Controlled evaluation of a community based injury prevention program in Australia.

OBJECTIVE To evaluate the effects of a community based, all age, all injury prevention program, the Safe Living Program, on injury risk and injury rates. DESIGN A quasiexperimental population based evaluation using an intervention and comparison community design. SETTING The intervention community (Shire of Bulla, n = 37,257) is an outer metropolitan area of Melbourne, Australia. The demographically matched comparison community (Shire of Melton, n=33,592) is located nearby. SUBJECTS AND METHODS The Safe Living Program in the Shire of Bulla targeted injury reduction in all settings with a focus on high risk groups. Strategies included program publicity, education and training, injury hazard reduction, and environmental change. Baseline and follow up measures of program reach, risk factors, and injury rates in both communities were used to evaluate program process, impact, and outcome. RESULTS Increase in program awareness was moderate and similar to other community based programs. The program achieved injury hazard reduction on the road, in schools, and, to a more limited extent, in the home. Other changes in injury risk factors could not necessarily be attributed to the program as similar changes were observed in the comparison community. No significant changes were found in rates of injury deaths, hospitalisations, or emergency department presentations in the Shire of Bulla after six years. Self reported household injuries, mostly minor, were reduced in the intervention community, but had been higher at program launch than in the comparison community. CONCLUSIONS The Safe Living Program was unable to replicate the significant reductions in injuries reported in other community based interventions. Replication of apparently successful community based injury prevention programs in different settings and populations requires evidence based interventions, sustained and effective program penetration, reliable data systems to measure change, at least one control community, and sufficient budget and time for effects to be observable.