Bidirectional influences between parents and children in smoking behavior: a longitudinal full-family model.

BACKGROUND: The present study investigated longitudinal associations and bidirectional influences between family members in smoking behavior using a longitudinal full-family design. Family systems provide a powerful social context in which modeling and imitation take place. In current literature, however, bidirectional associations between parents and children in smoking behavior are seldom considered.
METHODS: Participants were 426 families, including mothers, fathers, and 2 adolescent children. Associations were assessed over 5 waves with yearly intervals using a cross-lagged model in structural equation modeling.
RESULTS: Findings demonstrate that families resemble an interactive system affording smoking contagion across family members. Results suggest that associations between parents and children are bidirectional, that is, parental smoking behavior influences adolescent smoking behavior and adolescent smoking behavior influences parental smoking behavior. There is insufficient evidence to conclude that longitudinal associations between family members are generally bidirectional, as only unidirectional longitudinal associations were found between between siblings and partners.
CONCLUSIONS: The present study extends previous research on the intergenerational transmission of smoking behavior by demonstrating bidirectional influences between parents and children in smoking behavior. Moreover, the present study suggests that family members may be susceptible to adjust their smoking behavior across time in response to smoking behavior within the family.