Behavioural Science 549 Familial Aggregation of Leisure-Time Physical Activity - a Three Generation Study * M. Aarnio ', T. Winter1, U. M. Kujo/02, j. Koprio ' Department of Public Health, University of Helsinki, Finland Unit for Sports and Exercise Medicine, Institute of Biomedicine. University of Helsinki, Finland M. Aarnio, 1 Winter, U. M. Kujala, J. Kapno. Familial Aggregation of Leisure-Time Physical Activity - a Three Generation Study, Int.]. Sports Med., Vol. 18, pp. 549-556.1997. Accepted after revision: March 17,1997 Studies of parental influence on children's physical activ- ity have had different results. Parental effect on physical activ- ity during adolescence is less studied, and three generation studies have not been carried out. The purpose of our study was t o examine intra- and intergenerational associations of lei- sure time physical activity among family members in three gen- erations. Due t o the major changes in society during this time, we also took into consideration the socioeconomic status of the adult subjects. The material consisted of 3254 twins at the age of 16, their parents and grandparents. Twins and their par- ents received a questionnaire in 1991 -1993. which included questions about the health and lifestyle, socioeconomic status and leisure time physical activity. The parents' questionnaire also included questions about thelr own parents' leisure time physical activity and socioeconomic status. Based on these questions adolescents, parents and grandparents were classi- fied into physical activity classes. The socioeconomic classifica- tion of parents and grandparents was based on their occupa- tion. lntrageneratlonal physical activity patterns were signifi- cantly associated with each other. Among adolescents the strongest correlation were between monozygotic boys (0.720) and monozygotic girls (0.638). Physical activity patterns were not associated between generations, but there was a signifi- cant difference between very active and inactive mothers con- cerning their daughters' physical activity. Farmers had the low- est proportion of very active subjects only among the parental generation. Because physical activity patterns do not appear t o be transmitted from one generatlon to the next, it is probable that by constant training and education we can obtain the ben- efits of physical activity. Key words: Leisure-time, physical activity, adolescents, fam- ily. socioeconomic status Introduction Several studies reveal that both parents' exercise patterns and encouragement have an effect on children's exercise behavior (4,8,14,23,24,26,27), According to other studies parental ex- ercise is not associated with a child's exercise habits (7). Paren- tal effect on adolescents' physical activity patterns has been studied less. According to Rossow and Rise (22). the father's physical activity was positively associated with the adoles- cent's physical activity, but mother's physical activity was not. Lau et al. (15) reported no significant association between the mother's and the adolescent's physical activity. The relative magnitude of genetic and environmental compo- nents of aerobic and anaerobic performance have been studied before (5). Aerobic performance seems to have low genetic var- iance, the correlations of vOzmax between parents and chil- dren (16) as well as between monozygotic (MZ) and dizygotic (DZ) twins (5) being low. The results from the latter study sug- gest that genetic effects might be inflated by shared environ- mental conditions. because DZ twins were more similar than normal brothers even though both types of sibpairs have about 50% of their genes identical by descent. Perusse et al. (20) sug- gested that children can acquire from their parents certain customs regarding exercise behaviour and that the propensity toward being spontaneously active could partly be influenced by the genotype. The data also revealed that persons of the same generation, whether genetically related or not, tend to be more similar in their physical activity habits than the per- sons of two different generations. These results suggest that familial resemblance is probably associated with familial en- vironmental factors shared by persons of the same generation. but not transmitted from generation to generation (20). Gottlieb and Chen (8) found that parental exercise had a stronger influence on the frequency of exercise among girls than boys, but according to the srudy of Finnish school chil- dren (27) both parents' sport activities were correlated with boys' sports activities, but not with those of girls. Also accord- ing to Greendorfer and Lewko (9) only the father served as a significant influence on children's sport involvement. Int. J. Sports Med. 18 (1997) 549-556 0 Georg Thieme Verlag Stuttgart - New York * Supported by grants from U. S. P. H. S./AA-8315 Ministry of Education, Finland and the Paavo Nurmi Foundation Downloaded by: Queen's University. Copyrighted material.