Open Access Open Badges Research article

From childhood socio-economic position to adult educational level – do health behaviours in adolescence matter? A longitudinal study

Leena Kristiina Koivusilta12*, Patrick West3, Vesa Markus Antero Saaristo4, Tapio Nummi5 and Arja Hannele Rimpelä6

Author affiliations

1 University Consortium of Seinäjoki, Seinäjoki Technology and Innovation Centre,, University of Tampere, School of Health Sciences, Kampusranta 9C, Seinäjoki FIN-60320, Finland

2 Department of Social Research, University of Turku, Turku, Finland

3 MRC Social & Public Health Sciences Unit, Glasgow, Great Britain

4 National Institute for Health and Welfare, Finn-Medi 3, Tampere, Finland

5 School of Health Sciences, University of Tampere, Tampere, Finland

6 Department of Adolescent Psychiatry, Tampere University Hospital, Tampere, Finland

For all author emails, please log on.

Citation and License

BMC Public Health 2013, 13:711  doi:10.1186/1471-2458-13-711

Published: 2 August 2013



Our interest was in how health behaviours in early and late adolescence are related to educational level in adulthood. The main focus was in the interplay between school career and health behaviours in adolescence. Our conceptual model included school career and health-compromising (HCB) and health-enhancing (HEB) behaviours as well as family background. Two hypotheses were tested: 1) the primary role of school career in shaping educational level in adulthood (an unsuccessful school career in adolescence leads to HCB and not adopting HEB and to low educational level in adulthood); 2) the primary role of health behaviours (HCB and not adopting HEB in adolescence leads to a school career with low education in adulthood).


Mailed surveys to 12 to18 year-old Finns in 1981–1991 (N=15,167, response rate 82%) were individually linked with the Register of Completed Education and Degrees (28 to 32-year-olds). We applied structural equation modeling to study relations of latent variables (family SEP, family structure, school career, HCB, HEB) in adolescence, to the educational level in adulthood.


Standardized regression coefficients between school career and health behaviours were equally strong whether the direction was from school career to HEB (0.21-0.28 for 12–14 years; 0.38-0.40 for 16–18 years) or from HEB to school career (0.21-0.22; 0.28-0.29); and correspondingly from school career to HCB (0.23-0.31; 0.31-0.32) or from HCB to school career (0.20-0.24; 0.22-0.22). The effect of family background on adult level of education operated mainly through school career. Only a weak pathway which did not go through school career was observed from behaviours to adult education.


Both hypotheses fitted the data showing a strong mutual interaction of school achievement and adoption of HCB and HEB in early and late adolescence. Both hypotheses acknowledged the crucial role of family background. The pathway from health behaviours in adolescence to adult education runs through school career. The interplay between behaviours and educational pathways in adolescence is suggested as one of the mechanisms leading to health inequalities in adulthood.

Adolescence; Health behaviours; School career; Educational level; Structural equation modeling