Injuries claim the lives of more children each year than the next six leading pediatric disorders combined, and produce injuries that require medical attention for one in three children. In the preschool age group, 91 per cent of these accidents and over one-half the resultant fatalities occur in the home. This paper reports the results of a controlled clinical trial conducted to evaluate the implementation of a health education program intended to reduce the risk of childhood household injuries. The study population was randomly assigned into two demographically comparable groups. Only the experimental group mothers received an educational intervention consisting of a tutorial, home safety-proofing assignments, and follow-up. The homes of the two groups were later assessed for hazards during an unannounced visit by an interviewer who did not know to which group each home belonged. A home safety score mean for the two groups was almost identical. The program stimulated heightened interest and stated intent to improve, but did not result in actual reduction of household hazards. Active health education, as used and evaluated in this study, appears to have limited effectiveness when applied to home safety. Approaches such as "passive" measures may offer greater potential for household injury reduction.