The potential carryover effect of physical activity during childhood and adolescence into adult activity is of major interest in the context of the recent report of the Surgeon General on physical activity and health (21). Study of the effects of childhood and adolescent physical activity per se and its determinants as predictors of adult physical activity is a major research priority (18). However, evidence is limited to a relatively small number of longitudinal studies and generally low inter-age correlations for estimates of physical activity during childhood and adolescence and estimates in adulthood (12).
It is commonly assumed that habits of physical activity developed during childhood and adolescence continue into adulthood (11,13,20). The data, however, are inconsistent with this assumption. Correlations between estimates of physical activity at several ages during adolescence (13-18 yr) and during young adulthood (21-35 yr) are generally low, and as the age interval increases, inter-age correlations decline (2,16,22-24). Nevertheless, youth rated by teachers as having above average ability in sport at 13 yr of age and as extremely energetic at 15 yr of age had a greater likelihood of participating in active leisure pursuits at 36 yr of age (10). Across a broader age span, physical activity during the second and third decades of life was not related to indicators of physical activity and performance at 76 yr of age (8).
What is often lacking in the preceding studies of inter-age relationships is a distinction between childhood (preteens) and adolescent (teenage) experiences in specific physical activities, e.g., individual versus team sports, factors related to activity pursuits, and perceptions of physical activities. In the domain of specific health habits (e.g., smoking and food habits), some evidence indicates that preteen experiences may be more influential than teenage experiences (6). Also, it is suggested that participation in individual sports, sometimes labeled "lifetime sports," such as swimming, running, and skiing, has more carryover effect for adult physical activity than team sports participation (17). Further, it is common to criticize the focus on competitive sport skills as contrary to health and fitness goals (9,11,20). It is important to distinguish between youth or community sports and highly specialized sport for the elite because sport activities, competitive or recreational, are probably the major form of physical activity for the majority of children and adolescents.
There is a need to better understand the relationship, if any, between patterns of physical activity during childhood and adolescence and habits of physical activity in adulthood in a broader perspective. The purpose of the present study, therefore, is to evaluate the relationship between specific components of physical activity during childhood and adolescence and exercise habits in adulthood. Components of physical activity during youth include preteen and teenage experiences, participation in individual and team sports, and several psychosocial factors. Current adult physical activity habits were operationalized in the context of activity level and physical fitness.
The study was retrospective. Subjects completed questionnaires about their current and historic physical activity. The study was approved by the appropriate committee at the Cooper Institute for Aerobics Research in Dallas, TX. Informed consent was obtained from each subject.
Subjects. Subjects were 105 male volunteers who were clients at the Cooper Clinic. The men were generally middle aged (45.0 ± 7.5, range 32-60 yr), married (85%), well educated (83% completed a college or advanced degree), and were largely of European ancestry (87%). Each subject completed the clinic's medical history questionnaire and several physical fitness tests including components of cardiorespiratory endurance, lower back flexibility, and muscular strength. The inclusion criterion for the present study was the ability to exercise and complete a maximal treadmill test. In addition, the subjects completed a questionnaire dealing with physical activity and related factors during youth.
Medical examination. As a part of an overall medical evaluation, which included height, weight, and several skinfolds, a treadmill stress test was completed. The treadmill test was used as an indicator of current level of physical fitness.
Medical history questionnaire. The form includes a profile of demographic data and personal habits related to weight, diet, alcohol and tobacco use, exercise, and general health and lifestyle satisfaction. Relative to the present study, the subjects were asked to recall their weekly exercise during the past 3 months. Workouts per week, miles per workout, time per mile, or heart rate (HR) during exercise were reported for a variety of activities: walking, jogging/running, treadmill, bicycling, stationary cycling, swimming laps, aerobic dance/floor exercises, racquet sports, and other vigorous sports. Activities were converted to an activity level based on daily kilocalorie expenditure. The calculation included intensity (MET value) multiplied by the frequency and duration of the activity, following the protocol of the American College of Sports Medicine (1). The reliability and validity of the exercise questionnaire has been previously established (5). Also, there was a significant but low correlation between estimated kilocalorie expenditure (activity level) and treadmill time (fitness level) (r = 0.30, P = 0.001).
Childhood and Adolescent Physical Activity Patterns Questionnaire. An instrument was developed to estimate types of physical activity, habitual patterns of activity, and psychosocial factors related to physical activity during childhood (preteens, 6-12 yr) and adolescence (teens, 13-18 yr). Several items were adapted from previously developed instruments (J. F. Sallis, personal communication, 1990). The instrument included prompts to visualize the setting of activities (e.g., home, neighborhood, school, family, and friends) to facilitate recall. For each age period, the frequency of participation (none, infrequently, frequently) was recorded for physical education classes, organized youth sports (e.g., Little League baseball, football, etc.), specific sport lessons (e.g., swimming, tennis, etc.), and informal physical activities (e.g., neighborhood play, backyard badminton, etc.). Subjects were asked to specify whether they participated primarily in individual or team sports in childhood and adolescence. Subjects, also, were asked to rate on a five-point scale their physical activity level compared to age peers during the preteens (1 = much less active, 5 = much more active).
Psychosocial questions were based on Bandura's Social Cognitive Theory (3), which includes two important constructs: motivation (intrinsic and extrinsic) and behavioral capability (skill level). In the context of Social Cognitive Theory, individuals continue to participate in an activity that they enjoy and that they can do well; conversely, individuals discontinue an activity that they dislike and an activity that they feel incompetent or feel pressured to perform (3). From this perspective, the subjects rated how often they were encouraged and forced to be physically active (never, rarely, sometimes, often, very often) and whether they enjoyed (yes, no) each type of activity (physical education, organized sports, sport lessons, informal physical activities) during the preteen and teen years. On seven point scales, the subjects rated their general attitude toward physical activity during the preteens and teens (1 = very unfavorable, 7 = very favorable), and their overall ability in sport and ability in their favorite sport during the preteens and teens (1 = limited, 4 = average, 7 = one of the best).
The Childhood and Adolescent Physical Activity Patterns Questionnaire's 34 items were pretested for clarity and understanding. The internal consistency of the overall scale and all subscales was good. Cronbach alpha scores were ≥ 0.70 (Table 1), which is the criterion for new scales (14). The validity of long-term recall of physical activity has been previously reported (4).
Analysis. The criterion variable was adult level of physical activity (estimated daily kcal expenditure). Preteen and teenage frequency of physical activity, self-rated skill level, individual or team sports participation, general attitude, enjoyment, encouragement, and forced participation were predictor variables. Descriptive statistics and Pearson correlations between each predictor variable and adult level of physical activity were initially calculated. The predictor variables were then regressed on the criterion variable. Since current level of physical activity may be influenced by other factors, several potentially confounding variables, current age, health status, fitness, the body mass index, skinfolds, and level of general health and lifestyle satisfaction (15), were regressed on the criterion variable. Only current fitness (treadmill run time in minutes, β = 16.0 ± 5.1, T = 3.16, P < 0.001) and fatness (sum of skinfolds, β = −1.54 ± 0.62, T = 2.48, P = 0.02) were significantly related, the former positively and the latter negatively, to current level of physical activity. The predictor variables were then regressed on the criterion variable controlling for treadmill run time and sum of skinfolds.
Descriptive statistics for current physical activity and fitness and for childhood and adolescent indicators of physical activity and related factors are summarized in Table 2. The sample is quite heterogeneous in the major variables of the study, indicating a wide range of variation in current physical activity and in activity patterns and related psychosocial factors during childhood and adolescence. Overall, the sample is healthy with a rather small number of health conditions per subject (3.3).
Correlations between current adult physical activity and activity experiences and related psychosocial factors during youth are summarized in Table 3. All correlations are low; two are significant, while three are of borderline significance. Teen skill in physical activity (r = 0.17) and participating in team sports during the preteen years (r = 0.16) are positively related to adult physical activity, but being forced to exercise during the preteen (−0.20) and the teen (−0.15) years and being encouraged to exercise during the preteen years (−0.15) are inversely related to adult physical activity.
Regressions of physical activity experiences and related psychosocial factors during childhood and adolescence on adult physical activity without and with control of potentially confounding variables (treadmill run time and sum of skinfolds) are summarized in Table 4. When potential confounders are not controlled, only the frequency of being forced to exercise during the preteen years is negatively related to adult physical activity (P = 0.04), while teen level of skill in physical activity is positively related to adult physical activity (borderline significance, P = 0.08). When confounding variables related to current physical activity are controlled, only the frequency of being forced to exercise in the preteen years (P = 0.05) and the frequency of being encouraged to exercise during the preteen years (borderline significance, P = 0.06) are related to adult physical activity.
Variance analyses of differences in adult physical activity by type of preteen and teenage participation in organized sports without and with control of potentially confounding variables (treadmill run time and sum of skinfolds) are summarized in Table 5. There is no apparent association between level of physical activity in adulthood and type of sport participation (team, individual, both, neither) during childhood and adolescence.
Since being forced to exercise during the preteen years is inversely associated with adult physical activity (Tables 3 and 4), the association between being forced to exercise and type of sport participation was considered in a post-hoc analysis. Two questions were combined: 1) during your preteen years, how often were you forced to exercise, to be physically active, or to play sports; and 2) during your preteen years, did you primarily participate in team sports or individual sports. Being forced to exercise in the preteens is apparently more related to participation in individual sports than to participation in team sports or both (χ2 = 22.55, P = 0.03). Of the 19 adults who indicated that they participated in individual sports, 11 (58%) were forced to participate in exercise sometimes, often, or very often during their preteen years. In contrast, only 7 of 45 (16%) men who indicated they participated in team sports and 7 of 28 (25%) men who indicated they participated in both team and individual sports were forced to participate in exercise sometimes, often, or very often. Being encouraged to exercise in the preteens is more related to participation in both team and individual sports than to participation in team or individual sports (χ2 = 23.53, P = 0.02). Of the 29 adults who indicated that they participated in team and individual sports, 23 (79%) were encouraged to exercise sometimes, often, or very often during their preteen years. In contrast, only 25 of 46 (54%) men who indicated they participated in team sports and 10 of 19 (53%) men who indicated they participated in individual sports were encouraged to exercise sometimes, often, or very often.
Relationships between physical activity experiences and related psychosocial factors during childhood and adolescence and physical activity (estimated EE) during adulthood were analyzed. Overall, the relationships are weak, but several potentially interesting directions are indicated. Correlations (Table 3) indicate that self-reported teen level skill in physical activity (r = 0.17, P < 0.05) and participation in team sports during the preteen years (r = 0.16) are positively related to adult physical activity; however, being forced to exercise during the preteen (−0.20, P < 0.05) and the teen (−0.15) years and being encouraged to exercise during the preteen years (−0.15) are inversely related to adult physical activity. Results of the regression analyses, first without control of confounders of adult physical activity (fitness-treadmill run time, and fatness-sum of skinfolds) and then with control of the confounders, highlight a significant negative role for being forced to exercise during the preteen years for adult physical activity (Table 4). Only 29 of 102 subjects (28%) indicated that they were sometimes, often, or very often forced to participate in physical activity during the preteen years. This percentage is lower than the 69% of high school boys who, in a recent survey of the determinants of activity, indicated that they were forced to exercise (19). Nevertheless, these observations indicate the potential for establishing negative attitudes toward physical activity by forcing children and adolescents to exercise. The present study also indicates the negative role of being encouraged to exercise during the preteen years (Tables 3 and 4). The results thus seem to emphasize the need to give children a voice or choice in their physical activity and sport participation. Being forced and being encouraged to exercise during childhood may have potentially negative consequences for later activity.
Results of the present study are not consistent with the suggestion that participation in individual sports, sometimes labeled lifetime sports, such as swimming, running, and skiing, has more carryover effect for adult physical activity than team sports participation (17). Being forced to exercise in the preteens was more related to participation in individual sports than to participation in team sports or both. Assuming that the educational background of the subjects reflects the socioeconomic conditions under which they were reared, access to individual sports (tennis, swimming, etc.) was perhaps more readily available, and being active in such sports may have been viewed by their parents as socially desirable.
A potential role for self-rated skill in physical activity during the teenage years in adult physical activity (r = 0.17, P < 0.05) is suggested. This finding is consistent with the observation that a significantly greater percentage of boys rated by their teachers as above average in ability at games at 13 yr of age were in the most active group (participation in sports and recreational activities) at 36 yr of age (54%). In comparison, only 41% of boys rated as average and 35% of boys rated as below average in ability at age 13 were in the most active group at age 36. Generally similar trends, also, were evident in girls, 42%, 33%, and 22%, respectively (10).
Although some associations in the present study indicate an important influence of participation in activity during childhood and adolescence on adult physical activity, the associations are low. They are also similar in magnitude to inter-age correlations between estimates of physical activity in adolescence and adulthood in males and females (12). Correlations between a sport activity score in Belgian males at 13 and 18 yr of age and at 30 and 35 yr of age vary between 0.09 and 0.31 (23). Correlations between weekly habitual physical activity at 13 and 27 and 16 and 27 yr of age are 0.05 and 0.09, respectively, while those for estimated EE in organized sports are 0.09 and 0.16, respectively, in Dutch males (22). Corresponding correlations for Dutch girls are 0.17 and 0.11 between 13 and 27 and 0.16 and 0.13 between 16 and 27 yr of age (22).
In contrast to the weak associations indicated above, recalled preteen and teen frequency and amount of participation in physical activity, enjoyment of activity, and general attitude toward physical activity are not related to current level of physical activity in adulthood. When asked why they participate in sports, responses of children and adolescents are rather straightforward: to have fun, to improve skills and to learn new skills, to be with friends or to make new friends, for thrills and excitement, and to become physically fit (7). They may not perceive sport as physical activity. It may be essential to differentiate between physical activity and sport in future studies. The two are not equivalent. Physical activity can occur in a variety of contexts, one of which is in organized sports. Children and adolescents perhaps do not distinguish among different contexts of physical activity, although most derive a major portion of their daily physical activity from sports participation.
Interpretation of the results of the present study warrants caution. The estimation of historic physical activity during childhood and adolescence is a major challenge in the epidemiology of physical activity. The retrospective design required the subjects to recall physical activity and sport experiences and motivations 20 and more years in the past. Many events have occurred during the intervening period, which may influence earlier perceptions. The extent to which there is differential memory distortion and recall bias related to current physical activity habits is unknown. It is possible that the subjects interpreted being forced to exercise in childhood as a convenient rationale to account for current low levels of physical activity.
The present study is limited to middle aged male European American volunteers from middle class or privileged educational and socioeconomic circumstances. Generalizability to other ethnic, educational, and socioeconomic groups of men and to women is not known, and should be done with caution. Nevertheless, the results indicate the need for further study of the determinants of physical activity in childhood, adolescence, and adulthood, as well as their interrelationships.
1. American College of Sports
Medicine. Guidelines for Exercise Testing and Prescription.
Philadelphia: Lea & Febiger, 1991, pp. 102-107.
2. Andersen, L. B. and J. Haraldsdottir. Tracking of cardiovascular disease risk factors including maximal oxygen uptake and physical activity from late teenage to adulthood: an 8-year follow-up study. J. Intern. Med.
3. Bandura, A. Social Foundations of Thought and Action.
Englewood Cliffs, NJ: Prentice-Hall, Inc., 1986, pp. 18-28, 228-282.
4. Blair, S. N., M. Dowda, R. R. Pate, et al. Reliability of long-term recall of participation in physical activity by middle-aged men and women. Am. J. Epidemiol.
5. Blair, S. N., W. B. Kannel, H. W. Kohl, N. Goodyear, and P. W. F. Wilson. Surrogate measures of physical activity and physical fitness
. Am. J. Epidemiol.
6. Cohen, R. Y., K. D. Brownell, and M. R. J. Felix. Age and sex differences in health habits and beliefs of schoolchildren. Health Psychol.
7. Ewing, M. E. and V. D Seefeldt. American Youth and Sports Participation.
North Palm Beach, FL: Athletic Footwear Association, 1990, pp. 4-5.
8. Frandin, K., D. Mellstrom, V. Sundh, and G. Grimby. A life span perspective on patterns of physical activity and functional performance at the age of 76. Gerontology
9. James, W. P. T. A public health approach to the problem of obesity. Int. J. Obes.
19(Suppl. 3):S37-S45, 1995.
10. Kuh, D. J. L. and C. Cooper. Physical activity at 36 years: patterns and childhood predictors in a longitudinal study. J. Epidemiol. Commun. Health
11. Livingstone, M. B. E. EE and physical activity in relation to fitness in children. Proc. Nutr. Soc.
12. Malina, R. M. Tracking of physical activity and physical fitness
across the lifespan. Res. Q. Exerc. Sport
67(Suppl. 3):S48-S57, 1996.
13. Meredith, C. M. and J. T. Dwyer. Nutrition and exercise
: effects on adolescent health. Ann. Rev. Public Health
14. Nunnally, J. C. and I. H. Bernstein. Psychometric Theory.
3rd Ed. New York: McGraw-Hill, 1994, pp. 264-265.
15. Powell, K. E. and W. Dysinger. Childhood participation in organized school sports
and physical education as precursors of adult
physical activity. Am. J. Prev. Med.
16. Raitakari, O. T., K. V. K. Porkka, S. Taimela, R. Temala, L. Rasanen, and J. S. A. Vikari. Effects of persistent physical activity and inactivity on coronary risk factors in children and young adults. Am. J. Epidemiol.
17. Sallis, J. F. and T. L. McKenzie. Physical educator's role in public health. Res. Q. Exerc. Sport
18. Sallis, J. F., B. G. Simons-Morton, E. J. Stone, et al. Determinants of physical activity and interventions in youth. Med. Sci. Sports Exerc.
24(Suppl. 6):S248-S257, 1992.
19. Sallis, J. F., J. M. Zakarian, M. F. Howell, and C. R. Hofstetter. Ethnic, socioeconomic, and sex differences in physical activity among adolescents. J. Clin. Epidemiol.
20. Simons-Morton, B. G., G. S. Parcel, N. M. O'Hara, S. N. Blair, and R. R. Pate. Health-related physical fitness
in childhood: status and recommendations. Ann. Rev. Public Health
21. U. S. Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General.
Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, 1996, pp. 61-259.
22. van Mechelen, W. and H. C. G. Kemper. Habitual physical activity in longitudinal perspective. In: The Amsterdam Growth Study: A Longitudinal Analysis of Health, Fitness, and Lifestyle.
H. C. G. Kemper (Ed.). Champaign, IL: Human Kinetics, 1995, pp. 135-158.
23. Vanreusel, B., R. Renson, G. Beunen, et al. Adherence to sport from youth to adulthood: a longitudinal study on socialization. In: Youth Sport: A Social Approach.
Brussels: Vrije Universiteit Brussel Press, 1993, pp. 99-109.
24. Vanreusel, B., R. Renson, G. Beunen, et al. Involvement in physical activity from youth to adulthood: a longitudinal analysis. In: World-Wide Variation in Physical Fitness.
Leuven: Institute of Physical Education, Katholieke Universiteit Leuven, 1993, pp. 187-195.