The study was conducted between April 2024 and July 2024 at a UK university that serves approximately 2000 medical students annually. A mixed methods pre-post educational intervention design was used to evaluate a co-designed QI educational workshop [11]. The workshop was developed and delivered by a group of medical students working under the mentorship of a consultant clinician with expertise in QI. These individuals constituted the organizing committee responsible for designing and implementing the educational intervention.

We adopted Kern’s 6-step approach to curriculum development, informed by principles from Kolb’s Experiential Learning Theory, to guide the co-design and development of the workshop [12,13]. Each step is described in detail later to show how learner co-design informed the development process. To measure the impact of the workshop, we used a pre-post educational evaluation design with surveys and interviews. Quantitative data were used to assess changes in participants’ self-reported understanding, confidence, and attitudes, whereas qualitative data were used to explore how participants experienced the workshop and why particular elements were perceived as useful or challenging. The 2 data strands were collected sequentially and integrated at the interpretation stage using a convergent mixed methods approach, whereby the quantitative findings showed what changed after the workshop, and the qualitative findings helped explain why those changes may have occurred.

Participants included medical students from the University of Birmingham Medical School and resident physicians from the University Hospitals Birmingham National Health Service Foundation Trust who voluntarily registered to attend the workshop. Recruitment used a convenience sampling approach [14]. The event was advertised through university mailing lists, social media platforms, and institutional group messaging channels. Snowball sampling also occurred [15], as potential participants shared information about the workshop within their peer networks to encourage attendance.

This project underwent institutional review at the time of submission and was approved to proceed without the requirement for full ethics committee review, as it was deemed not to raise substantial ethical concerns (ERN_2545-May2024). Informed consent was obtained electronically from participants before completion of the preworkshop and postworkshop surveys. Verbal consent was also confirmed at the start of each interview recording. To protect privacy and confidentiality, data were pseudoanonymized with participant ID codes and stored securely in password-protected cloud storage accessible only to the research team. No participant compensation was provided, and participation was entirely voluntary.

A targeted literature search was conducted by an author (CP) to identify previously reported needs for QI education in medical training. The search strategy is summarized in Multimedia Appendix 1. The search identified 134 records, of which 8 (6) were deemed relevant following screening [8,16-22]. The literature shows that students often lack understanding of QI and that targeted educational interventions can improve learners’ knowledge of and attitudes toward QI participation [8,16,17,19,21].

The organizing committee considered several potential approaches to addressing these educational gaps. Rather than proposing formal curriculum reform, which may require substantial institutional resources, the group identified student-led workshops as a feasible and scalable strategy for introducing QI principles to learners [22]. However, the group also recognized potential limitations associated with peer-led education, including the possibility that participants may provide more favorable feedback when educational sessions are delivered by peers. To mitigate this risk, all workshop content was reviewed by a consultant clinician with expertise in QI and clinical audit prior to delivery to ensure scientific accuracy and educational quality.

Individuals who registered for the workshop were invited to complete a preworkshop survey (Multimedia Appendix 2). Participants provided informed consent prior to completing the survey through an electronic participant information and consent form. The survey included 10-point Likert-scale items assessing participants’ baseline knowledge, confidence, and attitudes toward QI activities, as well as their prior experience with audits and QIPs, and their expectations for the workshop. Open-ended questions were included to allow participants to elaborate on their responses and provide additional insights into their experiences and perceived learning needs. Responses to open-ended questions were analyzed thematically to identify common themes informing the development of the workshop curriculum (Multimedia Appendix 3).

Participants who completed the preworkshop survey were invited to participate in a semistructured interview to explore their experiences and perceptions of QI in greater depth (Multimedia Appendix 4). The interview guide was specifically developed by the research team to ensure that key topics were consistently addressed across participants. Interviews were conducted either via Zoom (Zoom Communications Inc) or in person, depending on participants’ accessibility. Audio recordings were obtained with participant consent prior to the start of each interview. Among the 27 participants who completed the preworkshop survey, 12 (44.4%) agreed to participate in the interview (n=10, 83.3% medical students and n=2, 16.7% resident physicians). Interviews lasted approximately 20 minutes. All recordings were transcribed verbatim while ensuring accuracy. Transcripts were reviewed and analyzed independently by 2 study authors (ALJY and MI) using thematic analysis [23]. Data were stored securely in password-protected institutional cloud storage accessible only to members of the research team.

Two researchers collaboratively developed a coding framework after familiarizing themselves with the interview transcripts. Discrepancies in coding were discussed and resolved through consensus to ensure consistency. Qualitative analysis was conducted using NVivo (version 14; Lumivero LLC). Intercoder reliability was assessed using the Cohen kappa coefficient [24]. Codes were inductively grouped into broader themes, which were subsequently reviewed and refined.

On the basis of the findings from the literature review and targeted needs assessment, the program goal and workshop learning objectives were developed to ensure that the final curriculum directly addressed the integration of learner co-design into the workshop’s development. The overall program goal was to equip early-career health care learners with foundational knowledge, skills, and confidence to engage in QI and medical audit activities in clinical practice.

The workshop was designed to achieve the following learning objectives: (1) to improve participants’ understanding of medical audits and QIPs, (2) to increase participants’ confidence in engaging with QI activities, and (3) to strengthen participants’ awareness of practical, ethical, and implementation issues relevant to QI work.

These outcomes were evaluated using self-reported preworkshop and postworkshop survey responses and qualitative interviews, which together addressed the study’s second objective.

The workshop curriculum was designed using Kolb’s Experiential Learning Theory [13], which conceptualizes learning as a cyclical process involving 4 stages: concrete experience, reflective observation, abstract conceptualization, and active experimentation. Educational activities were structured to approximate these stages within the constraints of a 1-day workshop. The workshop agenda can be found in Multimedia Appendix 5.

Participants were invited to complete a postworkshop survey (Multimedia Appendix 6) using the same 10-point Likert-scale measures as the preworkshop survey to evaluate changes in self-reported understanding, confidence, and attitudes toward QI. Participants also rated individual workshop components using a 5-point scale and provided an overall workshop rating using a 10-point Likert scale. Participants who completed the survey were invited to participate in follow-up interviews to explore their experiences in greater depth (Multimedia Appendix 7). Of the 31 participants who completed the postworkshop survey, 15 (48.4%) agreed to be interviewed. The interview adopted a semistructured format similar to the preworkshop interviews and lasted approximately 15 minutes.

Quantitative survey data were analyzed using Prism (version 10.3.0; GraphPad Software LLC). Participants who completed both preworkshop and postworkshop surveys were included in the analysis. Descriptive statistics for nonparametric continuous data are presented as median (IQR), while categorical data are summarized as n (%). Ordinal data derived from Likert scales were described with medians and IQRs, and the significance of the differences was assessed using the Wilcoxon matched-pairs signed-rank test. The Kruskal-Wallis test was used to identify the most efficacious aspects of the workshop, given the nonparametric nature of the data. The Fisher exact test was used to calculate P values for categorical data. Statistical significance was set at P<.05. Responses to the open-ended questions and postevent interviews were analyzed using the same approach as for the preworkshop survey (Multimedia Appendix 8).

In total, 31 delegates attended the event (n=29, 93.5% medical students and n=2, 6.5% resident physicians; n=24, 88.9% female). Attendees represented a range of ethnic backgrounds (Asian: n=17, 63%, White: n=6, 22.2%, and Black: n=4, 14.8%). Overall, 87.1% (27/31) participants completed the preworkshop survey (median age 21, IQR 21-22 years), and 77.4% (24/31) participants completed both preworkshop and postworkshop surveys.

Co-design within a structured curriculum development model is feasible and may help produce a workshop that learners perceive as relevant, practical, and effective in building confidence. To our knowledge, this is among the few reported learner–co-designed QI workshops that integrate Kolb’s Experiential Learning Theory [13] into Kern’s curriculum development framework [12]. The workshop was associated with improved participants’ self-reported understanding and confidence, suggesting greater perceived readiness to engage in QIPs. Learner involvement in the design process may have helped ensure that the workshop addressed real-world learning needs, improved relevance, and promoted active engagement, which may be less evident in traditional didactic approaches.

Although experiential learning is fully realized through implementation of QI initiatives in real clinical environments, educational workshops can approximate key elements of this cycle through small-group activities, structured project design exercises, and facilitated reflection. In this workshop, participants engaged in simulated QIP planning and peer discussions, allowing them to apply theoretical concepts and reflect on potential implementation challenges before undertaking real-world QI initiatives.

Our findings are consistent with the broader literature synthesized by Peiris-John et al [8], which suggests that structured QI teaching can improve learner-level outcomes such as self-reported knowledge, confidence, and engagement. The workshop was similar to the interventions identified in that review because it combined multiple educational strategies, including didactic teaching, experiential learning, and small-group discussion. However, this study differed in that it evaluated a 1-day workshop that was co-designed with learners within Kern’s curriculum development framework and complemented by a qualitative evaluation of participants’ experiences [12].

Kern’s model was chosen for its practical [12] and stepwise approach to curriculum development, which aligns well with medical education, particularly when an intervention must progress from needs assessment to implementation and evaluation within a clearly structured and reportable framework. Its flexibility also supports refinement of educational programs in response to identified learner needs and feedback, and it has been widely applied in medical education [25]. Alternative approaches were considered but were less suitable for the aims of this study. For example, backward design is useful for aligning learning outcomes with assessment [26], but it places less emphasis on needs assessment and stakeholder-informed problem definition, both of which were central to the co-design process. Design-based research could also have been used because it supports iterative refinement in authentic settings [27]; however, it typically requires multiple cycles of testing and redesign and was therefore less feasible for a single workshop delivered within a limited time frame. Similarly, intervention development approaches primarily grounded in implementation science may have been useful for later scale-up [28], but they were less well suited to reporting the educational curriculum development process itself.

A convergent mixed methods design was chosen because it enabled triangulation of quantitative and qualitative findings, providing complementary insights into the educational intervention [29]. In this study, the quantitative component identified what changed after the workshop, particularly in participants’ self-reported understanding, confidence, and attitudes, whereas the qualitative component helped explain why these changes may have occurred by exploring participants’ experiences of the workshop. Alternative mixed methods approaches, including an explanatory sequential design, would typically require quantitative data collection and analysis to be completed before qualitative follow-up to explain specific findings, which was less practical within the time frame of a single workshop evaluation. An exploratory sequential design was also less appropriate, as this is generally used when qualitative findings are intended to inform the subsequent development of a quantitative instrument or intervention, which was not the purpose of this study.

This structured curriculum development process, based on Kern’s framework [12], supports transparency and reproducibility, which is a strength of the study. The workshop was co-designed with learners, which helped ensure that the educational content addressed identified learner needs and was perceived as relevant to practice. The mixed methods design strengthens the evaluation by combining quantitative and qualitative data, allowing the study to show not only what changed after the workshop, but also why participants perceived the intervention as useful and applicable. Finally, the workshop provides a practical, resource-efficient educational model that can be adapted to similar training contexts.

However, although the institution serves a large medical student population, this study evaluated a single voluntarily attended workshop using convenience and snowball sampling. As such, the sample was not intended to be statistically representative of the wider student population. These sampling approaches also limit the transferability of the findings to other settings and learner groups. A further limitation is that the evaluation primarily assessed self-reported understanding, confidence, and attitudes rather than objectively measuring knowledge acquisition or practical QI skills. The workshop placed greater emphasis on some QI approaches, particularly PDSA cycles, which meant that other methodologies were explored less extensively. In addition, longer-term follow-up examining whether participants subsequently implemented QI projects was not conducted due to logistical constraints and the limited study time frame.

Future research should explore the scalability of this model across multiple medical schools, assess its long-term impact on learners’ QI engagement in practice, and incorporate perspectives from a wider range of stakeholders, including faculty and institutional leaders. Future studies should also include more objective educational outcomes, such as preintervention and postintervention knowledge tests, case-based assessments, observed structured scenarios, rubric-based evaluations, or workplace-based assessments. In addition, future work could examine whether longitudinal one-to-one mentorship models support sustained engagement and facilitate the development of tangible scholarly outputs such as conference posters or oral presentations.

The learner co-designed a QI workshop that improved understanding of QI concepts and increased self-reported confidence in engaging in QI activities, with positive perceptions of the workshop’s relevance and usefulness. Future research is needed to examine longer-term outcomes and assess whether this co-designed model can be adapted effectively across broader educational settings.