Purpose: To examine multistage medical decision-making (DM) and the impact of vicarious experience (VE) on breast and prostate cancer screening decisions. Most real life decisions involve multiple stages, especially medical DM. Researchers have examined single-stage DM but little multistage. Similarly, patients rely heavily on others' experiences (VE), yet it remains understudied.

Methods: We conducted a 3 (Type of VE: Positive, Negative, None) x 2 (Type of DM: Multistage, Single-stage) between-subjects experiment. Positive VE scenario emphasized spirituality and unexpected benefits of cancer. Negative VE scenario emphasized fatalism and cancer death. “None” had no VE. Multistage DM subjects experienced three decisions, whether a hypothetical patient, John (for men)/Mary (for women), (1) should undergo uncertain screening tests (PSA or mammogram), (2) have follow-up biopsies, and (3) after ambiguous results, take an experimental drug to reduce cancer risk. Single-stage decisions presented a summarized version of the information with no sequential DM process. Subjects reported their personal and vicarious experiences with cancer.

Results: 601 patients (298 women; 303 men) were randomly assigned to conditions. Multistage subjects were more likely to believe cancer patients gain a new perspective on life (F=4.01, p<0.05), compared with Single-stage subjects. Subjects in both VE conditions were more likely to recommend not taking the drug compared with No VE (Exp(B)=0.41, p<0.05), and positive VE subjects were 2.5 times more likely to do so than negative VE subjects (p<0.05).

The more actual VE subjects reported and the greater its impact, the more subjects (1) recommended not taking the drug (Exp(B)=0.62, p<0.05); (2) believed, despite everything that happened, John/Mary would still want screening tests (R²=0.06, p<0.05) and not become afraid and avoid future tests (R²=0.03, p<0.05); (3) believed it important to do something to treat cancer (R²=0.06, p<0.05); and (4) were not fatalistic if someone were to be diagnosed with cancer (R²=0.03, p<0.05). There was no effect of personal experience on any factor.

Conclusion: We are the first study to unpack experience and find that (1) experiencing the screening DM process can affect subjects' perceptions, and (2) both the VE experimental manipulation and subjects' actual VE could systematically influence the DM process. These findings suggest that to have truly shared DM, we must understand both patients' experiences and how they acquire and use VE in cancer screening decisions.