AI-tool for conversation training in health care

Longer description

The focus of the conversation training is on practicing person-centered dialogues in areas that healthcare professionals may find challenging. Based on the results from the project's planning phase, there is potential for later inclusion of students in professional education programs (both undergraduate and advanced/specialist training) as well as care staff within elderly care services. Initially, development will be conducted within three healthcare areas: psychiatric care, palliative care, and primary care.

The development was initially based on ChatGPT, and over time, the technical foundation has been further developed into a prototype, which has shown promising results in initial tests. The current development focuses on enhancing functionality, flexibility, and usability, which includes avatars linked to patient cases, optimization of the avatars' voices, and feedback to healthcare professionals after completing conversations. The possibility of making the tool openly accessible is being explored, with the aim of supporting broader utilization and further development through additional collaborations.

The innovation and research processes are integrated and proceed iteratively. 

The project is structured into three phases:

I. Initial Phase: This phase involves the development of patient cases for conversation training (approximately five per healthcare area), prompts for feedback to participants undergoing conversation training and assessment of the relevance and authenticity of the cases and the conversation training, conducted by an expert group.

II. Feasibility Testing: This phase evaluates the relevance and feasibility of the conversation training tool and the planned evaluation design. Healthcare professionals from psychiatric care, palliative care, and primary care will test the AI conversation training tool for approximately two weeks per group. 

After the training, participants will be invited to a focus group discussion to share their experiences. The collected data will be analyzed using interpretative descriptive methods. To assess feasibility for a subsequent effectiveness evaluation, participants will also complete self-assessment questionnaires, including a scale for self-efficacy (SE12), reflective practice (RPQ), and person-centred practices (PCPI-S). 

III. Evaluation of Training Effects: This phase will assess the effects of the conversation training on healthcare professionals, focusing on their confidence in conducting conversations and their attitudes toward reflection. The specific evaluation design will be developed based on the results from Phase II.