Neighbourhood Socioeconomic Status and Childhood Injury In Toronto:

Background: This cross-sectional, ecological, retrospective study identifies the extent to which neighbourhood socioeconomic trends are related to intentional and unintentional child injuries, using Toronto, Ontario, as a case study. Children living in lower socioeconomic status (SES) neighbourhoods have often been found to face a higher injury death and morbidity rate than more well-off children. A likely explanation is an increase in the unequal exposure to injury-promoting environments on the basis of the income polarization (a declining middle income group). However, the strength of the inverse relationship between SES and injury is related to a number of factors, including the SES indicator chosen by the researcher. Hence, a major goal of the present study is to determine whether neighbourhood socioeconomic trends toward income polarization have predictive power in explaining variation in injury rates in young children aged 0-6, over and above more typical measures of SES.

Methods: Census data from 1971 to 2006 was used to determine socioeconomic trends. Specifically, neighbourhoods (census tracts) were divided into three distinct categories based on neighbourhood change in average individual income: neighbourhoods that have been improving, declining, and those displaying mixed trends. This analysis of neighbourhoods was merged with geo-coded hospital-based emergency department data to calculate rates of overall injuries, falls, burns and poisoning to children aged 0-6. The predictive power of neighbourhood socioeconomic trends on injury was compared to more typical measures of socioeconomic status such as income level (high, medium, low), neighbourhood employment rates, education levels, and housing quality from the 2006 census.

Results: Socioeconomic trends contributed significantly to injury outcomes, but the contribution of other neighbourhood disadvantage indicators was higher. Housing in need of repair and individuals with no university degree in a neighbourhood were positively correlated with all four outcomes. A high immigrant population in a neighbourhood was negatively correlated with overall injuries, falls, and poisoning injuries. Neighbourhood socioeconomic trends had slightly more predictive power than the more typical measure of SES (high, medium or low income) for all four outcomes.

Conclusions: Results indicate that socioeconomic trends are a valid predictor of injury rates. Researchers should carefully consider the quality of their socioeconomic status measures when predicting injury outcomes. This study can help health and social service practitioners identify the specific risk factors related to injury in lower-SES communities and develop possible means of prevention.

Acknowledgements: This research would not be possible without the data made available and the assistance provided by the Injury Prevention Research Office, St. Michael's Hospital, Michael Cusimano, MD, PhD, FRCSC, FACS, Director; and the Neighbourhood Change SSHRC CURA at the Cities Centre, University of Toronto, J.D. Hulchanski, PhD, Principal Investigator.