Longitudinal associations between adherence to 24-h movement guidelines and physical fitness and body composition in pediatric overweight and obesity Authors: Ning Su, Guifang Liu, Junming Lyu, Lin Zhou, Jie Zhu, Yuxiu He, Ryan E. Rhodes, Xueyun Shao, Wei Liang Published: 2026 ABSTRACT This study investigated the longitudinal associations between adherence to the 24-h Movement Guidelines (24-HMG), specifically, meeting the recommendations for moderate-to-vigorous physical activity (MVPA) (≥ 60 min/day), recreational screen time (≤ 2 h/day), and sleep (9–11 h/night), and one-year changes in physical fitness and body composition among overweight and obese children. Particularly, the study identified whether the number and specific combinations of 24-HMG adherence were associated with more favorable health outcomes in this high-risk pediatric group. A prospective cohort of 337 children aged 6–12 years from six public schools in China was followed for one year. Physical activity, sedentary screen time, and sleep were assessed using triaxial accelerometers, validated questionnaires, and parental reports. Physical fitness was evaluated via the Chinese National Student Physical Fitness Battery, while body composition i.e., body fat percentage (BFP), fat-free mass (FFM), and skeletal muscle mass (SMM), was measured using bioelectrical impedance analysis. Covariates included demographic characteristics and parental support for healthy movement behaviors. Over the follow-up, sleep duration decreased and sedentary time increased, while MVPA showed no significant change. All physical fitness and body composition indicators increased over the one-year period. Adherence to two guidelines, especially those including MVPA, was associated with reductions in BFP and gains in FFM. Combinations of MVPA with reduced sedentary time or adequate sleep showed the strongest associations with improvements in physical fitness and FFM and decreases in BFP. Full adherence to all three guidelines was associated only with improvements in physical fitness and FFM. This study demonstrates that adherence to the 24-HMG, particularly combinations involving MVPA, is associated with improved physical fitness and body composition in overweight and obese children. The underlying mechanisms and intervention evidence warrant further investigation.Level of evidence Level III-well-designed cohort study.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-53270-4. Introduction Over recent decades, rapid economic development and lifestyle transitions have led to a dramatic global rise in childhood overweight and obesity1. This escalating public health concern has attracted considerable scientific attention due to its profound and wide-ranging impact on children’s health. Excess weight in childhood is associated with an elevated risk of chronic conditions, including metabolic syndrome, cardiovascular disease, and musculoskeletal disorders2,3. In addition to general obesity measured by BMI, abdominal obesity assessed by waist circumference is closely related to metabolic risks in children and adolescents4. Moreover, overweight and obese children often exhibit lower levels of cardiorespiratory endurance, muscular strength, speed, agility, coordination, and flexibility compared with their normal-weight peers4–6. Physical fitness and body composition are key indicators of pediatric health. Physical fitness underpins functional capacity and athletic performance, while a healthy body composition supports physiological functions and metabolic regulation, jointly contributing to overall well-being7. Understanding modifiable determinants of these outcomes, such as physical activity, sedentary behavior, and sleep, is essential for informing effective prevention and intervention strategies for overweight and obese children. Evidence consistently demonstrates that individual movement behaviors (e.g., physical activity, sedentary behavior, and sleep) play a critical role in shaping health. Higher levels of moderate-to-vigorous physical activity (MVPA) are linked to reduced adiposity and improved cardiometabolic fitness8,9, whereas excessive sedentary behavior, particularly sedentary screen time (SST), is associated with greater obesity risk and less favorable body composition10,11. Inadequate sleep duration has also been related to adverse metabolic outcomes and increased adiposity12,13. While these behaviors have been studied independently, they are interdependent elements of a finite 24-h period. Recognizing this interrelationship, researchers have advocated for a more integrated framework. The Canadian 24-h Movement Guidelines (24-HMG) were developed to capture the integrated effects of physical activity, sedentary behavior, and sleep within a single daily cycle14. The guidelines recommend that children aged 6–12 accumulate at least 60 min of MVPA per day, spend no more than 2 h on recreational screen time per day, and sleep 9–11 h per night14. Adherence to these guidelines has been associated with favorable cardiometabolic profiles, improved physical fitness, and healthier body composition, particularly in children with overweight or obesity9,14,15. Despite growing interest in the integrated approach, much of the existing literature remains limited in scope. Most studies have examined movement behaviors in isolation rather than in combination, and the majority are cross-sectional, restricting causal inference and failing to capture behavior changes over time16,17. In addition, reliance on self-reported measures introduces recall and social desirability biases; small or unrepresentative samples further limit generalizability, particularly for high-risk groups such as overweight and obese children. Few investigations have tracked adherence to the 24-HMG beyond one year, leaving important gaps in understanding the long-term implications for physical development. The present prospective cohort study addresses these gaps by exploring the longitudinal associations between adherence to the 24-HMG and changes in physical fitness and body composition in overweight and obese children. Specifically, it is hypothesized that the number of guidelines adherence would be associated with favorable changes in physical fitness and body composition indicators among overweight and obese children. In addition, the specific combinations of guidelines adherence would be associated with improvements in these health outcomes. By combining a longitudinal approach with objective measurement and an integrated analytic framework, this study provides timely and robust evidence on how combinations of daily movement behaviors affect physical development in an underrepresented, high-risk pediatric population. The findings aim to refine our understanding of behavior–health linkages and inform the design of realistic, culturally relevant interventions to promote healthier growth trajectories. Methods Study design, participants and procedure This study employed a two-wave prospective longitudinal design. A stratified random sampling strategy was used to select six public primary schools from three provinces in China (Hebei, Shandong, and Hunan). The selected schools were comparable in class size (approximately 30–40 students per class), curriculum structure, teaching quality, and cultural background. Within each school, one or two classes were randomly selected from grades 1 to 6. Eligible participants were overweight or obese school-aged children (6–12 years)18,19 without physical mobility limitations and without a diagnosis of cognitive or physical disorders within the preceding six months. Exclusion criteria included physical disabilities, recent surgical procedures, or documented intellectual impairments. Baseline assessments were conducted between April and May 2022, prior to the end of the spring semester, and follow-up assessments took place approximately one year later. The one-year interval aligns with common practice in behavioral longitudinal research. All procedures complied with the ethical principles of the Declaration of Helsinki and were approved by the Medical and Research Ethics Committee of the authors’ affiliated institution (Ref. No. 2021LLSC051). Written informed consent was obtained from both children and their legal guardians before participation. A total of 1,467 students were approached, and 1,370 responded (response rate: 93.4%). Of these, 428 met eligibility criteria and completed baseline measures. At the one-year follow-up, 337 participants provided valid data and were included in the final analyses (Appendix 1, Fig. S1). Measurements 24-h movement behaviors Children’s movement behaviors were objectively measured using ActiGraph GT3X+ triaxial accelerometers (ActiGraph LLC, Pensacola, FL, USA). Participants wore the device over the right hip for seven consecutive days, except during water-based activities. Valid data were defined as at least three weekdays and one weekend day with ≥ 10 h of wear time during waking hours. Non-wear time was defined as ≥ 20 consecutive minutes of zero counts19,20. Data were sampled at 80 Hz and aggregated into 15 sec epochs. Physical activity intensity was classified using the cut-points: sedentary behavior ≤ 25 counts/15 s; light physical activity 25–574 counts/15s; MVPA ≥ 574 counts/15 s21. SST was assessed using parent-reported items from the Health Behaviour in School-aged Children (HBSC)22, which recorded average daily hours spent in recreational screen-based sedentary behaviors (e.g., television, internet use, video games) on weekdays and weekends22. Parents were provided with written examples of recreational screen-based activities in the questionnaire to standardize reporting. Average daily SST was calculated as (weekday hours × 5 + weekend hours × 2)/7. Sleep duration was estimated using the ActiLife Sleep+ algorithm (ActiGraph LLC, Pensacola, FL, USA), which served as the primary measure for guideline classification. Parental reports of typical bedtimes and wake-up times on weekdays and weekends were collected for cross-validation and to supplement missing or invalid device data. To assess agreement between the two measures, we randomly selected 34 participants (approximately 10% of the sample) and calculated the intraclass correlation coefficient (ICC), which was 0.81, indicating satisfactory agreement; this subsample size was considered sufficient for a reliable ICC estimate23. When both measures were available, the device-derived value was prioritized for determining adherence to the sleep guideline. Based on the Canadian 24-Hour Movement Guidelines14 and Chinese expert consensus24, adherence for children aged 6–12 was defined as: (a) ≥ 60 min of MVPA/day; (b) ≤ 2 h/day of recreational sedentary screen time; (c) 9–11 h/night of sleep for 6–12 children. Following previous studies25,26, adherence was classified in two ways: (a) total number of guidelines met: 0 = met none, 1 = met one, 2 = met two, 3 = met al.l three; (b) specific adherence combinations: 0 = met none of the guidelines, 1 = met only the MVPA guideline, 2 = met only SST guideline, 3 = met only the sleep guideline, 4 = met both MVPA and SST guidelines, 5 = met both MVPA and sleep guidelines, 6 = met both SST and sleep guidelines, and 7 = met al.l three guidelines. Physical fitness and body composition Physical fitness was assessed using the 2014 Chinese National Student Physical Fitness Battery (MEPRC, 2014), which consists of seven tests, including BMI (kg/m²), forced vital capacity, 50-m sprint, sit-and-reach flexibility, 1 min rope-skipping, 1 min sit-ups, and a 50 m×8 shuttle run (administered only to Grades 5–6). Scores for each test were standardized for age and sex, then weighted and summed to yield a total score ranging from 0 to 100, with higher scores indicating better physical fitness. Physical fitness tests were conducted during the general physical education classes. For younger children (Grades 1–4) who did not complete the shuttle-run test, the composite score was based on the remaining six tests, with the weighting adjusted according to the national standards. Body composition, including body fat percentage (BFP, %), fat-free mass (FFM, kg), and skeletal muscle mass (SMM, kg), was measured using a multi-frequency bioelectrical impedance analyzer (InBody230, Seoul, South Korea)20,24. Measurements were conducted in the morning under fasting conditions, with participants barefoot and in light clothing. Parents and teachers were provided with written instructions beforehand to ensure standardized testing conditions, including avoiding vigorous physical activity the previous day and maintaining adequate hydration. Covariates and quality control Covariates included child age, sex, weight status, and primary caregivers’ characteristics such as age, sex, educational attainment, and household income were collected based on previous studies (Cronbach’s α = 0.87)16,27. Quality control measures included standardized training for all research assistants, adherence to uniform measurement protocols, and on-site supervision during data collection. Accelerometer compliance was monitored through caregiver-maintained wear logs, and devices were checked midweek for functionality. Physical fitness and body composition tests were conducted by the same trained personnel under standardized conditions, with equipment calibrated daily. All questionnaires were reviewed for completeness, and data were double entered with discrepancies resolved by cross-checking original records. Covariates were assessed only at baseline, whereas all other variables were measured at both baseline and follow-up. The study was conducted in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) checklist (https://www.strobe-statement.org/checklists/). Statistical analysis All statistical analyses were performed using IBM SPSS Statistics 28.0 (Armonk, NY, USA). Outliers, defined as values exceeding ± 3 standard deviations from the mean, were excluded for both independent and dependent variables28,29. Descriptive statistics, including means, standard deviations (SD), frequencies, and percentages, were calculated for all variables. Following previous research, analyses proceeded in two stages: first, intraindividual changes from baseline to follow-up were examined; second, potential predictors of these changes were explored. Paired-sample t-tests compared movement behaviors, physical fitness, and body composition between time points, while McNemar–Bowker tests assessed longitudinal changes in adherence patterns. Generalized linear mixed‑effects models with a random intercept for each participant were used to examine the associations between baseline adherence to the 24-HMG (both the number of guidelines met and the specific combinations) and one‑year changes in physical fitness and body composition. Change scores for each outcome were calculated as follow‑up value minus baseline value. Fixed effects included baseline adherence, the baseline value of the respective outcome (to adjust for regression to the mean and initial differences), and all predefined covariates (child age, sex, weight status, primary caregiver’s age, sex, educational attainment, household income, and parental support). The random intercept accounted for individual‑level variability, following previous longitudinal studies on 24‑hour movement guidelines25. A retrospective power analysis (α = 0.05, 1 − β = 0.08) indicated that the final sample size was sufficient to detect a small effect size (f2 = 0.02) in the regression models. All tests were two-tailed with significance set at P < 0.05. E-values were calculated for primary outcomes as part of sensitivity analyses to assess robustness to unmeasured confounding19,30. E-values greater than 1 suggest that only confounders with substantial associations with both the exposure and outcome could nullify the observed effects, thereby supporting the credibility of the findings. Results Descriptive characteristics of the study sample at baseline Baseline characteristics of the study sample, including children’s demographic variables (age, sex, and weight status), primary caregivers’ demographics (age, sex, educational level, and household income), 24 h movement behaviours, adherence to the 24 h movement guidelines (24-HMG), and indicators of physical fitness and body composition, are summarized in Table 1. The mean age of participants was 10.07 years (SD = 1.19), and 40.7% were girls. Most primary caregivers were female (73.9%), and 50.1% had an education level of college and above. Regarding guideline adherence, 33.5% of children met one of the three behavioural recommendations and 43.3% met two. In contrast, 12.8% of participants did not meet any guidelines, and only 10.4% complied with all three. Table 1Descriptive characteristics of study sample at baseline (n = 337).VariablesValueAge (Mean ± SD), years10.07 ± 1.19 Sex, n (%) Girls137 (40.7) Boys200 (59.3)Weigt status, n (%) Overweight96 (28.5) Obese241 (71.5)BMI (kg/m2) (Mean ± SD)21.64 ± 3.01Main caregiver’s age (Mean ± SD), years37.25 ± 3.34Main caregiver’s sex, n (%) Females249 (73.9) Males88 (26.1)Main caregiver’s educational level, n (%) Primary and below18 (5.3) High/middle school150 (44.5) College and above169 (50.1)Household income, n (%) Below average85 (25.2) Average150 (44.5) Above average102 (30.3)Parental support (Mean ± SD)3.21 ± 0.7424-hour movement behaviors (Mean ± SD) MVPA (min/day)49.83 ± 21.21 SST (hr/day)1.68 ± 0.83 SLP (hr/day)9.05 ± 0.74Adherence to 24-HMG The number of guideline adherence, n (%) None43 (12.8) One133 (33.5) Two146 (43.3) All three35 (10.4) Combinations of guideline adherence, n (%) None43 (12.8) Only MVPA26 (7.7) Only SST40 (11.9) Only SLP47 (13.9) MVPA + SST62 (18.4) MVPA + SLP72 (21.4) SST + SLP117 (34.7) MVPA + SST + SLP35 (10.4)Physical fitness (Mean ± SD)66.58 ± 11.18Body composition (Mean ± SD) BFP (%)31.23 ± 7.10 FFM (kg)29.41 ± 5.74 SMM (kg)15.53 ± 3.68BMI, body mass index; BFP, body fat percentage; FFM, fat-free mass; SMM, skeletal muscle mass,; MVPA, moderate-to-vigorous physical activity; SST, sedentary screen time; SLP, Sleep; 24-HMG, 24-hour movement guidelines. Table 1Descriptive characteristics of study sample at baseline (n = 337).VariablesValueAge (Mean ± SD), years10.07 ± 1.19 Sex, n (%) Girls137 (40.7) Boys200 (59.3)Weigt status, n (%) Overweight96 (28.5) Obese241 (71.5)BMI (kg/m2) (Mean ± SD)21.64 ± 3.01Main caregiver’s age (Mean ± SD), years37.25 ± 3.34Main caregiver’s sex, n (%) Females249 (73.9) Males88 (26.1)Main caregiver’s educational level, n (%) Primary and below18 (5.3) High/middle school150 (44.5) College and above169 (50.1)Household income, n (%) Below average85 (25.2) Average150 (44.5) Above average102 (30.3)Parental support (Mean ± SD)3.21 ± 0.7424-hour movement behaviors (Mean ± SD) MVPA (min/day)49.83 ± 21.21 SST (hr/day)1.68 ± 0.83 SLP (hr/day)9.05 ± 0.74Adherence to 24-HMG The number of guideline adherence, n (%) None43 (12.8) One133 (33.5) Two146 (43.3) All three35 (10.4) Combinations of guideline adherence, n (%) None43 (12.8) Only MVPA26 (7.7) Only SST40 (11.9) Only SLP47 (13.9) MVPA + SST62 (18.4) MVPA + SLP72 (21.4) SST + SLP117 (34.7) MVPA + SST + SLP35 (10.4)Physical fitness (Mean ± SD)66.58 ± 11.18Body composition (Mean ± SD) BFP (%)31.23 ± 7.10 FFM (kg)29.41 ± 5.74 SMM (kg)15.53 ± 3.68BMI, body mass index; BFP, body fat percentage; FFM, fat-free mass; SMM, skeletal muscle mass,; MVPA, moderate-to-vigorous physical activity; SST, sedentary screen time; SLP, Sleep; 24-HMG, 24-hour movement guidelines. Descriptive characteristics of study sample at baseline (n = 337). BMI, body mass index; BFP, body fat percentage; FFM, fat-free mass; SMM, skeletal muscle mass,; MVPA, moderate-to-vigorous physical activity; SST, sedentary screen time; SLP, Sleep; 24-HMG, 24-hour movement guidelines. Changes in studied variables between baseline and follow-up Over the one-year follow-up, significant changes were observed in most variables (all P < 0.05), except for MVPA (t336=1.64, P = 0.10) (Table 2). Sleep duration significantly declined (t336=2.18, P = 0.03), while SST significantly increased (t336 = − 6.79, P < 0.001). Increases were also observed in physical fitness (t336=12.79, P < 0.001), BFP (t336=12.79, P < 0.001), FFM (t336=43.89, P < 0.001), and SMM (t336=41.23, P < 0.001). The proportion of children adhering to all three movement guidelines decreased from 14.2 to 12.18%, whereas the proportion meeting none increased from 11.30% to 15.23%. Significant shifts were observed in the percentages of specific adherence combinations (χ²=41.77, P = 0.046). As shown in Fig. 1, the proportion of children meeting none of the guidelines increased from 12.8 to 16.0%, while the proportion meeting both MVPA and SST decreased from 8.0% to 3.6% and the proportion meeting both SST and sleep decreased from 24.3 to 18.1%. Conversely, the proportions of children meeting only SST (+ 8%) or only sleep (+ 11%) increased slightly. Changes in the total number of guidelines met were not significant (χ²=11.26, P = 0.080). Table 2Changes in 24-HMB, guideline adherence, physical fitness, and body composition between baseline and follow-up (n = 337).VariablesBaselineFollow-upPaired differencesMean (SD)/n (%)Mean (SD)/n (%)Mean/χ2SD95%CI P 24-h movement behaviors MVPA (min/day)49.83 (21.21)48.05 (19.31)1.7819.89− 0.35,3.910.100 SST (hr/day)1.68 (0.83)2.03 (0.90)0.110.91− 0.45,-0.25< 0.001 SLP (hr/day)9.05 (0.74)8.94 (0.83)-0.350.940.01,0.200.030Adherence to 24-HMG The number of guideline adherence None43 (12.80)54 (16.00)11.26NA0.080 Meeting one guideline133 (33.50)131 (38.90) Meeting two guidelines146 (43.30)113 (33.50) Meeting all three guidelines35 (10.40)39 (11.60) Combinations of guideline adherence None43 (12.80)54 (16.00)41.77NA0.046 Only MVPA26 (7.70)25 (7.40) Only SST40 (11.90)48 (14.20) Only SLP47 (13.90)58 (17.20) MVPA + SST27 (8.00)12 (3.60) MVPA + SLP37 (11.00)40 (11.90) SST + SLP82 (24.30)61 (18.10) MVPA + SST + SLP35 (10.40)39 (11.60)Physical fitness66.58 ± 11.1868.78 (11.02)− 2.209.68− 3.24,1.16< 0.001Body composition BFP (%)31.22 (7.10)31.68 (7.45)− 0.453.96− 0.88~− 0.030.036 FFM (kg)29.41 (5.74)31.71 (5.88)− 2.311.41− 2.46~− 2.15< 0.001 SMM (kg)15.53 (3.68)17.22 (4.01)− 1.701.17− 1.82~− 1.57< 0.001BFP, body fat percentage; FFM, fat-free mass; SMM, skeletal muscle mass; MVPA, moderate-to-vigorous physical activity; SST, sedentary screen time; SLP, sleep duration; CI, confidence interval; 24-HMG, 24-hour movement guidelines. Table 2Changes in 24-HMB, guideline adherence, physical fitness, and body composition between baseline and follow-up (n = 337).VariablesBaselineFollow-upPaired differencesMean (SD)/n (%)Mean (SD)/n (%)Mean/χ2SD95%CI P 24-h movement behaviors MVPA (min/day)49.83 (21.21)48.05 (19.31)1.7819.89− 0.35,3.910.100 SST (hr/day)1.68 (0.83)2.03 (0.90)0.110.91− 0.45,-0.25< 0.001 SLP (hr/day)9.05 (0.74)8.94 (0.83)-0.350.940.01,0.200.030Adherence to 24-HMG The number of guideline adherence None43 (12.80)54 (16.00)11.26NA0.080 Meeting one guideline133 (33.50)131 (38.90) Meeting two guidelines146 (43.30)113 (33.50) Meeting all three guidelines35 (10.40)39 (11.60) Combinations of guideline adherence None43 (12.80)54 (16.00)41.77NA0.046 Only MVPA26 (7.70)25 (7.40) Only SST40 (11.90)48 (14.20) Only SLP47 (13.90)58 (17.20) MVPA + SST27 (8.00)12 (3.60) MVPA + SLP37 (11.00)40 (11.90) SST + SLP82 (24.30)61 (18.10) MVPA + SST + SLP35 (10.40)39 (11.60)Physical fitness66.58 ± 11.1868.78 (11.02)− 2.209.68− 3.24,1.16< 0.001Body composition BFP (%)31.22 (7.10)31.68 (7.45)− 0.453.96− 0.88~− 0.030.036 FFM (kg)29.41 (5.74)31.71 (5.88)− 2.311.41− 2.46~− 2.15< 0.001 SMM (kg)15.53 (3.68)17.22 (4.01)− 1.701.17− 1.82~− 1.57< 0.001BFP, body fat percentage; FFM, fat-free mass; SMM, skeletal muscle mass; MVPA, moderate-to-vigorous physical activity; SST, sedentary screen time; SLP, sleep duration; CI, confidence interval; 24-HMG, 24-hour movement guidelines. Changes in 24-HMB, guideline adherence, physical fitness, and body composition between baseline and follow-up (n = 337). BFP, body fat percentage; FFM, fat-free mass; SMM, skeletal muscle mass; MVPA, moderate-to-vigorous physical activity; SST, sedentary screen time; SLP, sleep duration; CI, confidence interval; 24-HMG, 24-hour movement guidelines. Fig. 1Adherence to the 24-HMG over time. Key shifts: meeting none ↑, MVPA ~stable, SST ↑, sleep ↓, with redistribution across combinations (e.g., MVPA + SST and SST + SLP decreased, only SST and only SLP increased). Adherence to the 24-HMG over time. Key shifts: meeting none ↑, MVPA ~stable, SST ↑, sleep ↓, with redistribution across combinations (e.g., MVPA + SST and SST + SLP decreased, only SST and only SLP increased). Associations between meeting 24-HMG and changes in physical fitness and body composition over 12 months Table 3 presents the associations between baseline adherence to the 24-HMG and changes in physical fitness and body composition. After adjusting for covariates, meeting one, two or three guidelines were positively associated with improvements in physical fitness (β = 3.09 to 5.79, all P < 0.05). For body composition indicators, meeting two guidelines was associated with significant reductions in BFP (β=−1.40, P = 0.039), while meeting one or three guidelines showed no significant association with changes in BFP (all P > 0.05). Improved FFM was significantly associated with meeting two or three guidelines (β = 0.63 to 0.97, all P < 0.05), whereas adherence to a single guideline was not significant (β = 0.50, P = 0.052). Meeting two guidelines was positively associated with an increase in SMM (β = 0.47, P = 0.020), with no significant associations for meeting one or three guidelines (all P > 0.05). Table 3Associations between the number of guideline adherence at baseline and changes in physical fitness and body composition over one year (n = 337).PredictorsPhysical fitnessBFPFFMSMMβ (95%CI) P β (95%CI) P β (95%CI) P β (95%CI) P Intercept38.08 (25.34,50.83)< 0.0019.03 (3.42,14.65)0.0020.47 (− 1.50,2.44)0.637− 1.11 (− 2.68,0.46)0.166CovariatesAge− 0.59 (− 1.53,0.34)0.2120.09 (− 0.33,0.52)0.6720.04 (− 0.13,0.20)0.669− 0.10 (− 0.23,0.04)0.166Sex (Ref: boys) Girls1.44 (− 0.45,3.34)0.134− 0.95 (− 1.79,-0.12)0.026− 0.15 (− 0.46,0.16)0.3470.02 (− 0.23,0.27)0.896Weight status (Ref: overweight) Obese− 3.35 (− 5.43,− 1.27)0.0021.54 (0.34,2.74)0.012− 0.02 (− 0.39,0.35)0.919− 0.11 (− 0.41,0.20)0.492Main caregiver’s age− 0.15 (− 0.47,0.18)0.368− 0.08 (− 0.23,0.06)0.2670.03 (− 0.03,0.08)0.3030.07 (0.02,0.11)0.003Main caregiver’s sex (Ref: male) Female1.78 (− 0.30,3.86)0.093− 0.51 (− 1.46,0.44)0.294− 0.05 (− 0.41,0.30)0.7720.17 (− 0.11,0.46)0.237Main caregiver’s education level (Ref: primary school or below) College or above− 0.04 (− 4.20,4.12)0.985− 0.42 (− 2.32,1.49)0.6670.02 (− 0.68,0.73)0.9470.15 (− 0.42,0.72)0.600 High/middle school1.02 (− 3.12,5.16)0.6270.34 (− 1.55,2.23)0.724− 0.08 (− 0.78,0.62)0.816− 0.10 (− 0.66,0.47)0.735Household income (Ref: low) Average1.96 (− 0.37,4.28)0.098− 0.93 (− 1.99,0.14)0.0870.15 (− 0.25,0.54)0.471− 0.04 (− 0.36,0.28)0.801 Above average2.09 (− 0.42,4.59)0.103− 0.25 (− 1.40,0.89)0.6650.44 (0.01,0.86)0.0430.09 (− 0.25,0.44)0.600Dependent variable at baseline− 0.45 (− 0.53,− 0.36)< 0.001− 0.18 (− 0.25,− 0.10)< 0.001− 0.01 (− 0.04,0.02)0.6600.06 (0.02,0.09)0.006The number of guideline adherence (Ref: none) Meeting one3.09 (0.09,6.10)0.0430.14 (− 1.23,1.50)0.8430.50 (− 0.01,1.01)0.0520.28 (− 0.13,0.69)0.185 Meeting two5.13 (2.22,8.04)< 0.001− 1.40 (− 2.72,− 0.07)0.0390.63 (0.14,1.13)0.0120.47 (0.08,0.87)0.020 Meeting all three5.79 (2.02,9.56)0.003− 1.33 (− 3.04,0.39)0.1290.97 (0.33,1.61)0.0030.25 (− 0.26,0.77)0.337BFP, body fat percentage (%); FFM, fat-free mass (kg); SMM, skeletal muscle mass (kg); β, unstandardized standardized coefficient; CI, confidence interval; Ref, reference group. Change scores were calculated as follow-up value minus baseline value. Table 3Associations between the number of guideline adherence at baseline and changes in physical fitness and body composition over one year (n = 337).PredictorsPhysical fitnessBFPFFMSMMβ (95%CI) P β (95%CI) P β (95%CI) P β (95%CI) P Intercept38.08 (25.34,50.83)< 0.0019.03 (3.42,14.65)0.0020.47 (− 1.50,2.44)0.637− 1.11 (− 2.68,0.46)0.166CovariatesAge− 0.59 (− 1.53,0.34)0.2120.09 (− 0.33,0.52)0.6720.04 (− 0.13,0.20)0.669− 0.10 (− 0.23,0.04)0.166Sex (Ref: boys) Girls1.44 (− 0.45,3.34)0.134− 0.95 (− 1.79,-0.12)0.026− 0.15 (− 0.46,0.16)0.3470.02 (− 0.23,0.27)0.896Weight status (Ref: overweight) Obese− 3.35 (− 5.43,− 1.27)0.0021.54 (0.34,2.74)0.012− 0.02 (− 0.39,0.35)0.919− 0.11 (− 0.41,0.20)0.492Main caregiver’s age− 0.15 (− 0.47,0.18)0.368− 0.08 (− 0.23,0.06)0.2670.03 (− 0.03,0.08)0.3030.07 (0.02,0.11)0.003Main caregiver’s sex (Ref: male) Female1.78 (− 0.30,3.86)0.093− 0.51 (− 1.46,0.44)0.294− 0.05 (− 0.41,0.30)0.7720.17 (− 0.11,0.46)0.237Main caregiver’s education level (Ref: primary school or below) College or above− 0.04 (− 4.20,4.12)0.985− 0.42 (− 2.32,1.49)0.6670.02 (− 0.68,0.73)0.9470.15 (− 0.42,0.72)0.600 High/middle school1.02 (− 3.12,5.16)0.6270.34 (− 1.55,2.23)0.724− 0.08 (− 0.78,0.62)0.816− 0.10 (− 0.66,0.47)0.735Household income (Ref: low) Average1.96 (− 0.37,4.28)0.098− 0.93 (− 1.99,0.14)0.0870.15 (− 0.25,0.54)0.471− 0.04 (− 0.36,0.28)0.801 Above average2.09 (− 0.42,4.59)0.103− 0.25 (− 1.40,0.89)0.6650.44 (0.01,0.86)0.0430.09 (− 0.25,0.44)0.600Dependent variable at baseline− 0.45 (− 0.53,− 0.36)< 0.001− 0.18 (− 0.25,− 0.10)< 0.001− 0.01 (− 0.04,0.02)0.6600.06 (0.02,0.09)0.006The number of guideline adherence (Ref: none) Meeting one3.09 (0.09,6.10)0.0430.14 (− 1.23,1.50)0.8430.50 (− 0.01,1.01)0.0520.28 (− 0.13,0.69)0.185 Meeting two5.13 (2.22,8.04)< 0.001− 1.40 (− 2.72,− 0.07)0.0390.63 (0.14,1.13)0.0120.47 (0.08,0.87)0.020 Meeting all three5.79 (2.02,9.56)0.003− 1.33 (− 3.04,0.39)0.1290.97 (0.33,1.61)0.0030.25 (− 0.26,0.77)0.337BFP, body fat percentage (%); FFM, fat-free mass (kg); SMM, skeletal muscle mass (kg); β, unstandardized standardized coefficient; CI, confidence interval; Ref, reference group. Change scores were calculated as follow-up value minus baseline value. Associations between the number of guideline adherence at baseline and changes in physical fitness and body composition over one year (n = 337). BFP, body fat percentage (%); FFM, fat-free mass (kg); SMM, skeletal muscle mass (kg); β, unstandardized standardized coefficient; CI, confidence interval; Ref, reference group. Change scores were calculated as follow-up value minus baseline value. Table 4 presents the associations between specific guideline adherence combinations and changes in health outcomes. Meeting the MVPA guideline alone was significantly associated with improvements in physical fitness (β = 6.08, P = 0.004) and FFM (β = 0.93, P = 0.007), as well as reductions in BFP (β=−1.83, P = 0.049). Meeting only the SST guideline was associated with a significant increase in BFP (β = 1.71, P = 0.039) yet showed no significant associations with other health outcomes (all P > 0.05). Meeting only the sleep guideline was associated with increased FFM (β = 0.80, P = 0.007) but showed no significant associations with other health outcomes (all P > 0.05). Meeting both the MVPA and SST guidelines was associated with increases in physical fitness (β = 4.81, P = 0.021), FFM (β = 1.19, P < 0.001), and SMM (β = 0.92, P = 0.001), as well as decreases in BFP (β=−2.03, P = 0.028) over the one-year period. Similarly, adherence to the combination of the MVPA and sleep guidelines was associated with favorable changes in physical fitness (β = 6.57, P < 0.001), BFP (β=−2.77, P = 0.001), and FMM (β = 1.02, P = 0.001), but showed no significant association with SMM (β = 0.37, P = 0.151). Adherence to both the SST and sleep guidelines, or to all three guidelines, was associated only with improvements in physical fitness (β = 4.42 to 5.72, P < 0.01). Table 4Associations between the combinations of guideline adherence at baseline and changes in physical fitness and body composition over one year (n = 337).PredictorsPFBFPFFMSMMβ (95%CI) P β (95%CI) P β (95%CI) P β (95%CI) P Intercept39.72 (26.98,52.45)< 0.0018.21 (2.73,13.69)0.0031.01 (− 0.91,2.93)0.299− 1.01 (− 2.58,0.57)0.210CovariatesAge− 0.63 (− 1.57,0.30)0.1820.14 (− 0.28,0.55)0.5170.02 (− 0.14,0.18)0.825− 0.10 (− 0.23,0.04)0.156Sex (Ref: boys) Girls1.87 (− 0.04,3.77)0.055− 1.31 (− 2.13,− 0.48)0.002− 0.02 (− 0.33,0.28)0.8770.04 (− 0.22,0.29)0.774Weight status (Ref: overweight) Obese− 3.42 (− 5.51,− 1.32)0.0021.61 (0.43,2.79)0.0070.02 (− 0.35,0.39)0.909− 0.11 (− 0.41,0.20)0.500Main caregiver’s age− 0.19 (− 0.51,0.14)0.255− 0.05 (− 0.20,0.09)0.4610.02 (− 0.03,0.07)0.4600.07 (0.02,0.11)0.003Main caregiver’s sex (Ref: male) Female2.12 (0.04,4.20)0.046− 0.72 (− 1.65,0.20)0.1260.01 (− 0.33,0.36)0.9520.17 (− 0.11,0.46)0.235Main caregiver’s education level (Ref: primary school or below) College or above0.30 (− 3.86,4.45)0.889− 0.60 (− 2.45,1.26)0.5260.03 (− 0.65,0.71)0.9290.13 (− 0.43,0.70)0.647 High/middle school1.27 (− 2.87,5.39)0.5470.18 (− 1.66,2.02)0.849− 0.09 (− 0.77,0.58)0.786− 0.13 (− 0.69,0.43)0.649Household income (Ref: low) Average1.99 (− 0.53,4.51)0.122− 0.75 (− 1.78,0.29)0.1580.08 (− 0.31,0.46)0.796− 0.06 (− 0.38,0.26)0.706 Above average1.82 (− 0.50,4.14)0.124− 0.09 (− 1.21,1.03)0.8720.33 (− 0.09,0.74)0.1190.03 (− 0.31,0.38)0.848Dependent variable at baseline− 0.45 (− 0.54,− 0.37)< 0.001− 0.19 (− 0.27,− 0.12)< 0.001− 0.01 (− 0.04,0.02)0.5320.06 (0.02,0.09)0.006Combinations of guideline adherence (Ref: none) Only MVPA6.08 (1.97,10.18)0.004− 1.83 (− 3.66,− 0.01)0.0490.93 (0.26,1.61)0.0070.27 (− 0.29,0.83)0.346 Only SST0.73 (− 2.94,4.40)0.6971.71 (0.08,3.34)0.039− 0.17 (− 0.77,0.43)0.5810.07 (− 0.43,0.57)0.775 Only SLP3.32 (− 0.18,6.82)0.063− 0.02 (− 1.57,1.53)0.9770.80 (0.22,1.37)0.0070.44 (− 0.03,0.92)0.069 MVPA + SST4.81 (0.74,8.88)0.021− 2.03 (− 3.84,-0.22)0.0281.19 (0.52,1.85)< 0.0010.92 (0.36,1.48)0.001 MVPA + SLP6.57 (2.86,10.28)< 0.001− 2.77 (− 4.41,− 1.12)0.0011.02 (0.41,1.63)0.0010.37 (− 0.14,0.88)0.151 SST + SLP4.42 (1.27,7.57)0.006− 0.48 (− 1.88,0.92)0.4970.24 (− 0.28,0.76)0.3560.37 (− 0.06,0.80)0.089 MVPA + SST+ SLP5.72 (1.97,9.46)0.003− 1.29 (− 2.95,0.38)0.1300.94 (0.33,1.56)0.0030.25 (− 0.26,0.76)0.338BFP, body fat percentage (%); FFM, fat-free mass (kg); SMM, skeletal muscle mass (kg); β , unstandardized standardized coefficient; CI, confidence interval; Ref, reference group; MVPA, moderate-to-vigorous physical activity; SST, sedentary screen time; SLP, sleep duration. Change scores were calculated as follow-up value minus baseline value. Table 4Associations between the combinations of guideline adherence at baseline and changes in physical fitness and body composition over one year (n = 337).PredictorsPFBFPFFMSMMβ (95%CI) P β (95%CI) P β (95%CI) P β (95%CI) P Intercept39.72 (26.98,52.45)< 0.0018.21 (2.73,13.69)0.0031.01 (− 0.91,2.93)0.299− 1.01 (− 2.58,0.57)0.210CovariatesAge− 0.63 (− 1.57,0.30)0.1820.14 (− 0.28,0.55)0.5170.02 (− 0.14,0.18)0.825− 0.10 (− 0.23,0.04)0.156Sex (Ref: boys) Girls1.87 (− 0.04,3.77)0.055− 1.31 (− 2.13,− 0.48)0.002− 0.02 (− 0.33,0.28)0.8770.04 (− 0.22,0.29)0.774Weight status (Ref: overweight) Obese− 3.42 (− 5.51,− 1.32)0.0021.61 (0.43,2.79)0.0070.02 (− 0.35,0.39)0.909− 0.11 (− 0.41,0.20)0.500Main caregiver’s age− 0.19 (− 0.51,0.14)0.255− 0.05 (− 0.20,0.09)0.4610.02 (− 0.03,0.07)0.4600.07 (0.02,0.11)0.003Main caregiver’s sex (Ref: male) Female2.12 (0.04,4.20)0.046− 0.72 (− 1.65,0.20)0.1260.01 (− 0.33,0.36)0.9520.17 (− 0.11,0.46)0.235Main caregiver’s education level (Ref: primary school or below) College or above0.30 (− 3.86,4.45)0.889− 0.60 (− 2.45,1.26)0.5260.03 (− 0.65,0.71)0.9290.13 (− 0.43,0.70)0.647 High/middle school1.27 (− 2.87,5.39)0.5470.18 (− 1.66,2.02)0.849− 0.09 (− 0.77,0.58)0.786− 0.13 (− 0.69,0.43)0.649Household income (Ref: low) Average1.99 (− 0.53,4.51)0.122− 0.75 (− 1.78,0.29)0.1580.08 (− 0.31,0.46)0.796− 0.06 (− 0.38,0.26)0.706 Above average1.82 (− 0.50,4.14)0.124− 0.09 (− 1.21,1.03)0.8720.33 (− 0.09,0.74)0.1190.03 (− 0.31,0.38)0.848Dependent variable at baseline− 0.45 (− 0.54,− 0.37)< 0.001− 0.19 (− 0.27,− 0.12)< 0.001− 0.01 (− 0.04,0.02)0.5320.06 (0.02,0.09)0.006Combinations of guideline adherence (Ref: none) Only MVPA6.08 (1.97,10.18)0.004− 1.83 (− 3.66,− 0.01)0.0490.93 (0.26,1.61)0.0070.27 (− 0.29,0.83)0.346 Only SST0.73 (− 2.94,4.40)0.6971.71 (0.08,3.34)0.039− 0.17 (− 0.77,0.43)0.5810.07 (− 0.43,0.57)0.775 Only SLP3.32 (− 0.18,6.82)0.063− 0.02 (− 1.57,1.53)0.9770.80 (0.22,1.37)0.0070.44 (− 0.03,0.92)0.069 MVPA + SST4.81 (0.74,8.88)0.021− 2.03 (− 3.84,-0.22)0.0281.19 (0.52,1.85)< 0.0010.92 (0.36,1.48)0.001 MVPA + SLP6.57 (2.86,10.28)< 0.001− 2.77 (− 4.41,− 1.12)0.0011.02 (0.41,1.63)0.0010.37 (− 0.14,0.88)0.151 SST + SLP4.42 (1.27,7.57)0.006− 0.48 (− 1.88,0.92)0.4970.24 (− 0.28,0.76)0.3560.37 (− 0.06,0.80)0.089 MVPA + SST+ SLP5.72 (1.97,9.46)0.003− 1.29 (− 2.95,0.38)0.1300.94 (0.33,1.56)0.0030.25 (− 0.26,0.76)0.338BFP, body fat percentage (%); FFM, fat-free mass (kg); SMM, skeletal muscle mass (kg); β , unstandardized standardized coefficient; CI, confidence interval; Ref, reference group; MVPA, moderate-to-vigorous physical activity; SST, sedentary screen time; SLP, sleep duration. Change scores were calculated as follow-up value minus baseline value. Associations between the combinations of guideline adherence at baseline and changes in physical fitness and body composition over one year (n = 337). BFP, body fat percentage (%); FFM, fat-free mass (kg); SMM, skeletal muscle mass (kg); β , unstandardized standardized coefficient; CI, confidence interval; Ref, reference group; MVPA, moderate-to-vigorous physical activity; SST, sedentary screen time; SLP, sleep duration. Change scores were calculated as follow-up value minus baseline value. Results of sensitivity analysis Sensitivity analyses excluding 5% of participants at random (n = 17) showed that all associations remained robust and consistent with the primary findings, with two exceptions: the association between adherence to the SST guideline alone and BFP (β = 1.71, 95% CI = − 0.01 to 3.41) was no longer statistically significant, and the association between adherence to the sleep guideline alone and physical fitness became statistically significant (β = 3.49, 95% CI = 0.02 to 6.95) (Appendix 2, Tables S1–S2). E-values indicated that the observed associations between baseline 24-HMG adherence, both overall and by specific combinations, and subsequent changes in physical fitness and body composition indicators are unlikely to be fully explained by unmeasured confounding, as all E-values exceeded 1 (Appendix 2, Table S3). Discussion Primary findings This longitudinal cohort study examined the associations between baseline adherence to the 24-HMG and one-year changes in physical fitness and body composition indicators among overweight and obese children in China. The findings partially support the initial hypotheses. Specifically, adherence to one, two, or three guidelines was associated with improvements in physical fitness, while adherence to two guidelines was significantly associated with all body composition indicators. Adherence to all three guidelines was associated with improvements in FFM but not BFP or SMM. In addition, meeting the MVPA guideline alone was associated with all health outcomes except SMM. Particularly, combinations of guideline adherence, especially those including MVPA, were significantly associated with improvements in physical fitness, BFP, FFM, and SMM. These findings extend existing research by suggesting that combinations of structured behaviors, rather than full adherence to all three guidelines, may be sufficient to yield significant health improvements in this population. The low prevalence of full adherence to the 24-HMG observed here (10.4%–11.6%) is consistent with previous evidence in children and adolescents, which ranges from 1.7% to 26.1%14,17,26,31. The decline in adherence over the follow-up aligns with established trends of decreasing MVPA, increasing SST, and shortening sleep duration during childhood and adolescence22,32,33. In the present study, MVPA alone, or in combination with SST or sleep, was associated with improvements in physical fitness and FFM, as well as reductions in BFP, aligning with previous evidence that MVPA plays a central role in adiposity control and the improvement of metabolic and functional health markers34,35. These associations may be explained by both physiological and behavioral mechanisms. MVPA enhances body composition by increasing total energy expenditure, promoting fat oxidation, and stimulating muscle hypertrophy36,37. However, the MVPA–adiposity association may be modified by dietary behaviors and psychosocial factors38,39. Conversely, prolonged sedentary behavior is linked to lower energy expenditure and increased fat storage40. The finding that the combination of MVPA and SST adherence was associated with improvements in physical fitness and all body composition indicators suggests that MVPA’s anabolic effects may offset the detrimental impact of sedentary time on the physical health of overweight and obese children, potentially by enhancing muscle mass, improving fat metabolism, and promoting overall metabolic health. Similarly, the synergy between MVPA and adequate sleep in enhancing physical fitness, FFM, and reducing BFP may be attributable to sleep’s role in hormonal regulation of appetite and metabolism, complementing the metabolic benefits of MVPA41,42. No significant associations were found between SMM changes and other types of guideline adherence, in line with previous studies suggesting SMM’s limited sensitivity to short-term behavioral changes. This may be because skeletal muscle mass typically requires longer, targeted interventions like resistance training. In our study, MVPA was assessed as time spent in moderate-to-vigorous activity, while the type of MVPA, such as strength training, may be more critical for muscle growth. Without focusing on these types, MVPA time alone may not provide enough stimulus for noticeable muscle development, especially in children during rapid growth36,42. Additionally, individual variations and the influence of diet and environment could also contribute to the lack of significant association37,43. The absence of additional benefit from full adherence in BFP and SMM may reflect the small proportion of children achieving this level of compliance, as well as the difficulty of sustaining multiple behavioral changes simultaneously in a high-risk group facing psychosocial and environmental challenges. Additionally, other determinants such as dietary behaviors and nutrition intake may also play a critical role26,44,45. E‑value analyses further supported the robustness of the key findings. For the main significant associations (e.g., those involving MVPA), the E‑values ranged from 2.41 to 3.71 (see Appendix 2, Table S3), indicating that an unmeasured confounder would need to be associated with both the exposure and the outcome by risk ratios of at least 2.4–3.7 to fully explain away these effects. Such effect sizes are substantially larger than those of the measured covariates and are implausible in the context of this study. In contrast, associations that were not robust (e.g., SST‑only with BFP) had E‑values close to 1 (e.g., 1.07–1.35). Therefore, the main significant findings are unlikely to be explained by residual confounding, whereas the borderline findings should be interpreted with caution. Results of sensitivity analyses (excluding 5% of participants) largely supported these main conclusions, with two exceptions. First, the association between meeting the SST guideline alone and BFP (P = 0.039 in the main analysis) became borderline significant (P = 0.050), indicating instability. This counterintuitive finding may be explained by behavioral substitution (e.g., increased snacking or other sedentary activities after reducing screen time), measurement error, or residual confounding, and should be interpreted with caution. Second, the association between meeting the sleep guideline alone and SMM was also not robust. Apart from these two exceptions, the sensitivity analyses were consistent with the main findings. Strength and limits This study offers several strengths. Its prospective cohort design and the use of triaxial accelerometers for objective assessment of MVPA and sleep reduce measurement bias. The focus on overweight and obese children, a population underrepresented in movement behavior research, provides novel insights into the behavioral determinants of health in this high-risk group. The Chinese setting also adds cultural diversity to a literature base largely dominated by Western data. Nonetheless, limitations should be acknowledged. First, SST was parent-reported and did not distinguish between active and passive screen activities, which may lead to misclassification bias. Second, residual confounding is possible, as factors such as dietary intake, family dynamics, and school-level factors were not measured. Third, the study sample was restricted to school-aged children in China, limiting generalizability to other regions and cultural contexts. Additionally, some specific guideline adherence combinations (e.g., meeting only MVPA and SST) had relatively small sample sizes at baseline (n = 27), which may reduce the precision of estimates for those subgroups; results for these rare combinations should be interpreted with caution. Similarly, the proportion of children meeting all three guidelines was modest (10.4%, n = 35), which may have limited statistical power to detect significant associations with BFP and SMM. Moreover, although our composite physical fitness score was standardized for age and sex to ensure comparability, we did not perform age-stratified analyses. Future studies with larger samples should examine whether the associations between 24-HMG adherence and physical fitness vary across age groups. Furthermore, in our study, adherence patterns shifted over the follow-up period, reflecting the dynamic nature of movement behaviors in childhood. Although exposure instability may raise questions about using baseline adherence as a predictor, baseline patterns capture children’s initial behavioral status and remain informative for predicting subsequent health changes46,47. The robustness of our findings is supported by sensitivity analyses. Finally, the one-year follow-up period may be insufficient to capture the long-term effects of guideline adherence; extended follow-up is warranted to evaluate sustained health impacts. From a practical perspective, these results suggest that intervention strategies should prioritize MVPA, either alone or in combination with reduced SST or adequate sleep, rather than focusing solely on comprehensive adherence to all three 24-HMG components. For overweight and obese children, promoting achievable and high-impact behavior combinations may be a more realistic and sustainable approach. Implementation efforts should integrate school-based physical education policies with family-level strategies to reduce screen exposure and support healthy sleep, thereby creating a supportive environment for lasting improvements in movement behaviors and overall health. Conclusion This longitudinal study demonstrates that partial adherence to the 24-HMG, particularly combinations involving MVPA, is associated with favorable changes in physical fitness and body composition among overweight and obese children. Specifically, adherence to two guidelines, especially those incorporating MVPA, was most consistently linked to improvements in health outcomes. Full adherence to all three guidelines was associated with improvements in physical fitness and FFM, but not with changes in BFP or SMM. These findings suggest that prioritizing achievable behavior combinations, particularly those involving MVPA, may be more effective and sustainable than striving for full adherence to all three guidelines. Future research should explore the long-term effects of such behaviors, incorporating dietary and environmental factors, and evaluate the effectiveness of tailored interventions across more diverse populations. Long-term studies with extended follow-up periods are needed to determine whether these short-term improvements are sustained, and further investigation into the underlying mechanisms of behavior-health interactions is essential. Electronic Supplementary Material Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2