Does physical activity influence health behavior, mental health, and psychological resilience under the moderating role of quality of life?

The study investigates whether and how physical activity influences health behavior, mental health, and psychological resilience, and whether quality of life moderates these associations. Prior research indicates PA is linked to reduced chronic disease risk, improved mental well-being, and greater resilience. The authors posit that regular PA promotes beneficial health behaviors and psychological outcomes and that QoL may strengthen or alter these relationships. They develop hypotheses H1–H6 for direct effects among PA, HeB, MeH, and PsR, and H7–H12 for QoL’s moderating effects on these links.

The paper synthesizes evidence that regular PA benefits mental health, cognitive function, and subjective well-being, and is associated with resilience across age groups, including older adults. Studies suggest vigorous PA may confer stronger effects and that aerobic and strength training improve mental health and brain function. Health behaviors are linked to better mental health outcomes, with evidence from COVID-19 era studies showing associations between activity levels and psychological well-being in various populations. Regarding quality of life, literature indicates PA relates positively to health-related QoL across age groups, and QoL is associated with self-efficacy and mental health. Research suggests potential moderating roles of QoL and resilience in relationships among PA, mental health, and well-being, though mechanisms and moderators require further elucidation.

Design and ethics: Cross-sectional online survey conducted in China from April 15, 2023 to October 15, 2023, approved by Hunan City University Ethics Committee, adhering to the Helsinki Declaration. Informed consent obtained. Sampling and participants: Convenience sampling; inclusion criteria were age ≥21 and being active in PA to improve health and well-being. Of >1,000 responses, 743 were retained after data quality checks. Demographics (n=743): 52.09% male, 47.91% female; broad age distribution (21–25 to 51+); varied education, marital status, and occupations (self-employed, government employees, students). Survey development: Self-administered questionnaire based on prior literature; pilot-tested with 25 participants and refined. Included closed/open questions and Likert-type items. Measures: - Physical activity (PA): 5-item frequency-based 5-point Likert (from almost every day to sometimes) covering light, moderate, strenuous activities, walking, and social activities (adopted from Saqib et al., 2020). - Health behavior (HeB): 5-point Likert (strongly disagree to strongly agree) assessing health-related behaviors (from Fan et al., 2023). - Quality of life (QoL): 5 items on 5-point scale (poor to excellent) including mental health, physical health over past four weeks, psychiatric consultation, social/community/civic participation, overall well-being. - Mental health (MeH): 7 items on 5-point frequency scale (none of the time to all the time) including optimism, usefulness, relaxation, problem-solving, clarity, social closeness, decisiveness (from Fan et al., 2023). - Psychological resilience (PsR): Multiple items on 5-point agreement scale assessing confidence, adaptability, perseverance, recovery from setbacks, coping with unexpected problems, self-appreciation, multitasking, and self-belief (based on De Terte et al., 2014). Data analysis: Descriptive statistics via SPSS-25. PLS-SEM (SmartPLS 4.0) tested two structural models: Model 1 (direct effects without QoL moderation) and Model 2 (including six QoL moderation paths). Model fit indices (SRMR, d_ULS, d_G, Chi-square, NFI) were acceptable. Reliability and validity were assessed via Cronbach’s alpha, composite reliability (rho_a, rho_c), AVE, convergent validity (AVE), discriminant validity (Fornell-Larcker, HTMT), and multicollinearity (VIF<5). Factor loadings for constructs were adequate, and reliability thresholds (CR>0.70; AVE>0.50; alpha>0.70) were met.

Direct effects (Model 1): - PA → HeB: β=0.660, t=19.031, p<0.001 - PA → MeH: β=0.554, t=4.967, p<0.001 - PA → PsR: β=0.196, t=6.195, p<0.001 - HeB → MeH: β=0.539, t=12.232, p<0.001 - HeB → PsR: β=0.476, t=12.775, p<0.001 - MeH → PsR: β=0.164, t=4.793, p<0.001 Direct effects (Model 2, with QoL moderation included): - PA → HeB: β=0.512, t=11.472, p<0.001 - PA → MeH: β=0.159, t=3.762, p<0.001 - PA → PsR: β=0.175, t=5.237, p<0.001 - HeB → MeH: β=0.460, t=9.861, p<0.001 - HeB → PsR: β=0.447, t=11.347, p<0.001 - MeH → PsR: β=0.166, t=4.262, p=0.000 Moderation by QoL (Model 2): - QoL × PA → HeB: β=0.062, t=2.444, p=0.015 (significant) - QoL × PA → MeH: β=-0.054, t=1.394, p=0.164 (ns) - QoL × HeB → MeH: β=0.047, t=1.579, p=0.114 (ns) - QoL × HeB → PsR: β=0.084, t=2.041, p=0.041 (significant) - QoL × MeH → PsR: β=-0.116, t=2.336, p=0.020 (significant) - QoL × PA → PsR: β=-0.087, t=2.443, p=0.015 (significant) Summary: All hypothesized direct paths (H1–H6) were supported. Four of six moderation hypotheses were significant (H7, H10, H11, H12), while two (H8, H9) were not.

The findings support that regular physical activity enhances health behavior, mental health, and psychological resilience. Health behavior further contributes to both mental health and resilience, and mental health is positively associated with resilience, aligning with prior literature. Incorporating QoL as a moderator showed that QoL significantly alters several relationships—strengthening PA’s effect on health behavior and HeB’s impact on resilience, while moderating (negatively in this sample) the effects of PA and MeH on resilience. These results indicate that the benefits of PA on psychological outcomes are context-dependent, with QoL shaping the magnitude and direction of effects. Practically, fostering PA alongside strategies to enhance QoL may yield broader improvements in mental health, resilience, and health-promoting behaviors.

Physical activity exerts beneficial direct effects on health behavior, mental health, and psychological resilience. Quality of life moderates multiple pathways among these constructs, underscoring the complex, multidimensional nature of these relationships. The study highlights the importance of promoting consistent PA participation and considering QoL in the design of health promotion strategies to optimize psychological and behavioral outcomes. Future work should clarify mechanisms of moderation and evaluate these relationships longitudinally and across diverse populations to strengthen causal inference and generalizability.

- Limited generalizability due to convenience sampling; the sample may not represent the broader population. - Potential sampling bias from over- or under-representation of certain groups, affecting external validity. - Homogeneity of participants focused on individuals already engaging in PA for health and well-being, limiting applicability to those with different motivations or activity profiles. - Limited exploration of other influential factors (e.g., pre-existing medical conditions, socioeconomic status, psychological issues) that may confound associations.