Residential mobility and potentially avoidable hospitalisations in a population-based cohort of New Zealand children

Abstract

Background Residential moves are common in early childhood and associations have been found between residential mobility and adverse child health and well-being outcomes. Although there are studies on potentially avoidable hospitalisations (PAH) in children, few have examined PAH in relation to residential mobility. Our aim, therefore, was to investigate residential mobility and PAH in a population-based cohort of New Zealand children. Methods Using a retrospective cohort design, we analysed data from the Integrated Data Infrastructure for a cohort of 314 283 children born since the start of 2004, who had at least one residential address recorded by 2 years of age. Residential mobility was derived from address data and PAH were determined from hospital discharge data. Results Half of the cohort children (52%) experienced at least one residential move by 2 years of age, and 22% experienced two or more moves. Fifteen per cent of the cohort experienced one or more PAH between 2 and 4 years of age. A linear association between residential mobility and PAH was found (relative risk (RR)=1.18, CI 1.17 to 1.19) and this remained robust when adjusting for several covariates. Sensitivity analyses for ambulatory care sensitive hospitalisations (ACSH) and PAH attributable to the housing/physical environment (PAH-HE) produced results very similar to those for PAH (ACSH: adjusted RR (aRR)=1.10, CI 1.09 to 1.11; PAH-HE: aRR=1.11, CI 1.10 to 1.12). Conclusion This study found a linear association between higher residential mobility and an increased likelihood of PAH in young children. Avenues for further investigation are suggested.