Partnership, living arrangements, and low birth weight: evidence from a population-based study on Spanish mothers

Background Birth weight is considered a crucial indicator of individual and population health, as it determines a newborn’s growth and development. An extensive body of research has explored various determinants of perinatal health, including the impact of living arrangements. This population-based study analyzes the relationship between mothers’ partnership status and household structure and children’s low birth weights. It addresses two basic research objectives: on one hand, how living/not living in a couple affects birth weight; on the other, how partnership status impact on birthweight when mothers live in extended households with other non-nuclear members. Methods A novel database provided by the Spanish Office for National Statistics (INE), which links the 2011 census with births registered from 2011 to 2015 (sample size 22,433) is used. Llogistic regression models are estimated tto obtain adjusted odds ratios (OR) for the relative effects of living arrangements and other covariates such as characteristics of births and mothers’ socioeconomic profiles, on birth weight. Results Differences in low-birth-weight rates may be attributed to the dissimilar socio-demographic characteristics of the groups of mothers in the different coresidential situations. Although our models revealed that the impact of the covariates on birth weight was similar to that shown by previous studies, this was not the case for the effect of the main explanatory variable. Contrary to expectations, the presence/absence of a male partner in nuclear or in extended households does not reveal significant protection against low birth weight. Children born in households in which the male partner was absent were not more likely to have a low birth weight. On the other hand, analyzing the possible protective effect of extended households, we did not detect significant differences in the likelihood of low birth weight between single mothers without and with non-nuclear coresidents in their households. Conclusions Our analysis provides novel evidence regarding the effect of partnership status and household type on perinatal health in Spain. First, contrary to what has been observed in previous studies in Spain and elsewhere, our study shows that living without a partner has no effect on low birth weight. Second, we reveal that households including non-nuclear coresidents are associated with low birth weight suggesting that even in a basically familist societal context such as the Spanish one, the extended family does not fully protect against poor perinatal outcomes.