NTRODUCTION: Latinas face disparities in cancer screening rates compared with non-Latina whites. The Tepeyac Project aims to reduce these disparities by using a church-based approach to increase breast cancer screening among Latinas in Colorado. The objective of this study was to compare the effect of two Tepeyac Project interventions on the mammogram rates of Latinas and non-Latina whites enrolled in the Medicaid fee-for-service program. METHODS: Two intervention groups were compared: 209 churches in Colorado that received educational printed materials in Spanish and English (the printed statewide intervention) and four churches in the Denver area that received personalized education from promotoras, or peer counselors (the promotora intervention), in addition to the printed statewide intervention. Biennial Medicaid mammogram claim rates in Colorado before the interventions (1998-1999) and after (2000-2001) were used to compare the effect of the interventions on mammogram use among Latinas and non-Latina whites aged 50 to 64 years who were enrolled in the Medicaid fee-for-service program. Adjusted rates were computed using generalized estimating equations. RESULTS: Small, nonsignificant increases in screening were observed among Latinas exposed to the promotora intervention (from 25% at baseline to 30% at follow-up [P = .30]) as compared with 45% at baseline and 43% at follow-up for the printed statewide intervention (P = .27). Screening among non-Latina whites increased by 6% in the promotora intervention area (from 32% at baseline to 38% at follow-up [P = .40]) and by 3% in the printed statewide intervention (from 41% at baseline to 44% at follow-up [P = .02]). No significant disparities in breast cancer screening were detected between Latinas and non-Latina whites. After adjustment for the confounders by generalized estimating equations, the promotora intervention had a marginally greater impact than the printed statewide intervention in increasing mammogram use among Latinas (generalized estimating equation, P = .07). CONCLUSION: A personalized community-based education was only modestly effective in increasing breast cancer screening among Medicaid-insured Latinas. Education alone may not be the answer for this population. The barriers for these Medicaid enrollees must be investigated so that interventions can be tailored to address their needs.