Ecological analyses of the associations between injury risk and socioeconomic status, geography and Aboriginal ethnicity in British Columbia, Canada

BACKGROUND: The current study examines what factors contribute to higher injury risk among Aboriginal peoples, compared to the total British Columbia (BC) population. We explore socioeconomic, geographic, and cultural factors, and combinations of these factors, that contribute to increased injury risk for Aboriginal peoples. This follows from our previously reported findings of improvements in injury risk over time for both the total and Aboriginal populations. DATA AND METHODS: We use provincial population-based linked health care databases of hospital discharge records. We identify three population groups: total BC population, and Aboriginal populations living off-reserve, or on-reserve. For each group we calculate age and gender-standardized relative risks (SRR) of injury-related hospitalization, relative to the total population of BC, for two 5-year time periods (1999–2003, and 2004–2008). We use custom data from the 2001 and 2006 long-form Censuses that described income, education, employment, housing conditions, proportion of urban dwellers, proportion of rural dwellers, and prevalence of Aboriginal ethnicity. We use multivariable linear regression to examine the associations between the census characteristics and SRR of injury. RESULTS: The best-fitting model was an excellent fit (R(2) = 0.905, p < 0.001) among the three population groups within Health Service Delivery Areas of BC. We find indicators in all three categories (socioeconomic, geographic, and cultural) are associated with disparity in injury risk. While the socioeconomic indicators (income, education, housing, employment) were shown to be highly correlated, only living in housing that needs major repair and occupational hazardousness, along with rural residence and Aboriginal ethnicity, remained in the final model. Our data show that cultural density is not associated with injury risk for Aboriginal peoples, and that living off-reserve is associated with reduced injury by improving socioeconomic and geographic conditions (compared to living on-reserve). Finally, our analyses show that Aboriginal status itself is associated with injury risk. CONCLUSIONS: Our findings confirm previous research indicating that geographical differences differentiate injury risk, including for Aboriginal populations, and that socioeconomic determinants are associated with health risks. Our analyses showing that Aboriginal status itself contributes to injury risk is new, but we can only speculate about pathway, and whether the causes are direct or indirect.