Tracking of physical activity and sport from childhood and adolescence to adulthood: a systematic review and meta-analysis

Introduction

Physical activity, which includes any bodily movement produced by skeletal muscles that requires energy expenditure, plays a vital role in promoting health and well-being across the lifespan (1). Engaging in regular physical activity during childhood and adolescence has been associated with numerous physical, psychological, and social benefits (2,3), which can persist into adulthood (4). Furthermore, sports participation refers to organized activities such as team sports, individual sports, or structured physical activities that involve systematic training, competition, and adherence to specific rules. Sports participation also offers a range of immediate and long-term benefits that can positively impact physical, mental, and social well-being during childhood, adolescence (5), and adulthood (6).

The transition from children and adolescents to adulthood often brings about significant changes in lifestyle behaviors, including shifts in physical activity patterns (7). Understanding the tracking of physical activity and sports levels, defined as the degree to which these behaviors remain consistent over time, from childhood through adolescence into adulthood is crucial for developing effective strategies to promote lifelong physical activity and reduce the risk of chronic diseases (8). Numerous researchers have explored the continuity of physical activity over time (7). However, despite significant findings indicating a positive association between physical activity levels in childhood and adolescence and physical activity in adulthood, the precise strength of this relationship remains unclear. For instance, several reviews suggest that while there is a tendency for physical activity (9-12) and sports participation levels (13) to remain stable at low to moderate levels from childhood through various ages into adulthood, the extent of this tracking is not fully elucidated. The findings also show that consistency in tracking decreases over longer periods, but tracking improves with higher initial age levels (9,11). Contradictory findings regarding the strength of tracking were noted between males and females (9,11).

Although prior studies have explored tracking in various contexts, few have quantitatively synthesized the stability of both physical activity and sports participation across such an extended developmental period. Additionally, the role of follow-up duration and initial age in moderating this relationship remains underexplored. Our systematic review and meta-analysis aimed to address this knowledge gap by synthesizing existing evidence on the tracking of physical activity from childhood through both physical activity and sports participation. By comprehensively examining studies spanning from childhood and adolescence to adulthood, we aim to elucidate the tracking of physical activity behaviors across different life stages. Through this synthesis of evidence, we aimed to provide insights that can inform the development of targeted interventions and policies aimed at promoting sustained physical activity and sports participation throughout the lifespan. We present this article in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2025-89/rc).

Methods

The protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO) (CRD42024527591). In this meta-analysis, we used publicly available anonymized data from original studies, thus, ethical approval was not required. The two authors (A.G.H., Y.E.) independently conducted each step of the systematic review, including literature selection and data extraction.

Selection criteria

Longitudinal cohort studies, either prospective or retrospective, that measured physical activity and sport participation during childhood and/or adolescence (ages 6–18 years) and its association in young adulthood (ages 19–30 years) or adulthood (>30 years old) were eligible. Any measure of physical activity (subjective, such as self-report or parent-report, and objective, such as accelerometers or pedometers) and sport participation was considered.

Information sources and search strategy

Articles were retrieved from the electronic databases MEDLINE, Embase, and Web of Science, spanning from their inception to July 2024. Detailed search strategies can be found in the supplementary online content (Appendix 1). Search terms included terms such as “tracking”, “longitudinal study”, “physical activity”, “exercise”, “fitness”, “motor activity”, “sport participation”, “sports involvement”, and “athletic performance”.

Selection process and data collection process

The two authors independently collected the following information from each article: (I) study characteristics (including authors, publication year, sample size, sex, age of first-time collection point, country, and years of follow-up); (II) details on physical activity and sport participation (such as specific and comprehensive descriptions of the methodologies and tools used to measure the primary variables of interest); and (III) analysis and results.

Risk of bias in individual studies

Risk of bias assessment was performed independently by two authors using the Joanna Briggs Institute (JBI) Critical Appraisal checklist (14) developed for analytical cross-sectional and for cohort studies. This checklist is used to evaluate the methodological quality and risk of bias. It comprises eight items for analytical cross-sectional studies and eleven items for cohort studies, each tailored to evaluate specific study aspects. To assess each checklist item, a score of “yes” or “no” is assigned if the criterion is or is not met, ‘unclear’ if information is insufficient, or ‘not applicable’ if the item is not pertinent to the study design or context. Following item scoring, the overall appraisal score is determined by tallying “yes” responses. This total score provides an indication of the extent to which the study meets the methodological criteria.

Statistical analysis

Effect measures

The primary effect size in this study was the correlation coefficient (r). We transformed alternative estimations (such as unstandardized regression coefficients, standardized mean differences, and odds ratios) into correlation coefficients using their respective formulas (15-17).

Synthesis methods

Correlation coefficients, along with their corresponding standard errors or sample sizes, were entered into the software. We used STATA v17.0 (STATA Corp., College Station, TX, USA) to calculate pooled r values with a 95% confidence interval (CI), employing a random-effects inverse-variance model with the Hartung-Knapp-Sidik-Jonkman adjustment. The pooled effect size for r was classified as small (<0.30), moderate (0.30–0.60), or large (>0.60) (12).

Several key points regarding our statistical analyses require clarification: (I) meta-analyses were conducted separately for physical activity and sport participation; (II) when a study included data for both males and females, the data were analyzed separately by sex. However, studies that did not report data by sex, or combined data for both sexes without disaggregation, were excluded from the analysis; and (III) when multiple articles utilize the same subject cohort, we prioritized the cohort with a longer follow-up period and/or a larger number of subjects, to prevent overlap. If a single article reports separate studies on physical activity and sports participation, each study will be treated as a separate unit of analysis.

To address the expected heterogeneity between studies, the total variance (Q) and the inconsistency index (I²) (18) were utilized. I² values were classified into three categories: low inconsistency (<25%), moderate inconsistency (25–75%), and high inconsistency (>75%) (19).

Risk of bias across studies

We conducted Doi plots and calculated the Luis Furuya-Kanamori (LFK) index to evaluate small-study effects, which assess publication bias. The asymmetry of the LFK was categorized as follows: a value greater than 1 or less than −1 indicates minor asymmetry, while a value greater than 2 or less than −2 indicates major asymmetry (20). To further explore the potential impact of publication bias, we performed a nonparametric trim-and-fill analysis using a random-effects model with the restricted maximum likelihood (REML) method.

Additional analysis

Subgroup analysis was performed whenever possible, based on the study design (i.e., cross-sectional, retrospective, and longitudinal), as well as by sex (i.e., we removed studies that include males and females together), age group at baseline (i.e., from childhood or adolescence), and follow-up (i.e., to young adulthood or adulthood). We also evaluated whether differences existed among these subgroups.

Finally, mixed-effects meta-regression (random-effects inverse-variance model with the Hartung-Knapp-Sidik-Jonkman adjustment) was conducted to determine whether length of follow-up (years) was a moderator in these associations.

Results

Study selection

After eliminating duplicates, a total of 1,670 articles were screened, from which 38 reports were found to meet our inclusion criteria, as shown in Table S1. The identification, screening, and inclusion process is depicted in Figure 1. Detailed information on excluded studies can be found in Appendix 2.

Figure 1 PRISMA flow diagram illustrating the process of identification, screening, eligibility, and inclusion of studies in the systematic review and meta-analysis. PRISMA, Preferred Reporting Items for Systematic reviews and Meta-Analyses.

Study characteristics

Table S1 summarizes the key characteristics of studies investigating the relationship between physical activity and/or sports participation from childhood or adolescence into adulthood. These studies encompass diverse geographic settings, including the following 16 countries: Australia, Belgium, Brazil, Canada, Denmark, England-Scotland-Wales, Finland, Germany, Israel, Japan, the Netherlands, New Zealand, Northern Ireland, Norway, Sweden, and the United States. A total of 63,158 participants were followed up for an average of 20.9 years. Various study designs were employed, primarily prospective, with some retrospective or cross-sectional approaches. Measurement methods for physical activity varied from self-reports to accelerometer and pedometer usage. The sample included participants from childhood through older age, with follow-up periods ranging from 5 to 55 years.

Risk of bias within studies

The results of the JBI critical appraisal tool are shown in Table S2 in the supplement. To account for variations in study design in cohort studies, the percentage of “yes” responses was calculated for each study, reflecting the proportion of applicable checklist items met. The follow-up time assessment items showed that while most studies reported sufficient durations for outcomes, there were discrepancies in follow-up completeness, with some studies experiencing losses to follow-up. Similarly, the measurement tools for physical activity were diverse, with some studies employing valid and reliable methods such as accelerometers, pedometers, or validated questionnaires, while others used other approaches such as simple questions or diaries, potentially biasing and undermining study reliability. The mean percentage of “yes” responses across all included cohort and cross-sectional studies was approximately 51.91%, and 62.5%, respectively.

Synthesis of results

Physical activity

Our meta-analysis reveals that physical activity exhibits low tracking from childhood and/or adolescence to adulthood (r=0.14; 95% CI: 0.11 to 0.16; I²=85.41%) (Figure 2). Similar tracking patterns were observed across sexes (males: r=0.14; 95% CI: 0.10 to 0.18; I²=78.60%; females: r=0.13; 95% CI: 0.09 to 0.17; I²=78.28%; P=0.72) (Figure S1) and by age groups at baseline (childhood: r=0.10; 95% CI: 0.06 to 0.14; I²=70.29%; adolescence: r=0.15; 95% CI: 0.12 to 0.18; I²=87.51%; P=0.06) (Figure S2). However, the tracking of physical activity varied during follow-up stages, showing stronger associations in young adulthood (r=0.20; 95% CI: 0.16 to 0.24; I²=59.02%) compared to adulthood (r=0.10; 95% CI: 0.07 to 0.13; I²=86.51%; P<0.001) (Figure S3).

Figure 2 Forest plot of the tracking of physical activity from childhood or adolescence to adulthood. The squares represent the point estimate of the correlation for each individual study, the horizontal lines represent the confidence interval for the correlation, and the diamond represents the overall correlation estimate. CI, confidence interval.

The meta-regression analysis showed that the length of follow-up moderates this association (B=−0.002; 95% CI: −0.004 to −0.001, P=0.04) (Figure S4).

Major asymmetry suggestive of small-study effects was observed (LFK =2.85) (Figure S5). However, a nonparametric trim-and-fill analysis (random-effects model, REML method) did not impute any missing studies, and the pooled effect size remained unchanged (observed =0.136; observed + imputed =0.136; 95% CI: 0.109–0.163), suggesting limited impact of publication bias on the results.

Sport participation

In our meta-analysis, we found that sports participation shows low tracking from childhood and/or adolescence into adulthood (r=0.26; 95% CI: 0.20 to 0.31; I²=88.57%) (Figure 3). Similar tracking patterns were observed across sexes (males: r=0.26; 95% CI: 0.18 to 0.33; I²=82.23%; females: r=0.25; 95% CI: 0.16 to 0.33; I²=89.08%; P=0.85) (Figure S6) and by age groups at baseline (childhood: r=0.22; 95% CI: 0.13 to 0.32; I²=87.02%; adolescence: r=0.27; 95% CI: 0.20 to 0.33; I²=87.09%; P=0.39) (Figure S7). However, we observed variability in tracking during follow-up stages, with stronger associations in young adulthood (r=0.30; 95% CI: 0.22 to 0.38; I²=78.42%) compared to adulthood (r=0.20; 95% CI: 0.13 to 0.26; I²=86.54%; P=0.03) (Figure S8).

Figure 3 Forest plot of the tracking of sport participation in early years and physical activity tracking at adulthood. The squares represent the point estimate of the correlation for each individual study, the horizontal lines represent the confidence interval for the correlation, and the diamond represents the overall correlation estimate. CI, confidence interval.

The meta-regression analysis showed that the length of follow-up moderates this association (B=−0.006; 95% CI: −0.012 to −0.001, P=0.04) (Figure S9).

Minor asymmetry suggestive of small-study effects was observed (LKF index =−1.12) (Figure S10). The nonparametric trim-and-fill analysis did not impute any missing studies (imputed =0), and the pooled effect size remained unchanged (observed =0.262; 95% CI: 0.209–0.316), suggesting minimal influence of publication bias.

Differences in tracking between physical activity and sports participation

The global analysis shows that the association is stronger in sports participation compared to physical activity (P<0.001) (Figure S11).

Discussion

Our meta-analysis reveals a consistent trend of low tracking from childhood or adolescence to adulthood in both physical activity and sports participation, with a slightly stronger association in the latter. This pattern persists across sex and age groups at baseline, with differences noted at follow-up. Additionally, the length of follow-up appears to moderate this association, i.e., there is a general decline in correlation coefficients over time.

Despite all the benefits associated with regular physical activity for health (2,3), only one in five adolescents worldwide (21) meet the recommendations set by specialized institutions such as the World Health Organization. Therefore, the findings of this study emphasize that, despite its low tracking, physical activity and sport participation undertaken from early ages predicts physical activity later in life. The present findings confirm previous reviews indicating that the tracking of physical activity from childhood to early adulthood has been consistently low or non-significant (10,11). Other studies have investigated trajectories of physical activity (22) or sports participation (23,24) over time using accelerometry assessments, which corroborate our findings, particularly in females (23). Overall, our findings suggest that the relationship between physical activity and sports participation during childhood or adolescence and their persistence into adulthood is not always strong. While the habit formation hypothesis implies that some physical behaviors become automatic with repetition, the observed low continuity suggests other factors also influence physical behavior across time (25).

Indeed, it appears that physical inactivity is more stable than regular physical activity (26), which could explain its higher frequency over time (10). The self-selection hypothesis emphasizes the influence of genetic predisposition toward physical activity across the lifespan (27,28), although lifestyle changes and external factors may also seem to play a more prominent role. This fact becomes evident when analyzing the tracking of physical fitness from childhood and adolescence to adulthood (29), where a moderate tracking is observed over the years. Although influenced by both genetic components and behaviors, the former carries more weight than in physical activity.

Despite our results showing a slightly higher correlation coefficient among males compared to females, these differences are not significant. This confirms the contradictory findings reported in the existing literature, as highlighted in the review by Hayes et al. (9), which included the study conducted by Rauner et al. (30) as part of its meta-analysis. This particular study incorporated representative longitudinal data on physical activity from 947 German adolescents, collected using the Motorik-Modul (MoMo) physical activity questionnaire, and revealed inconsistent tracking results by sex across various physical activity indices. On the contrary, the study published by Fortier et al. (31) among Canadian population indicates sex differences in early adulthood, with males demonstrating greater tracking in physical activity levels than females. These discrepancies could reflect different factors such as cultural differences (12), varied assessments, and/or diverse follow-up protocols. Despite these differences, our work seems to indicate that overall, both for physical activity and sports participation, tracking is similar in both sexes.

Confirming other narrative studies (9-11), our meta-analysis also shows that tracking correlation tends to decrease as the follow-up period increases over time. This suggests that the longer the follow-up period, the lower pronounced or significant the observed association becomes. However, although this moderating effect was statistically significant, the effect was very small, indicating that the practical impact of follow-up duration on the strength of the association may be limited. As expected, there was a general decrease in activity levels observed throughout the measurement periods, a trend that is further corroborated when analyzing the differences between studies that examine activity levels up to young adulthood versus adulthood. This observation is consistent with existing literature indicating a decline in physical activity levels over the course of life, notably during adolescence (7). A study by Telama et al. (32) illustrates this decline clearly. It demonstrates that when participants were initially assessed at age 18 years using the same method to evaluate physical activity in a Finnish population, the correlation between their activity levels and subsequent follow-up time decreased over the years. A similar trend is observed in the work of Anderssen et al. (33) and Herman et al. (34). Therefore, the decrease in the coefficients tracking over the follow-up period appears to follow a linear trend. However, it is crucial to consider the potential impact of life changes during early adulthood, such as entering the workforce, pursuing higher education, or starting a family, which can significantly affect an individual’s physical activity levels. Moreover, as individuals transition through different life stages, such as children leaving the home and the potential increase in free/leisure time, it remains uncertain whether lower levels of physical activity become ingrained, whether individuals return to prior activity levels, or if new patterns of behavior emerge (35). This area warrants further investigation to better understand the long-term dynamics of physical activity across the lifespan.

Finally, another interesting finding from our meta-analysis is that the correlation coefficients are higher for sports participation than for general physical activity. Regarding the tracking and consistency of sports participation over time, a study reveals that certain activities like cycling and strength training tend to display moderately high tracking, while activities such as swimming, team sports, and racket sports exhibit more moderate tracking characteristics (36). It is important to note that in this manuscript, we define “sports” as activities involving structured rules and competition. Therefore, strength training may not traditionally be categorized as a sport, although it is included in the context of physical activities that are often performed in a competitive setting or as part of structured training regimens. Cycling, on the other hand, may be considered a sport depending on the context, whether recreational or competitive. Additionally, the plasticity of physical activity levels is noted, with the level of physical activity displaying greater adaptability compared to fitness (26). This suggests that physical activity levels are more prone to change over time, particularly if interest in voluntary physical activity is lacking during young adulthood, which may persist into later stages of life, particularly among specific demographic groups such as manual laborers, females, and individuals with initially poor perceived health (26). Another possible explanation as to why the tracking of sports participation is higher compared to general physical activity may be because structured and planned activities exhibit better consistency in tracking compared to informal recreational activities (32). Additionally, participation in sporting activities during school age has been identified as a reliable indicator of adult physical activity (37). This emphasizes the significance of regular and intense participation in youth sports for developing motor skills, abilities, attitudes, and motivation, all of which are crucial for maintaining physical activity throughout adulthood (13).

Perspective

The findings of this study provide important insights into the tracking of physical activity and sports participation from childhood to adulthood, contributing to the broader understanding of long-term physical activity behaviors within the field of sports medicine. The low tracking of physical activity and sports participation found in our study highlights the need for early interventions, particularly during adolescence, to improve sustained engagement in physical activity and reduce the risk of inactivity in adulthood.

These findings are aligned with existing literature emphasizing the critical importance of early-life physical activity not only in establishing lifelong habits but also in preventing chronic diseases and promoting overall health. Therefore, our results support the design and implementation of targeted public health strategies and pediatric policies that prioritize physical activity promotion from an early age, especially through school systems and community-based programs.

Furthermore, this evidence reinforces the need for long-term monitoring and multi-level interventions that can adapt to key life transitions (e.g., adolescence, entry into the workforce, parenthood). These transitions present both opportunities and challenges for maintaining physical activity, requiring tailored strategies to overcome age-specific barriers. Socioeconomic factors, access to resources, and cultural attitudes can strongly influence the continuity of physical activity across the lifespan, making it crucial to design policies that address these structural and sociocultural barriers.

Moreover, understanding why physical activity tends to be more variable than inactivity is essential in improving long-term engagement. The episodic nature of sports participation, along with shifting personal motivations and social influences, contributes to this variability. Therefore, interventions that are flexible, adaptable, and responsive to life-stage changes are critical in promoting sustained physical activity.

These insights may inform future research directions and clinical practices focused on mitigating the progressive decline in physical activity across the life course. By exploring how different interventions can be tailored to overcome individual, social, and environmental barriers, future studies can further refine our understanding of effective strategies to promote lifelong physical activity.

Limitations

The primary limitation of our research stems from the significant variety in measurement assessment and results concerning physical activity patterns and sport participation among the studies included, which could explain the heterogeneity observed in the different results. Despite this diversity, the results typically suggest backing for the persistence of both from childhood and adolescence to adulthood. Secondly, another limitation in many tracking studies is the lack of information about the validity and reliability of the assessment of physical activity. Thirdly, most of the included studies evaluated correlation coefficients or similar statistics, such as Spearman’s rank-order correlations, which does not allow for the role of covariates to be considered. Additionally, studies often do not distinguish between organized and non-organized sports participation, grouping self-organized yet physically active individuals with non-participants, which may introduce bias (38). Furthermore, most studies utilized self-reporting techniques, which often overlook intermittent habitual activities like active transportation or periods of sitting. This oversight can result in either underestimation or overestimation of the behaviors being examined. For example, female soccer players have been shown to get approximately 20 minutes of moderate to vigorous physical activity for every hour of game play or practice time in one study (39). In another study, both boys and girls playing in soccer games spent almost 50% of the match time sedentary (40). However, the two studies that included accelerometer (41) or pedometer (42) data confirmed low tracking across time. A significant risk of bias in the included studies is the reliance on retrospective self-reports, which can introduce recall bias, particularly in studies with long-term follow-up periods. Retrospective assessments of physical activity may lead to inaccurate reporting of past behaviors, potentially skewing the results and affecting the validity of the findings. This is an inherent limitation of the data available in physical activity research, as objective measurement tools are less frequently used over extended periods. Another important limitation of our analysis is that most of the included studies did not adjust for confounding factors such as socioeconomic status or life events, which prevents the possibility of conducting a multivariable analysis. Finally, it is important to note that the majority of the studies included in our meta-analysis were conducted in European countries. This limited geographic diversity could influence the generalizability of our findings.

Conclusions

Our meta-analysis uncovers a consistent pattern of low tracking from childhood or adolescence to adulthood in both physical activity and sports participation, with a slightly stronger tracking observed in the latter. Tracking of physical activity across the lifespan can aid in identifying critical periods for targeted interventions aimed at promoting lifelong healthy behaviors and elucidating the pathways through which physical activity in early life influences adult health status. Moving forward, we recommend that future initiatives not only strengthen early interventions but also prioritize programs and policies that encourage sustained physical activity habits in adolescents and adults. This holistic approach can maximize the public health impact of promoting active lifestyles across the lifespan.

Finally, considering the limitations mentioned above, further research is necessary to tackle these issues and enhance our comprehension of the long-term tracking of physical activity and sports participation. Specifically, future studies should prioritize standardizing measurement protocols, enhancing the validity and reliability of assessments, investigating the impact of covariates, and integrating objective measures of physical activity.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the PRISMA reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-89/rc

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-89/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2025-89/coif). A.G.H. serves as an unpaid editorial board member of Translational Pediatrics from August 2023 to July 2025. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Westerbeek H, Eime R. The Physical Activity and Sport Participation Framework-A Policy Model Toward Being Physically Active Across the Lifespan. Front Sports Act Living 2021;3:608593. [Crossref] [PubMed]
2. Bull FC, Al-Ansari SS, Biddle S, et al. World Health Organization 2020 guidelines on physical activity and sedentary behaviour. Br J Sports Med 2020;54:1451-62. [Crossref] [PubMed]
3. Piercy KL, Troiano RP, Ballard RM, et al. The Physical Activity Guidelines for Americans. JAMA 2018;320:2020-8. [Crossref] [PubMed]
4. Thompson AM, Humbert ML, Mirwald RL. A longitudinal study of the impact of childhood and adolescent physical activity experiences on adult physical activity perceptions and behaviors. Qual Health Res 2003;13:358-77. [Crossref] [PubMed]
5. Logan K, Cuff S. COUNCIL ON SPORTS MEDICINE AND FITNESS. Organized Sports for Children, Preadolescents, and Adolescents. Pediatrics 2019;e20190997. [Crossref] [PubMed]
6. Malm C, Jakobsson J, Isaksson A. Physical Activity and Sports-Real Health Benefits: A Review with Insight into the Public Health of Sweden. Sports (Basel) 2019;7:127. [Crossref] [PubMed]
7. Corder K, Winpenny E, Love R, et al. Change in physical activity from adolescence to early adulthood: a systematic review and meta-analysis of longitudinal cohort studies. Br J Sports Med 2019;53:496-503. [Crossref] [PubMed]
8. Lee IM, Shiroma EJ, Lobelo F, et al. Effect of physical inactivity on major non-communicable diseases worldwide: an analysis of burden of disease and life expectancy. Lancet 2012;380:219-29. [Crossref] [PubMed]
9. Hayes G, Dowd KP, MacDonncha C, et al. Tracking of Physical Activity and Sedentary Behavior From Adolescence to Young Adulthood: A Systematic Literature Review. J Adolesc Health 2019;65:446-54. [Crossref] [PubMed]
10. Telama R. Tracking of physical activity from childhood to adulthood: a review. Obes Facts 2009;2:187-95. [Crossref] [PubMed]
11. Evans JMM, Shelia CM, Kirk A, et al. Tracking of physical activity behaviours during childhood, adolescence and young adulthood: a systematic review. J Epidemiol Community Health 2009;63:9. [Crossref]
12. Malina RM. Physical activity and fitness: pathways from childhood to adulthood. Am J Hum Biol 2001;13:162-72. [Crossref] [PubMed]
13. Batista MB, Romanzini CLP, Barbosa CCL, et al. Participation in sports in childhood and adolescence and physical activity in adulthood: A systematic review. J Sports Sci 2019;37:2253-62. [Crossref] [PubMed]
14. Moola S, Munn Z, Sears K, et al. Conducting systematic reviews of association (etiology): The Joanna Briggs Institute's approach. Int J Evid Based Healthc 2015;13:163-9. [Crossref] [PubMed]
15. Nieminen P, Lehtiniemi H, Vähäkangas K, et al. Standardised regression coefficient as an effect size index in summarising findings in epidemiological studies. Epidemiol Biostat Public Health 2013;10: [Crossref]
16. Peterson RA, Brown SP. On the use of beta coefficients in meta-analysis. J Appl Psychol 2005;90:175-81. [Crossref] [PubMed]
17. Bring J. How to standardize regression coefficients. Am Stat 1994;48:209-13. [Crossref]
18. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ 2003;327:557-60. [Crossref] [PubMed]
19. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002;21:1539-58. [Crossref] [PubMed]
20. Furuya-Kanamori L, Barendregt JJ, Doi SAR. A new improved graphical and quantitative method for detecting bias in meta-analysis. Int J Evid Based Healthc 2018;16:195-203. [Crossref] [PubMed]
21. Garcia-Hermoso A, López-Gil JF, Ramírez-Vélez R, et al. Adherence to aerobic and muscle-strengthening activities guidelines: a systematic review and meta-analysis of 3.3 million participants across 32 countries. Br J Sports Med 2023;57:225-9. [Crossref] [PubMed]
22. Kwon S, Janz KF, Letuchy EM, et al. Developmental Trajectories of Physical Activity, Sports, and Television Viewing During Childhood to Young Adulthood: Iowa Bone Development Study. JAMA Pediatr 2015;169:666-72. [Crossref] [PubMed]
23. Yang X, Kukko T, Lounassalo I, et al. Organized Youth Sports Trajectories and Adult Health Outcomes: The Young Finns Study. Am J Prev Med 2022;63:962-70. [Crossref] [PubMed]
24. Scicluna N, Hamer M, Blodgett JM. Associations Between Adolescent Sport and Exercise Participation and Device-Assessed Physical Activity in Adulthood: Evidence From the 1970 British Cohort Study. J Phys Act Health 2023;20:812-22. [Crossref] [PubMed]
25. Verplanken B, Melkevik O. Predicting habit: The case of physical exercise. Psychol Sport Exerc 2008;9:15-26. [Crossref]
26. Hirvensalo M, Lintunen T. Life-course perspective for physical activity and sports participation. Eur Rev Aging Phys Act 2011;8:13-22. [Crossref]
27. Bouchard C, Dionne FT, Simoneau JA, et al. Genetics of aerobic and anaerobic performances. Exerc Sport Sci Rev 1992;20:27-58. [PubMed]
28. Beunen G, Thomis M. Genetic determinants of sports participation and daily physical activity. Int J Obes Relat Metab Disord 1999;23:S55-63. [Crossref] [PubMed]
29. García-Hermoso A, Izquierdo M, Ramírez-Vélez R. Tracking of physical fitness levels from childhood and adolescence to adulthood: a systematic review and meta-analysis. Transl Pediatr 2022;11:474-86. [Crossref] [PubMed]
30. Rauner A, Jekauc D, Mess F, et al. Tracking physical activity in different settings from late childhood to early adulthood in Germany: the MoMo longitudinal study. BMC Public Health 2015;15:391. Erratum in: BMC Public Health 2018;18:217. [Crossref] [PubMed]
31. Fortier MD, Katzmarzyk PT, Malina RM, et al. Seven-year stability of physical activity and musculoskeletal fitness in the Canadian population. Med Sci Sports Exerc 2001;33:1905-11. [Crossref] [PubMed]
32. Telama R, Leskinen E, Yang X. Stability of habitual physical activity and sport participation: a longitudinal tracking study. Scand J Med Sci Sports 1996;6:371-8. [Crossref] [PubMed]
33. Anderssen N, Wold B, Torsheim T. Tracking of physical activity in adolescence. Res Q Exerc Sport 2005;76:119-29. [Crossref] [PubMed]
34. Herman KM, Craig CL, Gauvin L, et al. Tracking of obesity and physical activity from childhood to adulthood: the Physical Activity Longitudinal Study. Int J Pediatr Obes 2009;4:281-8. [Crossref] [PubMed]
35. Corder K, Ogilvie D, van Sluijs EM. Invited commentary: Physical activity over the life course--whose behavior changes, when, and why? Am J Epidemiol 2009;170:1078-81; discussion 1082-3. [Crossref] [PubMed]
36. van der Zee MD, van der Mee D, Bartels M, et al. Tracking of voluntary exercise behaviour over the lifespan. Int J Behav Nutr Phys Act 2019;16:17. [Crossref] [PubMed]
37. Kraut A, Melamed S, Gofer D, et al. Effect of school age sports on leisure time physical activity in adults: The CORDIS Study. Med Sci Sports Exerc 2003;35:2038-42. [Crossref] [PubMed]
38. Suominen TH, Kukko T, Pahkala K, et al. Longitudinal associations of participation in organized and unorganized sports in youth with physical activity in mid-adulthood: The Young Finns Study. J Sports Sci 2024;42:1391-9. [Crossref] [PubMed]
39. Guagliano JM, Rosenkranz RR, Kolt GS. Girls' physical activity levels during organized sports in Australia. Med Sci Sports Exerc 2013;45:116-22. [Crossref] [PubMed]
40. Sacheck JM, Nelson T, Ficker L, et al. Physical activity during soccer and its contribution to physical activity recommendations in normal weight and overweight children. Pediatr Exerc Sci 2011;23:281-92. [Crossref] [PubMed]
41. Husøy A, Kolle E, Steene-Johannessen J, et al. Longitudinal changes in device-measured physical activity from childhood to young adulthood: the PANCS follow-up study. Int J Behav Nutr Phys Act 2024;21:29. [Crossref] [PubMed]
42. Raustorp A, Fröberg A A. 20-Year Follow-Up Study of Objectively Measured Physical Activity. Int J Environ Res Public Health 2021;18:3076. [Crossref] [PubMed]