Patterns of injury among residential construction workers before and after a safety intervention program, 1994-1998

This study served two purposes. It evaluated the effectiveness of the HomeSafe Pilot Program, a collaborative safety education and training program designed to reduce injuries among residential construction workers in the Denver-Metropolitan region in Colorado. In addition, the study described the patterns and characteristics of injuries among residential construction workers before and after the intervention program. Participating companies provided OSHA 200 logs, first reports of injury, workers' compensation loss runs, and total hours worked for each year of the study. Data from these sources were used to evaluate whether overall and lost workday injury incidence rates declined during the intervention period. Data were analyzed using injury incidence rates per 200,000 hours and Poisson regression to control for the effect of antecedent secular trend. Companies representing approximately 11,810 workers reported 1,971 injuries. The rate of injury pre-HomeSafe was 17.4/200,000 hours and the rate post-HomeSafe was 14.4. Lost workday injury rates were 5.8 pre-HomeSafe and 3.4 post-HomeSafe. Injury incidence rates declined significantly following the HomeSafe; however, this effect was not statistically significant once temporal variation was controlled. The Poisson regression model did, however, indicate positive effects of HomeSafe among specific trade groups, especially among severe injuries. The leading causes of injuries were cuts, punctures and scrapes, falls, and lifting or other movement injuries. The leading types of injury were strains, lacerations, contusions, punctures, foreign bodies, and sprains, and the body parts most commonly injured were the lower back, hand/wrist, ankle/foot, finger, leg/knee, arm/elbow, and eye. Trade specific analysis revealed distinct differences in the causes and types of injuries experienced by the different trade groups. The decline in injury rates following HomeSafe cannot be solely attributed to HomeSafe, however, it is likely that the program had an effect on injury rates, especially among specific trade groups. Programmatic and methodologic limitations were potential contributors to the inconclusive results and further research into the hazards, risk perceptions, and behaviors of residential construction workers is needed. Characteristics of injuries experienced by HomeSafe participants are consistent with those identified by prior research, though further investigation of the specific hazards faced by residential construction workers is warranted.