Examining the impact of a multimedia intervention on decisional conflict and psychological distress among early-stage breast cancer patients: results from a nationwide RCT

Lay Summary Healing Choices is a multimedia software program that provides information and decision-making support for women with early-stage breast cancer. We present the results of a randomized controlled trial that evaluated the impact of Healing Choices, compared with standard of care (National Cancer Institute's standard print material), on decisional conflict and psychological distress. In total, 388 participants (197 in the intervention and 191 in the control group) completed the 2-month post-intervention assessment. Results indicated that Healing Choices did not help with treatment decision-making but was associated with higher levels of psychological distress. Use among women assigned to Healing Choices, however, was low, at 41%. When comparing women who used the program with those who did not, we found that the effect of elevated distress disappeared, while program users felt more support than nonusers during the decision-making process. In the future, interventions such as Healing Choices should be regulated so as not to cause distress via information overload, a focus on monitoring and increasing engagement with the intervention is necessary, and, when engagement is low, as-treated analyses are critical to explore the efficacy of the intervention. According to intention-to-treat analyses, the Healing Choices software did not facilitate treatment decision-making among patients diagnosed with breast cancer and was associated with elevated levels of distress; however, as-treated analyses showed that Healing Choices improved decision support and the effect of elevated distress disappeared. We conducted a nationwide, randomized controlled trial to evaluate the impact of Healing Choices, a novel interactive education and treatment decision program rooted in the self-regulation theory framework, on decisional conflict and psychological distress at 2-month post-intervention in women with early-stage breast cancer. Patients were randomized to receive the National Cancer Institute's standard print material (control) or standard print material plus Healing Choices (the intervention). The final sample at 2-month post-intervention consisted of N = 388 participants (intervention: n = 197; control: n = 191). There were no significant differences in decisional conflict or its subscales; however, psychological distress was higher in the intervention group (16.09 & PLUSMN; 10.25) than in the control group (14.37 & PLUSMN; 8.73) at follow-up, B = 1.88, 95% CI [-0.03, 3.80], t(383) = 1.94, p = .05. Upon further examination, we found that engagement with the intervention was low-41%-prompting as-treated analyses, which showed no difference in distress between users and nonusers and a positive impact of Healing Choices on decisional conflict: decisional support subscale: users (35.36 & PLUSMN; 15.50) versus nonusers (39.67 & PLUSMN; 15.99), B = -4.31 (s.e. = 2.09), p = .04. Multiple recommendations for moving ahead stem from this work: (i) intent-to-treat analyses appeared to cause distress, cautioning against interventions that may lead to information overload; (ii) engagement with the intervention is low and future work needs to focus on increasing engagement and monitoring it throughout the study; and (iii) in studies with low engagement, as-treated analyses are critical.