ASPO Abstracts
Patient characteristics associated with participation in a web-based decision support trial for women at increased breast cancer risk: ENGAGED 2 trial
Category: Behavioral Science & Health Communication, Early Detection & Risk Prediction
Conference Year: 2018
Abstract Body:
PURPOSE: We are testing a personalized, web-based breast cancer risk decision support tool for women at increased risk. We evaluated patient characteristics associated with trial participation. METHODS: Eligible women are 40-69 years with a recent negative mammogram at Kaiser Permanente Washington. Women’s breast cancer risk is calculated based on risk factors reported at recent mammogram. Eligible women had either an intermediate 5-year risk of invasive breast cancer (1.67%—2.49%) and extremely dense breasts or a high 5-year cancer risk (2.50%) and either heterogeneously dense or extremely dense breasts. We used plain language to ensure all patient content was at a 6th grade reading level. Potential participants are mailed a recruitment letter and contacted via phone by trained interviewers within a few days to assess willingness to participate complete baseline interview. We calculated descriptive frequencies of patient characteristics of women who did and did not participate in the research study. We used multivariable logistic regression to calculate odds ratios (OR) with 95% confidence intervals (CI) for patient characteristics associated with participation adjusted for age, race, breast cancer family history, biopsy history, income, breast cancer risk, mammography facility, education, and menopause status. RESULTS: The study has contacted 2,263 eligible women and 710 (31.3%) have enrolled in the trial. Among participants, 71% were ages 60-69, 73% had >2.50% five-year risk of breast cancer, 48% had no prior breast biopsy, 45% had a first-degree family history of breast cancer, 72% were college graduates, and 95% identified as White. Higher education was statistically significantly associated with study participation; women with some college (OR=1.7, 95%CI 1.1-2.6) or college degree (OR=2.6, 95%CI 1.7-3.9) were more likely to participate than high school educated women. Patient characteristics associated with non-participation included: Asian race (OR=0.3, 95%CI 0.1-0.6) and no prior breast biopsy (OR=0.6, 95%CI 0.4-0.9). DISCUSSION: Use of plain language and representative sampling alone are not sufficient to ensure representative participation. Interpretation of trial results must include consideration of sample generalizability.
Keywords: breast cancer, participation, clinical trial, breast density, risk prediction