Copyright 2009 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2009
To compare the level of sports practice between adolescents with chronic health conditions (CHCs) and control peers and to examine the reasons given by adolescents with CHCs for not practicing any sports in comparison with the control group.
A total of 6790 students (3275 females) aged 16 to 20 years, grouped as adolescents with CHCs (355 females, 354 males) and control peers (2920 females, 3161 males).
Chronic health condition was defined using a noncategorical approach including adolescents with a chronic disease and/or a physical handicap.
Main Outcome Measures
Sports practice, barriers to sports practice among individuals not practicing any sports, and biological, psychological, socioeducative, and physical activity characteristics.
Males with CHCs were less likely than control males to practice sports, whereas no significant difference was observed for females. Chronically ill youth were significantly more likely to report having a CHC as a barrier for not practicing sports. However, the most frequently reported barrier was preference for other activities for males with CHCs and lack of time for control males and for females with and without CHCs.
Having a CHC seems to influence sports practice among males but not females. We recommend that practitioners dealing with adolescents remember to take into account sports practice as part of the care of young patients with CHCs.
Physical activity and sports practice are good for adolescents in terms of both somatic and mental health. There is evidence that regular physical activity decreases the risk of developing cardiovascular diseases, overweight, and type II diabetes and is important for skeletal development in the young, which may also prevent osteoporosis later in life.1,2 A positive effect on self-image and depression has also been described in adolescents.3,4
Sports practice or physical exercise is also an important issue for adolescents with chronic health conditions (CHCs) in terms of improving health in particular diseases5- 8 and on a psychosocial level by enhancing quality of life9 and expanding social networks through sports teams or clubs.10,11
Few articles have been published on sports practice among adolescents with CHCs, and these most often consider physical activity in general rather than sports practice. The literature mainly reports studies using clinical and mostly small samples12,13 or considering single diseases,13- 17 and some do not have a control group or use reports of healthy youths from the literature for comparison.13- 15,18 Moreover, the results are ambiguous overall as most of these studies show a tendency in chronically ill adolescents to be less physically active than their healthy counterparts,13,14,17- 20 whereas some demonstrate no difference15,16 or even higher activity levels than peers.21
Previous studies have shown the influence of several factors on sports practice during adolescence, such as age and sex (sports practice declines with increasing age, especially among females22), being overweight,23 low socioeconomic status,22,24- 26 living in a rural area,22 poor health perception, poor body image, and depression.23 Academic track can also influence sports practice as students practice more sports than apprentices.27 Belonging to a sports club also seems to be linked with higher sports participation.27,28 Furthermore, non-Swiss-born adolescent males seem to practice more sports than Swiss males, contrarily to females.27 However, available literature about sports practice in adolescents with CHCs rarely takes any of these potential confounders into account except for age, sex,12- 14,16,19,20 and sometimes body mass index12,14,20 or parents' education.19
The literature on reasons adolescents do not practice sports is also scarce, particularly among those with CHCs. To our knowledge, the literature about barriers to sports practice in CHCs is limited to a few specific diseases16,29 or to studies carried out among adults.30,31
To fill these gaps, in this study we use a noncategorical approach to CHCs to compare the level of sports practice of adolescents with and without CHCs in a representative national sample. We also examine the reasons given by adolescents with and without CHCs for not practicing sports. We hypothesize the following: (1) adolescents with CHCs practice less sports than their healthy peers; and (2) the main barrier to sports practice reported by adolescents with CHCs is their condition.
Data were collected from the Swiss Multicenter Adolescent Survey on Health 2002 database, a cross-sectional study conducted in Switzerland through a self-administered anonymous questionnaire within a representative national sample of 7548 adolescents aged 16 to 20 years attending postmandatory school. In Switzerland, after ending mandatory school by age 16 years, about one-third of adolescents enter high school (after which most students enter university), about two-thirds are enrolled as apprentices by companies to be trained for their future profession (these apprentices attend class at a vocational school 1 or 2 days per week), and about 10% interrupt or delay their education and therefore were not included in this study. The survey method has been described elsewhere.32 The study protocol was approved by the ethics committee of the University of Lausanne's Medical School.
From the 7548 individuals (3658 females) included in the Swiss Multicenter Adolescent Survey on Health 2002, we excluded 295 individuals with nonresponses to the questions about having a chronic disease and/or a physical handicap and another 94 with nonresponses to the question about sports practice. From the remaining, we also excluded 369 individuals with nonresponses to at least 1 covariable that was significant in a preliminary bivariate analysis. We compared excluded individuals with those included for age, academic track, sports practice, and physical activity by sex. There were no differences except for academic track (more apprentices among males and females excluded, data not shown). We controlled all of the analyses for academic track to reduce this possible bias. Our final sample included 3275 females and 3515 males.
We defined CHC using a noncategorical approach as described by Stein and Jessop33 that considers chronically ill young people as one population rather than specific disease classes. This approach is based on the life experiences shared by young patients with a wide range of chronic health problems regardless of the given diseases. This global approach has been used by other investigators34,35 and can be justified in this study by the numerous publications indicating that different illnesses have a similar negative influence on sports participation.13,14,19,20,36- 38
The CHC group included 355 females and 354 males reporting a chronic disease (“Do you have a chronic disease, that is to say a disease which lasts a long time (at least 6 months) and which may need regular care [eg, diabetes, scoliosis, etc]?”) and/or a physical handicap (“Do you have a physical handicap, that is to say an injury which affects your body integrity and limits its functioning in any way?”). These prevalence figures (females, 10.8%; males, 10.1%) are in agreement with those in the literature.35,39 The control group included all adolescents who answered both questions negatively, that is, 2920 females and 3161 males. To assure the homogeneity of the CHC group, we performed preliminary analysis comparing the 3 possible subgroups (chronic disease, physical handicap, both) for age, academic track, sports practice, and physical activity by sex. The 3 subgroups showed no significant differences between them (data not shown). Additionally, as asthma could represent a large proportion of CHCs, we assessed how many subjects in the CHC group also answered “quite often or very often” to the question “Have you had respiratory problems in the last 12 months?” One hundred eighteen males (33.3%) and 129 females (36.3%) were included in this subgroup. We compared them with the rest of the CHC group for age, academic track, sports practice, and physical activity by sex and found no significant differences between them (data not shown).
Sports practice was assessed by the question “Do you practice any extracurricular sports?” The question had 4 possible answers: no (reference category), about once per week, 2 to 3 times per week, and daily or almost daily.
We took a series of potential confounding factors to this relationship into account, namely biological, psychological, and socioeducative factors as well as the level of physical activity.
Biological factors included age (16-20 years) and being overweight. Overweight was assessed from the body mass index (calculated as weight in kilograms divided by height in meters squared from self-reported data on weight and height) using the cutoffs for age described by Cole et al.40
For psychological factors, health perception was assessed by the question “Do you think your health is excellent, very good, good, mediocre, or poor?” and was dichotomized into good (good to excellent) and poor (mediocre or poor). Body image was assessed by the statement “I am satisfied with my body” (yes or no). Depression was assessed with the Depressive Tendencies Scale with scores graduated from 0 (low) to 3 (high) and including 8 items about feelings of sadness, hopelessness, and unhappiness. This scale has proven to be a valid and reliable instrument.41,42 In this study, Cronbach α was 0.89.
Socioeducative factors included academic track (student or apprentice), nationality (Swiss-born or not), family structure (parents living together or not), place of residence (urban or rural), belonging to a sports club (yes or no), school grades (good or poor), and socioeconomic status. Socioeconomic status was assessed using the parents' education level as a proxy measure. Approximation of socioeconomic status by the parents' education level is supported in several publications.43,44 We dichotomized the mother's and father's education level into low (mandatory school or less) and high (more than mandatory school). We created a new variable called parents' education level that had 3 categories: low education level of both parents, low education level of 1 parent, and high education level of both parents (reference category).
The level of physical activity was evaluated using an ordinal scale from 0 to 7 days per week assessed by the question “During the last week, how many days did you engage in physical activity over a period of at least 20 minutes and making you sweat or breathe heavily?” as described by Sallis et al.45
Finally, those declaring that they were not practicing sports were also asked to answer a specific question about the barriers to sports practice. Twelve possibilities were listed and more than 1 answer was possible: I have an injury or physical handicap; I feel ill at ease with my body; I am not gifted; I do not like sports; I have an irregular work schedule; I already have a physically demanding job; I do not have time; there are not enough facilities or sports clubs; I am lazy; I prefer other activities; my friends do not practice sports either; or other.
All of the analyses were conducted separately by sex because there is evidence that females tend to practice less sports than males.22 In a first step, we performed a bivariate analysis to compare the 2 groups. We used χ2 tests to analyze categorical data and t tests for continuous data. Results are given as prevalences or means with 95% confidence intervals.
All of the significant variables in the bivariate analysis were considered possible explanatory factors in a binary logistic regression comparing the CHC and control groups. We also controlled for age and academic track even if these were not significant in bivariate analyses because these are important potential confounders. Standard procedures for automatic variable selection in regressions such as backward and stepwise selection are known to produce unstable and nonreproducible results,46 so we selected important factors using a bootstrap procedure. Ten thousand bootstrap samples were generated for females and for males, and a backward selection procedure was applied on each of them. Explanatory factors retained in at least 60% of the replications were selected for inclusion in the final model.47 Results are given as adjusted odds ratios (AORs) using the control group as the reference category.
In a second step, we performed analyses using only individuals reporting no sports practice (1076 females [139 with CHCs and 937 control peers] and 745 males [112 with CHCs and 633 control peers]). First, we looked for the prevalence (with 95% confidence interval) of each of the 12 barriers in both groups. To highlight potential confounders, we repeated a bivariate analysis of all variables in step 1 comparing individuals in both groups who reported no sports practice. All of the significant variables in the bivariate analysis were included in logistic regressions analyzing each barrier independently. We did not perform a variable selection in the regressions so that we could use the same model for each barrier. Results are given as AORs using the control group as the reference category.
We used Stata version 9.2 statistical software (StataCorp LP, College Station, Texas) for most analyses because this software allows for computing coefficient estimates and variances taking into account the sampling weights, clustering, and stratification procedure of the Swiss Multicenter Adolescent Survey on Health 2002. We used MATLAB 7 (MathWorks, Inc, Natick, Massachusetts) for the bootstrap variable selection procedure.
Bivariate analysis showed that females with CHCs were significantly more likely than control females to show a poor health perception, a negative body image, and a high depression level. Sports practice did not reach significance at the bivariate level (Table 1) and remained nonsignificant in the multivariate regression taking into account potential confounders (Table 2).
In the bivariate analysis, males with CHCs were significantly less likely than control males to practice sports, to have a high level of physical activity, and to belong to a sports club, and they were more likely to be older, to have a poor health perception, and to have a high depression level (Table 1). In the multivariate regression, males with CHCs were less likely than control males to practice sports daily or almost daily (AOR = 0.56), 2 to 3 times per week (AOR = 0.57), or once per week (AOR = 0.61) after controlling for potential confounders (Table 2).
The bivariate analysis including only individuals not practicing any sports showed that females with CHCs were significantly more likely than control females to have a poor health perception and a negative body image, whereas males with CHCs reported significantly better school grades than control males (Table 3).
As for barriers to sports practice, females with CHCs were significantly more likely than control females to report having an injury or physical handicap, feeling ill at ease with their body, and already having a physically demanding job and less likely to report preferring other activities in the bivariate analysis. When controlling for possible confounding variables, the difference remained significant only for having an injury or physical handicap (AOR = 10.79), having a physically demanding job (AOR = 2.26), and preferring other activities (AOR = 0.62). It is worth noting that, although not significant, the most frequently reported barrier is lack of time for both females with CHCs (54.4%) and control females (61.5%) (Table 4).
Among males, when comparing each barrier individually, the CHC group was significantly more likely than the control group to report having an injury or physical handicap as a barrier to sports practice. This was also the only significant barrier in the multivariate analysis (AOR = 15.61). The most frequently reported barrier was preference for other activities for the CHC group (47.8%) and lack of time for control males (48.7%) (Table 5).
Our findings indicate that contrarily to females, adolescent males with CHCs practice less sports than their peers, confirming our first hypothesis concerning males only. Contrarily to our second hypothesis, CHCs do not appear to be the main barrier to sports practice among adolescents with CHCs.
In our study, CHCs seem to have little influence on sports practice among females. A possible explanation is that there is evidence adolescent females in general practice less sports than males.18,22,24,48- 50 Similarly, our results do not show any negative effect of CHCs on females' physical activity either. On the other hand, males with CHCs in our study have a significantly lower weekly physical activity level compared with their peers. A positive correlation between physical activity and sports practice has already been shown,28 which our findings seem to confirm.
In fact, our results show a negative influence of CHCs on sports practice among males as described in the literature.51 Health conditions probably make it more difficult for youths to practice sports not only because of physical impairment but also because of the requirements of their treatment. Furthermore, sports in late adolescence become very competitive, leaving few opportunities to practice sports in a noncompetitive way.52 Thus, the less competitive youths or those who have less time available for sports practice drop out earlier.52 This might be the case for adolescents having a condition with even little effect on leisure physical activity. Moreover, males who are good at sports can dream of a professional sports career,53 which stimulates them to practice more sports. This is much less the case for chronically ill males because they might be less competitive.
Our findings indicate that having a CHC plays the role of a barrier to sports practice for both males and females with CHCs. However, it is not the most commonly cited barrier among adolescents with CHCs. The most frequently reported barriers among males and females are lack of time and preference for other activities. Nowadays, there is such a wide choice of possible leisure activities that sports practice should become more attractive in competing with such activities. There is evidence that enjoyment of sports and physical activity decreases with age in adolescents.54 Females frequently prefer activities other than sports practice, which is often considered as different from stereotypical feminine concepts in Western societies.49,50 Additionally, high-level sports for women remain insufficiently regarded and socially supported compared with sports for men.49,53 Thus, there is a need to promote the fun and informal side of sports practice, making it more attractive to females as well. Another major problem is raised by the fact that in this study, most adolescents with and without CHCs report not having enough time to include sports practice in their daily life. This finding is supported in the literature28,55 and seems to be a difficult barrier to overcome. Adolescents now have heavy schedules, and unfortunately sports practice is often the first leisure activity to be dropped, especially in the case of poor grades at school when adolescents need to spend more time studying.
Our results also show that significantly more females with CHCs than control females report uneasiness about their body as a reason for not practicing sports. This might be related to their health condition: females with CHCs appear to be less satisfied with their body,56 and in our findings they report a more negative body image. Another significant barrier reported by females with CHCs is having a physically demanding job, although females overall rarely choose physically demanding apprenticeships, contrarily to males. This finding probably indicates that some CHCs make most jobs more tiresome for them. Control females appear to have significantly more preference for other activities, although this is also an issue among females with CHCs, this barrier being reported by 1 in 4 females in the CHC group.
The similarities in barriers to sports practice between adolescents with and without CHCs raised in this study seem to indicate that lifestyles of youths with CHCs and their peers might be more alike than is usually supposed, which has already been described for sexuality or substance use.39
The main strength of our study is that to our knowledge this is the first study comparing sports practice among adolescents with and without CHCs using a nationally representative sample and controlling for potential confounders. This study also has several limitations. First, being a cross-sectional study, causality cannot be assumed. A second limitation is a possible response bias due to self-reported data. The literature suggests that a self-administered anonymous questionnaire encourages respondents to report the truth, especially about sensitive topics.57- 59 Additionally, it is less likely that respondents will intentionally give incorrect answers about sports practice than about more sensitive subjects like sex or drugs. Third, no information about the severity of the conditions was included in the survey. However, the fact that we controlled for health perception should minimize this bias. Fourth, we do not have information about the kinds of sports that youth with CHCs perform, which could be less strenuous than those practiced by adolescents without CHCs. Finally, adolescents with CHCs unable to follow mainstream schooling (ie, mentally handicapped or severely physically disabled individuals) were not included in the survey, although they represent only a small percentage of the CHC population. However, there is a risk of excluding the most extreme cases of adolescents with CHCs, possibly making our findings slightly more optimistic than the reality.
In conclusion, our results indicate that males with CHCs practice significantly less sports than their male control peers. It would therefore be worth encouraging them to participate in sports activities by promoting the practice of sports for fun rather than for competition. On the other hand, having a CHC has little influence on sports practice for females, suggesting that recommendations to these females about sports practice should be the same as for other adolescent females.
From the point of view of health professionals, practitioners dealing with adolescents should remember to encourage sports practice as part of the care of young patients with CHCs at a level appropriate for each specific health problem. Further research should be undertaken to investigate factors that would promote sports practice among adolescent females in general and among males with CHCs in particular.
Correspondence: Joan-Carles Surís, MD, MPH, PhD, Research Group on Adolescent Health, Institute of Social and Preventive Medicine, Bugnon, 17, 1005 Lausanne, Switzerland (email@example.com).
Accepted for Publication: October 15, 2008.
Financial Disclosure: None reported.
Author Contributions:Study concept and design: Pittet, Berchtold, Michaud, and Surís. Acquisition of data: Michaud. Analysis and interpretation of data: Pittet, Berchtold, Akré, and Surís. Drafting of the manuscript: Pittet. Critical revision of the manuscript for important intellectual content: Berchtold, Akré, Michaud, and Surís. Statistical analysis: Pittet and Berchtold. Obtained funding: Michaud. Administrative, technical, and material support: Akré. Study supervision: Surís.
Funding/Support: The Swiss Multicenter Adolescent Survey on Health 2002 was carried out with the financial support of the Swiss Federal Office of Public Health (contract 00.001721/2.24.02.-81) and the participating cantons.
Role of the Sponsor: The funding bodies had no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.
Additional Contributions: The survey was performed within a multicenter, multidisciplinary group from the Institute of Social and Preventive Medicine, University of Lausanne (Françoise Narring, MD, MSc [principal investigator], Véronique Addor, RN, MPH, Chantal Diserens, MA, André Jeannin, MA, Guy van Melle, PhD, and Pierre-André Michaud, MD), the Institute for Psychology, University of Bern (Françoise Alsaker, PhD, Andrea Bütikofer, MA, and Annemarie Tschumper, MD), and the Sezione Sanitaria, Dipartimento della sanità e della socialità, Canton Ticino (Laura Inderwildi Bonivento, MA).
Pittet I, Berchtold A, Akré C, Michaud P, Surís J. Sports Practice Among Adolescents With Chronic Health Conditions. Arch Pediatr Adolesc Med. 2009;163(6):565-571. doi:10.1001/archpediatrics.2009.49