METHOD: The study was conducted over a four-year period, 1996 and 2000. Participants were recruited using Computer Assisted Telephone Interviewing (CATI) software and random-digit dialing (RDD). Study eligibility criteria included living in the King/Drew Medical Center service area in Los Angeles, having an operable telephone, being female > or = 40 years old and not having had a screening mammogram in the past year. Four-hundred-thirty respondents were randomly assigned to the intervention and comparison groups. English and Spanish focus-group-tested tailored interventions were administered telephonically by trained interviewers. African Americans and Latinas constituted 83.0% of the sample at assignment and 83.8% at six-month follow-up, which is representative of the study area. RESULTS: The main outcome variable of interest in this study was having a screening mammogram during the time interval between baseline and the six-month follow-up assessment. Multiple logistic regressions that revealed factors predicting the outcome variable included: 1) age (p < or = 0.05, OR=2.22, CI 0.98-5.0); 2) study group (p < or = 0.05, OR=1.76, CI 1.06-2.92); 3) prior mammograms (p < or = 0.05, O0R=2.51, 1.39-4.56); and 4) and knowledge of the age when a woman should begin getting mammograms on a regular basis (p < or = 0.05, OR=0.55, 0.33-0.92). CONCLUSION: Tailored telephone counseling increased the instances of screening mammograms by nearly 8% in the intervention group at follow-up. The results of this study confirm previous findings regarding the impact of structural and behavioral factors related to screening mammography.

This study was conducted to determine the predictors of screening mammography among women 40 years old and older residing in South Central Los Angeles, California. The population is predominately African American and Hispanic. Using Computer Assisted Telephone Interview (CATI) software and the Random Digit Dialing (RDD) method, a 54-item, 20-min questionnaire was administered to 505 women. All interviews were conducted in English or Spanish. The Health Belief Model provided the conceptual framework for the design of the questionnaire. A majority (81.8%) of the participants reported having at least one mammogram in their lifetime, with African Americans reporting the lowest rate (74.7%). Multiple logistic regression analyses found that 4 of the 23 independent variables assessed were predictive of ever users of mammography (p < 0.05), while 6 independent variables were predictive of never users of mammography (p < 0.05). A significant finding of this study is the lower rate of screening mammography utilization in this sample compared to estimates for the general population. The results of this study also suggest that substantial improvements in the rate of screening mammography could be achieved if women in their 40s, who are without health insurance, were referred by their physicians to have affordable mammograms every year or two.