Exit Interviews in Health Professions Education: Protocol for a Scoping Review

Introduction

Background

Exit interviews are structured discussions with individuals leaving an organization, educational program, or training pathway to explore their experiences, reasons for departure, and suggestions for improvement [1]. They are widely used across sectors to inform organizational development [2], workforce planning [3], and retention strategies [4]. Health care workforce shortages, attrition, and staff retention remain international concerns, prompting growing interest in approaches that can improve workplace and training experiences [5].

Although organizational and policy contexts vary between countries, workforce attrition can affect service capacity, continuity of care, and organizational resilience [5-7]. Understanding why health care professionals leave their employing organizations is therefore important for developing effective workforce retention strategies [5,6,8]. Exit interviews represent one mechanism through which organizations can systematically explore employees’ reasons for leaving and identify potentially modifiable organizational factors [2,9]. However, their usefulness may be limited by challenges, including variable participation, employees’ willingness to disclose negative experiences, differences in how interviews are conducted, and whether findings are systematically analyzed and translated into organizational action [2,9,10]. International literature can therefore provide useful context for understanding both the potential value and limitations of exit interviews in health care [9,10]. Although the proposed review has an international scope, the United Kingdom provides a useful case example of the rationale for exit interviews, the issues hindering implementation, and why the UK-based authors chose to study them.

The United Kingdom’s National Health Service (NHS), where two of the authors of this paper are based, has conducted an electronic exit questionnaire for staff leaving the NHS. A small number of trusts have successfully implemented exit interviews to improve staff well-being and retention [11]. However, exit interviews are far from commonplace, and their limited implementation is a missed opportunity to learn about our workforce, workplaces, and training. Significant barriers to conducting exit interviews are the time required and the power gradient that often exists between the interviewer and the interviewee. In 2022, the British Medical Association passed a motion at its annual general meeting urging NHS employers to publish anonymized exit interview results through an open-access national database [12]. No such program has been implemented. More recently, the Royal College of Psychiatrists’ retention charter repeatedly mentions exit interviews as a central component of improving workforce issues in the specialty [13].

Insights from business literature, such as the Harvard Business Review, offer valuable lessons on how to leverage exit interviews effectively [10]. In business settings, exit interviews are widely used to gather feedback, understand organizational challenges, and improve retention strategies. However, the application of these principles to health care remains limited, raising questions about how best to adapt these methodologies to the unique challenges faced by the NHS workforce.

A 2014 Cochrane review on exit interviews to reduce turnover among health care professionals identified 8 papers, but none met its inclusion criteria, leading the authors to conclude that “evidence about the effectiveness of exit interviews to reduce turnover is currently not available” [9].

Within health professions education, exit interviews may provide valuable insights into learner, trainee, and practitioner experiences, including factors influencing program completion, training progression, specialty choice, career transitions, and attrition. This information can inform educational quality improvement, trainee support, and workforce planning. However, despite the apparent relevance of exit interviews to health professions education, the literature remains fragmented, with studies dispersed across professions, training stages, and settings. To date, no comprehensive synthesis has mapped how exit interviews are defined, conducted, and used within health professions education.

This scoping review will address this gap by systematically mapping the literature on exit interviews in health professions education.

The initial definitions for the scoping review protocol are as follows.

Exit interviews are defined as “An interview held with an employee about to leave an organisation, typically in order to discuss the employee’s reasons for leaving and their experience of working for the organisation” [14]. We will extend the definition to include exit interviews conducted after an individual leaves an organization or role.

Health professions education is defined as “the formal study and training for individuals pursuing careers in healthcare, including professions like medicine, nursing, and physiotherapy. It encompasses undergraduate and postgraduate programs, apprenticeships, and continuing professional development, focusing on equipping students and practitioners with the knowledge, skills, and values necessary to provide quality healthcare and improve health outcomes” [15-18].

In addition to the Cochrane review mentioned above [9], a preliminary search found no literature syntheses addressing our research question.

Review Questions

The primary review question is as follows: What is reported in the literature about how exit interviews have been used, implemented, and conceptualized in health professions education?

The subquestions are as follows:

1. What type of studies and papers make up the literature on exit interviews?
2. What definitions of “exit interview” are used in the literature, and how have they evolved?
3. What research questions were exit interviews used to address?
4. What questions were asked during the exit interviews?
5. What barriers or facilitators to implementing exit interviews have been described in the literature?

Methods

Overview

This review will use the JBI methodology for scoping reviews [19,20]. This protocol was developed using the JBI guidance on scoping review protocol development [21]. The protocol outlines methodologies that go beyond the JBI guidance, including AI checking of human screening and extraction; these are justified in the protocol. The review will be reported in accordance with PRISMA-ScR (Preferred Reporting Items For Systematic Reviews And Meta-Analyses Extension for Scoping Reviews) [22].

Inclusion Criteria

Participants

This scoping review will consider all papers in academic journals or policy documents related to health professions engaged in any type of training program. This includes both undergraduate and postgraduate education. The search will not be limited by geography.

Participants are those undertaking health professions education, as defined above, with the following clarifications.

For allied health professionals, this includes students or graduates in physiotherapy, occupational therapy, speech and language therapy, radiography, dietetics, paramedic science, or comparable disciplines. For mixed-population studies, data relating to eligible allied health professionals should be reported separately and be extractable. If the data cannot be separated, the study should only be included where the whole population meets the predefined health professions education criteria.

For physicians, this will include trainees (resident physicians in the United Kingdom as of 2024, specialty and core trainees in the United Kingdom as of 2018, and registrars and senior house officers historically) and those practicing outside of a recognized training program, including clinical fellows, locally employed physicians, and staff and associate specialists. We will also include those who have completed their training and are now practicing independently (consultants in the United Kingdom and attending physicians in the United States). For participants who are not enrolled in formal training programs, their required participation in continuing professional development is considered to constitute health professions education.

The participants will be leaving employment, transitioning to a new role or practice location, or completing a course or program of study. We expect further clarifications for participants to be generated throughout the review.

Concept

While the working definition of exit interviews for this protocol is listed above [14], the search concept will be “exit interviews” or a similar term listed in Textbox 1, as used in the identified papers.

Context

We will consider all health care professions and settings with training programs. Textbox 1 shows the health professions included in the pilot search. A 2022 review used the following list of health care workers: nurses, physicians, allied health workers, general practitioners, other primary care workers, social workers, pharmacists, midwives, health visitors, mental health workers, psychological professionals, psychiatrists, surgeons, paramedics, and students [23]. This list complements our pilot search and has influenced the planned search. Papers from all countries will be included in this review.

Types of Sources

Our preliminary search revealed no reason to exclude any types of literature. We will use qualitative and descriptive reports, along with gray literature, such as expert opinions, policy documents, and editorials. No date restrictions will be applied to the included papers. Only papers published in English will be reviewed because of the team’s language limitations.

Exclusion Criteria

Sources will be excluded if they do not include exit interviews and focus on health professions education. Exit interviews pertaining to dismissal or poor performance will be excluded. We will also exclude local guidelines from our gray literature search that only refer to national guidelines recommending that organizations conduct exit interviews.

Protocol Registration

This scoping review protocol was prospectively registered on OSF Registries on January 22, 2026.

Ethical Considerations

This scoping review will use publicly available data; therefore, it does not require ethics approval. The completed review will be submitted for publication in a peer-reviewed journal and presented at relevant conferences.

Patient and Public Involvement

Because the study will involve reviewing and reporting existing literature, there will be no patient and public involvement.

Study Timeline

This protocol describes a review scheduled to run from October through December 2026.

Search Strategy Development

To provide a comprehensive understanding of the use of exit interviews in health professions education, this scoping review aims to identify and present the available information on the use of exit interviews in all settings where health professions education occurs, focusing on the approaches to conducting exit interviews, the contexts in which they have been implemented, and the barriers and facilitators to their implementation.

This review will give equal weight to academic and gray literature; therefore, the search strategy involved iterative searches of sources likely to yield both types of literature. An initial limited search was conducted on CINAHL, Medline, and Embase by Olivia Schaff, clinical and outreach librarian at Manchester University Hospitals NHS Foundation Trust, on December 20, 2024. Two of the paper’s authors (DD and TG) conducted a preliminary search of UK-oriented gray literature sources, including NHS England, NHS Employers, The Health Foundation, the Nuffield Trust, The King’s Fund, the Association for the Study of Medical Education, and Skills for Health. Our gray literature search strategy was developed with advice from a librarian with a specialist interest in gray literature (John Barbrook).

For CINAHL, Medline, and Embase, the papers in the initial search were reviewed to identify keywords in the titles and abstracts, as well as index MeSH terms to inform the development of the formal search strategy, which will be tailored for each database with the help of a medical librarian. Key search terms include “exit interviews” and “health professions education,” with each term having several keywords authors may have used instead. Textbox 1 presents the initial terms.

Search Strategy

Sources for academic literature include CINAHL, MEDLINE, Embase, HMIC, Social Policy and Practice, the Cochrane Database of Systematic Reviews, and the British Medical Journal Collection via the NHS Knowledge and Library Hub. PsycInfo will be searched via EBSCOhost, accessed through Lancaster University. We will search business and management journals via Business Source Complete, ProQuest Business Database, and Emerald Business and Management Journals. The reference lists of all included papers will be hand-searched. No date, geographical, study design, or publication type filters will be applied to the bibliographic database searches. Searching multiple databases aims to reduce the risk of missing relevant studies [24]. The proposed search strategy for the NHS Knowledge and Library Hub is available in Multimedia Appendix 1.

Gray literature databases will be accessed. These include HMIC, Policy Commons, Overton, and Trip Medical Database. We will screen the first 100 search results, ranked by relevance, from each gray literature source. We will record this process using screenshots of search results or, if available, by exporting the search results.

Experts identified during the search process may be contacted directly for further information or clarification.

The initial search terms to be used in the gray literature search are exit interview*, exit-interview*, departure interview, leaving interview, employee exit discussion, and end-of-service interview.

For both the academic and gray literature, the reviewers will contact authors for further information if required.

Screening and Selection

Overview

Following the search, all identified records will be collated and uploaded to Zotero (Corp for Digital Scholarship) [25], where duplicates will be removed. After deduplication, we will upload the academic literature to Rayyan (Rayyan systems Inc) [26] to facilitate title and abstract screening.

Academic Literature

Our screening and selection approach is based on the JBI guidance, with some modifications described herein. This is required primarily to manage the large volume of irrelevant titles. This review will add a prescreening stage, as a large number of studies, many of which are likely to be irrelevant, are anticipated from the initial search. The prescreening stage will involve reviewing titles and removing papers that are clearly irrelevant. This will be followed by the usual 2-stage JBI process with screening of titles and abstracts to identify papers that might be relevant to the study question. After title and abstract screening, the full text of all included papers will be sourced. After reading each paper, we will remove any additional irrelevant papers. The remaining papers will be relevant to the study question and will be selected for data extraction.

For both the title and abstract screening and full-paper screening stages, an initial pilot phase, involving screening 10% of retrieved titles or abstracts, will be completed by 2 independent reviewers. Discrepancies will be discussed before proceeding. We will calculate a kappa score to assess agreement during screening, with substantial agreement (0.61-0.80) required to continue. If this threshold is not initially met, we will screen additional increments of 10% until substantial agreement is achieved. After that, a single reviewer will conduct the remaining title and abstract screening using Rayyan.

We will present a summary of the study selection process as a PRISMA (Preferred Reporting Items for Systematic Review and Meta-Analysis) flowchart [27].

Gray Literature

The iterative nature of the gray literature search is likely to make the selection process more straightforward. It is anticipated that the identified documents will be reviewed and that irrelevant documents will be discarded. The lead author will complete the initial search and compile a list of studies for consideration. This will be reviewed independently by 2 reviewers, with any disagreements resolved as described above. We will document all searches. We will document decisions about inclusion and exclusion after the title screening phase.

Data Extraction

Data will be extracted to facilitate answering the review questions and subquestions listed above. Initially, this will involve tabulating the data to enable organization and transparency of any findings generated. Table 1 shows the planned initial data extraction. This was developed from the JBI data extraction tool [19], following a pilot extraction of the documents identified during the pilot searches described above. The draft data extraction tool will be modified as required throughout the review, as described in the JBI Reviewer’s Manual [19]. These modifications will be documented in the full scoping review report.

One reviewer will perform data extraction, and a second reviewer will check it. Reviewers will resolve initial disagreements by consensus. If consensus cannot be reached, an independent panel of research team members will review the extraction. We will perform further AI-powered verification of data extraction, as reported below.

We anticipate needing some flexibility for gray literature extraction, either through iterative development of the data extraction tool, the potential need to develop a bespoke gray literature extraction tool, or extracting and reporting specific gray literature in alternative formats such as narrative.

AI Support for Evidence Selection and Data Extraction

We will use AI to verify full-text screening. This approach aims to support and improve researchers’ accuracy rather than replace them with AI trained to perform these tasks. Human decisions and data extraction will serve as the gold standard, and AI decisions and extraction will be used only to check for errors. For evidence selection, we will apply the protocol to each item; extract verbatim evidence for each research question and subquestion; classify each as meets, unclear, or no; provide a confidence score for this decision; and generate an overall decision of include, exclude, or unclear [28]. For data extraction, Microsoft 365 Copilot (Microsoft) [29] will extract evidence verbatim and organize it by research question. The human reviewer will also complete data extraction using the reviews’ template. The researcher and Copilot will then compare the AI- and human-extracted data iteratively and complementarily to develop the data extraction table to be presented in the review.

For data extraction, Copilot will run 2 identical sequential passes [30]. The first pass will use criteria-aware keyword and semantic matching to screen titles and abstracts, followed by targeted section scanning and semantic retrieval-based question answering for data extraction. The second pass will cross-validate these preliminary decisions and flag uncertainties, with any discrepancies sent to arbitration. Researchers remain responsible for final decisions. Disagreements and uncertainties found by AI will be reviewed and resolved by the human researcher. We will develop prompts in Copilot using a human-double-verified 10% gold standard dataset from both stages to create a single prompt command. We will then blind-test this command for accuracy and calibrate it against the gold standard dataset. We will report these stages in the final review.

Analysis of Evidence

Remembering “that scoping reviews do not synthesize the results/outcomes of included sources of evidence” [31] but rather aim to describe or map the available evidence, we outline below our plan for analyzing the anticipated evidence in relation to the review question and each subquestion.

Analysis will be guided by the review objective and research questions and will follow the JBI guidance for scoping reviews [19]. The analysis will organize, describe, and map the available evidence rather than undertake statistical or qualitative evidence synthesis. We will summarize quantitative and categorical characteristics of included sources descriptively using counts and frequencies. Textual data will be descriptively categorized using basic qualitative content analysis, where appropriate. Categories will be developed iteratively from the extracted data and applied consistently across included sources. The resulting analysis will map how exit interviews have been defined, used, and implemented across health professions, educational contexts, and settings and identify areas where evidence is limited or absent.

For the primary review question, exit interview tools and approaches will be descriptively categorized according to their characteristics and context of use. For subquestion 1, study and source characteristics will be summarized using descriptive counts and frequencies. For subquestions 2 to 4, definitions, purposes, and questions used in exit interviews will be grouped into descriptive categories using basic qualitative content analysis. For subquestion 5, reported barriers and facilitators will similarly be categorized and mapped according to relevant contextual characteristics, such as profession, educational setting, and exit interview approach. No meta-analysis or formal qualitative evidence synthesis is planned.

Presentation of Results

The results will be presented in accordance with the review questions and the analytic approach described above. A summary table will provide the principal characteristics of included sources and relevant findings for each review question. The study selection process will be presented using a PRISMA flow diagram. Additional tables and figures will be included where they improve the clarity and accessibility of the mapped evidence. The final format will be informed by the volume and characteristics of the evidence identified.

For the primary review question, the findings will be presented to show the range and characteristics of exit interview approaches identified across health professions education. A summary table will describe the contexts in which exit interviews have been used and the key characteristics of the approaches reported. Where supported by the available evidence, findings will be presented across relevant contextual characteristics, such as health profession, educational or career stage, setting, country, timing of the exit interview, and interview format. These characteristics will be included where they are reported consistently enough across sources to allow meaningful comparison.

The characteristics of the included literature will be presented using tables and descriptive counts. These will include, where available, publication year, country, health profession, population, setting, and source or study type. Simple graphical displays may be used where these improve interpretation, for example, to illustrate the distribution of evidence across professions or changes in the volume or type of evidence over time.

Definitions and descriptions of exit interviews will be presented according to the descriptive categories developed during analysis. Where the available evidence permits, these may also be presented chronologically to demonstrate how definitions and their component characteristics have been reported over time. Similarly, the purposes and research questions for which exit interviews have been used will be presented according to the descriptive categories developed during analysis and, where appropriate, mapped against relevant contextual characteristics, such as health profession, educational setting and stage of training or practice. Counts may be used to illustrate how frequently particular categories are represented within the included literature.

The content of exit interviews will be presented by grouping reported questions or areas of inquiry according to the descriptive categories identified during analysis. Where complete exit interview tools or schedules are reported, these will be identified and their principal characteristics summarized. The presentation will distinguish, where possible, between sources reporting individual interview questions, those reporting broader topic areas or domains, and those describing an exit interview without reporting its content.

Finally, reported barriers and facilitators to implementing exit interviews will be presented according to the descriptive categories developed during analysis. Where sufficient evidence is available, these will be mapped against relevant contextual characteristics, such as profession, educational setting, interview approach, or interviewer characteristics. This presentation will describe the contexts in which barriers and facilitators have been reported rather than assess their relative importance or effectiveness.

Results

The protocol was prospectively registered on OSF Registries on January 22, 2026. At the time of manuscript submission, search strategies and data extraction procedures had been developed and pilot-tested. The review is scheduled to run from October through December 2026. The completed review is expected to be submitted for publication in late 2026.

Discussion

This scoping review will map how exit interviews have been used, implemented, and conceptualized across health professions education. Although exit interviews are well established as an organizational practice and have been proposed as a means of understanding health care workforce turnover, their role within health professions education is less clearly defined. This review does not aim to determine whether exit interviews are effective but to establish the nature and extent of the available evidence, describe how they have been used across professions and educational contexts, and identify areas where evidence is limited or absent.

This distinction between use and effectiveness is important. Exit interviews may identify why learners or health professionals leave a program, workplace, specialty, or profession, but this does not demonstrate that conducting them subsequently improves retention, educational quality, working conditions, or other outcomes. A previous Cochrane review found no eligible evidence from which to determine the effectiveness of exit interviews in reducing turnover among health care professionals [9]. Our review addresses a different question by examining the broader ways in which exit interviews have been used and conceptualized, including their application within educational and professional transitions.

The intersection between education and employment is particularly relevant to health professions education. Learners and trainees may simultaneously occupy educational and employment roles and may move between programs, employers, clinical environments, geographic locations, and stages of professional practice. Departure may therefore provide information not only about workforce turnover but also about educational experiences, training environments, transition into professional practice, career decision-making, and the relationship between educational preparation and workplace expectations [32]. Mapping the purposes for which exit interviews have been used may help clarify whether and how these functions are represented in the existing literature.

Existing organizational literature suggests that exit interviews can provide information about employees’ reasons for leaving and their experiences within organizations while also highlighting challenges related to participation, willingness to disclose negative experiences, interviewer-interviewee power differentials, and whether the findings subsequently result in organizational action [2,10]. International concern about health care workforce attrition and retention strengthens the rationale for understanding how organizations collect and use information from departing staff.

The proposed review extends this previous work by examining exit interviews across health professions education and by including both academic and gray literature. It encompasses exit interviews associated with education and training, transitions between roles or locations, and departure from employment where these experiences are relevant to health professions education. The international scope is also important. Although examples motivating the review include UK initiatives and policy discussions, health care employment arrangements, professional training systems, and relationships between educational institutions and employers vary internationally. Mapping this literature may therefore identify approaches and contexts that are not represented in the UK literature.

A principal strength is the breadth of the planned search across health and care, business and management, and gray literature sources, recognizing that relevant evidence may lie outside conventional health professions education literature. The search strategy has been developed iteratively with specialist librarian input. Prospective registration on the OSF, use of established JBI methodology, and planned reporting using PRISMA-ScR provide an additional framework for transparency.

There are several anticipated limitations. First, terminology is likely to present a challenge. Relevant activities may not always be described as “exit interviews,” while the same term may refer to different practices. Despite using multiple related search terms, we may miss relevant evidence using unanticipated terminology. Substantial heterogeneity may also exist across health professions, educational stages, countries, and training systems. While such heterogeneity is informative for a mapping review, it may make direct comparison between approaches inappropriate.

Second, gray literature searching necessarily involves pragmatic limits. Screening the first 100 results from each source, ranked by relevance, may miss relevant documents that appear later in the results. Gray literature searching may also be less reproducible than searching bibliographic databases because website content, indexing, and relevance algorithms can change. Search procedures will therefore be documented to maximize transparency.

Third, the screening strategy represents a pragmatic modification of the usual JBI approach. Following an initial phase of independent screening and demonstration of substantial agreement, a single reviewer will undertake the remaining screening. This approach cannot eliminate the possibility of erroneous exclusion or other human error. AI-supported verification will provide an additional check, but human reviewers will remain responsible for all decisions about study inclusion, exclusion, and data extraction.

Finally, we will review only English-language evidence, which may limit the international representativeness of the findings. No formal assessment of methodological quality or risk of bias is planned, consistent with the purpose of this scoping review. Consequently, the review will describe the extent, characteristics, and content of the literature rather than make conclusions about the strength of evidence supporting particular exit interview approaches.

This review may provide a foundation for more systematic study and use of exit interviews within health professions education. It may identify areas of consistency and variation in purpose, content, and implementation, alongside gaps related to specific professions, career stages, educational settings, countries, or types of transition. These gaps could help prioritize subsequent primary research.

The findings may also be relevant to educational institutions, health care employers, training organizations, professional bodies, and policymakers considering the development or refinement of exit interview processes. Potential questions for future research include when exit interviews should be conducted, who should conduct them, how confidentiality and power differentials should be managed, which domains should be explored, and how information can be translated into measurable educational or organizational improvement. However, this review will not establish a standardized model or determine which approaches are most effective. Any recommendations will need to distinguish practices reported in the literature from practices supported by evidence of effectiveness.

Ultimately, an important future step may be to move from describing exit interview practice to evaluating it. If this review identifies candidate approaches or commonly explored domains, subsequent research could examine their acceptability, feasibility, and effectiveness, including whether collecting and responding to exit interview information produces measurable improvements in education, transition to practice, workplace experience, or workforce retention [33].