How can A.I. patients effectively develop pharmacy students’ consultation skills?
Billy Smith
Keywords- Artificial intelligence; Pharmacy education; consultation skills; communication skills; design-based research; healthcare education
Rationale
Effective consultation skills are vital in pharmacy practice, and a core standard for General Pharmaceutical Council registration (GPhC, 2017). Traditionally, consultation and communication skills training have been undertaken using simulated patients (either actors, staff or peers) in a controlled environment prior to any patient facing practice (ElGeed et al., 2021). The advent of generative AI chat bots has led to huge implications for education; studies have shown that using AI patients to practice consultations can produce realistic discussions in medical education (Yamamoto et al., 2024). There is currently limited research around using AI patients to support consultation skills in healthcare education, therefore their use in pharmacy education has not been standardised nor effectively evaluated. Due to costs, environmental impact and the time associated with creating AI patients, research is necessary to highlight whether they can be beneficial to the curriculum and how.
This presentation will describe proposed research towards AI patients in pharmacy consultation education- specifically focussing on consultation skills development. The research aims to consider the views and concerns of educators, students and stakeholders on using AI patients, while designing and evaluating an AI based teaching programme alongside all stakeholders.
Utilising a Design-Based Research approach (Wang and Hannafin, 2005) , initial research will be gathered on appropriate software alongside discussions with stakeholders to identify requirements. Semi-structured interviews with stakeholders will gather data around concerns and beliefs which will support curriculum design and training. Practitioners will use an online forum to provide feedback and discuss the programme. This will be available through all stages of the research to ensure continued support. Qualitative data from this forum will be collected and analysed regularly using thematic coding to investigate areas such as attitude towards the programme, ways to improve it and barriers experienced.
Once the programme has been designed, teaching sessions will be undertaken over a semester, with student feedback gathered by qualitative analysis of surveys followed by focus groups with students to discuss feedback in more detail.
Exam results from formative and summative practice-based examinations (known as OSPEs and OSCEs) will be analysed against previous cohorts to identify any statistically significant differences.
Feedback and analysis will influence design changes in the curriculum as well as determine the types of data needing to be gathered moving onwards. Programme redesign will then be undertaken alongside practitioners, followed by further teaching, data gathering, then review. This process will be repeated multiple times to gather sufficient data and create an effective curriculum.
While Design-Based Research allows for continued adaptations and additions of further research questions throughout the project (based on stakeholder and participant feedback); the initial intention is for results to answer three overarching questions:
- What do educators perceive as barriers towards introducing AI patients to the curriculum
- How effective are AI patients compared to traditional methods in supporting consultation skill development
- How best can AI patients be used to effectively support development?
Research findings will be reported back to the stakeholders through presentations and disseminated to other institutions through openly published journals. Findings will determine the future use of AI patients locally plus influence similar research in the education of other healthcare professionals worldwide.
By highlighting perceived barriers, appropriate training plans and support can be designed for practitioners. Knowing if there is a benefit and how best to utilise AI patients will ultimately reduce costs, save time and reduce environmental impact (reducing unnecessary use of high energy consuming software) in educational institutions around the world.
References
ElGeed, H. et al. (2021) ‘The utilization of simulated patients for teaching and learning in the pharmacy curriculum: exploring pharmacy students’ and recent alumni’s perceptions using mixed-methods approach’, BMC Medical Education, 21(1), pp. 1–13. Available at: https://doi.org/10.1186/S12909-021-02977-1/TABLES/5.
GPhC (2017) ‘Standards for the initial education and training of pharmacy technicians’. Available at: https://www.pharmacyregulation.org/initial-PT (Accessed: 30 March 2023).
Wang, F. and Hannafin, M.J. (2005) ‘Design-Based Research and Technology-Enhanced Learning Environments’, Educational Technology Research and Development, 53(4), pp. 5–23. Available at: https://openurl.ebsco.com/contentitem/edsjsr:edsjsr.30221206?sid=ebsco:plink:crawler&id=ebsco:edsjsr:edsjsr.30221206&crl=c (Accessed: 16 February 2025).
Yamamoto, A. et al. (2024) ‘Enhancing Medical Interview Skills Through AI-Simulated Patient Interactions: Nonrandomized Controlled Trial’, JMIR Med Educ 2024;10:e58753 https://mededu.jmir.org/2024/1/e58753, 10(1), p. e58753. Available at: https://doi.org/10.2196/58753.
6 responses to “How can A.I. patients effectively develop pharmacy students’ consultation skills?”
This isn’t something that I know much about so I’m looking forward to learning more at the conference. I have a couple of simple questions:
1. With the traditional approach to this type of training, would the actors/peers/staff be following some sort of script/briefing document for the way they describe their symptoms etc?
2. How much time would pharmacy students spend on this consultation/communication training? Do you think it would be beneficial to have more time on this if AI could provide a more convenient/less costly alternative to actors? Or do you think the amount of training is sufficient but AI could offer a more efficient/cost effective way of achieving it?
Hi Billy, I am looking forward to attending your session as this appears to be a helpful aid. My question for you is: Do you feel that using AI patients, the human reactions and emotions could be stilted?
Hi Billy,
AI is really penetrating all aspects of our lives. It is an interesting approach to maximising CPD for health care providers. Do you have concerns over how data generated from the use of AI in these contexts might be used by third parties? Looking forward to your presentation.
An innovative focus exploring and stretching the possibilities of AI for professional development! It would be fascinating to understand how AI patients are ‘manifested’. Would they be chatbots? Given that AI output is governed by human input, how might diversity within patients be considered?
If you have any questions or would like to see an example of an AI patient in practice, I would be happy to respond via email at billy.smith@uea.ac.uk
Hi Billy, I am an alumni of the Master’s and I really enjoyed your presentation today. Your delivery was very confident and engaging.
As a skeptic of AI you helped me see genuine value in teaching. I hope it will give us all more access to healthy self-care information as it progresses in apps like the Zoe.com UK epidemiologist, Tim Specter the researcher behind the app and his work might be useful. They have a brillant podcast/youTube channel.
It was so interesting to a really practical application of how AI could save money on ‘actors’, reduce the repetition fatigue and provide consistency, as well as provide big data for individuals to stay well and extend wellness – health span not just living longer. Well done!