Presentation
Human-AI Teaming in Rural Hospitals: Human Factors Evaluation and Co-Design of an AI System for a Patient Movement and Flow Unit
SessionPoster Session 1
DescriptionIn this study, we are conceptualizing and modelling a patient access and flow (PAF) unit in a rural hospital system in Ontario through observation and interviews with staff. Building on these initial findings, this project will aim to deploy a user-centred design collaboration with PAF staff to understand the potential for AI or other technologies, the impact the implementation that such tools can have on workflow and the design of these tools to minimize workflow disruption while optimizing their function for PAF in this rural hospital system.
Patient flow can be described as the movement of patients within and through healthcare systems and facilities [1]. Hospital departments are the most fundamental units in healthcare systems, and they must be designed to be well-defined in their independent function, while permitting necessary overlap and high levels of communication across units to ensure the proper functioning of the overall system [2]. This aspect of hospital units is highly important in the context of patient flow management for healthcare systems, as each unit needs to simultaneously ensure proper flow and timing within their defined unit but also to ensure limited disruptions to the flows and schedules in units that are receiving or transferring patients [2]. Patient flow management is a critical component that needs to be considered for solutions aimed at optimizing healthcare [3]. Patient flow is key to improving the quality of care through reductions in wait times and costs and ensuring proper planning of hospital staffing and resources [3]. Inefficiencies that exist in patient flow systems are directly linked to negative clinical outcomes and contribute to the overcrowding of health centres [4]. Technologies, especially artificial intelligence (AI), are being investigated as potential catalysts to help optimize patient flow and address current issues faced in this area.
Out of all health systems in Canada, rural hospital systems face unique challenges that must be considered when assessing patient flow. The Rural Ontario Municipal Association (ROMA) recently reported that significant challenges are still being faced by rural communities accessing healthcare, including ED closures and staff shortages [5]. With current research focusing on systems design for patient flow units is ongoing [6], [7], effort must be made to ensure rural health systems are included in the design, testing and implementation of potential technologies being developed as part of the system redesign. This effort will strive to maintain equitable health access for rural communities and has the potential to mitigate against unique challenges faced by rural hospital patient flow units.
To get familiarized with the domain of work for the PAF in the rural hospital system in Owen Sound, Ontario, a pilot study was conducted through observation sessions in the PAF unit and unstructured interviews with PAF staff. Detailed notes of observation sessions of all levels of staff in the PAF unit were recorded over a three-day observation session. Unstructured interview data and responses from questions posed to staff to describe or clarify observed processes were obtained. Upon completion of this observation period, the findings were reported and discussed with the PAF manager to determine which areas of the PAF domain should be the focus for technology and potential AI design and implementation. The findings from this initial report will be further used to guide the next study steps.
Continuation steps will utilize a user-centred, or human-centred, design protocol, which is an accepted practice to test the implementation and acceptability of various healthcare technologies[8], [9], [10]. Once the areas of PAF workflow is identified and agreed upon by both researchers and PAF staff as the ideal area of focus for technology and AI exploration, the next step will be exploring various technology or AI options being deployed in larger health systems that align with the area of focus. If similar tools have been designed and implemented in other Canadian hospitals, our team will instigate collaboration with those centres to discuss the possibility for collaborating on the re-design of this tool to tailor it to a rural PAF hospital setting. If collaboration is not possible or if it is deemed that no similar system of interest is available, the research team will explore ground-up design of an in-house technology designed directly for the rural PAF unit. The design protocol regardless of through collaboration with other centres or in-house will be initiated through information-gathering processes to inform design practices by understanding how the tool is intended to be utilized in practice and the needs of its intended users [11]. The intended users (PAF staff) will be asked to consent to participate in this process, and upon consent will be interviewed to understand their needs for using the tool and how they imagine it being implemented into existing workflows. Information gathered in the interview phase will guide the development of the proposed initial prototype (the first version of the designed tool to be used in practice) [11]. Following an introduction of this prototype to the participating users, the prototype will continue to be improved based on user feedback and in practice testing cycles until an acceptable prototype is produced [11].
Key takeaway points:
• This project aims to collaborate with PAF and healthcare workers in the iterative design process of new technologies and AI systems being adopted into rural hospitals.
• Identifying and addressing barriers to adoption can increase the likelihood of successful adoption in care. The findings of this research project can guide future research focused on AI adoption practices with specific focus and implication for rural hospitals.
References
Al Harbi, S., Aljohani, B., Elmasry, L., Baldovino, F. L., Raviz, K. B., Altowairqi, L., & Alshlowi, S. (2024). Streamlining patient flow and enhancing operational efficiency through case management implementation. BMJ Open Quality, 13(1), e002484. https://doi.org/10.1136/bmjoq-2023-002484
Burns, C. (2018). Human-centred design. In eHealth Research, Theory and Development (pp. 207–227). Routledge. https://doi.org/10.4324/9781315385907-10
Canada’s Drug and Health Technology Agency. (2024). CADTH Horizon Scan Artificial Intelligence for Patient Flow. Canadian Journal of Health Technologies, 4(5).
De Vito Dabbs, A., Myers, B. A., Mc Curry, K. R., Dunbar-Jacob, J., Hawkins, R. P., Begey, A., & Dew, M. A. (2009). User-centered design and interactive health technologies for patients. Computers, Informatics, Nursing : CIN, 27(3), 175–183. https://doi.org/10.1097/NCN.0b013e31819f7c7c
Francisco, K. M., & Burns, C. M. (2024). An Approach to Potentially Increasing Adoption of an Artificial Intelligence–Enabled Electronic Medical Record Encounter in Canadian Primary Care: Protocol for a User-Centered Design. JMIR Research Protocols, 13, e54365. https://doi.org/10.2196/54365
Hall, R. (2013). Patient flow. AMC, 10(12), 4.
Melles, M., Albayrak, A., & Goossens, R. (2021). Innovating health care: key characteristics of human-centered design. International Journal for Quality in Health Care : Journal of the International Society for Quality in Health Care, 33(Supplement_1), 37–44. https://doi.org/10.1093/intqhc/mzaa127
Nguyen, Q., Wybrow, M., Burstein, F., Taylor, D., & Enticott, J. (2022). Understanding the impacts of health information systems on patient flow management: A systematic review across several decades of research. PLOS ONE, 17(9), e0274493. https://doi.org/10.1371/journal.pone.0274493
Paterson, E., Chari, S., McCormack, L., & Sanderson, P. (2024). Application of a Human Factors Systems Approach to Healthcare Control Centres for Managing Patient Flow: A Scoping Review. Journal of Medical Systems, 48(1), 62. https://doi.org/10.1007/s10916-024-02071-1
Paterson, E., & Sanderson, P. M. (2025). Applying Work Domain Analysis to a Healthcare Control Centre for Patient Flow Management. Journal of Cognitive Engineering and Decision Making, 19(3), 250–282. https://doi.org/10.1177/15553434251330114
Rural Ontario Municipal Association. (2024). Fill the Gaps Closer to Home Improving Access to Health Services for Rural Ontario.
Patient flow can be described as the movement of patients within and through healthcare systems and facilities [1]. Hospital departments are the most fundamental units in healthcare systems, and they must be designed to be well-defined in their independent function, while permitting necessary overlap and high levels of communication across units to ensure the proper functioning of the overall system [2]. This aspect of hospital units is highly important in the context of patient flow management for healthcare systems, as each unit needs to simultaneously ensure proper flow and timing within their defined unit but also to ensure limited disruptions to the flows and schedules in units that are receiving or transferring patients [2]. Patient flow management is a critical component that needs to be considered for solutions aimed at optimizing healthcare [3]. Patient flow is key to improving the quality of care through reductions in wait times and costs and ensuring proper planning of hospital staffing and resources [3]. Inefficiencies that exist in patient flow systems are directly linked to negative clinical outcomes and contribute to the overcrowding of health centres [4]. Technologies, especially artificial intelligence (AI), are being investigated as potential catalysts to help optimize patient flow and address current issues faced in this area.
Out of all health systems in Canada, rural hospital systems face unique challenges that must be considered when assessing patient flow. The Rural Ontario Municipal Association (ROMA) recently reported that significant challenges are still being faced by rural communities accessing healthcare, including ED closures and staff shortages [5]. With current research focusing on systems design for patient flow units is ongoing [6], [7], effort must be made to ensure rural health systems are included in the design, testing and implementation of potential technologies being developed as part of the system redesign. This effort will strive to maintain equitable health access for rural communities and has the potential to mitigate against unique challenges faced by rural hospital patient flow units.
To get familiarized with the domain of work for the PAF in the rural hospital system in Owen Sound, Ontario, a pilot study was conducted through observation sessions in the PAF unit and unstructured interviews with PAF staff. Detailed notes of observation sessions of all levels of staff in the PAF unit were recorded over a three-day observation session. Unstructured interview data and responses from questions posed to staff to describe or clarify observed processes were obtained. Upon completion of this observation period, the findings were reported and discussed with the PAF manager to determine which areas of the PAF domain should be the focus for technology and potential AI design and implementation. The findings from this initial report will be further used to guide the next study steps.
Continuation steps will utilize a user-centred, or human-centred, design protocol, which is an accepted practice to test the implementation and acceptability of various healthcare technologies[8], [9], [10]. Once the areas of PAF workflow is identified and agreed upon by both researchers and PAF staff as the ideal area of focus for technology and AI exploration, the next step will be exploring various technology or AI options being deployed in larger health systems that align with the area of focus. If similar tools have been designed and implemented in other Canadian hospitals, our team will instigate collaboration with those centres to discuss the possibility for collaborating on the re-design of this tool to tailor it to a rural PAF hospital setting. If collaboration is not possible or if it is deemed that no similar system of interest is available, the research team will explore ground-up design of an in-house technology designed directly for the rural PAF unit. The design protocol regardless of through collaboration with other centres or in-house will be initiated through information-gathering processes to inform design practices by understanding how the tool is intended to be utilized in practice and the needs of its intended users [11]. The intended users (PAF staff) will be asked to consent to participate in this process, and upon consent will be interviewed to understand their needs for using the tool and how they imagine it being implemented into existing workflows. Information gathered in the interview phase will guide the development of the proposed initial prototype (the first version of the designed tool to be used in practice) [11]. Following an introduction of this prototype to the participating users, the prototype will continue to be improved based on user feedback and in practice testing cycles until an acceptable prototype is produced [11].
Key takeaway points:
• This project aims to collaborate with PAF and healthcare workers in the iterative design process of new technologies and AI systems being adopted into rural hospitals.
• Identifying and addressing barriers to adoption can increase the likelihood of successful adoption in care. The findings of this research project can guide future research focused on AI adoption practices with specific focus and implication for rural hospitals.
References
Al Harbi, S., Aljohani, B., Elmasry, L., Baldovino, F. L., Raviz, K. B., Altowairqi, L., & Alshlowi, S. (2024). Streamlining patient flow and enhancing operational efficiency through case management implementation. BMJ Open Quality, 13(1), e002484. https://doi.org/10.1136/bmjoq-2023-002484
Burns, C. (2018). Human-centred design. In eHealth Research, Theory and Development (pp. 207–227). Routledge. https://doi.org/10.4324/9781315385907-10
Canada’s Drug and Health Technology Agency. (2024). CADTH Horizon Scan Artificial Intelligence for Patient Flow. Canadian Journal of Health Technologies, 4(5).
De Vito Dabbs, A., Myers, B. A., Mc Curry, K. R., Dunbar-Jacob, J., Hawkins, R. P., Begey, A., & Dew, M. A. (2009). User-centered design and interactive health technologies for patients. Computers, Informatics, Nursing : CIN, 27(3), 175–183. https://doi.org/10.1097/NCN.0b013e31819f7c7c
Francisco, K. M., & Burns, C. M. (2024). An Approach to Potentially Increasing Adoption of an Artificial Intelligence–Enabled Electronic Medical Record Encounter in Canadian Primary Care: Protocol for a User-Centered Design. JMIR Research Protocols, 13, e54365. https://doi.org/10.2196/54365
Hall, R. (2013). Patient flow. AMC, 10(12), 4.
Melles, M., Albayrak, A., & Goossens, R. (2021). Innovating health care: key characteristics of human-centered design. International Journal for Quality in Health Care : Journal of the International Society for Quality in Health Care, 33(Supplement_1), 37–44. https://doi.org/10.1093/intqhc/mzaa127
Nguyen, Q., Wybrow, M., Burstein, F., Taylor, D., & Enticott, J. (2022). Understanding the impacts of health information systems on patient flow management: A systematic review across several decades of research. PLOS ONE, 17(9), e0274493. https://doi.org/10.1371/journal.pone.0274493
Paterson, E., Chari, S., McCormack, L., & Sanderson, P. (2024). Application of a Human Factors Systems Approach to Healthcare Control Centres for Managing Patient Flow: A Scoping Review. Journal of Medical Systems, 48(1), 62. https://doi.org/10.1007/s10916-024-02071-1
Paterson, E., & Sanderson, P. M. (2025). Applying Work Domain Analysis to a Healthcare Control Centre for Patient Flow Management. Journal of Cognitive Engineering and Decision Making, 19(3), 250–282. https://doi.org/10.1177/15553434251330114
Rural Ontario Municipal Association. (2024). Fill the Gaps Closer to Home Improving Access to Health Services for Rural Ontario.
Event Type
Poster Presentation
TimeMonday, March 234:45pm - 6:15pm EDT
LocationRhinelander Gallery
Hospital Environments