Cluster randomized controlled trial of decision support for breast cancer chemoprevention, MiCHOICE
Katherine D Crew, Garnet L Anderson, Kathryn B Arnold, Sam Dzingle, Alissa Michel, Michael DeLucie, Cynthia W Law, Sandhya Pruthi, Ana C Sandoval Leon, Rachel Shirley, Maria T Grosse Perdekamp, Sarah V Colonna, Stacy L Krisher, Tari A King, Lisa D Yee, Tarah J Ballinger, Christa Braun-Inglis, Debra A Mangino, Kari B Wisinski, Claudia A DeYoung, Masey Ross, Justin D Floyd, Andrea Kaster, Lindi H VanderWalde, Thomas J Saphner, Corrine Zarwan, Shelly Lo, Cathy Graham, Alison K Conlin, Kathleen Yost Butler, Doreen M Agnese, Cheryl Jernigan, Dawn L Hershman, Marian L Neuhouser, Jason A Zell, Banu Arun, Rita KukafkaAbstract
Introduction
Breast cancer chemoprevention is underutilized among high-risk women. We examined whether decision support could increase informed decision-making.
Methods
We conducted a cluster randomized controlled trial to evaluate the effect of a patient-facing decision aid (RealRisks) and provider tool (BNAV) compared to standard education on the number of women with high-risk breast lesions making informed choices. Questionnaires were administered at baseline, 6 and 12 months. The primary outcome was chemoprevention informed choice (defined as adequate knowledge and attitudes congruent with decision) at 6 months. Secondary endpoints included breast cancer risk perceptions, worry, chemoprevention knowledge, decision conflict, and chemoprevention decision.
Results
Across 31 randomized sites, we enrolled 210 providers, who were primarily breast health specialists. Twenty-four sites (14 intervention, 10 control) enrolled 412 patients. Among 287 patients evaluable for the primary outcome, there were no significant differences in informed choice between the intervention and control arms at 6 months (35% vs 27%, respectively; p = 0.20) and 12 months (37% vs 25%, respectively; p = 0.05). At 6 months, women in the intervention compared to control arm were more likely to have accurate breast cancer risk perceptions (31% vs 21%, respectively; p = 0.03) and adequate chemoprevention knowledge (33% vs 23%, respectively; p = 0.04). At 12 months, 52% of women in the control and 50% in the intervention arm self-reported initiating chemoprevention.
Conclusion
Decision support led to modest improvements in accurate risk perceptions and chemoprevention knowledge, but not informed choice. Relatively high chemoprevention uptake was achieved among women with high-risk breast lesions managed mainly by breast health specialists.
Trial Registration
NCT04496739