Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Feb 11;31(3):208–216. doi: 10.1111/camh.70074

Associations of Fitbit measured physical activity and sedentary behavior with mental health in U.S. youth: a quantile regression analysis

Li Niu 1,2,#, Jiayi Ji 3,#, Donglan Zhang 4, Lihua Li 3, Mi Xiang 5, Yan Li 3,5,✉
PMCID: PMC13474284  PMID: 41670396

Abstract

Background

Physical activity and sedentary behavior are associated with adolescent mental health. However, prior studies have not assessed whether these associations differ across varying levels of mental health severity. This study uses objectively measured physical activity and sedentary behavior to examine their associations with adolescent mental health and to determine how these associations vary across the distribution of mental health symptoms.

Methods

This longitudinal prospective study examined data from the Adolescent Brain Cognitive Development (ABCD) Study, a large cohort of children and adolescents recruited from 21 study sites in the United States. This analysis included 5640 participants whose physical activity and sedentary behavior data were recorded using Fitbit wearable devices when aged 11–12 years. Outcomes were internalizing and externalizing symptoms assessed a year later by the caregiver using the Child Behavior Checklist.

Results

More daily steps (b = −0.43, 95% CI [−0.58, −0.29]), longer moderate‐to‐vigorous physical activity (MVPA) (b = −0.34, 95% CI [−0.47, −0.22]) and less sedentary time (b = 0.32, 95% CI [0.15, 0.48]) were each associated with fewer internalizing symptoms a year later. The associations were stronger for youth at higher internalizing symptom quantiles. Limited or no effects were found for externalizing symptoms. Sex differences were also observed.

Conclusion

Objectively measured daily steps, MVPA, and sedentary behavior are prospectively associated with adolescent internalizing symptoms. The associations differ substantially among adolescents with varying levels of symptom severity and between sexes. Findings may inform targeted physical activity interventions to enhance adolescent mental health.

Keywords: Physical activity, sedentary time, adolescent, mental health

Key Practitioner Message.

What is currently known?

  • Physical activity and sedentary behavior are linked to adolescent mental health, but prior studies examined relationships between physical activity and psychopathology in general, without examining how this relationship may differ across different levels of severity of psychopathology, potentially underestimating effects among youth with greater mental health needs and limiting clinical guidance on who may benefit most

What has been shown?

  • In this cohort study including 5640 adolescents, more daily steps, longer moderate‐to‐vigorous physical activity, and less sedentary time were each associated with fewer internalizing symptoms a year later

  • These associations were stronger for youth who experienced greater internalizing symptoms

What is the significance of this for clinical practice?

  • Clinicians should consider incorporating physical activity recommendations into mental health interventions for youth with internalizing disorders, focusing on activities that increase physical activity and reduce sedentary time

  • Our findings further suggest that interventions targeting adolescents with higher levels of internalizing symptoms may be especially effective

Introduction

Adolescent mental health has become a public health crisis globally in recent years (Gunnell, Kidger, & Elvidge, 2018; Lu & Keyes, 2023). According to a recent meta‐analysis, the global prevalence of depression and anxiety symptoms has doubled compared to the rates observed before the COVID‐19 pandemic, reaching 25.2% and 20.5%, respectively, with higher prevalence among girls than boys (Racine et al., 2021). Internalizing problems, such as depression and anxiety, and externalizing problems, such as attention problems and conduct disorders, are among the leading causes of illness and disability among adolescents (Erskine et al., 2015). Given the rising occurrence and debilitating effects of mental health problems in adolescents, it is crucial to identify modifiable factors that could be targeted with policy and intervention strategies, and determine who would most benefit from such interventions.

Adolescence is a critical period for establishing healthy behaviors important for mental wellbeing. However, during the period of adolescence, physical activity often decreases while sedentary time increases due to more frequent use of digital devices (e.g., mobile phones) and more time devoted to education and self‐learning (Armstrong et al., 2018; Felez‐Nobrega, Raine, Haro, Wijndaele, & Koyanagi, 2020; van Sluijs et al., 2021; Xiang et al., 2020). It was estimated that 81% of adolescents aged 11 to 17 years did not meet the level of physical activity recommended by the World Health Organization (Guthold, Stevens, Riley, & Bull, 2020), which specified that adolescents should have at least 60 min of moderate‐to‐vigorous intensity physical activity (MVPA) like cycling or jogging every day (World Health Organization, 2020).

There is an abundance of evidence on the association between more physical activity and better mental health in children and adolescents (Biddle, Ciaccioni, Thomas, & Vergeer, 2019). Several previous studies—including those using the Adolescent Brain Cognitive Development (ABCD) Study sample (e.g., Damme, Vargas, Walther, Shankman, & Mittal, 2024; Sampasa‐Kanyinga et al., 2021)—have demonstrated that higher levels of MVPA are associated with lower risks of internalizing and externalizing problems in youth. Even engaging in light activity or walking, or simply reducing sedentary time, has been found to be associated with decreased risks of mental health problems (Kandola, Lewis, Osborn, Stubbs, & Hayes, 2020; Liu, Wu, & Yao, 2016; Rodriguez‐Ayllon et al., 2019; Toseeb et al., 2014). However, most previous studies examined average effects across the sample and typically modeled outcomes as normally distributed. This can obscure important variations in associations, particularly at the extremes of the mental health spectrum, where needs are greatest (Lê Cook & Manning, 2013). As a result, such studies may underestimate the benefits of physical activity for youth with elevated symptoms and fail to identify subgroups most likely to benefit from the intervention. By contrast, quantile regression allows for the estimation of effects across the full distribution of mental health outcomes, offering clinically relevant insights into which youth may benefit most. Another limitation is that previous studies often rely on self‐reported physical activity, which is prone to recall and social desirability bias (Adamo, Prince, Tricco, Connor‐Gorber, & Tremblay, 2009), often leading to overestimation of physical activity and underestimation of sedentary time (Matthews, Moore, George, Sampson, & Bowles, 2012). These biases may be larger among youth with mental health challenges; for example, adolescents at clinical high risk for psychosis showed poorer concordance between self‐reported and objective activity levels compared to their healthy peers (Damme et al., 2021), suggesting symptom‐related distortions. These concerns highlight the importance of using objective activity measures when studying mental health across varying symptom levels.

In this study, we assessed the prospective associations of objectively measured physical activity and sedentary behavior with subsequent internalizing and externalizing symptoms over a one‐year period. We used quantile regressions to assess how these associations differed across varying levels of mental health severity. We also conducted stratified analyses to examine associations of physical activity and sedentary behavior with mental health problems separately for boys and girls.

Methods

Data source and study sample

We used longitudinal data from the baseline, two‐year follow‐up (Year 2) and three‐year follow‐up (Year 3) of the Adolescent Brain Cognitive Development (ABCD) Study (Data Release 5.0). The ABCD Study is the largest study examining brain development and health longitudinally in the United States (US). At baseline, 11,875 children aged 9–10 years and their caregivers were recruited across 21 research sites in the US from September 10, 2018, to September 29, 2020. The recruitment process aimed to ensure that the demographic composition of participants closely mirrored that of the US youth population (Garavan et al., 2018). The study sample, recruitment, procedures, and measures have been described elsewhere (Garavan et al., 2018). All children gave their assents and their parents/caregivers signed informed consents. The study was approved by the University of California, San Diego, centralized Institutional Review Board (IRB), and local IRBs from each study site also gave their approvals. The current study followed the Strengthening the Reporting of Observational Studies in Epidemiology (Appendix S1) reporting guidelines (Elm et al., 2007).

Among the initial 11,875 ABCD Study participants, 7371 were distributed Fitbit Charge 2 devices to be worn for 21 days in Year 2. A rolling wear protocol was used where each site sequentially assigned and retrieved devices to participants at different times. We excluded 1189 (16.1%) participants with missing information on mental health outcomes at Year 3 and 542 (7.3%) participants with missing data on key covariates such as household income, parental educational attainment, pubertal status, and body mass index (BMI; Figure S1). The final analytical sample included 5640 participants.

Measures

Fitbit activity measures (Year 2)

Data on physical activity and sedentary behavior were extracted from Fitbit devices, which had been shown to offer reliable and valid measurements of physical activity over extended durations in adolescents (Bagot et al., 2018; Godino et al., 2020). Fitbit data were collected and processed by the ABCD Study team, and the curated weekly fitness summaries from Data Release 5.0 were used for the present analyses. We adhered to the established recommendations by the ABCD Study for extracting, filtering, and processing the data (Bagot et al., 2018; Godino et al., 2020). We included all days with >599 min of waking wear time and a minimum of 1000 steps, consistent with previous studies (Bagot et al., 2018; Nagata et al., 2023). Based on metabolic equivalents (METS) intensity thresholds, recorded time was classified as sedentary (<1.5 METS), moderately active (3–5.9 METS), and vigorously active (>6 METS) time (Bagot et al., 2018). Step counts, sedentary, and moderately and vigorously active minutes were averaged across the days of each week. These daily metrics were then averaged across the 3 weeks, correcting for the number of valid days in a given week (Bagot et al., 2018; Nagata, Alsamman, et al., 2023). Daily minutes spent moderately and vigorously active were summed to create MVPA.

Mental health (Year 3)

Adolescent mental health was assessed by the Child Behavior Checklist (CBCL) reported by parents/caregivers. The CBCL consisted of 113 questions assessing emotional and behavioral problems in children and adolescents that occurred in the preceding 6 months (Achenbach, 2001). Each question was rated on a three‐point Likert scale (0 = not true; 1 = sometimes true/somewhat true; 2 = very true/often true). The questions were grouped into two categories: internalizing behaviors, which summed syndrome scores of anxious/depressed, withdrawn/depressed, and somatic complaints, and externalizing behaviors, which summed syndrome scores of rule‐breaking and aggressive behaviors (Achenbach, 2001). These composite measures have shown excellent internal consistency (Cronbach's α = 0.91 for internalizing and α = 0.92 for externalizing behaviors) and validity (Achenbach, 2001; Achenbach, Ivanova, Rescorla, Turner, & Althoff, 2016). To account for developmental and sex‐related differences, we analyzed raw behavior scores and included sex and age as covariates, which were consistent with previous studies using data from the ABCD Study (Barch et al., 2021; McNeilly, Saragosa‐Harris, Mills, Dahl, & Magis‐Weinberg, 2022). Higher scores indicated more internalizing or externalizing problems.

Covariates

Covariates were selected based on prior studies demonstrating their associations with adolescent physical activity and mental health. These included age in years, race and ethnicity (Asian, Black, Latinx, White, and other/multiracial), biological sex (male or female), annual household income categories (≤$50 K, $50 K–$100 K, or ≥$100 K), and highest level of parental education (less than high school diploma, high school diploma or equivalent, some college education, bachelor's degree, or post graduate degree) reported by the parents/caregivers of the adolescent. We also adjusted for adolescent BMI due to its correlation with physical activity levels and mental health symptoms (Nagata et al., 2023; Patalay & Hardman, 2019). We included parent‐reported pubertal development (categorized as pre‐puberty, early puberty, mid‐puberty, and late/post‐puberty) to adjust for potential effects of pubertal maturation on adolescent mental health (Niu, Sheffield, & Li, 2023).

Statistical analysis

We used quantile regression to examine the associations between Fitbit‐measured activity behaviors (daily steps, MVPA, sedentary behavior) and mental health outcomes across the symptom distribution. Unlike linear regression, which estimates average effects, quantile regression models conditional associations at different points (quantiles) of the outcome distribution (Buchinsky, 1998; Koenker, 2005). It is also less sensitive to outliers compared to linear regression (Wei, Kehm, Goldberg, & Terry, 2019). This approach enabled us to capture varying associations across symptom severity without categorizing participants into arbitrary subgroups. We selected the 10th, 25th, 50th, 75th, and 90th percentiles to examine how associations vary across a broad range of mental health severity. The choice of quantiles is consistent with prior quantile regression studies (Bann, Johnson, Li, Kuh, & Hardy, 2018; Zhang et al., 2015). Unlike subgroup or moderator analyses, which rely on arbitrary stratification or predefined interaction terms, quantile regression examines how associations vary across the full distribution of the outcome, providing a flexible, distribution‐sensitive approach to detecting differential effects. We estimated the regression coefficients and their 95% confidence intervals (CIs) at each specified quantile. For all models, we adjusted for baseline familial and individual covariates. Given sex differences in physical activity patterns during adolescence (Nagata, Alsamman, et al., 2023), we conducted quantile regression analyses separately for boys and girls. In addition to our primary prospective models using Year 3 mental health outcomes, we conducted a supplementary analysis using Year 2 symptoms (concurrent with Fitbit assessment) to assess whether similar patterns were evident cross‐sectionally.

We reported unstandardized regression coefficients (b) and their 95% CIs for all analyses. All predictor variables were standardized (z‐scores) prior to analysis to enable comparability across activity measures; outcome variables were retained in their original units. Accordingly, we report unstandardized regression coefficients that represent the change in CBCL raw scores associated with a one‐standard deviation (SD) increase in the predictor. A p‐value < .05 from a two‐sided test was considered statistically significant. All statistical analyses were conducted using R version 4.3.2, with quantile regression models run using the “quantreg” R package. To promote transparency and reproducibility, the analysis code is available on the Open Science Framework (OSF): https://osf.io/z2apj/.

Results

The sample included 5640 youth (2745 girls [48.7%]; 105 Asian [1.9%], 516 Black [9.1%], 982 Hispanic [17.4%], 3483 White [61.8%], and 554 children of other/multiracial ethnicity [9.8%], see Table 1). The mean (SD) age of the study sample at baseline was 9.93 (0.63) years. The Fitbit data showed that adolescents had an average of 9467.3 (SD = 3248.2; median = 8949.3, IQR = 6937.3–11191.1) steps, 159.3 (SD = 133.1; median = 120.8, IQR = 55.8–212.1) minutes of MVPA, and 554.4 (SD = 98.7; median = 553.9, IQR = 496.5–615.5) minutes of sedentary time per day over the 21‐day period of data collection. Internalizing and externalizing symptoms 1 year after Fitbit data collection had a mean score of 5.2 (SD = 5.9; median = 3.0, IQR = 1.0–7.0) and 3.9 (SD = 5.3; median = 2.0, IQR = 0.0–5.0), respectively, with positive skewness (range = 2.01–2.49; Figures S2 and S3).

Table 1.

Description of the analytic sample at baseline (N = 5640)

Variables M (SD) Median (IQR)
Race and ethnicity, n (%)
White 3483 (61.8)
Black 516 (9.1)
Hispanic 982 (17.4)
Asian 105 (1.9)
Other race 554 (9.8)
Male, n (%) 2895 (51.3)
Household income, n (%)
≤50 K 1285 (22.8)
50 K–100 K 1738 (30.8)
≥100 K 2617 (46.4)
Parent highest education, n (%)
< High School Diploma 128 (2.3)
High School Diploma or equivalent 324 (5.7)
Some College 1323 (23.5)
Bachelor 1637 (29.0)
Post Graduate Degree 2228 (39.5)
Pubertal status, n (%)
Pre‐puberty 3062 (54.3)
Early puberty 1325 (23.5)
Mid‐puberty 1184 (21.0)
Late or post‐puberty 69 (1.2)
Body mass index, mean (SD) 18.48 (3.96)
Fitbit measures (2‐year follow‐up)
Daily steps, mean (SD) 9467.29 (3248.18)
MVPA in minutes, mean (SD) 159.27 (133.09)
Sedentary behavior in minutes, mean (SD) 554.41 (98.70)
Mental health outcomes (3‐year follow‐up)
CBCL internalizing problems, mean (SD) 5.22 (5.89) 3 (1–7)
CBCL externalizing problems, mean (SD) 3.85 (5.33) 2 (0–5)
Open in a new tab

Both the mean (SD) and the median (IQR) were presented to reflect the non‐normal distribution of the outcome variables. CBCL, Child Behavioral Checklist; GED, General Education Degree; HS, high school; IQR, interquartile range; MVPA, moderate‐to‐vigorous intensity physical activity; SD, standard deviation.

The quantile regression model (Table 2) suggested that at the 50th percentile, fewer daily steps (b = −0.43; 95% CI, −0.58 to −0.29, p < .001), less MVPA (b = −0.34; 95% CI, −0.47 to −0.22, p < .001), and more sedentary time (b = 0.32; 95% CI, 0.15–0.48, p < .001) were each significantly associated with greater internalizing symptoms a year later among adolescents. These associations became stronger at higher percentiles of internalizing symptoms (all ps < .001). For instance, a one‐SD increase in daily steps corresponded to a decrease of 1.43 (95% CI, 0.99–1.87, p < .001) in internalizing symptom score in the 90th percentile group, while the same increase in daily steps was associated with a decrease of 0.21 (95% CI, 0.05–0.37, p < .01) in internalizing symptom score in the 25th percentile group. Tables S1–S3 present the full model results for internalizing symptoms, including coefficients for all control variables.

Table 2.

Quantile regression of the association between Fitbit measures and adolescent mental health a year later

Outcome Quantile Daily steps MVPA Sedentary behavior
b (SE) 95% CI b (SE) 95% CI b (SE) 95% CI
Internalizing symptoms
Q10
 −0.00 (0.00) [−0.00, 0.00] −0.00 (0.00) [−0.00, 0.00] −0.00 (0.00) [−0.00, 0.00]
Q25
 −0.21 (0.08)** [−0.37, −0.05] −0.14 (0.08) [−0.30, 0.01] 0.00 (0.07) [−0.14, 0.14]
Q50
 −0.43 (0.07)*** [−0.58, −0.29] −0.34 (0.06)*** [−0.47, −0.22] 0.32 (0.09)*** [0.15, 0.48]
Q75
 −0.80 (0.10)*** [−1.01, −0.58] −0.63 (0.11)*** [−0.88, −0.37] 0.60 (0.18)*** [0.32, 0.88]
Q90
 −1.43 (0.23)*** [−1.87, −0.99] −0.96 (0.20)*** [−1.36, −0.57] 0.86 (0.30)** [0.21, 1.50]
Externalizing symptoms
Q10
 0.00 (0.01) [0.00, 0.00] −0.00 (0.00) [−0.00, 0.00] −0.00 (0.00) [−0.00, 0.00]
Q25
 0.00 (0.00) [−0.01, 0.01] −0.00 (0.01) [−0.01, 0.01] −0.00 (0.00) [−0.00, 0.00]
Q50
 0.08 (0.03) [−0.03, 0.19] 0.00 (0.05) [−0.10, 0.11] −0.07 (0.06) [−0.19, 0.04]
Q75
 0.07 (0.10) [−0.18, 0.32] −0.25 (0.13) [−0.53, 0.02] −0.24 (0.14) [−0.51, 0.03]
Q90
 0.19 (0.11) [−0.22, 0.60] −0.09 (0.23) [−0.52, 0.35] −0.63 (0.18)*** [−0.95, −0.30]
Open in a new tab

*p < .05, **p < .01, ***p < .001. Unstandardized regression coefficients, standard errors (SE), and 95% confidence intervals (CI) were shown. All models controlled for baseline participant age, sex, race/ethnicity, parental education, family household income, and family history of psychopathology (N = 5640). MVPA, moderate‐to‐vigorous intensity physical activity.

Quantile regressions stratified by sex suggested stronger associations of daily steps, sedentary behavior, and MVPA with internalizing symptoms in girls than boys (Figure 1). For instance, in the 90th percentile group, a one‐SD increase in MVPA corresponded to a 1.51 (95% CI, 0.84–2.19, p < .001) decrease in internalizing symptom score among girls, compared to a 0.90 (95% CI, 0.31–1.48, p < .01) decrease among boys (Figure 1, Table S4). In addition, there were significant associations between daily steps and MVPA and internalizing symptoms in girls in the 25th percentile group, while the associations were not significant in boys in the same percentile group. Moreover, more sedentary time was significantly associated with increased internalizing symptoms in girls from the 25th to 90th percentiles, while the association was not significant in boys across any quantiles.

Figure 1.

Figure 1

Open in a new tab

Unstandardized regression coefficients of the association between Fitbit measures (daily steps, MVPA, sedentary behavior) and adolescent internalizing symptoms by biological sex. Coefficients are statistically significant at p < .05 where 95% CIs do not cross the 0 line

We did not find any significant associations of daily steps and MVPA with externalizing symptoms across different quantile groups (Table 2). More sedentary time was significantly associated with few externalizing symptoms only in the 90th percentile group. Tables S5–S7 present the full model results for externalizing symptoms. Sex‐stratified analyses showed that lower MVPA was significantly associated with increased risks of externalizing symptoms in the 75th and 90th percentiles in girls (Figure 2, Table S1). In boys, more sedentary time was significantly associated with fewer externalizing symptoms in the 75th and 90th percentile groups.

Figure 2.

Figure 2

Open in a new tab

Unstandardized regression coefficients of the association between Fitbit measures (daily steps, MVPA, sedentary behavior) and adolescent externalizing symptoms by biological sex. Coefficients are statistically significant at p < .05 where 95% CIs do not cross the 0 line

Cross‐sectional analyses using Year 2 mental health symptoms showed similar patterns to the longitudinal models (Table S8). More daily steps and MVPA, and less sedentary time were significantly associated with fewer internalizing symptoms, particularly at the 75th and 90th percentiles. Associations with externalizing symptoms were limited and observed only at the 90th percentiles.

Discussion

In this longitudinal study, we examined associations between objectively measured physical activity and sedentary behavior and mental health a year later across adolescents with varying levels of symptom severity. Increased daily step counts and MVPA, and reduced sedentary time were each independently associated with fewer internalizing symptoms but were not associated with externalizing symptoms. The associations became stronger for youths at higher quantiles of internalizing symptoms. We also identified sex differences, with stronger associations of MVPA, daily steps, and sedentary behavior with internalizing symptoms in girls than boys. This study is among the first to combine a longitudinal design with device‐based measures of physical activity and sedentary behavior and to apply quantile regression to identify differential associations across the distribution of adolescent mental health symptoms.

Our results are consistent with previous studies that have shown clear associations of physical activity with adolescent mental health (Damme et al., 2024; Kandola et al., 2020; Rodriguez‐Ayllon et al., 2019; Sampasa‐Kanyinga et al., 2021). We extend the literature by showing that the associations differ substantially across individuals. While most studies have shown that physical activity including MVPA and daily steps has mental health benefits in adolescents, these benefits may be larger in adolescents already presenting internalizing symptoms. For adolescents with no or few internalizing symptoms, these benefits became smaller or non‐significant. This unique pattern of associations between physical activity and mental health has been identified in the adult population. Based on a recent meta‐analysis that estimated the effect of physical activity interventions in improving mental health outcomes in adults, the greatest benefits of physical activity were observed among adults diagnosed with depression (Singh et al., 2023). Our supplementary cross‐sectional analysis using concurrent mental health outcomes reinforced the primary findings, suggesting that physical activity may confer both immediate and sustained benefits for youth with elevated internalizing symptoms. This consistency across cross‐sectional and longitudinal models enhances confidence in the robustness of the observed associations. One possible explanation for the larger benefits of physical activity among adolescents with internalizing symptoms is that they may be more sensitive to the mechanisms through which activity supports mental health. Physical activity can improve internalizing symptoms through multiple pathways, including neurobiological changes in brain regions implicated by depression and anxiety, reductions in systemic inflammation, enhanced self‐esteem and social connectedness, and healthier sleep patterns (Kandola, Ashdown‐Franks, Hendrikse, Sabiston, & Stubbs, 2019; Zhang et al., 2019). Adolescents with more severe symptoms often show greater neurobiological vulnerabilities, such as dysregulation in the hypothalamic–pituitary–adrenal (HPA) axis (Gujral, Aizenstein, Reynolds, Butters, & Erickson, 2017), which may make them particularly responsive to these effects. Further research is needed to determine the role of biological factors in the association between physical activity and improvements in psychopathology among adolescents with and without internalizing symptoms.

We found that longer sedentary time was associated with increased internalizing symptoms in adolescents, consistent with several previous studies (Kandola et al., 2020; Liu et al., 2016; Rodriguez‐Ayllon et al., 2019). Again, larger effects were observed in youths who experience above median internalizing symptoms, which may reflect greater sensitivity to mental health improvements. Our findings, based on objectively measured sedentary time (<1.5 METS) using wearable technology over 21 days, provide high‐quality longitudinal evidence of the potential harm of sedentary behavior on internalizing symptoms in adolescents, particularly for those at the highest risk for depression and anxiety.

Physical activity (MVPA and daily steps) had stronger effects on internalizing symptoms in girls than boys. Prior studies suggest that girls experience a steeper decline in physical activity during adolescence compared to boys (Armstrong et al., 2018; Guthold et al., 2020; Pate, Saunders, Taverno Ross, & Dowda, 2022), likely due to both biological factors (e.g., pubertal maturation) and psychosocial barriers (e.g., body image concerns, gender norms) (Baker, Birch, Trost, & Davison, 2007; Gallant et al., 2023; Rosselli et al., 2020). Our study suggests that despite these challenges, girls who do engage in physical activity may experience greater mental health benefits. Additional research is needed to fully understand the sex‐specific pathway to mental health benefits of physical activity. In addition, the association between sedentary behavior and internalizing symptoms were found only in girls. Given the rising sedentary behavior in adolescents (Felez‐Nobrega et al., 2020), our findings underscore the urgent need for addressing high levels of sedentary behavior in this population.

In contrast to internalizing symptoms, physical activity (daily steps and MVPA) was not significantly associated with externalizing symptoms across youth with varying externalizing symptoms. However, sex differences exist in the association between MVPA and externalizing symptoms, such that greater MVPA has benefits for girls (but not boys) who experience above median symptoms. This finding indicates that MVPA could be an effective intervention target for girls with externalizing symptoms (Damme et al., 2024; Wu, Bastian, Ohinmaa, & Veugelers, 2018). We also found that more sedentary time was associated with fewer externalizing symptoms for boys who experience more externalizing symptoms. This contrasts previous research that has shown either a positive association between sedentary time and externalizing behavior (Wu et al., 2018) or no association at all (Damme et al., 2024). It is possible that sedentary activities, such as playing video games or watching television, may offer a structured environment that reduces the likelihood of behavior that typically leads to externalizing symptoms. However, this finding should be interpreted with caution, as it does not imply that sedentary behavior is beneficial in a broader context.

Implication

The findings imply that increasing physical activity and reducing sedentary time could yield more benefits in youths with more severe symptoms of psychopathology and in girls. Clinicians should consider incorporating physical activity recommendations into mental health interventions, focusing on activities that increase daily steps and MVPA. Personalized interventions targeting adolescents with higher levels of internalizing symptoms may be especially effective. The observed sex differences suggest the need for tailored approaches, taking into account the unique needs of both sexes. Schools and community settings should provide more opportunities for physical activity, including promoting safe environments and access to recreational spaces.

Strengths and limitations

Strengths of this study include accelerometry measurement of physical activity and sedentary behavior using a 21‐day protocol, prospective design over 1 year, large population‐based sample, and rich data on covariates for model adjustments. The study extended previous knowledge about physical activity and mental health by incorporating quantile‐dependent associations that account for varying levels of mental health severity in the adolescent population.

The study also had several limitations. First, sedentary behavior was measured by accumulating all time below a MET threshold of 1.5, which precluded us from distinguishing effects of various types of sedentary behavior. Future studies should aim to disentangle the effects of total sedentary time from specific activities such as screen exposure or social media use, as these may have additional negative effects on mental health outcomes due to factors like increased social comparison, cyberbullying, or disrupted sleep patterns (Boers, Afzali, Newton, & Conrod, 2019). Relatedly, we did not control for sleep, which is influenced by daytime physical activity patterns and may, in turn, affect emotional regulation and risk for internalizing and externalizing symptoms (Lang et al., 2016; Wang & Yip, 2020). Future research should explicitly examine the dynamic interplay between physical activity, sleep, and mental health, potentially using mediation or longitudinal network models to better capture their complex relationships. Second, while Fitbit devices offer a convenient and accessible method for monitoring physical activity on a large scale, the accuracy of step count data may be compromised compared to research‐grade accelerometers. Specifically, Fitbit may overestimate daily steps due to non‐walking arm movements and underestimate steps during slow walking or activities with limited wrist motion (Feehan et al., 2018; Toth et al., 2018). This potential measurement error may attenuate the observed associations. Third, although this analysis controlled for the quality of the Fitbit device by including only days with more than 599 min of waking wear time, variations in participants' total wear time were not controlled for. Fourth, we relied on parent‐reported CBCL to assess internalizing and externalizing symptoms. While parent reports may underestimate internalizing problems—especially in boys (Ford & McCoy, 2022)—prior studies suggest that associations with physical activity are generally consistent across parent‐ and youth‐reported measures (Wu et al., 2025; Zhang, Schwandt, Vines, & Volkow, 2024). Future research should incorporate multi‐informant assessments to measure adolescent mental health symptoms and validate the observed associations. Fifth, we examined effects of physical activity assessed between 12 and 13 years on mental health outcomes over 1 year, which limited our ability to capture the effect of physical activity across the full range of adolescence. The longevity of the association needs to be examined when more waves of data from the ABCD Study became available.

Conclusion

In summary, using objective measures, we found that increased MVPA and daily step counts and decreased sedentary behavior between the ages of 12 and 13 years were associated with fewer internalizing symptoms 1 year later, with stronger associations observed for adolescents who experienced greater internalizing symptoms. Findings underscore the need for more direct and specific consideration of adolescents' pre‐existing mental health symptoms in public health interventions aimed at reducing mental health problems.

Conflict of interest statement

The authors have declared that they have no competing or potential conflicts of interest.

Funding information

This project was supported in part by the National Natural Science Foundation of China (#32300888) and the Fundamental Research Funds for the Central Universities (#01900310400209543). The funder played no role in the study design, data analysis, writing, or decision to submit the manuscript for publication.

Ethical considerations

The study used data from the Adolescent Brain Cognitive Development (ABCD) Study. Written informed consent was obtained from a parent or legal guardian, and assent was obtained from all child participants prior to enrollment. The ABCD Study was reviewed and approved by the Institutional Review Board at the University of California, San Diego (UCSD; IRB #160091; approved September 13, 2016) as the central IRB; participating study sites also obtained approval from their local IRBs consistent with the central IRB (Garavan et al., 2018).

Supporting information

Figure S1. Flowchart of study sample inclusion and exclusion.

Figure S2. Histogram of CBCL internalizing scores at the 3‐year follow‐up.

Figure S3. Histogram of CBCL externalizing scores at the 3‐year follow‐up.

Table S1. Full model results for the association between daily steps and adolescent internalizing problems.

Table S2. Full model results for the association between MVPA and adolescent internalizing problems.

Table S3. Full model results for the association between sedentary behavior and adolescent internalizing symptoms.

Table S4. Quantile regression of the association between Fitbit measures and adolescent mental health a year later, stratified by biological sex.

Table S5. Full model results for the association between daily steps and adolescent externalizing symptoms.

Table S6. Full model results for the association between MVPA and adolescent externalizing symptoms.

Table S7. Full model results for the association between sedentary behavior and adolescent externalizing symptoms.

Table S8. Quantile regression of the concurrent associations between Fitbit measures and adolescent mental health at Year 2.

Appendix S1. STROBE Statement—Checklist of items that should be included in reports of cohort studies.

CAMH-31-208-s001.docx (268.4KB, docx)

Acknowledgements

L.N., J.J., and Y.L. conceptualized the study. L.N. acquired the study data and performed the data processing, analysis, and visualization. L.N. and J.J. wrote the initial manuscript draft. D.Z., L.L., M.X. and Y.L. assisted in reviewing the results and critically reviewed the manuscript. All authors approved the final submitted version. L.N. is the guarantor. Data used in the preparation of this article were obtained from the Adolescent Brain Cognitive DevelopmentSM (ABCD) Study (https://abcdstudy.org), held in the NIMH Data Archive (NDA). This is a multisite, longitudinal study designed to recruit more than 10,000 children age 9‐10 and follow them over 10 years into early adulthood. The ABCD Study® is supported by the National Institutes of Health and additional federal partners under award numbers U01DA041048, U01DA050989, U01DA051016, U01DA041022, U01DA051018, U01DA051037, U01DA050987, U01DA041174, U01DA041106, U01DA041117, U01DA041028, U01DA041134, U01DA050988, U01DA051039, U01DA041156, U01DA041025, U01DA041120, U01DA051038, U01DA041148, U01DA041093, U01DA041089, U24DA041123, U24DA041147. A full list of supporters is available at https://abcdstudy.org/federal‐partners.html. A listing of participating sites and a complete listing of the study investigators can be found at https://abcdstudy.org/consortium_members/. ABCD consortium investigators designed and implemented the study and/or provided data but did not necessarily participate in the analysis or writing of this report. This manuscript reflects the views of the authors and may not reflect the opinions or views of the NIH or ABCD consortium investigators. The ABCD data repository grows and changes over time. The NDA study for this project can be found at doi: 10.15154/7k3a‐qb76.

Data availability statement

The data are from the ABCD Study Curated Annual Release 5.0 and are available on request from the NIMH Data Archive (https://data‐archive.nimh.nih.gov/abcd). Restrictions apply to the availability of these data, which were used under license for this study.

References

  1. Achenbach, T.M. (2001). Manual for ASEBA school‐age forms & profiles. Burlington, VT, USA: University of Vermont, Research Center for Children, Youth & Families. [Google Scholar]
  2. Achenbach, T.M. , Ivanova, M.Y. , Rescorla, L.A. , Turner, L.V. , & Althoff, R.R. (2016). Internalizing/externalizing problems: Review and recommendations for clinical and research applications. Journal of the American Academy of Child and Adolescent Psychiatry, 55, 12. [DOI] [PubMed] [Google Scholar]
  3. Adamo, K.B. , Prince, S.A. , Tricco, A.C. , Connor‐Gorber, S. , & Tremblay, M. (2009). A comparison of indirect versus direct measures for assessing physical activity in the pediatric population: A systematic review. International Journal of Pediatric Obesity: An Official Journal of the International Association for the Study of Obesity, 4, 2–27. [DOI] [PubMed] [Google Scholar]
  4. Armstrong, S. , Wong, C.A. , Perrin, E. , Page, S. , Sibley, L. , & Skinner, A. (2018). Association of physical activity with income, race/ethnicity, and sex among adolescents and young adults in the United States: Findings from the National Health and nutrition examination survey, 2007–2016. JAMA Pediatrics, 172, 732–740. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bagot, K.S. , Matthews, S.A. , Mason, M. , Squeglia, L.M. , Fowler, J. , Gray, K. , … & Patrick, K. (2018). Current, future and potential use of mobile and wearable technologies and social media data in the ABCD study to increase understanding of contributors to child health. Developmental Cognitive Neuroscience, 32, 121–129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Baker, B.L. , Birch, L.L. , Trost, S.G. , & Davison, K.K. (2007). Advanced pubertal status at age 11 and lower physical activity in adolescent girls. The Journal of Pediatrics, 151, 488–493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bann, D. , Johnson, W. , Li, L. , Kuh, D. , & Hardy, R. (2018). Socioeconomic inequalities in childhood and adolescent body‐mass index, weight, and height from 1953 to 2015: An analysis of four longitudinal, observational, British birth cohort studies. The Lancet. Public Health, 3, e194–e203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Barch, D.M. , Albaugh, M.D. , Baskin‐Sommers, A. , Bryant, B.E. , Clark, D.B. , Dick, A.S. , … & Xie, L. (2021). Demographic and mental health assessments in the adolescent brain and cognitive development study: Updates and age‐related trajectories. Developmental Cognitive Neuroscience, 52, 101031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Biddle, S.J. , Ciaccioni, S. , Thomas, G. , & Vergeer, I. (2019). Physical activity and mental health in children and adolescents: An updated review of reviews and an analysis of causality. Psychology of Sport and Exercise, 42, 146–155. [Google Scholar]
  10. Boers, E. , Afzali, M.H. , Newton, N. , & Conrod, P. (2019). Association of Screen Time and Depression in adolescence. JAMA Pediatrics, 173, 853–859. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Buchinsky, M. (1998). Recent advances in quantile regression models: A practical guideline for empirical research. The Journal of Human Resources, 33, 88–126. [Google Scholar]
  12. Damme, K.S.F. , Sloan, R.P. , Bartels, M.N. , Ozsan, A. , Ospina, L.H. , Kimhy, D. , & Mittal, V.A. (2021). Psychosis risk individuals show poor fitness and discrepancies with objective and subjective measures. Scientific Reports, 11, 9851. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Damme, K.S.F. , Vargas, T.G. , Walther, S. , Shankman, S.A. , & Mittal, V.A. (2024). Physical and mental health in adolescence: Novel insights from a transdiagnostic examination of FitBit data in the ABCD study. Translational Psychiatry, 14, 75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Elm, E. , Altman, D.G. , Egger, M. , Pocock, S.J. , Gøtzsche, P.C. , & Vandenbroucke, J.P. (2007). The strengthening the reporting of observational studies in epidemiology (STROBE) statement: Guidelines for reporting observational studies. The Lancet, 370, 1453–1457. [DOI] [PubMed] [Google Scholar]
  15. Erskine, H.E. , Moffitt, T.E. , Copeland, W.E. , Costello, E.J. , Ferrari, A.J. , Patton, G. , … & Scott, J.G. (2015). A heavy burden on young minds: The global burden of mental and substance use disorders in children and youth. Psychological Medicine, 45, 1551–1563. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Feehan, L.M. , Geldman, J. , Sayre, E.C. , Park, C. , Ezzat, A.M. , Yoo, J.Y. , … & Li, L.C. (2018). Accuracy of Fitbit devices: Systematic review and narrative syntheses of quantitative data. JMIR mHealth and uHealth, 6, e10527. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Felez‐Nobrega, M. , Raine, L.B. , Haro, J.M. , Wijndaele, K. , & Koyanagi, A. (2020). Temporal trends in leisure‐time sedentary behavior among adolescents aged 12‐15 years from 26 countries in Asia, Africa, and the Americas. International Journal of Behavioral Nutrition and Physical Activity, 17, 102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Ford, S.H. , & McCoy, T.P. (2022). Minding the gap: Adolescent and parent/caregiver reporter discrepancies on symptom presence, impact of covariates, and clinical implications. Journal of Pediatric Health Care: Official Publication of National Association of Pediatric Nurse Associates & Practitioners, 36, 225–230. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Gallant, F. , Hebert, J.J. , Thibault, V. , Mekari, S. , Sabiston, C.M. , & Bélanger, M. (2023). Puberty timing and relative age as predictors of physical activity discontinuation during adolescence. Scientific Reports, 13, 13740. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Garavan, H. , Bartsch, H. , Conway, K. , Decastro, A. , Goldstein, R.Z. , Heeringa, S. , … & Zahs, D. (2018). Recruiting the ABCD sample: Design considerations and procedures. Developmental Cognitive Neuroscience, 32, 16–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Godino, J.G. , Wing, D. , de Zambotti, M. , Baker, F.C. , Bagot, K. , Inkelis, S. , … & Tapert, S.F. (2020). Performance of a commercial multi‐sensor wearable (Fitbit charge HR) in measuring physical activity and sleep in healthy children. PLoS One, 15, e0237719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Gujral, S. , Aizenstein, H. , Reynolds, C.F. , Butters, M.A. , & Erickson, K.I. (2017). Exercise effects on depression: Possible neural mechanisms. General Hospital Psychiatry, 49, 2–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Gunnell, D. , Kidger, J. , & Elvidge, H. (2018). Adolescent mental health in crisis. BMJ, 361, 2608. [DOI] [PubMed] [Google Scholar]
  24. Guthold, R. , Stevens, G.A. , Riley, L.M. , & Bull, F.C. (2020). Global trends in insufficient physical activity among adolescents: A pooled analysis of 298 population‐based surveys with 1·6 million participants. The Lancet Child & Adolescent Health, 4, 23–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kandola, A. , Ashdown‐Franks, G. , Hendrikse, J. , Sabiston, C.M. , & Stubbs, B. (2019). Physical activity and depression: Towards understanding the antidepressant mechanisms of physical activity. Neuroscience and Biobehavioral Reviews, 107, 525–539. [DOI] [PubMed] [Google Scholar]
  26. Kandola, A. , Lewis, G. , Osborn, D.P.J. , Stubbs, B. , & Hayes, J.F. (2020). Depressive symptoms and objectively measured physical activity and sedentary behaviour throughout adolescence: A prospective cohort study. The Lancet Psychiatry, 7, 262–271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Koenker, R. (2005). Quantile regression. Cambridge: Cambridge University Press. [Google Scholar]
  28. Lang, C. , Kalak, N. , Brand, S. , Holsboer‐Trachsler, E. , Pühse, U. , & Gerber, M. (2016). The relationship between physical activity and sleep from mid adolescence to early adulthood: A systematic review of methodological approaches and meta‐analysis. Sleep Medicine Reviews, 28, 32–45. [DOI] [PubMed] [Google Scholar]
  29. Lê Cook, B. , & Manning, W.G. (2013). Thinking beyond the mean: A practical guide for using quantile regression methods for health services research. Shanghai Archives of Psychiatry, 25, 55–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Liu, M. , Wu, L. , & Yao, S. (2016). Dose–response association of screen time‐based sedentary behaviour in children and adolescents and depression: A meta‐analysis of observational studies. British Journal of Sports Medicine, 50, 1252–1258. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lu, W. , & Keyes, K.M. (2023). Major depression with co‐occurring suicidal thoughts, plans, and attempts: An increasing mental health crisis in US adolescents, 2011–2020. Psychiatry Research, 327, 115352. [DOI] [PubMed] [Google Scholar]
  32. Matthews, C.E. , Moore, S.C. , George, S.M. , Sampson, J. , & Bowles, H.R. (2012). Improving self‐reports of active and sedentary behaviors in large epidemiologic studies. Exercise and Sport Sciences Reviews, 40, 118–126. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. McNeilly, E.A. , Saragosa‐Harris, N.M. , Mills, K.L. , Dahl, R.E. , & Magis‐Weinberg, L. (2022). Reward sensitivity and internalizing symptoms during the transition to puberty: An examination of 9‐and 10‐year‐olds in the ABCD study. Developmental Cognitive Neuroscience, 58, 101172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Nagata, J.M. , Alsamman, S. , Smith, N. , Yu, J. , Ganson, K.T. , Dooley, E.E. , … & Pettee Gabriel, K. (2023). Social epidemiology of Fitbit daily steps in early adolescence. Pediatric Research, 94, 1838–1844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Nagata, J.M. , Smith, N. , Alsamman, S. , Lee, C.M. , Dooley, E.E. , Kiss, O. , … & Gabriel, K.P. (2023). Association of Physical Activity and Screen Time with Body Mass Index among US adolescents. JAMA Network Open, 6, e2255466. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Niu, L. , Sheffield, P. , & Li, Y. (2023). Pubertal timing, neighborhood income, and mental health in boys and girls: Findings from the adolescent brain cognitive development study. Social Science & Medicine, 334, 116220. [DOI] [PubMed] [Google Scholar]
  37. Patalay, P. , & Hardman, C.A. (2019). Comorbidity, Codevelopment, and temporal associations between body mass index and internalizing symptoms from early childhood to adolescence. JAMA Psychiatry, 76, 721–729. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Pate, R.R. , Saunders, R.P. , Taverno Ross, S.E. , & Dowda, M. (2022). Patterns of age‐related change in physical activity during the transition from elementary to high school. Preventive Medicine Reports, 26, 101712. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Racine, N. , McArthur, B.A. , Cooke, J.E. , Eirich, R. , Zhu, J. , & Madigan, S. (2021). Global prevalence of depressive and anxiety symptoms in children and adolescents during COVID‐19: A meta‐analysis. JAMA Pediatrics, 175, 1142–1150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Rodriguez‐Ayllon, M. , Cadenas‐Sánchez, C. , Estévez‐López, F. , Muñoz, N.E. , Mora‐Gonzalez, J. , Migueles, J.H. , … & Esteban‐Cornejo, I. (2019). Role of physical activity and sedentary behavior in the mental health of preschoolers, children and adolescents: A systematic review and meta‐analysis. Sports Medicine, 49, 1383–1410. [DOI] [PubMed] [Google Scholar]
  41. Rosselli, M. , Ermini, E. , Tosi, B. , Boddi, M. , Stefani, L. , Toncelli, L. , & Modesti, P.A. (2020). Gender differences in barriers to physical activity among adolescents. Nutrition, Metabolism, and Cardiovascular Diseases: NMCD, 30, 1582–1589. [DOI] [PubMed] [Google Scholar]
  42. Sampasa‐Kanyinga, H. , Colman, I. , Goldfield, G.S. , Janssen, I. , Wang, J. , Tremblay, M.S. , … & Chaput, J.‐P. (2021). 24‐hour movement behaviors and internalizing and externalizing behaviors among youth. Journal of Adolescent Health, 68, 969–977. [DOI] [PubMed] [Google Scholar]
  43. Singh, B. , Olds, T. , Curtis, R. , Dumuid, D. , Virgara, R. , Watson, A. , … & Maher, C. (2023). Effectiveness of physical activity interventions for improving depression, anxiety and distress: An overview of systematic reviews. British Journal of Sports Medicine, 57, 1203–1209. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Toseeb, U. , Brage, S. , Corder, K. , Dunn, V.J. , Jones, P.B. , Owens, M. , … & Goodyer, I.M. (2014). Exercise and depressive symptoms in adolescents: A longitudinal cohort study. JAMA Pediatrics, 168, 1093–1100. [DOI] [PubMed] [Google Scholar]
  45. Toth, L.P. , Park, S. , Springer, C.M. , Feyerabend, M.D. , Steeves, J.A. , & Bassett, D.R. (2018). Video‐recorded validation of wearable step counters under free‐living conditions. Medicine and Science in Sports and Exercise, 50, 1315–1322. [DOI] [PubMed] [Google Scholar]
  46. van Sluijs, E.M.F. , Ekelund, U. , Crochemore‐Silva, I. , Guthold, R. , Ha, A. , Lubans, D. , … & Katzmarzyk, P.T. (2021). Physical activity behaviours in adolescence: Current evidence and opportunities for intervention. Lancet (London, England), 398, 429–442. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Wang, Y. , & Yip, T. (2020). Sleep facilitates coping: Moderated mediation of daily sleep, ethnic/racial discrimination, stress responses, and adolescent well‐being. Child Development, 91, e833–e852. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Wei, Y. , Kehm, R.D. , Goldberg, M. , & Terry, M.B. (2019). Applications for quantile regression in epidemiology. Current Epidemiology Reports, 6, 191–199. [Google Scholar]
  49. World Health Organization . (2020). WHO guidelines on physical activity and sedentary behaviour. Washington, DC: WHO. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Wu, J. , Zhao, X. , Shao, Y. , Zang, W. , Jun, H. , & Yu, W. (2025). The impact of physical exercise on internalizing and externalizing problem behaviors among middle school students: Correlation and regression prediction analysis. Child and Adolescent Psychiatry and Mental Health, 19, 45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Wu, X. , Bastian, K. , Ohinmaa, A. , & Veugelers, P. (2018). Influence of physical activity, sedentary behavior, and diet quality in childhood on the incidence of internalizing and externalizing disorders during adolescence: A population‐based cohort study. Annals of Epidemiology, 28, 86–94. [DOI] [PubMed] [Google Scholar]
  52. Xiang, M.‐Q. , Tan, X.‐M. , Sun, J. , Yang, H.‐Y. , Zhao, X.‐P. , Liu, L. , … & Hu, M. (2020). Relationship of physical activity with anxiety and depression symptoms in Chinese college students during the COVID‐19 outbreak. Frontiers in Psychology, 11, 582436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Zhang, R. , Schwandt, M.L. , Vines, L. , & Volkow, N.D. (2024). Changes in rest‐activity rhythms in adolescents as they age: Associations with brain and behavioral changes in the ABCD study. Journal of the American Academy of Child and Adolescent Psychiatry, 12, 1105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Zhang, Y. , Niu, L. , Zhang, D. , Ip, P. , Ho, F. , Jiang, Y. , … & Jiang, F. (2019). Social‐emotional functioning explains the effects of physical activity on academic performance among Chinese primary school students: A mediation analysis. The Journal of Pediatrics, 208, 74–80. [DOI] [PubMed] [Google Scholar]
  55. Zhang, Y. , Zhang, D. , Jiang, Y. , Sun, W. , Wang, Y. , Chen, W. , … & Jiang, F. (2015). Association between physical activity and teacher‐reported academic performance among fifth‐graders in Shanghai: A quantile regression. PLoS One, 10, e0115483. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1. Flowchart of study sample inclusion and exclusion.

Figure S2. Histogram of CBCL internalizing scores at the 3‐year follow‐up.

Figure S3. Histogram of CBCL externalizing scores at the 3‐year follow‐up.

Table S1. Full model results for the association between daily steps and adolescent internalizing problems.

Table S2. Full model results for the association between MVPA and adolescent internalizing problems.

Table S3. Full model results for the association between sedentary behavior and adolescent internalizing symptoms.

Table S4. Quantile regression of the association between Fitbit measures and adolescent mental health a year later, stratified by biological sex.

Table S5. Full model results for the association between daily steps and adolescent externalizing symptoms.

Table S6. Full model results for the association between MVPA and adolescent externalizing symptoms.

Table S7. Full model results for the association between sedentary behavior and adolescent externalizing symptoms.

Table S8. Quantile regression of the concurrent associations between Fitbit measures and adolescent mental health at Year 2.

Appendix S1. STROBE Statement—Checklist of items that should be included in reports of cohort studies.

CAMH-31-208-s001.docx (268.4KB, docx)

Data Availability Statement

The data are from the ABCD Study Curated Annual Release 5.0 and are available on request from the NIMH Data Archive (https://data‐archive.nimh.nih.gov/abcd). Restrictions apply to the availability of these data, which were used under license for this study.

Articles from Child and Adolescent Mental Health are provided here courtesy of Wiley

RESOURCES