The European Journal of Public Health Advance Access originally published online on September 12, 2005
The European Journal of Public Health 2006 16(3):331-335; doi:10.1093/eurpub/cki189
Health-Related Behaviours |
Do adolescent leisure-time physical activities foster health and well-being in adulthood? Evidence from two British birth cohorts
Amanda Sacker and Noriko CableDepartment of Epidemiology and Public Health, University College London, UK
Correspondence: Amanda Sacker, Department of Epidemiology and Public Health, University College London, 1-19 Torrington Place, London WC1E 6BT, UK, tel: +44 0 20 7679 1820, fax: +44 0 20 7813 0242, e-mail: a.sacker{at}ucl.ac.uk
Received May 7, 2005, accepted August 18, 2005
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsStatistical analysisResultsDiscussionReferences |
|---|
Background: Calls for public health initiatives to increase adolescent leisure-time physical activity suggest that increasing activity in this age group will reduce social inequalities in health. While the public health benefits of exercise are undisputed, there is little evidence on its role in reducing health inequalities. The paper examines the extent to which adolescent leisure-time physical activity promotes adult health and well-being and explores whether adolescent leisure-time physical activity can act to reduce health inequalities arising from material deprivation during childhood. Methods: This is a longitudinal study of the 1958 British birth cohort followed from age 16 to age 33 years (N = 15 452) and the 1970 British birth cohort followed to age 30 years (N = 14 018). Adult self-rated general health and Malaise Inventory scores are regressed on self-reports of leisure time physical activity. Analyses are conducted separately for men and women controlling for adolescent body mass index (BMI) and psychosocial problems. Results: There was a consistent relationship between leisure-time physical activity in adolescence and psychological well-being
15 years later for both the cohorts. This relationship was independent of adolescent BMI and psychosocial problems. More physical activity in adolescence predicted better adult self-assessed health in the 1958 cohort only. Leisure-time physical activity did not affect inequalities in health. Conclusions: Policies aimed at increasing participation in leisure-time physical activities in youth may improve population health but are unlikely to prevent the development of social inequalities in health.
Keywords: BCS70, longitudinal, NCDS, physical activity
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsStatistical analysisResultsDiscussionReferences |
|---|
Evidence showing the health benefits of leisure-time physical activity is mounting,1 leading to calls for public health initiatives to increase physical activity.2 A focus on adolescent leisure-time activity is highlighted in several initiatives with the belief that increasing activity in this age group will lead to better population health and to a reduction in social inequalities in health.3,4 However, the evidence base is not as firm as it might be supposed. The aims of this study are to examine whether adolescent leisure-time physical activity promotes adult physical and psychological well-being and to investigate whether adolescent leisure-time physical activity can act to reduce inequalities in health arising from material deprivation during childhood.
The evidence that leisure-time activity during adult life is related to better health is well established. Reports show negative associations with coronary heart disease (CHD) and CHD risk factors,5 cancer morbidity and mortality6 and depression.7,8 The relationship between leisure-time activity and psychological well-being during adolescence is also supported in many studies,9,10 but the evidence in support of its effect on physical health is much weaker.11
Generally, there is an assumption that because adult activity is inversely related to some of the risk factors for poor health,12,13 encouraging physical activity in adolescence will improve adult health. However, a longitudinal study found no relationship between sports participation in adolescence and adult CHD risk factors14 and the direct link between activity in youth and adult health is rarely made.
If exercise is a habit that can be maintained throughout the life course, then catching people young may still be the most effective policy. Research suggests that this is not necessarily the case. Several studies support tracking of habits acquired in adolescence into adulthood, but more have shown that participation in physical activities drops dramatically as young people leave education.15–17 Can policies to increase exercise in childhood have health benefits in later in life? One of our aims is to examine whether activity in adolescence can continue to have positive effects on health and well-being 15 years later. The role of leisure-time activity in reducing health inequalities is less clear. While deprivation may act as a barrier to participation in leisure activities during childhood and adulthood,18–20 the assumption that social inequalities in adult health can be affected by this process has rarely been put to the test.
We examine these issues using data from two British Birth Cohorts, the National Child Development Study (NCDS) and the 1970 British Cohort Study (BCS70) who were interviewed when they were 16 years of age in 1974 and 1986, respectively, and then again in their early 30s in 1991 and 2000. The current study has the potential to illuminate the relationship between social disadvantage and leisure-time physical activity. In 1980, the British Conservative government encouraged local authorities to rationalise their assets. Thousands of playing fields were sold for housing and business. The data from the British cohort studies span this period, allowing for a natural experiment of the effects of reducing provision on the leisure-time physical activity for young men and women.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsStatistical analysisResultsDiscussionReferences |
|---|
Study population
The NCDS is an ongoing longitudinal study targeting British individuals who were born between 3rd and 9th March 1958, while the BCS70 is a study surveying British subjects born between 5th and 11th April 1970.21 NCDS data collected at 16 and 33 years (N = 15 452) and BCS70 data at ages 16 and 30 (N = 14 018) are used for the analysis.
Measures
Material disadvantage at 16 years: A material index is derived from questions regarding the existence of poor housing facilities and financial problems. Responses to items regarding social class, housing tenure, housing type, lack of amenities, receiving supplementary benefit, and experiencing financial difficulty were dichotomised and summed. A score of seven indicates the maximum disadvantage as opposed to a score of zero indicating no material disadvantages.
Leisure-time physical activity at 16 years: In the NCDS, leisure-time physical activity is measured using three questions, on participation in indoor sports, outdoor sports and swimming. Cohort members responding ‘like to but no chance’ are included in ‘never or hardly ever’. The original coding on each item is reversed to give a score from 0 for ‘never or hardly ever’ to 2 for ‘often’, and the sum of the three items used in the analysis (range 0–6). For the BCS70, two questions on playing sports at clubs or centres and playing sports on the street or in a park or playground are used. Items are rated on a 4-point ordinal scale. These are recoded so that zero indicated rare involvement and three indicated being involved more than once a week. Again, a summed score of the items is used with a range from 0 to 6.
Psychological distress in adulthood: This is measured by the 24-item Malaise Inventory.22 A total symptom score, ranging from 0 to 24, is derived from a count of the number of items eliciting a positive response. Psychological distress is indicated by a score >7 (labelled Malaise cases).
Poor self-assessed health in adulthood: Cohort members were asked to rate their overall health on a 4-point scale representing excellent, good, fair or poor health. This is dichotomised into 0 for excellent/good health and 1 for fair/poor health.
Body mass index at 16 years: BMI is calculated from the weight and height measures taken at the school medical examination.
Adolescent psychosocial problems: A modified version of the Rutter (A) Scale, omitting items about sleeping and eating behaviour, was completed by a parent of the cohort member (usually the mother) as part of the home interview.22
|
Statistical analysis |
|---|
|
TopAbstractIntroductionMethodsStatistical analysisResultsDiscussionReferences |
|---|
All modelling is carried out using Stata version 8.2. Five logistic regression models assess the effect of adolescent leisure-time physical activities on adult health and well-being and their ability to ameliorate the effects of material disadvantage on later health and well-being. Model 1 estimates the unadjusted odds of poor self-assessed health or Malaise caseness associated with a unit increase in the physical activities score. Model 2 additionally controls for BMI and adolescent psychosocial problems. Model 3 estimates the unadjusted odds of poor self-assessed health or Malaise caseness associated with a unit increase in the material disadvantage score. Model 4 adds the confounders. Model 5 includes physical activity, material disadvantage, BMI, and psychosocial problems as independent variables predicting poor self-assessed health or Malaise caseness. The models are estimated separately for the NCDS and BCS70 by gender because of known differences in participation in leisure-time physical activity and rates of poor health and well-being in adulthood.
Missing values are filled in by multiple imputation using the MICE programs implemented in Stata by Royston.23 Five replicates of the data were created for the NCDS and ten for BCS70, giving 95% efficiency. The Stata program MICOMBINE is used to calculate the average regression estimates over the set of replicates, adjusting the standard errors according to Rubin's rule. All the details are given in Royston.23 Multiple imputation is recommended for epidemiological studies with extensive missing values, as long as these amount to <60% of the total.24,25 In the present study, the extent of missingness is well within this limit.
|
Results |
|---|
|
TopAbstractIntroductionMethodsStatistical analysisResultsDiscussionReferences |
|---|
Table 1 shows the distribution of the variables averaged over the filled-in datasets. They show that material conditions have improved in the 12 years between the two surveys. The physical activities scales for the NCDS and BCS70 are similar but not directly comparable. Nevertheless, gender differences in leisure-time physical activities have remained proportionally the same. Small increases in mean BMI and Rutter scores are also evident for the more recently born cohort. The BCS70 cohort at the age 30 had slightly higher rates of poor self-assessed health but a much higher proportion of Malaise cases than the NCDS at the age 33. In the NCDS, there was an excess of both poor psychological well-being and health in the females, but in the year 2000, more BCS70 men reported poor self-assessed health than women did, although the traditional female propensity for poor psychological well-being remained. Checks on the distribution of variables in the filled-in datasets against the raw data showed that imputation assigned values in line with known characteristics of responders who dropped out of the survey or refused to complete the questionnaires.
|
Physical activities reduced the likelihood of poor self-assessed health for the earlier born cohort but not the 1970 cohort (see table 2, model 1). The inverse relationship between health and physical activity for the 1958 cohort persisted after controlling for BMI and psychosocial adjustment at the age 16. The association between physical activity and psychological well-being was stronger than that between physical activity and self-assessed health. For both young men and young women in the two cohorts, more adolescent physical activity was related to reduced odds of Malaise caseness in adulthood (table 3). These effects survived adjustments for BMI and psychosocial adjustment in model 2, although they were of marginal significance for BCS70 women.
|
|
Model 3 shows the effect of material disadvantage on poor self-assessed health and Malaise caseness. Men and women in both cohorts had an increased likelihood of poor health in adulthood if they experienced material disadvantage in adolescence. Adjustment for the two confounders attenuated the effect, but it was still statistically significant (see model 4). The final column in tables 2 and 3 shows the effect of material disadvantage on adult health after controlling for physical activity levels and the confounders. Physical activity had no impact on the estimates relating material disadvantage to either health measure. Nevertheless, independent of the negative relationship between disadvantage and health, physical activities continued to be related better self-assessed health and well-being of the NCDS men and women and to the greater well-being of BCS70 men.
To understand the relationship between material disadvantage and physical activity in adolescence, activity was regressed on the material disadvantage score after adjustment for BMI and psychosocial problems. Males in the NCDS took part in more leisure-time activities if they were living in more disadvantaged circumstances (b = 0.07; 95% CI 0.03 to 0.11). In contrast, disadvantaged young women were as likely to participate in physical activities as their more advantaged peers were. Twelve years later there was a general shift, with disadvantaged young women now significantly less likely to participate in physical activities than their more fortunate peers in BCS70 (b = –0.06; 95% CI –0.11 to 0.01), while material disadvantage had no effect on the levels of leisure-time physical activity of young men. However, the significant relationships between disadvantage and physical activity are substantively very weak.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsStatistical analysisResultsDiscussionReferences |
|---|
Based on two cohorts born 12 years apart, we observed a consistent relationship between leisure-time physical activity in adolescence and psychological well-being
15 years later. This relationship was independent of BMI and psychosocial problems. In contrast, the relationship between physical activity in adolescence and self-assessed health was only seen for the earlier born cohort. There was no evidence that leisure-time physical activity reduced the adverse sequelae of a disadvantaged background. The rates of poor psychological well-being in this study differ somewhat from those reported in an earlier paper.26 Here we have used multiple imputation models that included additional variables known to predict missing values. These results suggest that the prevalence of poor psychological well-being in the BCS70 cohort may be downwardly biased when dropout and wave non-response are not taken into account and only respondents with complete histories are used. The results of this study are based on the data from self-reports of both physical activity and later health and thus may be more prone to measurement errors than other more objective measures. However, this is counter-balanced by the use of two large cohorts of fully representative samples of the British population. Our findings may have been different if we had measures of physical fitness instead of leisure-time physical activities. Correlations between measures of physical activity and fitness tend to be modest, with independent benefits from physical activity and physical fitness.27,28 Given the increasing levels of sedentary behaviour in industrialised societies, policies to increase frequency of physical activity of any sort may have greater success at improving population health than those aiming to increase the amount of aerobic exercise.
The development of inequalities in health over the life-course appears to be difficult to be eradicated through the modification of health behaviour, such as leisure-time physical activity. Our results show that even at a time when disadvantaged young men were more likely to engage in leisure-time activities than their more advantaged peer group, the long-term positive effects of physical activity were too small to dent the negative influences of adverse living conditions. However, the public health benefits of greater participation in leisure-time physical activities in adolescence are confirmed. Although only circumstantial, long-term effects on self-assessed health were found at a time before the closure of many school playing fields but not after this policy had been enacted. These findings are consistent with others showing that older adolescents favour formal activities over informal exercise and that boys use school facilities for leisure-time activities more than girls.29,30 Research on ways to encourage young people to exercise has produced conflicting results.31,32 Nevertheless, greater availability of school facilities for out-of-hours use by young people and older adults should be encouraged. It has also been suggested that local opportunities for physical activity may help women in particular to adopt a healthy lifestyle.33 Disadvantage was associated with young women in the 1970 cohort taking part in fewer sessions of leisure-time physical activities, emphasising the importance of providing opportunities for everyone independent of their means.
Key points
|
|
Acknowledgments |
|---|
This work was funded by the UK Economic and Social Research Council Grant No. L326253061. Data from the National Child Development Study and the 1970 British Cohort Study were supplied by the ESRC Data Archive. Those who carried out the original collection and analysis of the data bear no responsibility for its further analysis and interpretation.
|
References |
|---|
|
TopAbstractIntroductionMethodsStatistical analysisResultsDiscussionReferences |
|---|
1 World Health Organization. The World Health Report 2002: reducing risks, promoting healthy life. Geneva: WHO, 2002.
2 World Health Assembly. World Health Assembly Resolution WHA57.17: global strategy on diet, physical activity and health. Geneva: World Health Organization, 2004.
3 Baranowski T, Bar-Or O, Blair S, et al. Centers for Disease Control and Prevention: Guidelines for school and community programs to promote lifelong physical activity among young people. Morb Mortal Wkly Rep 1997;46:1–36.[Medline]
4 American Public Health Association. Encourage healthy behavior by adolescents. Am J Public Health 2001;91:508–10.[Medline]
5 Batty GD. Physical activity and coronary heart disease in older adults: a systematic review of epidemiological studies. Eur J Public Health 2002;12:171–6.
6 Lee IM. Physical activity and cancer prevention—data from epidemiologic studies. Med Sci Sports Exerc 2003;35:1823–7.[CrossRef][ISI][Medline]
7 France C, Lee C, Powers J. Correlates of depressive symptoms in a representative sample of young Australian women. Aust Psychol 2004;39:228–37.[CrossRef]
8 Goodwin RD. Association between physical activity and mental disorders among adults in the United States. Prev Med 2003;36:698–703.[CrossRef][ISI][Medline]
9 Kirkcaldy BD, Shephard RJ, Siefen RG. The relationship between physical activity and self-image and problem behaviour among adolescents. Soc Psychiatry Psychiatr Epidemiol 2002;37:544–50.[CrossRef][ISI][Medline]
10 Motl RW, Birnbaum AS, Kubik MY, Dishman RK. Naturally occurring changes in physical activity are inversely related to depressive symptoms during early adolescence. Psychosom Med 2004;66:336–42.
11 Biddle SJH, Gorely T, Stensel DJ. Health-enhancing physical activity and sedentary behaviour in children and adolescents. J Sports Sci 2004;22: 679–701.[CrossRef][ISI][Medline]
12 Hasselstrom H, Hansen SE, Froberg K, Andersen LB. Physical fitness and physical activity during adolescence as predictors of cardiovascular disease risk in young adulthood. Danish Youth and Sports Study. An eight-year follow-up study. Int J Sports Med 2002;23 (Suppl 1):S27–31.
13 Hernelahti M, Levalahti E, Simonen RL, et al. Relative roles of heredity and physical activity in adolescence and adulthood on blood pressure. J Appl Physiol 2004;97:1046–52.
14 Lefevre J, Philippaerts R, Delvaux K, et al. Relation between cardiovascular risk factors at adult age, and physical activity during youth and adulthood: The Leuven Longitudinal Study on Lifestyle, Fitness and Health. Int J Sports Med 2002;23 (Suppl 1):S32–8.
15 Kelder SH, Perry CL, Klepp KI, Lytle LL. Longitudinal tracking of adolescent smoking, physical activity, and food choice behaviors. Am J Public Health 1994;84:1121–6.
16 Brown WJ, Trost SG. Life transitions and changing physical activity patterns in young women. Am J Prev Med 2003;25:140–3.[CrossRef][ISI][Medline]
17 Twisk J, Kemper H, van Mechelen W, Post G. Tracking of risk factors for coronary heart disease over a 14-year period: a comparison between lifestyle and biologic risk factors with data from the Amsterdam Growth and Health Study. Am J Epidemiol 1997;145:888–98.
18 Droomers M, Schrijvers CTM, Mackenbach JP. Educational level and decreases in leisure time physical activity: predictors from the longitudinal GLOBE study. J Epidemiol Community Health 2001;55:562–8.
19 Huurre T, Aro H, Rahkonen O. Well-being and health behaviour by parental socioeconomic status. A follow-up study of adolescents aged 16 until age 32 years. Soc Psychiat Pscychiatr Epidemiol 2003;38:249–55.
20 Tammelin T, Nayha S, Laitinen J, et al. Physical activity and social status in adolescence as predictors of physical inactivity in adulthood. Prev Med 2003;37:375–81.[CrossRef][Medline]
21 Economic and Social Data Service. http://www.esds.ac.uk Accessed March 3 2005.
22 Rutter M, Tizard J, Whitmore K. Education, health and behaviour. London: Longmans, 1970.
23 Royston P. Multiple imputation of missing values. Stata J 2004;4:227–41.
24 Barzi F, Woodward M. Imputations of missing values in practice: results from imputations of serum cholesterol in 28 cohort studies. Am J Epidemiol 2004;160:34–45.
25 Greenland S, Finkle WD. A critical look at methods for handling missing covariates in epidemiologic regression analyses. Am J Epidemiol 1995;142:1255–64.
26 Sacker A, Wiggins RD. Age-period-cohort effects on inequalities in psychological distress, 1981-2000. Psychol Med 2002;32:977–90.[Medline]
27 Lakka TA, Venalainen JM, Rauramaa R, et al. Relation of leisure-time physical activity and cardiorespiratory fitness to the risk of acute myocardial infarction in men. N Engl J Med 1994;330:1549–54.
28 Fox KR. The influence of physical activity on mental well-being. Public Health Nutr 1999;2:411–8.[Medline]
29 Mota J, Esculcas C. Leisure-time physical activity behavior: structured and unstructured choices according to sex, age, and level of physical activity. Int J Behav Med 2002;9:111–21.[CrossRef][ISI][Medline]
30 McKenzie TL, Marshall SJ, Sallis JF, Conway TL. Leisure-time physical activity in school environments: An observational study using SOPLAY. Prev Med 2000;30:70–7.[CrossRef][ISI][Medline]
31 Taylor WC, Blair SN, Cummings SS, et al. Childhood and adolescent physical activity patterns and adult physical activity. Med Sci Sports Exerc 1999;31:118–23.[ISI][Medline]
32 Trudeau F, Laurencelle L, Shephard RJ. Tracking of physical activity from childhood to adulthood. Med Sci Sports Exerc 2004;36:1937–43.[Medline]
33 Rutten A, Abel T, Kannas L, et al. Self reported physical activity, public health, and perceived environment: results from a comparative European study. J Epidemiol Community Health 2001;55:139–46.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||