Abstract
BACKGROUND: Childhood family structure is considered to play a role in person's health and welfare. This study investigated the relationships between the longitudinal changes of adult health behaviours and childhood family structure. METHODS: From Northern Finland Birth Cohort 1966 questionnaires, we collected data on childhood family structure at the age of 14 ('two-parent family', 'one parent not living at home/no information on father', and 'father or mother deceased'), and on health behaviours (smoking, alcohol consumption and physical activity status) at the ages of 31 and 46. We used the multinomial logistic regression model to estimate the unadjusted and adjusted associations between childhood family structures and the longitudinal changes between 31 and 46 years of health behaviours (four-category variables). RESULTS: Of the study sample (n = 5431; 55.5% females), 7.1% of the offspring were represented in the 'One parent not living at home/no information on father' subgroup, 6.3% in the 'Father or mother deceased' subgroup and 86.6% in the 'Two-parent family'. 'One parent not living at home/no information on father' offspring were approximately twice as likely to smoke (adjusted OR 2.19, 95% CI 1.70-2.81) and heavily consume alcohol (adjusted OR 1.99, 95% CI 1.25-3.16) at both times in adulthood, relative to not smoking or not heavily consume alcohol, and compared with 'two-parent family' offspring. We found no statistically significant associations between childhood family structure and physical activity status changes in adulthood. CONCLUSIONS: Our findings suggest that the offspring of single-parent families in particular should be supported in early life to diminish their risk of unhealthy behaviours in adulthood.