O’Flaherty, M., Kalucza, S., & Bon, J. (2021). ‘Does Anyone Suffer from Teenage Motherhood? Mental Health Effects of Teen Motherhood in the UK are Small and Homogenous’, Life Course Centre Working Paper Series, 2021-20. Institute for Social Science Research, The University of Queensland.
Martin O’Flaherty, Sara Kalucza, and Joshua Bon.
Teen motherhood is associated with many poor long-term outcomes, including poorer mental health. Important questions remain, however, about the extent to which these relationships reflect effects of teen motherhood per se, as distinct from high levels of disadvantage experienced before becoming a mother. Moreover, emerging evidence suggests that economic effects of teen motherhood may be largest among the (relatively advantaged) women who are least likely to experience teen motherhood, but no previous work addresses this issue in relation to mental health outcomes.
Drawing on data from the 1970 British Cohort Study, we investigate the effects of becoming a teen mother (in comparison to motherhood aged 20-30) on the mother’s mental health at ages 30, 34, and 42. We apply a flexible Bayesian estimation approach with important technical and substantive advantages. In particular, we are able to identify possible differences in the effects of teen motherhood dependent on a wide array of background characteristics that may not have been suggested by theory.
Our results show that overall effects of teen motherhood on later life mental health are consistently small across different outcome measures and ages. Furthermore, we show that effects of teen motherhood are also largely homogeneous – in addition to there being minimal overall effects, there are also no identifiable subgroups in the data for whom teen motherhood appears to have larger effects on mental health. This contrasts with previous findings that suggest heterogeneous effects of teen motherhood on economic outcomes.
We conclude that associations between teen motherhood and later life mental health reflect primarily selection into teen motherhood among disadvantaged adolescent girls, rather than effects of teen motherhood. This indicates that efforts to prevent teen motherhood are unlikely to yield substantial benefits to mental health. It is, however, likely valuable for services to continue targeting mental health support to teen mothers as a clearly disadvantaged group. We further suggest that our analytic method may be of broad utility to the social sciences, where analysts typically lack prior knowledge of the true relationships between variables, and heterogeneous effects are likely commonplace in many fields of study.