THE EFFECTS OF SHARED DECISION MAKING ON CANCER SCREENING: A SYSTEMATIC REVIEW

Wednesday, October 22, 2014
Poster Board # PS4-14

Candidate for the Lee B. Lusted Student Prize Competition

Sarah Lillie, PhD, MPH1, Melissa Partin, PhD1, Nancy Greer, PhD2, Angela Fabbrini, MPH3, Sagar Patel, MD4, Kathryn Rice, MD3 and Timothy Wilt, MD, MPH1, (1)Minneapolis VAHCS Center for Chronic Disease Outcomes Research, Minneapolis, MN, (2)Minneapolis VAHCS Chronic Disease Outcomes Research, Minneapolis, MN, (3)Minneapolis VAHCS Pulmonary Section, Minneapolis, MN, (4)Minneapolis VAHCS Section of General Internal Medicine, Minneapolis, MN
Purpose: To inform the VA Office of Health Promotion and Disease Prevention we conducted a systematic review to examine the effects of cancer screening shared decision making (SDM) interventions.

Method: We searched MEDLINE, CINAHL, PsycINFO, and relevant journals from 1995 through 2013 for randomized controlled trials (RCTs) of cancer screening SDM interventions in adults in clinical settings. We extracted key study population and intervention characteristics and the effect of SDM on three outcomes of interest. Outcomes of interest drawn from the Ottawa Decision Support Framework included: (1) Decision Quality (informed, values-based, patient involvement), (2) Decision Action (screening preference/intention, screening behavior), and (3) Decision Impact (decisional conflict, health services use, decision satisfaction).

Result: We identified twenty-two eligible RCTs evaluating SDM interventions for: breast (k=2), colorectal (k=3), and prostate (k=17) cancer screening. More than half of all SDM interventions included a values clarification exercise (k=15), but few used a theoretical framework (k=7). SDM intervention effects on outcomes varied. For Decision Quality, knowledge was assessed in the majority of studies (n=19), and in all cases SDM increased knowledge. Studies measured values and patient involvement less often (k=5 and k=11, respectively) and found no consistent effect. Eleven studies assessed Decision Action with no consistent SDM effect of SDM. Screening intention was evaluated for breast (k=2) and prostate cancer (k=9). Three studies assessed screening preference, all colorectal cancer. Screening behavior was assessed in 16 studies, ranging from 2 weeks to 1 year. SDM enhanced Decision Impact as measured by a consistent decrease in decisional conflict (k=12). However, SDM had no consistent effect on health services use (k=6) or decision satisfaction (k=2), though outcomes were infrequently reported.  

Conclusion: SDM interventions for cancer screening consistently increase patient knowledge and often decrease decisional conflict. However, there have been varying intervention effects on Decision Action or Decision Impact. Research is needed to guide future SDM intervention development to enhance the effect on decision action and impact.