DO TESTIMONIALS IN A COLON CANCER SCREENING DECISION AID INFLUENCE PATIENT DECISION MAKING?

Purpose: In video-based patients’ decision aids (DAs), filmed “testimonials” are often used to describe: a) the reasons why different individuals make different choices; b) their experiences while undergoing their chosen option; and/or c) their experiences with the outcomes of their chosen option. Our objective was to determine the impact of DA testimonials on patients’ decision to get screened for colon cancer and their subsequent screening activity.

Method: In a single-site, prospective randomized controlled trial, all 49-51 year old patients scheduled for a preventive medicine appointment received, prior to their visit, a video-based DA describing the options (no screen, fecal occult blood test, sigmoidoscopy, colonoscopy). Patients were randomized to receive one of two versions of the decision aid video: the original version that includes testimonials for and against screening and an edited version with these testimonials removed. A questionnaire captured participants’ screening intention before and after watching the video.  A chart review was conducted to determine if patients received colon cancer screening.

Result: From June 2008 – May 2009, 562 DAs were distributed; 225 patients returned the study materials. 142 patients were not eligible due to previous screening activity, resulting in 83 completed questionnaires.  Average age was 50 years; 66% were females. No differences in age, gender, education, or enrollment were observed between the study groups (p ≥ 0.57). Pre- to Post-DA shifts in screening intention. In both groups, a) a smaller proportion (11%) were unsure about their screening intention (versus 27% before; p < 0.001); b) a larger proportion (81%) favored one of the “screening” options (versus 65% before (p < 0.001); and c) a comparable proportion (7%) favored the “no screening” option (versus 8% before; p = 0.65)  Post-DA Screening intention.  After viewing the DA, there were no across-group differences in the proportional distribution of favored options (p = 0.52). Subsequent Screening Behaviors. Chart review revealed similar screening rates for both groups (Original DA: 55%, Edited DA: 54%, p = 0.89).

Conclusion: For these patients, the presence of testimonials for and against screening apparently does not affect viewers’ pre- to post-DA shifts in the decision to get screened, their post-DA intention, or their subsequent screening behaviors.