Development of a DCE-based classification tree to enhance shared decision-making in patients with advanced prostate cancer

Prostate cancer remains one of the most prevalent cancer types in the United States. The extensive range of available treatment options and diverse care pathways often creates significant uncertainty and anxiety for patients navigating the decision-making process. Each patient, along with their support network, tends to have unique priorities regarding treatment, such as financial stability, alleviating symptoms, or extending survival. There is a need for effective tools to help shared decision-making. The Cancer Health Aid to Manage Preferences and Improve Outcome through Navigation (CHAMPION) trial is a randomized controlled study designed to evaluate whether Community Patient Navigators (CPNs), equipped with mobile health tools, can improve patient confidence and satisfaction during cancer treatment decision-making, and enhance collaboration between patients and healthcare practitioners. Within the CHAMPION trial, a nested study, Cancer Preference Reflection and Elicitation for Family Support (C PREFS), utilizes discrete choice experiments (DCEs) to examine and quantify patient preferences when considering treatment options for advanced prostate cancer. DCEs involve presenting participants with hypothetical scenarios and asking them to indicate their preferred choices. The primary objective of this project is to develop a classification tree, which serves as a model for decision-making, informed by interview data collected from patient-supporter dyads. Advanced prostate cancer patient-supporter dyads were interviewed about disease and treatment history, beliefs about medical treatment, and symptomatic characteristics of the cancer. Patients continue to be recruited at the Johns Hopkins Sidney Kimmel Comprehensive Cancer Center. Eligible patients who consent to participate in the study will complete future visit surveys and DCEs that will be used to validate the classification tree. This study is thus contributing towards development of a classification tree capable of accurately predicting the treatment preferences of advanced prostate cancer patients. The resulting tool has the potential to serve clinicians and patients by supporting shared decision-making processes and ensuring that treatment recommendations are closely aligned with individual patient values and preferences.