Emergency hospital admissions associated with a non-randomised housing intervention meeting national housing quality standards: a longitudinal data linkage study

BACKGROUND: We investigated tenant healthcare utilisation associated with upgrading 8558 council houses to a national quality standard. Homes received multiple internal and external improvements and were analysed using repeated measures of healthcare utilisation. METHODS: The primary outcome was emergency hospital admissions for cardiorespiratory conditions and injuries for residents aged 60 years and over. Secondary outcomes included each of the separate conditions, for tenants aged 60 and over, and for all ages. Council home address and intervention records for eight housing cointerventions were anonymously linked to demographic data, hospital admissions and deaths for individuals in a dynamic cohort. Counts of health events were analysed using multilevel regression models to investigate associations between receipt of each housing improvement, adjusting for potential confounding factors and regional trends. RESULTS: Residents aged 60 years and over living in homes when improvements were made were associated with up to 39% fewer admissions compared with those living in homes that were not upgraded (incidence rate ratio=0.61, 95% CI 0.53 to 0.72). Reduced admissions were associated with electrical systems, windows and doors, wall insulation, and garden paths. There were small non-significant reductions for the primary outcome associated with upgrading heating, adequate loft insulation, new kitchens and new bathrooms. CONCLUSION: Results suggest that hospital admissions can be avoided through improving whole home quality standards. This is the first large-scale longitudinal evaluation of a whole home intervention that has evaluated multiple improvement elements using individual-level objective routine health data.