Geospatial hotspots and neighbourhood deprivation associated with unintentional home injuries in British Columbia, Canada

Despite existing injury prevention initiatives, preventable unintentional home injuries remain a significant public health concern in Canada and are often influenced by the social determinants of health. This study identified census subdivision-area-level hotspots of unintentional home injuries resulting in hospitalizations across British Columbia (B.C.), Canada, from 2015 to 2019, and examined their relationship with sociodemographic factors. Unintentional home injury hospitalization data from B.C., Canada (2015-2019) were obtained from the Discharged Abstract Database. These data were then aggregated at the census subdivision (CSD) level and linked to the social profiles of the 2016 for B.C. Spatial autocorrelation and hotspot analysis were performed using Anselin Moran's I statistics. Age-standardized injury rates for each CSD were calculated. Exploratory regression was conducted in ArcGIS Pro to identify the most suitable combination of predictor variables, and the best-fitting model was subsequently retained for geographically weighted regression (GWR) analysis. Between 2015 and 2019, the average age-standardized rate of unintentional home injuries leading to hospitalization in B.C. was 166.6 per 100,000 population. Statistically significant local spatial autocorrelation in age-standardized unintentional home injury hospitalization rates across B.C. was observed, with high-high clusters concentrated in the Lower Mainland. Among five models tested, the most stable, featuring low multicollinearity (Max VIF = 1.19) and a significant Koenker (BP) p value (p < 0.001), included no formal education, after-tax annual household income, and low-rise apartment dwellings. The GWR model explained 77% of the variation in age-standardized unintentional home injury hospitalization rates across 428 CSDs in B.C., with injury risk increasing by 7.9% per 1% rise in population without formal education, decreasing by 1.5% per $1000 increase in after-tax annual household income, and increasing by 8% for every additional 100 low-rise apartment dwellings. These findings identify geographic hotspots of unintentional home injuries and a strong association with key sociodemographic factors, which may serve as a basis for further investigation and inform public health planning and place-based intervention strategies in high-risk communities.