JMIR Publications

ABSTRACT

Background:

Effective shared decision making between patients and physicians is crucial for enhancing healthcare quality and reducing medical errors. It is shown in literature that detrimental patient engagement and decision outcomes will be resulted in the absence of an effective means to deliver effective shared decision-making.

Objective:

In this paper, we propose a Collaborative Decision Description Language (CoDel) to support the modelling of shared decision-making between patients and physicians, providing a theoretical foundation to study a variety of shared decision scenarios.

Methods:

The CoDel is built upon an extension to the interaction protocol language of Lightweight Social Calculs (LSC). Our language leverages Speech Acts to define the attitudes towards decision propositions held by shared decision maker as well as their semantic relationships in dialogues. Interactive argumentation among decision makers is supported via encapsulating clinical evidence into each dialogue piece of decision protocols. Moreover, the language supports personalized decision-making whereas the characteristics of persistency, critical thinking, and openness can be exhibited.

Results:

The feasibility of the approach is demonstrated using a case study of shared decision making in the disease domain of atrial fibrillation. Our experimental results shows that an integration of the proposed language with GPT can further empower its capabilities in interactive decision-making with improved interpretability.

Conclusions:

The proposed novel collaborative decision description language of CoDeL can help to enhance doctor-patient shared decision making in a rationale, personalized, and interpretable manner.