How do reproduction, parenting, and health cluster together? Exploring diverging destinies, life histories and weathering in two UK cohort studies.

Abstract

Life history theory researchers often assume reproductive, parenting and health behaviours pattern across a fast-slow continuum, with 'fast' life histories (typified by short lifespans, early maturation and investing in quantity over quality of children) favoured in poor quality environments and/or when resources are scarce. Some researchers further reduce this down to a simplistic 'fast' versus 'slow' dichotomy. Some of these ideas, with different theoretical motivations, are echoed in the 'diverging destinies' and 'weathering' frameworks developed in the social sciences. Whether clustering of reproductive, parenting and health traits exists has rarely been empirically tested, however. Using latent class analysis on data on mothers from the UK's Millennium Cohort (MCS) and Born in Bradford (BiB) studies, we explored whether reproduction and parenting traits clustered into 'diverging destinies', whether 'weathering' effects tied together health and reproduction, and whether all three domains were combined into either 'fast' vs 'slow' life histories, or into three groups more indicative of a fast-slow continuum. We leveraged ethnic diversity in these samples to examine four groups of mothers separately: 1. MCS White British/Irish (n = 15,423); 2. MCS Pakistani-origin (n = 923); 3. BiB White British (n = 3937); 4. BiB Pakistani-origin (n = 4351), and explored whether faster 'weathering' was evident amongst Pakistani-origin mothers. Both two and three class models emerged as potential descriptions of latent subgroups, potentially providing support for fast and slow life histories or a continuum of traits. However, response profiles provided only limited support for theoretical predictions of which traits should cluster together, with inconsistent and restricted clustering of traits both within and between the domains of reproduction, parenting, and health. In addition, trait clustering was more pronounced amongst White mothers and we found no clear evidence supporting faster 'weathering' amongst Pakistani-origin mothers; the observed clustering instead suggested that cultural constraints may influence linkages between traits. Our results therefore provide some limited support for models which suggest certain traits cluster together in predictable ways, but it is also clear that theoretical frameworks should not emphasise very rigid clustering of large numbers of traits and should allow for contextual influences on clustering.