Family formation patterns and physical health in Australia: a life course approach

Abstract

Listen

Translate article

Social science has long studied the relationships between processes of family formation and health, although empirical patterns in this domain of study are often variable, and their interpretation uncertain. Constrained by lack of suitable data, analytic techniques, and conceptual tools, work in this area has until recently been largely static, focussed on documenting and interpreting cross-sectional associations between occupancy of family roles and health. However, as data and methodology matured and developed economies continued their transition away from the once dominant ‘male breadwinner’ model of the family, more sophisticated approaches that acknowledge the dynamic relationships between family formation and health have become both more feasible and more urgent. This thesis applies a life course approach to the study of family formation and health in the Australian context. The sociological life course tradition highlights the importance of timing, context, and interdependence between multiple role trajectories, while the parallel epidemiological tradition emphasises accumulation of risk and later-life consequences of bodily insults experienced much earlier. Although previous Australian research has investigated associations between family roles and health status, in doing so it has mostly failed to incorporate the insights of life course theory, and there is consequently broad scope for life course research to add new insights. Data for the research are drawn from the Household Income and Labour Dynamics in Australia (HILDA) study (2001-2014), and form the primary material for four empirical chapters. The first empirical chapter utilizes multichannel sequence analysis and growth models to investigate how holistic patterns of family formation (incorporating both partnership and fertility trajectories from ages 18-50) are associated with trajectories of physical health at ages 51 and older. For men, family life course trajectories characterized by early family formation, failure to marry, or early divorce without subsequent remarriage are found to be predictive of poorer health outcomes, while for women only those who experienced high fertility paired with a disrupted marital history were found to be in poorer health than those who experienced a normatively ‘standard’ family life course. The second, third, and fourth empirical chapters focus on different aspects of the relationship between parental age at first birth and later life heath. The first of these contextualizes changes in first birth timing in historical context, presents a descriptive analysis of the characteristics of younger/older first time mothers and fathers, and uses multinomial logistic regression to assess the contribution of early-life family and socio-economic disadvantage, health, and cognitive and non-cognitive skills to first birth timing. Consistent with prior research, background family disadvantage and age at school leaving was found to have strong effects on birth timing, in particular for women. There was also some evidence of health selection into early first birth, with the largest effects again found for women. The third empirical chapter considers the effect of first birth timing on long-term health outcomes (ages 41 and older). In addition to the primary issue of whether there is an effect of birth timing on health, the analysis pays particular attention to several research questions: first, which pre-parenthood factors confound the relationship between birth timing and health? Second, is first birth timing consequential for processes of cumulative advantage in health? Third, what mid-life factors mediate the relationship between first birth timing and health? Fourth, does the strength of the effect of first birth timing depend on early-life disadvantage, education, marital history, parity, or birth cohort? Results indicate persistent positive effects of older first birth timing for both men and women, which are partially mediated by mid-life socio-economic status. For women, the magnitude of the effect increases over time, indicating cumulative advantage, although the analysis suggests that this may not be attributable to birth timing per se. Only scant evidence is found that the relationship depends on other factors, as only remaining never married alters the effect of birth timing for women. The final empirical chapter investigates short term changes across the parenthood transition at different ages, aiming to identify potential mechanisms for the long-run health effects of birth timing. Potential mechanisms considered included health-related behavioural changes (alcohol consumption and physical activity) and indicators of socio-economic hardship. Results indicated no support for health-related behaviour change or change in the likelihood of socio-economic hardship as mechanisms, as changes at the time of the parenthood transition and in the years thereafter did not differ depending on age at first birth. Younger parents do experience potentially harmful changes in alcohol use and socio-economic hardship before parenthood, suggesting that these factors may be part of the reason why young parents tend to be in poorer health in later life, however it is unlikely that alcohol use or socio-economic hardship form part of a causal pathway from fertility timing to later-life health status. Implications for the broader literature and directions for future research are discussed in the final chapter.