Patient-specific, indication-based prescribing: a mixed-methods evaluation of a clinical decision support tool for prescribers in UK secondary care
File(s)
Author(s)
Feather, Calandra
Type
Thesis
Abstract
Medication errors, particularly prescribing errors, pose significant risks in healthcare, leading to preventable harm and increased healthcare costs. This risk is accentuated in paediatric care due to complex dosing requirements and the vulnerability of the patient population. Despite the advent of electronic prescribing and clinical decision support systems, errors persist, highlighting the need for improved solutions. This thesis evaluates the development and implementation of Touchdose, a patient-specific, indication-based prescribing tool, aimed at enhancing prescribing safety and efficiency in UK secondary care.
Employing a convergent mixed-methods design grounded pragmatism, this research integrates qualitative and quantitative analyses across four interlinked studies. The first study explores healthcare professionals' perceptions of Touchdose v1 through thematic analysis of focus group data from a paediatric simulation setting. The second study conducts a systematic review and narrative synthesis of existing literature on indication-based prescribing, assessing its effectiveness and identifying barriers and facilitators to implementation. The third study uses rapid ethnographic observation to evaluate real-world prescribing practices at a major NHS Trust, focusing on the use of dosing support tools. The final study compares Touchdose v2 with current practice through a mixed-methods, randomised user testing study, measuring safety, performance, and user acceptability.
Findings demonstrate that Touchdose significantly reduces prescribing errors, enhances workflow efficiency, and is well-received by clinicians for its intuitive design. Building trust with users to foster confidence in the system is essential, and real-world clinical evaluation is necessary to validate these findings and ensure successful adoption. Recommendations include iterative refinement of decision support tools and initiatives that prioritise tailored training and engagement to build user confidence and support integration into existing practices.
Employing a convergent mixed-methods design grounded pragmatism, this research integrates qualitative and quantitative analyses across four interlinked studies. The first study explores healthcare professionals' perceptions of Touchdose v1 through thematic analysis of focus group data from a paediatric simulation setting. The second study conducts a systematic review and narrative synthesis of existing literature on indication-based prescribing, assessing its effectiveness and identifying barriers and facilitators to implementation. The third study uses rapid ethnographic observation to evaluate real-world prescribing practices at a major NHS Trust, focusing on the use of dosing support tools. The final study compares Touchdose v2 with current practice through a mixed-methods, randomised user testing study, measuring safety, performance, and user acceptability.
Findings demonstrate that Touchdose significantly reduces prescribing errors, enhances workflow efficiency, and is well-received by clinicians for its intuitive design. Building trust with users to foster confidence in the system is essential, and real-world clinical evaluation is necessary to validate these findings and ensure successful adoption. Recommendations include iterative refinement of decision support tools and initiatives that prioritise tailored training and engagement to build user confidence and support integration into existing practices.
Date Issued
2024-11-15
Date Awarded
01/06/2025
License URL
Advisor
Darzi, Ara
Dean Franklin, Bryony
Publisher Department
Department of Surgery & Cancer
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)