Family structure and socio-emotional wellbeing in the early years: a life course approach


  • Anna Pearce UCL Institute of Child Health
  • Steven Hope UCL Institute of Child Health
  • Hannah Lewis UCL Institute of Child Health; Psychosocial and Family Services Great Ormond Street Hospital for Children NHS Foundation Trust
  • Catherine Law UCL Institute of Child Health



family structure, child, socio-emotional behaviour, poverty, life course


Children living in reconstituted and lone parent families are at greater risk of poorer socio-emotional wellbeing than those in couple families. A life course approach can help us to understand the dynamics of family structure and how they influence child wellbeing, through consideration of sensitive and critical periods, accumulation, and trajectories of stability or change. We do this using data on 10,357 children from the UK Millennium Cohort Study from infancy to middle childhood. Family structure (natural couple, lone parent, reconstituted family) was measured at 9 months, 3, 5 and 7 years. We used a structured life course approach to examine how family structure might influence socio-emotional wellbeing throughout childhood. We also considered the role of early-life selection, and cumulative poverty (number of sweeps spent in income poverty). We found no evidence of sensitive or critical periods for exposure to certain family types. A measure capturing trajectories of family structure stability or change was as predictive of socio-emotional wellbeing at age 7 as a saturated measure representing all permutations of family structure over time. Compared to children living in a natural couple family throughout, all other groups were more likely to experience poor socio-emotional wellbeing, although children who were living in a natural couple family which transitioned to a lone parent family had a lower prevalence ratio (PR 1.80 [95% confidence interval: 1.54, 2.10]) than the other trajectory types, such as lone parent family throughout (PR=2.77 [2.34, 3.29]), or a lone parent family which transitioned to a reconstituted family (2.66 [1.99, 3.56]). Number of sweeps spent in a lone parent or reconstituted family was also as predictive of poor socio-emotional wellbeing as the saturated model, with the elevated risk increasing incrementally with every sweep spent in either of these family types (PR=1.86 [1.52, 2.26]) for 1 sweep, rising to 2.87 [2.46, 3.56] for 4). The association between both family structure measures and socio-emotional wellbeing were, in most cases, substantially attenuated after adjustment for early-life selection factors (such as maternal social class and separation of the mother’s parents in childhood) and cumulative poverty throughout childhood. This analysis confirms that policies to provide support to vulnerable families in the early years and to reduce poverty are likely to benefit child wellbeing.  

Author Biography

Anna Pearce, UCL Institute of Child Health