 18.7 3.44 ± 0.98 3.28 ± 1.04 2.83 ± 0.88 4.84 ± 1.77 Ball 203 13.1 3.43 ± 0.98 3.27 ± 1.06 2.88 ± 0.86 4.83 ± 1.74 Spinning 974 63.0 3.41 ± 0.97 3.35 ± 1.04 2.92 ± 0.85 4.99 ± 1.65 Aerial Yoga 239 15.5 3.41 ± 1.00 3.37 ± 1.03 2.83 ± 0.89 4.88 ± 1.72 Resistance Exercise Free weight 895 57.9 3.41 ± 0.98 3.32 ± 1.05 2.88 ± 0.87 4.95 ± 1.67 Fixed equipment 974 63.0 3.44 ± 0.96 3.37 ± 1.03 2.92 ± 0.85 4.93 ± 1.69 Bodyweight 1318 85.3 3.42 ± 0.97 3.32 ± 1.04 2.88 ± 0.86 4.93 ± 1.69 Elastic resistance 725 46.9 3.42 ± 0.98 3.30 ± 1.06 2.88 ± 0.87 4.95 ± 1.68 Frequency/ Weekly Less than 1 466 30.1 3.44 ± 0.98 3.29 ± 1.08 2.88 ± 0.88 4.89 ± 1.71 2–3 618 40.0 3.47 ± 0.96 3.37 ± 1.02 2.93 ± 0.85 4.93 ± 1.69 4–5 288 18.6 3.38 ± 0.96 3.39 ± 1.02 2.90 ± 0.86 4.96 ± 1.66 6+ 174 11.3 3.20 ± 1.01 3.19 ± 1.05 2.80 ± 0.87 5.05 ± 1.67 Duration/Time Less than 30 m 464 30.0 3.45 ± 0.96 3.37 ± 1.03 2.90 ± 0.86 5.03 ± 1.63 30 m–1 h 760 49.2 3.35 ± 0.98 3.28 ± 1.04 2.89 ± 0.86 4.92 ± 1.71 1–2 h 243 15.7 3.48 ± 0.97 3.37 ± 1.05 2.85 ± 0.87 4.91 ± 1.69 2 h+ 79 5.1 3.59 ± 0.95 3.48 ± 1.04 3.05 ± 0.81 4.66 ± 1.79 Open in a new tab Additionally, Table 1 presents a comparison of mean scores for the core study variables across participant demographic characteristics. For work stress, the highest mean scores were observed among participants who exercised for more than two hours per session (M = 3.59, SD = 0.95), followed by those who exercised for one to two hours (M = 3.48, SD = 0.97). Compared with males, females reported higher levels of work stress (M = 3.45, SD = 0.96) (M = 3.39, SD = 0.98), and individuals over the age of 40 reported higher levels of stress than their younger counterparts. High stress scores were also concentrated among those with a college degree or undergraduate degree. With respect to exercise choice, the highest levels of work stress were reported by those who engaged in swimming (M = 3.44, SD = 0.98) and machine-based resistance training (M = 3.44, SD = 0.96). In terms of sleep disturbance, the highest scores were reported for the group exercising for more than two hours (M = 3.48, SD = 1.04), followed by individuals with a monthly income exceeding 20,000 RMB (M = 3.44, SD = 1.03). Among specific activities, participants who chose dancing reported notably higher sleep disturbance scores (M = 3.40, SD = 1.01) than those who used machine-based training did (M = 3.37, SD = 1.03). With respect to anxiety and depression, the highest scores were reported by participants who exercised for more than two hours per session (M = 3.05, SD = 0.81), followed by those in the 41–55 age group (M = 2.95, SD = 0.84). The scores were largely consistent across different exercise types, with the highest levels concentrated among participants in dancing, spinning, and machine-based training (mean scores approximately 2.92). Finally, for physical activity enjoyment, the highest levels were reported by individuals who exercised six or more times per week (M = 5.05, SD = 1.67), followed by those whose sessions lasted less than 30 min (M = 5.03, SD = 1.63). Among aerobic activities, spinning was associated with markedly greater enjoyment (M = 4.99, SD = 1.65) than other options. In contrast, the enjoyment levels for resistance training were highly consistent, with mean scores ranging between 4.93 and 4.95. Furthermore, a cross-variable comparison revealed several overarching patterns. A consistent trend emerged for the adverse outcome variables: females, individuals aged 41–55 years, those with a college/undergraduate degree, and participants who exercised for more than two hours per session consistently reported the highest mean scores on work stress, sleep disturbance, and anxiety and depression. Interestingly, the patterns for two of these groups—females and individuals with a college/undergraduate degree—were consistent across all four constructs, as they reported high scores not only for the three negative outcomes but also for physical activity enjoyment. In contrast, the other demographic groups exhibited divergent trends. For instance, participants aged 31–40 years and those with a monthly income between 10,000 and 20,000 RMB reported higher levels of physical activity enjoyment than their respective counterparts in other age brackets and income brackets did. The descriptive statistical analysis in this study provides a realistic portrait of the mental health and exercise behaviors of professionals in China’s first-tier cities. Through preliminary mean comparisons across multiple dimensions, our findings reveal the complex distribution of the core variables among different sociodemographic and behavioral subgroups. Notably, the data reveal a compelling issue: individuals who might be considered highly active—specifically, females, those in the 41–55 age range, and those exercising for more than two hours per session—also reported higher average scores for work stress, sleep disturbance, and anxiety and depression. These findings suggest that the benefits of exercise duration do not follow a simple “more is better” paradigm; rather, the relationship appears to be more nuanced. This implies that the protective effects of physical activity may not be determined solely by quantitative metrics. Instead, they may be significantly influenced by qualitative dimensions, thereby providing preliminary, contextual evidence for the subsequent examination of physical activity enjoyment as a key moderator. Measurement models The measurement model was tested using confirmatory factor analysis (CFA) to assess its goodness-of-fit (see Table 2 ). The results for the proposed four-factor model indicated an excellent fit to the data, χ² = 591.349, df = 554, and χ²/df = 1.067 (within the acceptable range of 1–3). Additional fit indices were also strong: SRMR = 0.0146 (< 0.05), GFI = 0.979, AGFI = 0.976, NFI = 0.986, and CFI = 0.999 (all > 0.90). Furthermore, the RMSEA was 0.007 (< 0.05), providing further support for the strong fit of the model [ 76 , 77 ]. As shown in Table 2 , the fit indices for this four-factor model were substantially better than those for the alternative three-factor, two-factor, and single-factor models, confirming the distinctiveness of the four constructs. Table 2. Model fitness test Models χ2 dƒ χ2/ dƒ GFI AGFI RMSEA SRMR NFI CFI 4Factor WS, SD, AD, PAE 591.349 554 1.067 0.979 0.976 0.007 0.0146 0.986 0.999 3Factor WS + SD.AD.PAE 5731.686 557 10.290 0.699 0.660 0.078 0.0879 0.868 0.879 2Factor WS + SD+AD.PAE 14983.14 559 26.803 0.415 0.341 0.129 0.1828 0.654 0.662 1Factor WS + SD+AD + PAE 20559.14 560 36.713 0.374 0.296 0.153 0.1973 0.525 0.532 Criteria (Kline,2018) < 3 > 0.90 > 0.90 < 0.05 < 0.05 > 0.90 > 0.90 Open in a new tab WS Work Stress, SD Sleep Disturbance, AD Anxiety and Depression, PAE Physical Activity Enjoyment Furthermore, as presented in Table 3 , the four-factor model demonstrated strong construct validity. The composite reliability (CR) for each construct exceeded the established cutoff of 0.70, and the average variance extracted (AVE) for each construct surpassed the recommended threshold of 0.50, indicating convergent validity [ 78 , 79 ]. Table 3 also displays the Cronbach’s alpha coefficients, means, standard deviations, and bivariate correlations for all the study variables. The results show that all the constructs were significantly correlated. As hypothesized, physical activity enjoyment was negatively correlated with work stress, sleep disturbance, and anxiety and depression. Finally, all item factor loadings in the unconstrained four-factor model were significant and exceeded the benchmark of 0.50 [ 77 ]. Taken together, these results provide robust evidence for the reliability and validity of the measurement model. Table 3. Descriptive statistics and correlations for study variables Variables Mean ± SD CR AVE α 1 2 3 4 1. Work Stress 3.416 ± 0.973 0.931 0.573 0.931 1 2. Sleep Disturbance 3.331 ± 1.042 0.916 0.610 0.916 0.378 ** 1 3. Anxiety and Depression 2.892 ± 0.859 0.979 0.705 0.960 0.267 ** 0.261 ** 1 4. Physical Activity Enjoyment 4.936 ± 1.685 0.944 0.807 0.944 -0.269 ** -0.071 ** -0.172 ** 1 Open in a new tab Moderated mediation model First, the direct path effects among the variables were examined (Fig. 1 and Table 4 ). The analysis provided an initial depiction of the associations between work stress and the outcome variables of anxiety and depression and sleep disturbance. The results clearly revealed that work stress was significantly and positively associated with anxiety and depression (β = 0.184, p < 0.001). These findings align with the core tenets of the ERI model, which posits that negative emotional states can arise when an individual perceives a significant imbalance between their work-related efforts and rewards. Therefore, H1 was supported. Fig. 1. Open in a new tab Moderated mediation effect Table 4. Moderated mediation model analysis Predictors β(Coeff) SE t p LLCI ULCI SD(M) R 2 = 0.156 F = 94.957*** Work Stress (X) 0.431 0.026 16.413 0.000 0.379 0.482 Work Stress (X)⋅ PAE(W) -0.081 0.017 -4.684 0.000 -0.114 -0.047 AD(Y) R 2 = 0.210 F = 81.698*** 0.000 Work Stress (X) 0.184 0.023 7.933 0.000 0.139 0.230 SD(M) 0.126 0.020 6.206 0.000 0.086 0.166 Work Stress (X)⋅PAE(W) -0.166 0.016 -10.398 0.000 -0.197 -0.135 SD(M)⋅PAE(W) -0.031 0.012 -2.541 0.011 -0.055 -0.007 Conditional effects (W=Mean ± SD) Work Stress (X) to SD(M) at the values of PAE (W) (moderator) Direct effect SE t p LLCI ULCI M-1SD ( -1.685) 0.567 0.042 13.609 0.000 0.485 0.648 M (0) 0.431 0.026 16.413 0.000 0.379 0.482 M+1SD (1.685) 0.295 0.036 8.099 0.000 0.223 0.366 Work Stress (X) to AD(Y) at the values of PAE (W) (moderator) M-1SD ( -1.685) 0.464 0.040 11.704 0.000 0.386 0.541 M (0) 0.184 0.023 7.933 0.000 0.139 0.230 M+1SD (1.685) -0.095 0.031 -3.085 0.002 -0.156 -0.035 SD (M) to AD(Y) at the values of PAE (W) (moderator) M-1SD ( -1.685) 0.179 0.029 6.093 0.000 0.121 0.236 M (0) 0.126 0.020 6.206 0.000 0.086 0.166 M+1SD (1.685) 0.074 0.029 2.582 0.01 0.018 0.130 Indirect effect Boot effect Boot SE BootLLCI BootULCI Work Stress (X) on AD(Y) via SD (M) 0.071 0.011 0.049 0.094 PAE (W) Moderated mediation effect M-1SD ( -1.685) 0.101 0.012 0.079 0.126 M (0) 0.054 0.010 0.035 0.074 M+1SD (1.685) 0.022 0.011 0.002 0.044 Effect1 Effect2 Contrast BootSE BootLLCI BootULCI Pairwise contrasts between conditional indirect effects (Effect1 minus Effect2) 0.054 0.101 -0.047 0.010 -0.067 -0.028 0.022 0.101 -0.079 0.015 -0.108 -0.051 0.022 0.054 -0.033 0.005 -0.042 -0.023 Open in a new tab SD Sleep Disturbance, AD Anxiety and Depression, PAE Physical Activity Enjoyment Second, the results of the analysis revealed that work stress was significantly positively associated with sleep disturbance (β = 0.431, p < 0.001), and sleep disturbance, in turn, was positively associated with anxiety and depression (β = 0.126, p < 0.001). From a psychological perspective, excessive occupational stress can trap individuals in a ruminative thought pattern focused on work-related problems, leading to a prolonged state of cognitive and physiological hyperarousal. This persistent state of arousal conflicts with the biopsychosocial relaxation necessary for sleep onset, thus acting as a primary trigger for sleep disturbance. In turn, sleep disturbance impairs emotional regulation capabilities, and such dysregulation is a key factor in the development of anxiety and depressive symptoms. Therefore, H2 and H3 were supported. More importantly, the bootstrap analysis confirmed that sleep disturbance played a significant mediating role in the relationships between work stress and anxiety and depression. The indirect effect was statistically significant (β = 0.071, 95% CI [0.049, 0.094]), thus supporting H4. These results indicate that a portion of the adverse association between work stress and negative emotional outcomes is not direct; rather, it is indirectly channeled through the fundamental pathway of compromised sleep quality. In this model, sleep disturbance emerges as a core pathway through which stress is linked to anxiety and depression. The core objective of this study was to test whether physical activity enjoyment could effectively buffer the adverse associations of work stress with anxiety and depression and with sleep disturbance, thereby revealing its multipathway protective mechanism within the stress–health model. First, the results revealed that physical activity enjoyment significantly moderated the relationship between work stress and anxiety and depression (β = -0.166, p < 0.001, 95% CI [-0.197, -0.135]). An analysis of the interaction effect (Fig. 2 ) revealed that for individuals with low physical activity enjoyment, the positive association between work stress and anxiety and depression was particularly strong (β simple = 0.464, p < 0.001; 95% CI [0.386, 0.541]). In contrast, for individuals with high physical activity enjoyment, this association, while still significant, was substantially attenuated (β simple = -0.095, p < 0.01, 95% CI [-0.156, -0.035]). These findings indicate that the enjoyment of physical activity is more than mere physical exertion; it acts as a form of psychological immunization. Providing positive emotional experiences and a sense of accomplishment effectively counteracts the psychological depletion caused by work stress, helping individuals maintain emotional stability. Therefore, H5a was supported. Fig. 2. Open in a new tab The interaction of work stress and physical activity enjoyment on anxiety and depression Second, physical activity enjoyment significantly moderated the association between work stress and sleep disturbance (β = -0.081, p < 0.001, 95% CI [-0.114, -0.047]). The interaction plot (Fig. 3 ) indicated that while sleep may be compromised for everyone under high-stress conditions, the strength of the association between work stress and sleep disturbance was significantly weaker for individuals with high physical activity enjoyment (β simple = 0.295, p < 0.001; 95% CI [0.223, 0.366]). In contrast, for those with low physical activity enjoyment, the association was much stronger (β simple = 0.567, p < 0.001, 95% CI [0.485, 0.648]). This may be because enjoying physical activity facilitates psychological detachment from work, allowing individuals to fully immerse themselves in the pleasant sensations of exercise. This process could interrupt the persistent cognitive and emotional impact of work stress, effectively lowering hyperarousal and creating the preconditions for quality sleep. These findings support H5b. Fig. 3. Open in a new tab The interaction of work stress and physical activity enjoyment on sleep disturbance Finally, even when an individual’s sleep was already compromised, physical activity enjoyment significantly attenuated the association between sleep disturbance and anxiety and depression (β = -0.031, p < 0.05). As illustrated in the interaction plot (Fig. 4 ), as the severity of sleep disturbance increased, the corresponding increase in anxiety and depression was significantly smaller for individuals with high physical activity enjoyment (β simple = 0.074, p < 0.05, 95% CI [0.018, 0.130]) than for those with low physical activity enjoyment (β simple = 0.179, p < 0.001, 95% CI [0.121, 0.236]). This is a particularly crucial finding. It suggests that physical activity enjoyment functions as an effective emotion regulation strategy, counteracting the negative mood states induced by sleep deprivation through mechanisms such as boosting positive affect and enhancing self-efficacy. Doing so helps to disrupt the vicious cycle from sleep problems to anxiety and depression. Therefore, H5c was supported. Fig. 4. Open in a new tab The interaction of sleep disturbance and physical activity enjoyment on anxiety and depression Having established support for the individual moderation effects, a subsequent analysis was conducted to test the overall moderated mediation model. The objective was to determine whether physical activity enjoyment moderates the entire indirect effect of work stress on anxiety and depression through the mechanism of sleep disturbance. In other words, this analysis examined whether physical activity enjoyment could weaken the negative pathway linking work stress, sleep disturbance, anxiety and depression as a whole. As shown in Table 4 , the strength of this indirect effect varied significantly across levels of physical activity enjoyment. Specifically, for the low physical activity enjoyment group, the indirect association of work stress with anxiety and depression due to sleep disturbance was the strongest (indirect effect β = 0.101, 95% CI [0.079, 0.126]). In contrast, for the high physical activity enjoyment group, the magnitude of this indirect effect was reduced to its lowest point (indirect effect β = 0.022, 95% CI [0.002, 0.044]). Table 4 also presents pairwise comparisons of these conditional indirect effects. The results indicated that the indirect effect at a moderate level of physical activity enjoyment was significantly weaker than the effect at a low level (contrast = -0.047, 95% CI [-0.067, -0.028]). Furthermore, the indirect effect at a high level of physical activity enjoyment was not only significantly weaker than the effect at a low level (contrast = -0.079, 95% CI [-0.108, -0.051]) but also significantly weaker than the effect at a moderate level (contrast = -0.033, 95% CI [-0.042, -0.023]). Therefore, H6 was supported. Sensitivity Analyses: Given the distinct conceptual and clinical characteristics of anxiety and depression, we conducted sensitivity analyses by re-estimating the moderated mediation models with anxiety (HADS-A) and depression (HADS-D) as separate outcome variables. The results demonstrated that the overarching pattern of significance remained highly robust and consistent with the combined model. Specifically, physical activity enjoyment significantly moderated the pathways linking work stress, sleep disturbance, to both anxiety and depression individually, with only marginal variations in the magnitude of the coefficients. To maintain model parsimony and focus on the overarching mechanism of general psychological distress in the main text, the detailed results of these separate models are provided in Supplementary Appendix. Taken together, the findings of this study move beyond an analysis of discrete pathways to reveal the value of physical activity enjoyment from a more holistic and systematic perspective. Such enjoyment not only functions as a buffer at individual nodes within the model but also mitigates the entire core negative cascade from work stress to anxiety and depression. These findings suggest that enjoyable physical activity is not merely a simple coping strategy but rather a protective resource capable of reshaping the dynamic relationship between stress and health. When individuals derive intrinsic satisfaction from exercise, this experience can effectively buffer the adverse impact of work stress on emotional well-being, particularly the portion that is channeled through sleep-related problems. Discussion This study was designed to address the increasingly severe mental health challenges faced by professionals in China’s first-tier cities. The central objective was to elucidate the complex psychological pathways through which work stress is linked to anxiety and depression and to explore whether physical activity enjoyment can function as a protective resource for this process. The results provide clear empirical support for the proposed moderated mediation model. Work stress was not only directly positively associated with anxiety and depression but also indirectly associated with these outcomes via the mediating role of sleep disturbance. Crucially, the core contribution of this research lies in demonstrating the key moderating role that physical activity enjoyment plays in this adverse pathway. Work stress and the cascade of emotional distress The initial findings of this study align with the established literature in occupational health psychology. The positive association between work stress and anxiety and depression corroborates conclusions from cross-cultural and cross-occupational research, confirming that adverse work environments are a significant contributor to psychological distress among professionals [ 7 – 9 , 26 ]. This relationship, however, presents additional complexity within the context of China’s hypercompetitive professional landscape. Our study revealed that individuals who engaged in excessively long exercise sessions reported higher, not lower, levels of stress. This suggests that when the nature of exercise shifts from leisure to being task oriented, its positive psychological effects may be diminished. In high-pressure work environments, individuals may alienate physical activity from a form of self-care and leisure, transforming it into another obligatory, performance-driven task that must be completed. This phenomenon, which can be described as a compensatory‒compulsive exercise [ 80 , 81 ], extends the traditional boundaries of the ERI model. While use of the ERI model has historically focused on the occupational domain, our findings imply that this imbalanced psychological pattern can spill over into an individual’s nonwork domains, particularly their exercise habits. Similarly, our finding of a positive association between work stress and sleep disturbance is well supported by a broad body of research demonstrating that chronic occupational stress can significantly contribute to insomnia through mechanisms such as hyperarousal [ 13 , 31 , 34 ]. These results prompt a reexamination of the high levels of sleep disturbance reported by the subgroup engaging in prolonged single-session exercise. This observation suggests that when physical activity is undertaken in a high-stress context and lacks inherent relaxing or restorative qualities, it may paradoxically become a new source of stress. In these cases, exercise may fail to counteract the hyperarousal induced by work pressure and could thereby negatively impact sleep [ 20 ]. In conclusion, our findings showed that sleep disturbance was significantly related to anxiety and depression, thus demonstrating the basic importance of sleep as a resource for emotions. Consistent findings across multiple studies suggest that sleep quality deterioration is a significant factor in mood disorders [ 11 , 12 , 36 , 40 ]. These findings suggest that the effectiveness of exercise in managing emotions might be limited by sleep quality. When significant sleep disturbance was present, even long-duration exercise failed to effectively inhibit the pathway to anxiety and depression in our sample. This, once again, hints at the critical role of the subjective experience of exercise. Sleep disturbance as a core mechanism in the work stress pathway One of the central findings of this study is the significant mediating role of sleep disturbance in the relationship between work stress and anxiety and depression. This result is consistent with discoveries from previous research across diverse occupational groups [ 11 , 12 , 35 ], collectively revealing that the adverse impact of work stress is not always direct. Instead, its negative effects are frequently channeled indirectly through the deterioration of sleep. The core mechanism underlying this pathway is that chronic work stress prevents the body’s biopsychosocial systems from effectively calming down and relaxing. This state allows work-related thoughts to remain persistently active during nonwork hours, thereby interfering with normal sleep onset [ 34 ]. However, our findings, situated within the context of the current Chinese professional landscape, add a new layer of complexity to this established pathway. We found that this negative mediating chain persists even among individuals who are actively engaged in specific types of physical activity. This suggests that while different exercise modalities (e.g., aerobic vs. resistance) may offer distinct benefits for emotional well-being [ 15 ], mere participation in physical activity, in and of itself, appears insufficient to completely buffer the detrimental association between work stress and sleep disturbance. Furthermore, a significant advancement of the present study, compared with prior research that often focused on validating this single mediational pathway in more narrowly defined populations [ 34 , 39 ], lies in its application to a more representative sample. By extending the analysis from specific occupational groups to a broad cross-section of professionals in China’s first-tier cities, our research confirms the model’s applicability across diverse sociocultural and professional contexts. By replicating the pathway through which high-effort, low-reward stress patterns are associated with adverse mental health outcomes in this wider and more diverse sample, our study enhances the external validity and theoretical generalizability of the ERI model. Physical activity enjoyment: a theoretical extension of the stress-buffering model By examining the moderating effects of physical activity enjoyment across multiple pathways, this study expands the theoretical boundaries and explanatory power of the SBM. According to the first study, physical activity enjoyment was found to buffer the relationship between work stress and anxiety and depression. Thus, an enjoyable exercise process can effectively refill the psychological reserves that are emptied when dealing with occupational stress, as it provides intrinsically motivated feelings of pleasure and accomplishment [ 18 ]. With such a positive experience, an individual may experience a fundamental shift in mindset, changing from a passive stress sufferer to an active, effective stress cope [ 19 ]. These findings provide important information for exercise interventions in stressful working environments in China. People may not benefit much from aerobic exercise that alleviates anxiety and resistance training that reduces depression. For this reason, when sufferers from anxiety and depression does attempt such exercises, they should have a good experience if they are to benefit. That is, we suggest that the pleasantness of the experience of exercising may explain its effect on stress relief. This understanding fine-tunes what can be seen as a no-pain, no-gain task into an enjoyable leisure activity. Exercising, therefore, is a powerful psychological resource. In addition, physical activity enjoyment may promote exercise adherence and sustainability by weakening the relationship between work stress and sleep disturbance [ 20 ]. Notably, mind–body exercises such as yoga and tai chi, which are known for their capacity to bring individuals to focus on the task at hand and are enjoyable in themselves, have already been shown to reduce stress and improve sleep [ 48 , 49 ]. These findings provide collateral support for our results. This moderating effect is particularly significant in the context of China’s professional landscape, where long working hours and high job demands are prevalent. The state of persistent biopsychosocial tension induced by work stress is a key driver of insomnia [ 34 ]. Enjoyable physical activity can help mitigate this tension and promote relaxation, thereby effectively alleviating work-related sleep disturbances. Therefore, our study introduces a crucial boundary condition to the conventional wisdom that exercise improves sleep. While prior research has often emphasized the benefits of long-term, regular exercise [ 20 ], our findings reveal that the degree of enjoyment an individual derives from physical activity may be directly related to his or her ability to achieve a state of psychophysiological tranquility conducive to sleep. Finally, this study is the first to demonstrate that physical activity enjoyment can buffer the subsequent association between sleep disturbance and anxiety and depression. This implies that even when an individual’s sleep is already compromised, an enjoyable exercise experience still appears to exert a psychological buffering effect, helping to attenuate the spillover from sleep problems to negative emotional states. The underlying mechanism may be that physical activity itself is an independent antianxiety and antidepressant intervention [ 45 , 50 ]. Individuals who enjoy exercise are likely more inclined to use it as a proactive coping strategy to counteract the negative moods and energy depletion that result from insufficient sleep [ 82 ]. In the Chinese professional context, when employees face exhaustion and low spirits due to poor sleep, enjoyable physical activity can provide immediate positive affective feedback and a sense of accomplishment, which can effectively combat the negative mood states exacerbated by sleep problems. This discovery offers a new perspective on SBMs. This reveals a moderating mechanism that operates at a late stage of the stress process. Whereas previous models have often focused on how resources help individuals cope with an initial stressor, our research indicates that high-quality psychological experience can effectively mitigate the adverse consequences stemming from an intermediary factor, such as sleep disturbance. This study is the first to include physical activity enjoyment, viewed as a dynamic, actively cultivable subjective experience, within this framework. Thus, in creating the model, we have moved our theory from a static conception of having to understanding the process of experiencing. The model we developed has more rich and encompassing content. This extension is consistent with emerging ideas in exercise science, through the empirical validation of the relationship between a stressor (work stress), a mediator (sleep disturbance), and a moderator (physical activity enjoyment) on a health outcome (anxiety and depression), this study not only contributes to the theoretical richness of the model but also helps pave the way toward more humanistic workplace mental health interventions. Conclusion Drawing upon the theoretical foundations of the Effort–Reward Imbalance (ERI) and the Stress-Buffering Models (SBM), this study investigated the associations of work stress with sleep disturbance and anxiety and depression. We further examined the moderated mediating role of physical activity enjoyment in the relationship between work stress and anxiety and depression via sleep disturbance. The results demonstrated that this moderating effect significantly attenuated the strength of the entire indirect pathway, confirming that the critical protective value of physical activity enjoyment in mitigating the process through which work stress is linked to anxiety and depression. Consequently, this research makes important contributions to both theory and practice. Theoretical contributions First, this study moves beyond the generalized approach often used in the social science of sport literature concerning the mental health benefits of exercise. By focusing specifically on aerobic and resistance training in a more representative sample of general professionals in China’s first-tier cities, we dissected the adverse cascade linking work stress, sleep disturbance, and anxiety and depression. The core innovation of this research lies in its shift beyond the physical attributes of exercise to a deeper investigation of its qualitative dimensions. We demonstrated that the positive psychological experience derived from exercise—conceptualized as physical activity enjoyment—is a critical variable that can effectively intervene in the aforementioned stress process. This not only offers a more granular understanding of the psychological protection mechanisms of exercise but also fills a theoretical void in the literature concerning the role of the psychological attributes of exercise within specific stress–health models. Second, in terms of theory-building, this study innovatively extends two foundational models in occupational health psychology: the Effort–Reward Imbalance (ERI) model and the Stress-Buffering Model (SBM). Within a unified empirical framework, our results demonstrate that the adverse effects of stress, as articulated by the ERI model, can be effectively mitigated by the protective resources described in the SBM. More groundbreaking is our study’s introduction of physical activity enjoyment—a dynamic, actively cultivable subjective experience—as a core moderating variable within the SBM framework. This is the first study to incorporate this. This conceptual move propels the focus of the theory beyond relatively static resources, such as social support, toward an exploration of dynamic, process-oriented positive psychological experiences, thereby greatly enriching its theoretical substance and practical applicability. These findings open a new perspective for future interventions, suggesting a shift from merely identifying external risks to enhancing the internal experiences that build individual psychological resilience. In doing so, it lays a solid foundation for constructing more agentic and empowering workplace mental health support systems. Practical implications The practical implications of this study are clearly targeted, aiming to provide evidence-based intervention strategies for urban professionals facing high-stress work environments, as well as for the health service and management systems that support them. First, for individuals, the key takeaway is the need to redefine the goal of exercise: shifting the focus from completing a health-related task to discovering a source of intrinsic enjoyment. We encourage professionals to proactively explore a variety of physical activities. Whether it is a team-based sport or a spinning or dance class that combines music and social interaction, the primary criterion for selection should be “Do I enjoy this process?” rather than simply “How many calories am I burning?” Correspondingly, this places new demands on physical activity service providers, such as gyms, personal trainers, and online platforms. Their product design and instructional strategies must move beyond performance metrics to focus more on cultivating a positive and supportive social atmosphere, enhancing the enjoyability of their programs, and helping users discover and sustain their passion for movement through personalized plans. By doing so, they can guide users in making a profound shift from passive fitness to proactive well-being. Second, for organizations and public health bodies, our findings highlight the need to develop a broader, synergistic support system. For corporations, employee wellness programs should evolve beyond traditional offerings such as gym memberships. Instead, they should focus on organizing more diverse and inclusive team-based physical activities, such as outdoor hikes or recreational sports days. Moreover, companies should actively cultivate a workplace culture that encourages employees to take restorative breaks and psychologically detach themselves from work. Promoting employees’ enjoyment of physical activity should be viewed not as a perk but as a strategic investment in organizational psychological capital and productivity. For municipal public health and sports authorities, this study offers a new direction for policy-making. Public health campaigns should shift their messaging from a singular emphasis on the functional benefits of exercise to a dual focus that also champions its enjoyable aspects, thereby shaping a more positive societal conception of physical activity. More importantly, urban planning and development initiatives should prioritize the creation of highly accessible, aesthetically pleasing, and multifunctional recreational spaces. Doing so would provide citizens with both the robust physical infrastructure and the social support network necessary to truly enjoy being active. Limitations and avenues for future research Several limitations of the current study must be acknowledged. First, our use of a cross-sectional design inherently constrains our ability to draw causal inferences among the variables, as all the data were collected at a single point in time. Although our model is grounded in established theory, the correlational nature of the data cannot rule out the possibility of reverse causality or bidirectional relationships. Consequently, the findings should be interpreted as significant associations rather than confirmed causal links. The pathways specified in our model represent a theoretically plausible relational structure that requires further validation through more rigorous research designs. Second, the study sample was limited to professionals in China’s first-tier cities, and all the data were based on self-report measures. These factors may limit the generalizability of our conclusions and could introduce reporting biases. Therefore, caution should be exercised when extending these findings to broader or different populations. Finally, our measurement of the core construct of physical activity enjoyment was relatively broad. This approach may not have fully captured the nuanced effects of its specific subdimensions, such as social enjoyment versus immersive flow experiences. It is possible that these distinct facets of enjoyment play different buffering roles, a possibility that warrants further investigation. Indeed, a key finding from our descriptive statistics, namely, that longer exercise durations were paradoxically associated with poorer mental health outcomes serves as the primary impetus for both the present study and future explorations. This observation highlights the limitations of assessing psychological benefits based solely on physical metrics and provides a strong theoretical rationale for a more nuanced future research agenda. Therefore, we recommend that future studies employ longitudinal designs or randomized controlled trials (RCTs) to more clearly delineate the temporal sequencing and causal pathways among work stress, sleep, physical activity enjoyment, and mental health. To enhance external validity, subsequent research should broaden its sampling to include different city tiers and occupational groups while also incorporating objective measurement tools, such as data from wearable devices or physiological indicators, to validate and supplement self-report information. Furthermore, future inquiries could deconstruct the core construct of enjoyment to examine the differential effects of its subdimensions in buffering the associations between work stress and the outcome variables (sleep disturbance and anxiety and depression). Finally, to bridge the gap between research and practice, we advocate for action research or field experiments conducted in real-world workplace settings to evaluate the efficacy and challenges of corporate wellness programs designed to promote physical activity enjoyment. Such research must include careful ethical consideration: when enjoyable exercise is institutionally promoted, could it be unintentionally alienated into a new form of performance pressure? Attending to these real-world complexities will significantly increase the societal value and applied depth of future research in this area. Supplementary Information Supplementary Material 1. (364.8KB, docx) Acknowledgements Not applicable. Authors’ contributions Kunpeng Li : Writing – review & editing, Writing – original draft, Visualization, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Chih-Hung Pai : Writing – review & editing, Supervision, Project administration, Methodology, Conceptualization. Funding This research was supported by the 2024 Key Project for Philosophy and Social Sciences Research in Shaoxing City (14th Five-Year Plan) (Grant Number: 145D103). Data availability Data will be made available on request. Declarations Ethics approval and consent to participate All procedures performed in the study involving human participants were in accordance with the 1964 Declaration of Helsinki and its subsequent amendments or comparable ethical standards. The survey was approved by the Zhejiang Yuexiu University Institutional Review Board (IRB reference number: 2024ZYU0100231), Written informed consent was obtained from all participants and, where applicable, their legal guardians, prior to the start of the survey. All individuals were informed of the research objectives and voluntarily agreed to participate. Consent for publication Written informed consent was obtained from all participants and their legal guardians before their inclusion in the study. This study does not include any individual person’s data in any form (including individual details, images or videos). Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References 1. World Health Organization, WHO guidelines on mental health at work. 2022. Available: https://www.who.int/news-room/fact-sheets/detail/mentalhealth-at-work . [ PubMed ] 2. Labaran RS, Johnson S, Johnson DA. The association of stress and work hours with sleep duration and insomnia symptoms among U.S.-born and Foreign-born Black adults. Sleep Health. 2023;9(1):56–63. 10.1016/j.sleh.2022.09.011. [ DOI ] [ PMC free article ] [ PubMed ] 3. Virtanen M, et al. Long working hours and depressive symptoms: systematic review and meta-analysis of published studies and unpublished individual participant data. Scand J Work Environ Health. 2018;44(3):239–50. 10.5271/sjweh.3712. [ DOI ] [ PubMed ] 4. Li H, Wu X. ‘Research on the impact of career advancement on the mental health of young workers’. PLOS One . 2025 ; 20(8):e0328440. 10.1371/journal.pone.0328440 [ DOI ] [ PMC free article ] [ PubMed ] 5. Ma S, Jia N. ‘The Symptom Structure and Causal Relationships of Comorbid Anxiety and Depression Among Chinese Primary and Middle School Teachers: A Network Analysis’. Psychol Res Behav Manag . 2024 ; 17:3731–3747. 10.2147/PRBM.S483231. [ DOI ] [ PMC free article ] [ PubMed ] 6. Niu L, Yang Z, ‘Impact of Performance Climate on Overtime Behaviors of New Generation Employees. Dec., : The Moderating Effect of Perceived Employability and Mediating Role of Job Insecurity’. Psychol Res Behav Manag. 2022;15:3733–3749. 10.2147/PRBM.S390051. [ DOI ] [ PMC free article ] [ PubMed ] 7. Kiema-Junes H et al. Oct., ‘The association between leisure‐time physical activity and cardiorespiratory and muscular fitness and job strain—Study with a Northern Finland birth cohort of 1966’. Stress Health. 2024;40(5) 10.1002/smi.3477. [ DOI ] [ PubMed ] 8. Borrelli I, Melcore G, Perrotta A, Santoro PE, Rossi MF, Moscato U. ‘Workplace ethical climate and its relationship with depression, anxiety and stress’. Occup Med . 2024;74(6):449–454. 10.1093/occmed/kqae065. [ DOI ] [ PubMed ] 9. Bouche A, et al. Psychological Empowerment as a Mediator of Effort–Reward Imbalance Effects on Burnout, Anxiety and Depression Among Nursing Home Professionals: A Cross-Sectional Study. J Adv Nurs. 2025. 10.1111/jan.16709. [ DOI ] [ PMC free article ] [ PubMed ] 10. Laferton JAC, Schiller S, Conrad D, Fischer D, Zimmermann-Viehoff F. Stress beliefs moderate the impact of COVID‐19 related work stress on depressive, anxiety and distress symptoms in health care workers. Stress Health. 2024;40(4). 10.1002/smi.3410. [ DOI ] [ PubMed ] 11. Tomac P, Rapić IJ, Lugović-Mihić L, Macan J, Bjelajac AK. ‘Working hours indirectly affect anxiety symptoms through sleep and stress in dentists, physicians, and psychotherapists’. Arch Ind Hyg Toxicol. 2025;76(2):102–112. 10.2478/aiht-2025-76-3954. [ DOI ] [ PMC free article ] [ PubMed ] 12. Pachi A, et al. Stress, Anxiety and Depressive Symptoms, Burnout and Insomnia Among Greek Nurses One Year After the End of the Pandemic: A Moderated Chain Mediation Model. J Clin Med. 2025;14(4):1145. 10.3390/jcm14041145. [ DOI ] [ PMC free article ] [ PubMed ] 13. Rahman F, et al. Insomnia and job stressors among healthcare workers who served COVID-19 patients in Bangladesh. BMC Health Serv Res. 2023;23(1). 10.1186/s12913-023-09464-x. [ DOI ] [ PMC free article ] [ PubMed ] 14. Chekroud SR et al. Sep., Association between physical exercise and mental health in 1·2 million individuals in the USA between 2011 and 2015: a cross-sectional study. Lancet Psychiatry. 2018;5(9):739–746.10.1016/s2215-0366(18)30227-x. [ DOI ] [ PubMed ] 15. Yuk K, Lim J, Kim S, Kim TY, Moon HY. Exploring the effects of aerobic and resistance exercise on mood-related symptoms and EEG activity. Front Hum Neurosci. 2025;19:562702. 10.3389/fnhum.2025.1562702. [ DOI ] [ PMC free article ] [ PubMed ] 16. Higham SM, Mendham AE, Rosenbaum S, Allen NG, Smith G, Duffield R. Effect of Concurrent Exercise Training on Stress, Depression and Anxiety in Inactive Academics: Secondary Analysis of a Randomized Controlled Trial. Res Q Exerc Sport. 2025:1–10. 10.1080/02701367.2025.2461322. [ DOI ] [ PubMed ] 17. Harada K, et al. Cross-sectional associations of weekly time, social context, and motivation of exercise with mental health among workers. Int Arch Occup Environ Health. 2024;97(1):23–33. 10.1007/s00420-023-02021-3. [ DOI ] [ PubMed ] 18. Aziz S, Meier B, Wuensch K, Dolbier C. Take a break! Leisure participation moderates the workaholism–work stress relationship. Career Dev. Q. 2023;71(4):315–329. 10.1002/cdq.12336. [ DOI ] [ PMC free article ] [ PubMed ] 19. Roswag M, Abdel Hadi S, Häusser JA, Mojzisch A. Running toward my challenges: Day-level effects of physical activity before work on appraisal of the upcoming workday and employee well-being. J Occup Health Psychol. 2023;28(5):310–324. 10.1037/ocp0000360. [ DOI ] [ PubMed ] 20. Bjornsdottir E et al. Mar. Association between physical activity over a 10-year period and current insomnia symptoms, sleep duration and daytime sleepiness: a European population-based study. BMJ Open. 2024;14(3):e067197. 10.1136/bmjopen-2022-067197. [ DOI ] [ PMC free article ] [ PubMed ] 21. Siegrist J. Adverse Health Effects of High-Effort/Low-Reward Conditions. J Occup Health Psychol. 1996. Available: 10.1037/1076-8998.1.1.27. [ DOI ] [ PubMed ] 22. Alshehri FF, et al. Post-traumatic Stress Disorder and Its Associated Risk Factors Among Emergency Healthcare Workers: A Saudi Cross-Sectional Analytical Study. Cureus Aug. 2023. 10.7759/cureus.44327. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Borrelli I et al. Occupational exposure to work-related stress, a proposal of a pilot study to detect psychological distress in collar-workers. Ann Ig Med Prev E Comunita. 2023;35(5):521–533. 10.7416/ai.2023.2566. [ DOI ] [ PubMed ] 24. Obeidat NA, Dodin YI, Hawari FI, Albtoosh AS, Manasrah RM, Mansour AH. Mitigating psychological distress in healthcare workers as COVID-19 waves ensue: a repeated cross-sectional study from Jordan. Hum Resour Health Dec. 2022;20(1):32. 10.1186/s12960-022-00728-x. [ DOI ] [ PMC free article ] [ PubMed ] 25. Akova I, Hasdemir O, Kilic E. Evaluation of the relationship between burnout, depression, anxiety, and stress levels of primary health-care workers (Center Anatolia). Alex J Med. 2021;57(1):52–60. 10.1080/20905068.2021.1874632. 26. Maddock A. The Relationships between Stress, Burnout, Mental Health and Well-Being in Social Workers. Br J Soc Work. 2024;54(2):668–686. 10.1093/bjsw/bcad232. 27. Chaabouni A. The effect of job stress appraisals on mental health among health professionals: The mediating role of work engagement. Eur Psychiatry. 2021;64:S466–S466. 10.1192/j.eurpsy.2021.1244. 28. Gao L, Wang L-L, Yang R, Yang X-J, Zhou S-J. Anxiety, Depression, Perceived Stress, and Burnout Among Chinese Researchers: A Cross-Sectional Nationwide Study. Psychol Res Behav Manag. 2023;16:4209–20. 10.2147/PRBM.S431011. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Kaushik A, Ravikiran SR, Suprasanna K, Nayak MG, Baliga K, Acharya SD. Depression, Anxiety, Stress and Workplace Stressors among Nurses in Tertiary Health Care Settings. Indian J Occup Environ Med. 2021;25(1):27–32. 10.4103/ijoem.IJOEM_123_20. [ DOI ] [ PMC free article ] [ PubMed ] 30. Sun J, Sarfraz M, Ivascu L, Iqbal K, Mansoor A. How Did Work-Related Depression, Anxiety, and Stress Hamper Healthcare Employee Performance during COVID-19? The Mediating Role of Job Burnout and Mental Health. Int J Environ Res Public Health. 2022;19(16):10359. 10.3390/ijerph191610359. [ DOI ] [ PMC free article ] [ PubMed ] 31. Frazier C, Brown TH. Stress Work-Related. Psychosocial Resources, and Insomnia Symptoms Among Older Black Workers. J Aging Health. 2022;34(3):424–434. 10.1177/08982643221085899. [ DOI ] [ PMC free article ] [ PubMed ] 32. Gustavsson K, Wichniak A. Night work as a stressor: The role of sleep reactivity to stress in the relationship between night work and insomnia. J Sleep Res. 2022;31(2):e13468. 10.1111/jsr.13468. [ DOI ] [ PubMed ] 33. Wang M-F, et al. The relationship between occupational stressors and insomnia in hospital nurses: The mediating role of psychological capital. Front Psychol Feb. 2023;13:1070809. 10.3389/fpsyg.2022.1070809. [ DOI ] [ PMC free article ] [ PubMed ] 34. Gustavsson K, Wichniak A. Overcommitment to work as a mediator of the association between effort-reward imbalance and insomnia among shift working nurses. Eur Psychiatry. 2021;64:S817–S818. 10.1192/j.eurpsy.2021.2160. 35. Gustavsson K, Jernajczyk W, Wichniak A. Insomnia Partially Mediates the Relationship of Occupational Stress with Mental Health Among Shift Working Nurses and Midwives in Polish Hospitals. Nat Sci Sleep. 2022;14:1989–99. 10.2147/NSS.S375375. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 36. Shen Y, Zhan Y, Zheng H, Liu H, Wan Y, Zhou W. Anxiety and its association with perceived stress and insomnia among nurses fighting against COVID-19 in Wuhan: A cross-sectional survey. J Clin Nurs. 2021;30(17–18):2654–2664. 10.1111/jocn.15678. [ DOI ] [ PMC free article ] [ PubMed ] 37. Lindsäter E, et al. The mediating role of insomnia severity in internet-based cognitive behavioral therapy for chronic stress: Secondary analysis of a randomized controlled trial. Behav Res Ther. 2021;136:103782. 10.1016/j.brat.2020.103782. [ DOI ] [ PubMed ] 38. Ubara A, et al. How Does E-mail-Delivered Cognitive Behavioral Therapy Work for Young Adults (18–28 Years) with Insomnia? Mediators of Changes in Insomnia, Depression, Anxiety, and Stress. Int J Environ Res Public Health. Apr. 2022;19(8):4423. 10.3390/ijerph19084423. [ DOI ] [ PMC free article ] [ PubMed ] 39. Xinjie D, Runlian H, Ahmed O, Cho E, Chung S. Mediating Role of Viral Anxiety and Insomnia in Relationship Between Work-Related Stress and Depression Among Cold Chain Workers During COVID-19 Pandemic. J Korean Med Sci. 2023;38(43):e338. 10.3346/jkms.2023.38.e338. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 40. Anupama M, Kulkarni H, Nisarga V, Sushravya. Perceived Stress and Insomnia Among Medical Residents in South India: A Cross-Sectional Study. Ann Indian Psychiatry. 2022;6(2):142–148. 10.4103/aip.aip_140_21. 41. Omichi C, et al. Limited social support is associated with depression, anxiety, and insomnia in a Japanese working population. Front Public Health. 2022;10:981592. 10.3389/fpubh.2022.981592. [ DOI ] [ PMC free article ] [ PubMed ] 42. Cohen S, Wills TA. Stress, Social Support, and the Buffering Hypothesis. Psychol Bull. 1985. Available: 10.1037/0033-2909.98.2.310. [ PubMed ] 43. Bustamante EE, Brellenthin AG, Brown DR, O’Connor PJ. Up for Debate: Does Regular Physical Activity Really Improve Mental Health? Med Sci Sports Exerc. 2025;57(5):1056–66. 10.1249/MSS.0000000000003636. [ DOI ] [ PubMed ] 44. Cortes-Denia D, Lopez-Zafra E, Pulido-Martos M. Physical and psychological health relations to engagement and vigor at work: A PRISMA-compliant systematic review. Curr Psychol. 2023;42(1):765–80. 10.1007/s12144-021-01450-y. 45. Noetel M et al. Feb., Effect of exercise for depression: systematic review and network meta-analysis of randomised controlled trials. BMJ. 2024:e075847. 10.1136/bmj-2023-075847. [ DOI ] [ PMC free article ] [ PubMed ] 46. Liu R, Tang X. Effect of leisure-time physical activity on depression and depressive symptoms in menopausal women: a systematic review and meta-analysis of randomized controlled trials. Front Psychiatry. 2025;15:1480623. 10.3389/fpsyt.2024.1480623. [ DOI ] [ PMC free article ] [ PubMed ] 47. Rosel I, et al. The effect of exercise and affect regulation skills on mental health during the COVID-19 pandemic: A cross-sectional survey. Psychiatry Res. 2022;312:114559. 10.1016/j.psychres.2022.114559. [ DOI ] [ PMC free article ] [ PubMed ] 48. Novielli J et al. Tai Chi/Qigong in Adults with Depression and Anxiety: A Pilot Retrospective Study. Integr Med Rep. 2022;1(1):131–139. 10.1089/imr.2021.0023. 49. Singphow C, et al. Effect of Yoga on Stress, Anxiety, Depression, and Spinal Mobility in Computer Users with Chronic Low Back Pain. Int J Yoga. May 2022;15(2):114–21. 10.4103/ijoy.ijoy_9_22. [ DOI ] [ PMC free article ] [ PubMed ] 50. Correia EM et al. Mar. Analysis of the Effect of Different Physical Exercise Protocols on Depression in Adults: Systematic Review and Meta-analysis of Randomized Controlled Trials. Sports Health- Multidiscip Approach. 2024;16(2):285–294. 10.1177/19417381231210286 [ DOI ] [ PMC free article ] [ PubMed ] 51. Zhang Y, Li G, Liu C, Guan J, Zhang Y, Shi Z. Comparing the efficacy of di erent types of exercise for the treatment and prevention of depression in youths: a systematic review and network meta-analysis. Front Psychiatry. 2023;14:1199510. 10.3389/fpsyt.2023.1199510. [ DOI ] [ PMC free article ] [ PubMed ] 52. Maharaj SS, Nuhu JM. Effect of treadmill walking for anxiety, depression, and social well-being in women with diabetes type 2: A randomized controlled trial. Health Care Women Int. 2023;44(6):734–752. 10.1080/07399332.2022.2090566. [ DOI ] [ PubMed ] 53. Su Y, et al. The role of anxiety and depression in the relationship between physical activity and sleep quality: A serial multiple mediation model. J Affect Disord. 2021;290:219–26. 10.1016/j.jad.2021.04.047. [ DOI ] [ PubMed ] 54. Ye Y-P et al. Jun., The impact of physical exercise on university students’ life satisfaction: The chain mediation effects of general self-efficacy and health literacy. PLOS One. 2025;20(6):e0325835. 10.1371/journal.pone.0325835. [ DOI ] [ PMC free article ] [ PubMed ] 55. Shen X, et al. The dual roles of empathy in mediating structural empowerment and compassion fatigue among Chinese nurses. BMC Nurs. 2024;23(1):837. 10.1186/s12912-024-02499-3. [ DOI ] [ PMC free article ] [ PubMed ] 56. Yang Q, et al. Misconceptions, prejudice and social distance towards people with alcohol use disorders in China. BMC Psychiatry. 2025;25(1). 10.1186/s12888-025-06595-9. [ DOI ] [ PMC free article ] [ PubMed ] 57. Faul F, Erdfelder E, Buchner A, Lang A-G. Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behav Res Methods. 2009;41(4):1149–1160. 10.3758/BRM.41.4.1149. [ DOI ] [ PubMed ] 58. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale, N.J: L. Erlbaum Associates; 1988. [ Google Scholar ] 59. Jj E, Bartlett JW, Kotrlik CC, Higins. Organizational Research Determining Appropriate Sample Size in Survey Research. 60. American College of Sports Medicine, Eds. ACSM’s guidelines for exercise testing and prescription. 9th ed. Philadelphia: Wolters Kluwer/Lippincott Williams & Wilkins Health; 2014. [ Google Scholar ] 61. Dhahbi W, et al. Physical Activity to Counter Age–Related Cognitive Decline: Benefits of Aerobic, Resistance, and Combined Training—A Narrative Review. Sports Med - Open. 2025;11(1):56. 10.1186/s40798-025-00857-2. [ DOI ] [ PMC free article ] [ PubMed ] 62. Trujillo-Muñoz PJ, Sánchez-Ojeda MA, Rodríguez-Huamán EC, Mezyani-Haddu K, Hoyo-Guillot I, Navarro-Prado S. Effects of Physical Exercise on Symptoms and Quality of Life in Women in Climacteric: A Systematic Review and Meta-Analysis. Healthcare. 2025;13(6):644. 10.3390/healthcare13060644. [ DOI ] [ PMC free article ] [ PubMed ] 63. Cohen S, Kamarck T, Mermelstein R. A Global Measure of Perceived Stress. J Health Soc Behav. 1983;24(4)385. 10.2307/2136404. [ PubMed ] 64. Chaudhary FA, et al. Association of sociodemographic and occupational stress factors with smoking behavior among healthcare professionals: The mediating role of physical exercise. Tob Induc Dis. 2025;23:1–9. 10.18332/tid/204007. [ DOI ] [ PMC free article ] [ PubMed ] 65. Bastien C. Validation of the Insomnia Severity Index as an outcome measure for insomnia research. Sleep Med. 2001;2(4):297–307. 10.1016/s1389-9457(00)00065-4. [ DOI ] [ PubMed ] 66. Haghighi S, Ali M. Validity and Reliability of the Iranian Version of the Insomnia Severity Index. Malays J Medical Sci. 2012. [ PMC free article ] [ PubMed ] 67. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand. 1983;67(6):361–70. 10.1111/j.1600-0447.1983.tb09716.x. [ DOI ] [ PubMed ] 68. Zupkauskiene J, et al. Changes in health-related quality of life, motivation for physical activity, and the levels of anxiety and depression after individualized aerobic training in subjects with metabolic syndrome. Hellenic J Cardiol. 2022;66:41–51. 10.1016/j.hjc.2022.04.003. [ DOI ] [ PubMed ] 69. Deborah DeCarlo KD. Physical activity enjoyment scale: two validation studies. J Sport Exerc Psychol. 1991. 10.1123/jsep.13.1.50. [ Google Scholar ] 70. Chen C et al. Oct., A Short Version of the Physical Activity Enjoyment Scale: Development and Psychometric Properties. Int J Environ Res Public Health. 2021;18(21):11035. 10.3390/ijerph182111035. [ DOI ] [ PMC free article ] [ PubMed ] 71. Fritsch J, et al. A Study on the Psychometric Properties of the Short Version of the Physical Activity Enjoyment Scale in an Adult Population. Int J Environ Res Public Health. 2022;19(22):15294. 10.3390/ijerph192215294. [ DOI ] [ PMC free article ] [ PubMed ] 72. Hayes AF. Introduction to mediation, moderation, and conditional process analysis: a regression-based approach. Second edition. in Methodology in the social sciences. New York: Guilford Press; 2018. 73. Andrew F Hayes. PROCESS. A Versatile Computational Tool for Observed Variable Mediation, Moderation, and Conditional Process Modeling. Contract Publ Guilford Press; 2012. http://www.afhayes.com/ . 74. Igartua J-J, Hayes AF. Mediation, Moderation, and Conditional Process Analysis: Concepts, Computations, and Some Common Confusions. Span J Psychol. 2021;24. 10.1017/sjp.2021.46. [ DOI ] [ PubMed ] 75. Ozturk A, Yilmaz O. Determination of Environmental Risk Perception and Environmental Attitudes of Individuals Participating in Recreational Aimed Outdoor Sports Activities. J Environ Prot Ecol. 2020;21(6):2072–82. [ Google Scholar ] 76. Hu L, Bentler PM. ‘Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Struct Equ Model Multidiscip J. 1999;6(1):1–55. 10.1080/10705519909540118. 77. Kline RB. Response to Leslie Hayduk’s Review of Principles and Practice of Structural Equation Modeling, 4th Edition. Can Stud Popul. 2018;45(3–4):188. 10.25336/csp29418. 78. Fornell C, Larcker DF. Evaluating Structural Equation Models with Unobservable Variables and Measurement Error. J Mark Res. 1981;18(1):39. 10.2307/3151312. 79. Møller SF, Frese JV, Bro R. Robust methods for multivariate data analysis. J Chemom. 2005;19(10):549–563. 10.1002/cem.962. 80. Biber DD, Rice K, Ellis R. Self-compassion training within a workplace physical activity program: A pilot study. Work. 2021;68(4):1059–1067. 10.3233/WOR-213436. [ DOI ] [ PubMed ] 81. Bratland-Sanda S, Schmidt SK, Reinboth MS, Vrabel KA. Under pressure to exercise: a cross-sectional study of characteristics and predictors of compulsive exercise in early adolescents. J Eat Disord. 2022;10(1):156. 10.1186/s40337-022-00686-8. [ DOI ] [ PMC free article ] [ PubMed ] 82. Rösel I, et al. The effect of exercise and affect regulation skills on mental health during the COVID-19 pandemic: A cross-sectional survey. Psychiatry Res. 2022;312:114559. 10.1016/j.psychres.2022.114559. [ DOI ] [ PMC free article ] [ PubMed ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplementary Material 1. (364.8KB, docx) Data Availability Statement Data will be made available on request. Articles from BMC Psychology are provided here courtesy of BMC ACTIONS View on publisher site PDF (1.9 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top