OBJECTIVE: Trachoma is one of the leading preventable causes of blindness worldwide. We evaluate the impact of a health education program on the prevalence of active trachoma in children 3 to 9 years old. DESIGN: Community randomized intervention study. Data were collected by means of cross-sectional surveys before and after intervention. PARTICIPANTS: Within 40 rural Ethiopian communities, households were selected at random, and all 3- to 9-year-old children within households were invited for examination. METHODS: Health education messages broadcast by radio were received by all communities. Nongovernmental organization activities to prevent trachoma, based on the SAFE (surgery for trichiasis, antibiotic treatment, face washing, and environmental improvements) strategy, were received by 30 of the 40 communities. Ten of these communities received enhanced educational messages using videos. Cluster summary measures were compared across surveys and intervention arms. Active trachoma at follow-up was modeled using random-effects logistic regression, adjusting for baseline prevalence and study area variability, at the cluster and individual level. MAIN OUTCOME MEASURES: Active trachoma in 3- to 9-year-old children and adult knowledge and behavior related to the nature and transmission of trachoma infection. RESULTS: At baseline, 1410 of 1960 (72%) children examined and, at follow-up, 1289 of 2008 (64%) had active trachoma. The overall reduction in prevalence at cluster level was 8% (95% confidence interval [CI], 4%-12%; P<0.001). There was a statistically significant increase in the awareness of trachoma. After adjustment for area and cluster level baseline prevalence, the odds of active trachoma were reduced in both intervention arms, standard (odds ratio [OR], 0.78; 95% CI, 0.53-1.16) and enhanced (OR, 0.76; 95% CI, 0.48-1.21), compared with the control arm, but not significantly. CONCLUSIONS: Overall, there was a small but statistically significant reduction in the prevalence of active trachoma between surveys, but differences between the 3 intervention arms were not statistically significant. Awareness of trachoma control increased in all communities, but there was little change in behavior associated with the transmission of Chlamydia trachomatis. It is therefore unlikely that observed reductions in active trachoma were solely due to health education.