Abstract
Background: Healthcare teams face time constraints to work on projects aimed at tackling healthcare and integrated care challenges, and often receive limited formal training and practice in the use of useful planning tools such as the Theory of Change (ToC), Driver Diagrams (DD), and Logic Models (LM) - our custom Generative Pre-Trained Transformer (GPT) model aims to address this.
Audience: This workshop is designed for healthcare professionals, project managers, program evaluators, quality improvement teams, and healthcare system planners who are directly involved in the design, implementation, and evaluation of healthcare interventions. It will be particularly valuable for teams who wish to streamline their planning processes and utilise the capabilities of Large Language Models (LLMs) to construct ToC, DD, and LM that are tailored to their specific integrated care project needs. Whether participants are new to these planning frameworks or have prior experience, the workshop offers practical insights into accelerating these processes using AI technology.
Approach: The workshop will be structured to provide both theoretical knowledge and hands-on experience with our custom GPT model. The session will be as follows:
15 minutes: Introduction
The workshop will begin with a brief introduction to the common challenges faced by healthcare teams in project planning, particularly the time and effort required to develop well-structured ToC, DD, and LM. The rationale behind the creation of the custom GPT model will be explained, highlighting its potential to reduce planning time and improve quality of project design.
15 minutes: Demonstration Live demonstration of the GPT model with a pre-fabricated use case will be conducted, showing how it can be used to assist in the creation of the ToC, DD, and LM. Participants will see how the GPT model guides users through a series of directed questions to generate ToC, DD, and LM tailored to specific project needs.
20 minutes: Interactive activity
Participants will break into small groups, based on shared healthcare interests or challenges polled during the workshop registration. Each group will use the GPT model to develop their own ToC, DD, or LM based on their healthcare interest or challenge. Participants will direct experience how the GPT model can expedite the planning process and experience the collaborative potential of group model building.
10 minutes: Feedback session
Participants will reconvene and will share their experiences using the GPT model, strengths and areas for improvement of process using the GPT model, as well as reflections on how the planning tools developed during the exercise can be applied to real-world integrated care projects.
5 minutes: Closing
Participants will be encouraged to implement the tools and techniques learned in their own workplaces to drive more effective and efficient healthcare and integrated care project design.
Outcomes: Participants will leave with a practical understanding of using the custom GPT model to streamline planning processes, effectively constructing the ToC, DD and LM that lead to clearer project goals, better collaboration, and more impactful programme outcomes.