Abstract

STUDY OBJECTIVE: To assess the influences related to social mobility, particularly health related behaviours, as one potential explanation for the social class variation in health among adults. DESIGN: The study is based on questionnaire data from the Adolescent Health and Lifestyle Surveys of 1985, 1987, and 1989. SETTING: The whole of Finland. PARTICIPANTS: A representative sample of 8355 adolescents. The response rate was 79%. MEASUREMENT AND MAIN RESULTS: The relation between social mobility and health related behaviours among 16 and 18 year old young people was studied. The measure of social mobility was based on a combination of the social class of origin and achieved social position measured by the present educational status, educational attainment, and labour market position. Three mobility groups were constructed: the downwardly mobile, the upwardly mobile and the stable. Health related behaviours in an upwardly or downwardly mobile group were compared with a stable group from the same social class of origin by calculating relative risks (RR). RRs were assessed by calculating age and sex adjusted rate ratios approximating a Mantel-Haenszel estimate. In logistic regression analyses the independent effects of the social class of origin and the achieved social position were investigated. Most of the nine behaviours studied (smoking, alcohol use, heavy intoxication, coffee drinking, tooth brushing, consumption of sweets, lack of physical exercise, choice of bread spread, and consumption of milk) were related to the direction of mobility so that health compromising behaviours were more frequent among downwardly mobile and less frequent among upwardly mobile young people than their stable peers. Achieved social position proved to determine health related behaviours more strongly than class of origin, thus emphasising the way education facilitates both health values and behaviours as well as the future social position. CONCLUSIONS: The close relation between social mobility and health related behaviours is concluded to be a part of an explanation of social class differences in health observed among adults.