Quantifying the effect of a community-based injury prevention program in Queensland using a generalized estimating equation approach.

OBJECTIVE: To develop a generalized estimating equation (GEE) model of childhood injury rates to quantify the effectiveness of a community-based injury prevention program implemented in 2 communities in Australia, in order to contribute to the discussion of community-based injury prevention program evaluation. DESIGN: An ecological study was conducted comparing injury rates in two intervention communities in rural and remote Queensland, Australia, with those of 16 control regions. A model of childhood injury was built using hospitalization injury rate data from 1 July 1991 to 30 June 2005 and 16 social variables. The model was built using GEE analysis and was used to estimate parameters and to test the effectiveness of the intervention. RESULTS: When social variables were controlled for, the intervention was associated with a decrease of 0.09 injuries/10 000 children aged 0-4 years (95% CI -0.29 to 0.11) in logarithmically transformed injury rates; however, this decrease was not significant (p = 0.36). CONCLUSIONS: The evaluation methods proposed in this study provide a way of determining the effectiveness of a community-based injury prevention program while considering the effect of baseline differences and secular changes in social variables. [ABSTRACT FROM AUTHOR]