JMIR Publications

ABSTRACT

Background:

Breast cancer surgical and reconstructive decision-making is a complex, preference-sensitive process that requires patients to balance oncologic safety, aesthetic outcomes, recovery burden, and long-term quality of life. Despite the growing emphasis on shared decision making (SDM), existing patient decision aids (PDAs) in breast reconstruction are often static, text-heavy, and insufficiently responsive to individual patient values and emotional needs. Artificial intelligence (AI) offers an opportunity to develop adaptive, patient-centered decision-support tools that integrate clinical evidence, patient narratives, and personalized feedback.

Objective:

This study protocol describes the development and early feasibility testing of an AI-supported, narrative-driven digital decision aid designed to facilitate shared decision-making for patients considering breast cancer surgery and reconstruction.

Methods:

A multiphase mixed-methods design will guide development and preliminary evaluation. Phase I involves qualitative semi-structured interviews with breast cancer patients and clinical stakeholders to identify key informational needs, emotional challenges, and experiential factors influencing decision-making. Interview transcripts will undergo inductive thematic analysis to inform the conceptual framework, content structure, and narrative integration of the decision aid. Phase II is a pilot mixed-methods feasibility study involving 50 patients with early-stage breast cancer considering surgical and reconstructive options. Participants will use the digital decision aid and complete validated measures including the Decisional Conflict Scale and Decision Regret Scale, along with investigator-developed usability and acceptability assessments. Semi-structured exit interviews will provide qualitative feedback on usability and perceived value.

Results:

Findings will inform refinement of the decision aid and guide the design of future effectiveness trials.

Conclusions:

This study outlines a rigorous, stakeholder-informed framework for developing an AI-supported decision aid for breast cancer surgical decision-making. If successful, this approach may enhance shared decision-making and serve as a model for ethically grounded AI-supported decision tools in other preference-sensitive clinical contexts.