Objective Sport practice is widely encouraged, both in guidelines and in clinical practice, because of its broad range of positive effects on health. However, very limited evidence directly supports this statement among adolescents and the sport duration that we should recommend remains unknown. We aimed to determine sport durations that were associated with poor well-being.
Methods We conducted a survey including 1245 adolescents (16–20 years) from the general Swiss population. Participants were recruited from various settings (sport centres, peers of sport practicing adolescents, websites) and asked to complete a web-based questionnaire. Weekly sport practice was categorised into four groups: low (0–3.5 h), average (≈ recommended 7 h (3.6–10.5)), high (≈14 h (10.6–17.5)) and very high (>17.5 h). We assessed well-being using the WHO-5 Well-Being Index.
Results Compared with adolescents in the average group, those in the very high group had a higher risk of poor well-being (OR 2.29 (95% CI 1.11 to 4.72)), as did those in the low group (OR 2.33 (1.58 to 3.44)). In contrast, those in the high group had a lower risk of poor well-being than those in the average group (OR 0.46 (0.23 to 0.93)).
Conclusions We found an inverted, U-shaped relationship between weekly sport practice duration and well-being among adolescents. The peak scores of well-being were around 14 h per week of sport practice, corresponding to twice the recommended 7 h. Practicing higher sport durations was an independent risk factor of poor well-being.
- mental health
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
What is already known on this topic
Physicians encourage adolescents to have physical activity so as to improve their health. Guidelines recommend practicing physical activity at least 7 h per week.
Well-being is a positive predictor of health among adolescents and young adults.
Practicing sport more than the recommended 7 h per week is supposed to increase well-being, but very high sport practice may have a deleterious impact.
What this study adds
This study provides evidence of an inverted, U-shaped relationship between weekly sport practice duration and well-being among adolescents.
The peak scores of well-being laid around 14 h per week of sport practice, corresponding to twice the recommended 7 h.
Practicing higher sport durations was an independent risk factor for poor well-being.
Sport practice has a positive impact on physical and mental health.1–3 Physical activity has been associated with positive emotional well-being, reduced depressive, anxiety and stress disorders, and improved self-esteem and cognitive functioning in children and adolescents.1 ,2 Conversely, physical inactivity appears to be associated with psychological disorders.4 Thus, existing guidelines recommend practicing physical activity at least 7 h per week for adolescents.5–7 Practicing more than these recommended 7 h is supposed to increase well-being,4 ,8 which is a positive predictor of a better perceived health among adolescents and of fewer risky health behaviours among young adults.9 However, athletes with prolonged physical activity are more at risk for depression, irritability, anxiety, lack of mental concentration and release of pro-inflammatory cytokines.10 ,11 The aim of this research was to determine the association between sport duration and well-being. We hypothesised in particular that very high sport practice related to lower well-being.
We conducted a survey among 16–20-year-old adolescents living in the French-speaking part of Switzerland between February 2009 and January 2010. This study was part of a larger research project on the use of psychoactive and performance-enhancing substances among both sports-practicing and non-sports-practicing youths in the French-speaking part of Switzerland.12 The Ethics Committee of the University of Lausanne's School of Medicine approved the whole project. We enrolled participants through direct communication, emails, website and social networking services. Using a unique identifier, participating adolescents were also asked to recruit close peers with variable levels of sport practice. We used a pretested, anonymous, web-based questionnaire that included questions about demographics, height and weight, socioeconomic status, sport practice, sport injuries, and well-being, which needed at most 20 min for completion.12 ,13
Well-being was assessed using WHO-5 Well-Being Index, ranging from 0 to 25, for which a score below 13 indicates poor well-being.14 Weekly sport practice was categorised into four groups centred on multiples of the recommended 7 h (n×7 h per week±3.5): low (0–3.5 h), average (≈7; 3.6–10.5 h; reference), high (≈14; 10.6–17.5 h) and very high (>17.5 h). We used the same categorisation of sport practice duration across both genders, as WHO, European and American guidelines support the same duration of sport regardless of gender.5–7 We selected the most salient health variables related to sport and well-being on the basis of associations described in the literature: sex, age, overweight or obese status (defined by age- and sex-specific Body Mass Index (BMI) cut-offs according to the International Obesity Task Force,15 and calculated using self-reported height and weight), socioeconomic status (assessed using the Family Affluence Scale),16 sport type (team vs individual) and sport injury with more than 3 months of sport exemption (yes/no). We computed means for continuous variables, and tested their difference using a one-way analysis of variance test.
We compared adolescents reporting poor well-being (<13 on WHO-5 Index) with those reporting satisfactory well-being (≥13). Using χ2 test, we identified all variables that differed significantly between the two groups (namely sex and well-being), and included these variables in a logistic regression model in addition to the variable of age. We did not compute subgroup analysis in the absence of solid a priori hypotheses of interaction between sport practice duration and other statistically significant variables.17
Out of 1303 participants, 1245 (95.7%) had complete data and were included. Half of the respondents (50.4%) were male, with a mean age of 17.95 years (standard deviation (SD) 1.40), a mean BMI of 21.40 kg/m2 (SD 2.58) with 8.9% presenting as overweight or obese, and a mean WHO-5 Well-Being Index of 17.08 (SD 3.81). Regarding weekly sport practice (table 1), 35.2% of the respondents were categorised in the low group (0–3.5 h per week), 41.5% in the average (around 7 h (3.6–10.5)), 18.5% in the high (around 14 h (10.6–17.5)) and 4.8% in the very high (>17.5 h). The means of WHO-5 Well-Being Index were 15.95 in the low group, 17.58 in the average group, 18.13 in the high group and 17.00 in the very high group. According to sport practice duration, average self-reported well-being varied significantly (one-way analysis of variance: F(1241,3)=13.4, p<0.001), showing an inverted U-shaped relationship (figure 1). Compared with adolescents in the average group, adolescents in the very high group showed a higher risk of poor well-being (adjusted OR (aOR) 2.29, 95% CI 1.11 to 4.72, p=0.03), as did those in the low group (aOR 2.33, 95% CI 1.58 to 3.44, p<0.001). In contrast, those in the high group had a lower risk of reporting a poor well-being (aOR 0.46, 95% CI 0.23 to 0.93, p=0.03). The strength of the association was similar in univariable (data not shown) and multivariable analysis, which suggests the robustness of the results.
This study provides evidence of an inverted U-shaped relationship between weekly sport practice and adolescent well-being. Adolescents in the low and very high groups showed more than a twofold risk increase of reporting poor well-being compared with those practising the recommended 7 h per week. The association between sport duration and well-being showed a dose–response pattern. The peak scores of well-being were around 14 h per week of sport practice, which corresponds to twice the recommended 7 h. However, the association was reversed when practicing for more than 17.5 h per week.
To our knowledge, this is the first study assessing the association between poor well-being and very high sport duration among a general adolescent population. In the absence of randomised trials assessing this specific question, this study provides the best evidence, to date, on the threshold over which sport duration apparently ceases to be a protective factor and becomes an independent risk factor for poor well-being. This question is of major clinical interest for physicians who provide advice and counselling to adolescents about physical activity so as to improve their health.
Physicians may use our results to inform patients of this association: adolescents practising sport in accordance to once or twice the recommended duration have shown higher levels of well-being. In contrast, those practising lower or higher duration have shown poor well-being.
These results are all the more relevant when facing adolescents struggling with poor well-being and who have limited sport practice. Conversely, athletes suffering from poor well-being should be made aware that practicing more than twice the recommended sport duration is actually an independent risk factor of their poorer physical and mental health. Regardless of their decision to pursue their level of practice, these adolescents probably need a supportive and closer follow-up of their health and well-being. Finally, our findings can inform guideline panels who produce recommendations on sport practice for adolescents.
Among its strengths, this study demonstrated the magnitude of the association of weekly sport practice on well-being. Furthermore, it included a large, population-based sample of adolescents. However, the sample was not randomly selected and therefore some elements of selection bias might have impacted on the findings. Other limitations of this study are inherent to its observational design. Data were self-reported, which could result in measurement bias. However, asking the patients to integrate and report information, such as sport practice duration, corresponds to the reality of daily clinical practice during consultation. Finally, the direction of causality remains to be proven: does the level of physical activity (too high or too low) impact on well-being, or rather the other way around? Results from biological studies seem to favour the first direction, as higher levels of pro-inflammatory cytokines have been reported in very high and chronic sport practice, with a negative impact on physical and mental health.10 No randomised trial addressed this specific question. However, a systematic review has shown a positive impact of exercise on children and adolescent self-esteem.18 This body of indirect evidence may suggest a similar relationship between sport duration and well-being.
In conclusion, we found that the sport practice apparently ceased to be a protective factor and became an independent risk factor for poor well-being when practicing more than twice the seven recommended hours per week.
These results highlight the importance for physicians, caring for adolescents, to follow-up their level of sport practice and concurrently inquire about their well-being.
We thank Ms Abha Athale, MSc(c) for her editorial feedback, and the following teams for their help and support: the Adolescent Health Research Group (GRSA) and the Multidisciplinary Medicine and Health Unit for Adolescents (UMSA), University Hospital Center of Vaud (CHUV), Lausanne, Switzerland.
Contributors AM: conception and design, analysis and interpretation of data, drafting the article and final approval of the version to be published. AF and JCS: conception and design, analysis and interpretation of data, revising it critically for important intellectual content and final approval of the version to be published. REB: conception, data collection and interpretation of data, revising it critically for important intellectual content and final approval of the version to be published. PAM: conception and design, interpretation of data, revising it critically for important intellectual content and final approval of the version to be published.
Funding The Swiss National Science Foundation partially supported Dr Merglen (Grant PBGEP3-139828).
Competing interests None.
Patient consent Obtained.
Ethics approval The Ethics Committee of the University of Lausanne's School of Medicine, Lausanne, Switzerland.
Provenance and peer review Not commissioned; externally peer reviewed.