Background
Within the UK, the average length of time of an ear, nose and throat (ENT) placement is 7.3 days.1 Despite compulsory ENT teaching being provided by 37 of 47 (79%) UK medical schools, gaps in undergraduate ENT education persist.1 Recent graduates have nevertheless reported low confidence in managing common ENT conditions.2
In the UK, newly qualified doctors undertake Foundation training across a series of medical and surgical rotations and may provide first-line assessment of acute ENT presentations.3 As a result, many resident doctors will act as the ENT Senior House Officer (SHO), regardless of their career interests. Where senior ENT clinicians are non-resident out of hours, junior doctors may be the first clinicians to assess an acute ENT emergency.4 It is therefore important that newly qualified doctors are able to recognise and initiate management of common ENT emergencies whilst awaiting senior support. However, self-reported confidence should not be conflated with objective clinical competence.
Given the limited duration of undergraduate ENT placements, students may have few opportunities to observe or participate in the management of acute ENT emergencies. Simulation-based education provides a safe and reproducible environment in which learners can rehearse the recognition and initial management of acute clinical scenarios. Previous studies have reported improvements in learners’ self-reported confidence following simulation-based teaching, including within ENT.5,6
However, much of the published ENT simulation literature has focused on postgraduate specialty trainees, procedural skills or bootcamp-style training. There is comparatively limited evidence evaluating a structured, multi-scenario ENT emergency simulation programme among final-year medical students approaching transition into clinical practice. It therefore remains unclear whether a brief simulation intervention embedded within an undergraduate ENT placement can improve immediate self-reported confidence in recognising and initiating management of acute ENT emergencies.
The programme was developed as a structured undergraduate simulation intervention focussed on the recognition and management of selected acute ENT emergencies. The programme incorporated three scenarios: supraglottitis, epistaxis and post tonsillectomy haemorrhage. These conditions were selected because they represent clinically important acute presentations in which prompt recognition and initial management are required from the first attending clinician.
Collectively, they expose learners to both haemorrhagic and airway emergencies and align with General Medical Council outcomes requiring graduates to recognise and initiate management of acutely deteriorating patients.7
Aim
To evaluate the immediate change in final year medical students’ self-reported confidence in recognising and initiating management of selected acute ENT emergencies following participation in a structured simulation based teaching session.
Methods
This was a prospective pre- and post-intervention evaluation of a structured ENT emergency simulation programme delivered to final-year medical students during a mandatory undergraduate ENT placement. The simulation programme was a pre-existing educational intervention embedded within the undergraduate ENT placement. The present study prospectively evaluated this programme using anonymised pre- and post-intervention questionnaire data.
To improve transparency and reproducibility, the description of the educational intervention was structured with reference to relevant elements of the Guideline for Reporting Evidence-based practice Educational interventions and Teaching (GREET), including the educational setting, facilitator characteristics, intervention content, delivery, duration, standardisation and debriefing approach.8 The primary outcome was the immediate change in self-reported confidence following the simulation session.
Full scenario scripts, progression cues, expected learner actions and scenario end-points are provided in Supplementary Material 1.
Participants
All eligible final-year medical students rotating through the ENT placement during the 16-week study period were included (n=110). All 110 students completed both the pre- and post-intervention questionnaires, giving a response rate of 100% and no missing data for the three clinical-scenario items included in the primary analysis. No a priori sample-size calculation was performed because the study included all eligible students undertaking the ENT placement during the predefined 16-week study period.
UGs participated in the ENT simulation day during their mandatory five-day ENT placement; completion of the questionnaire was voluntary and had no bearing on academic assessment or progression. Students were organised into small groups of six. Within these groups, learners worked in pairs to manage each simulated emergency case, with the remaining members adopting the roles of patients or nursing colleagues.
This structure was intentionally designed to promote an appreciation of multidisciplinary teamwork, particularly the pivotal role of nursing staff in frontline ENT emergencies, and to reinforce the importance of clear, continuous communication during acute clinical encounters. The simulation session lasted approximately three hours and comprised three acute ENT scenarios delivered in a fixed order: supraglottitis, epistaxis and post-tonsillectomy haemorrhage. Each session was facilitated by two clinical teaching fellows, one in ENT and one in medicine.
One of the authors was involved in delivering the simulation sessions in their role as an ENT teaching fellow but had no role in the original design of the programme, did not act as an academic assessor or placement supervisor, and was not involved in decisions regarding students’ academic progression. The second teaching fellow involved in session delivery was not an author of the study. The same facilitators delivered the majority of sessions throughout the 16-week study period, with substitution only when required because of staff absence.
Facilitators were briefed regarding the structure and progression of each scenario before delivery.
Questionnaire
A questionnaire (Supplementary material 2) was administered immediately before and after the simulation session. The primary analysis focused on three separate five-point Likert items assessing self-reported confidence in recognising and initiating management of the three simulated clinical emergencies: post-tonsillectomy haemorrhage, epistaxis and supraglottitis. These three items were selected for the primary analysis because they corresponded directly to the clinical scenarios delivered during the intervention and to the study aim of evaluating confidence in recognising and initiating management of acute ENT emergencies.
Responses ranged from 1 (“strongly disagree”) to 5 (“strongly agree”), and each clinical scenario was analysed separately. The questionnaire also contained two ancillary educational-evaluation items relating to identification of ENT equipment and understanding of in-hospital emergency calls. These items assessed distinct educational domains, were not treated as components of a composite confidence measure, and were not included in the primary statistical analysis.
Descriptive response distributions for all five questionnaire items at both time points are provided in Supplementary Table 1.
The content of the simulation programme was developed through a two-stage consultation process. Recent ENT Senior House Officers (SHOs) first identified acute presentations that they frequently encountered during on-call duties and considered important for junior doctors to recognise and manage. These candidate scenarios were subsequently reviewed by the lead consultant for the undergraduate ENT rotation in consultation with two additional ENT colleagues, with consensus reached regarding the final selection of supraglottitis, epistaxis and post-tonsillectomy haemorrhage.
The questionnaire was subsequently piloted with postgraduate clinical fellows prior to implementation. Feedback identified ambiguity in the wording of individual items, which were revised for clarity before administration to the study cohort.
Questionnaire responses were collected anonymously, with no names, unique identifiers or other identifying information recorded. Although the same cohort completed both questionnaires, individual pre- and post-intervention responses could not therefore be matched at participant level. Anonymous data collection was selected deliberately to preserve participant anonymity and minimise any perception that individual responses could be linked to academic assessment.
Although facilitators could determine when the number of submitted questionnaires was lower than the number of students attending a session, no identifying information was collected and it was therefore not possible to determine which individual student had chosen not to complete the questionnaire.
ENT Simulation Programme and Scenarios
At the start of the simulation day, participants were asked to complete a pre-session questionnaire. This included a five-point Likert scale in which UGs self-rated their confidence in recognising and managing key ENT emergencies. Responses were recorded using individual five-point Likert items, ranging from 1 (“strongly disagree”) to 5 (“strongly agree”).
The three clinical conditions (namely supraglottitis, epistaxis, and post-tonsillectomy haemorrhage) were selected through the consultation process described above. The pre-session questionnaire was also helpful in encouraging self-reflection amongst the participants and establishing a baseline for comparison.
In this study, the supraglottitis scenario represented a progressive upper-airway emergency characterised by sore throat followed by stridor and impending airway compromise.
In addition to the questionnaire, participants were offered the opportunity to familiarise themselves with core practical skills prior to commencing the scenarios. These activities included supervised practice using common pieces of ENT equipment such as nasal specula, crocodile forceps, and Tilley’s forceps. The scenarios were delivered as low-fidelity simulations, with students adopting simulated patient roles rather than using simulation manikins.
Relevant clinical equipment was incorporated as props, including rapid rhinos and Magill forceps. The equipment-familiarisation component was intended to provide participants with exposure to commonly used ENT instruments before undertaking the emergency scenarios.
Each scenario followed a written script specifying the information available to the simulated patient and cues determining when additional information should be disclosed. Scenario progression was standardised using predefined verbal and behavioural triggers; participants were required to demonstrate or verbalise key actions, such as escalation to the ENT registrar, before subsequent stages of the scenario were introduced.
No formal checklist of predefined learning objectives or performance criteria was used. Participants were expected to manage each scenario within the scope of a newly qualified doctor, including recognising when escalation and senior support were required.
For each scenario, two participants acting as observers were assigned specific behavioural and organisational elements to monitor, including communication clarity, delegation strategy, and resource allocation. Each simulated case ran for ten minutes. Following each scenario, a structured facilitator-led debrief was conducted.
Observing students were allocated predefined behavioural and organisational domains, such as communication, delegation and resource allocation, on which to focus during the scenario. For each domain, participating students were first invited to reflect on their own performance, after which observers provided feedback supported by specific examples from the scenario, including observed actions or verbal communication. The facilitator then consolidated the discussion by highlighting effective aspects of performance and identifying areas for improvement.
Following this, the peers who had assumed the roles of “patient” and “nurse” provided feedback on their lived experience of the scenario, focusing on interpersonal dynamics, human factors and situational awareness. Observer feedback was then offered, with emphasis on specific objective behaviours witnessed during the scenario. The clinical reasoning and management of each emergency were subsequently reviewed by the ENT Teaching Fellow, relating the participants’ management approach to real-world clinical practice and reinforcing the key learning points from each scenario.
Participants were subsequently provided the opportunity to practise select high-yield skills relevant to the scenario, such as rapid rhino insertion or safe silver nitrate cautery, using simulation models.
The scenario content, sequence, scripts, progression triggers and debriefing structure were kept consistent throughout the 16-week study period. No modifications were made to the scenarios during data collection in order to maintain consistency of the intervention across participant groups.
Immediately following completion of the simulation session, participants completed an identical post-intervention questionnaire assessing the same five questionnaire items. Responses were collected using Microsoft Forms and handled in accordance with institutional GDPR requirements.
Statistical Analysis
Questionnaire responses were recorded using individual five-point Likert items and were treated as ordinal data on conceptual grounds; no normality testing was used to determine the analytical approach. Median values and median absolute deviations (MAD) were calculated to summarise response distributions for each of the three prespecified clinical-scenario outcomes, and differences between pre- and post-intervention group medians were reported descriptively. Although the same cohort completed the pre- and post-intervention questionnaires, responses were collected anonymously and could not be matched at participant level.
A paired analysis was therefore not possible, and the pre- and post-intervention response distributions were compared pragmatically using the two-sided Mann–Whitney U-test. This analysis should be interpreted as a comparison of the marginal pre- and post-intervention response distributions rather than a participant-level paired analysis. Rank-biserial correlation (r_rb) was calculated as the effect-size measure for each comparison, with positive values indicating higher post-intervention responses.
Ninety-five per cent confidence intervals for r_rb were estimated using non-parametric bootstrap resampling of the pre- and post-intervention response distributions, with the two groups resampled independently to reflect the unpaired analysis. To account for multiplicity across the three prespecified clinical-scenario comparisons, p-values were adjusted using the Holm method to control the family-wise error rate, and a two-sided Holm-adjusted p-value of <0.05 was considered statistically significant. All 110 pre-intervention and 110 post-intervention observations were included in each of the three primary comparisons.
There were no missing data for these outcomes; therefore, no imputation or other missing-data procedure was required. No a priori sample-size calculation was performed because the evaluation included all final-year medical students undertaking the ENT placement during the predefined 16-week study period.
Results
Analysis of the pre- and post-intervention responses showed significantly higher self-reported confidence scores at post-intervention assessment across all three clinical scenarios. All 110 students completed both the pre- and post-intervention questionnaires, resulting in a 100% questionnaire response rate. There were no missing data for the three clinical-scenario items included in the primary analysis.
Figure 1 presents the distribution of pre- and post-intervention Likert responses for the three clinical-scenario items. Exact response frequencies for all five questionnaire items at both time points are provided in Supplementary Table 1. Pre-intervention median scores were 3 for post-tonsillectomy haemorrhage and 2 for both epistaxis and supraglottitis, compared with a post-intervention median of 4 for each scenario.
Post-intervention responses were concentrated toward the upper end of the five-point Likert response range, suggesting a potential ceiling effect. Figure 1Distribution of pre- and post-intervention self-reported confidence scores for the three acute ENT emergency scenarios. Boxes represent the interquartile range, horizontal lines indicate the median, and whiskers represent the minimum and maximum observed responses.
Confidence was assessed using individual five-point Likert items, ranging from 1 (“strongly disagree”) to 5 (“strongly agree”).A box and whisker plot showing pre and post confidence scores across three clinical scenarios.
Table 1 presents the pre- and post-intervention median Likert responses and median absolute deviations (MAD), together with differences in group medians, Mann–Whitney U statistics, rank-biserial effect sizes with 95% confidence intervals, and Holm-adjusted p-values for each clinical scenario.Table 1Pre- and Post-Intervention Self-Reported Confidence in Recognising and Managing Acute ENT EmergenciesOutcomePre nPre Median(MAD)Post nPost Median(MAD)MedianDifferenceMann–Whitney URank-biserialCorrelation, r_rb (95% CI)Holm-Adjustedp-valuePost-tonsillectomy haemorrhage1103 (1)1104 (0)+111,187.50.849 (0.785–0.898)1.17 × 10−30Epistaxis1102 (1)1104 (0)+211,739.00.940 (0.891–0.970)7.93 × 10−36Supraglottitis1102 (0)1104 (0)+211,786.00.948 (0.899–0.975)1.95 × 10−36Notes: Data are presented as median (median absolute deviation [MAD]). Pre- and post-intervention responses were compared using the Mann–Whitney U-test because anonymous data collection prevented participant-level matching. P-values were adjusted for the three prespecified clinical-scenario comparisons using the Holm method.
Rank-biserial correlation (r_rb) represents the effect size, with positive values indicating higher post-intervention responses. Ninety-five per cent confidence intervals for r_rb were estimated using non-parametric bootstrap resampling of the pre- and post-intervention response distributions, with groups resampled independently to reflect the unpaired analysis.
For post-tonsillectomy haemorrhage, the median self-reported confidence score was 3 (MAD 1) pre-intervention and 4 (MAD 0) post-intervention, representing a one-point difference in group medians. Post-intervention responses were significantly higher than pre-intervention responses (Mann–Whitney U=11,187.5; Holm-adjusted p=1.17×10−30), with a rank-biserial correlation of 0.849 (95% CI 0.785–0.898).
For epistaxis, the median self-reported confidence score was 2 (MAD 1) pre-intervention and 4 (MAD 0) post-intervention, representing a two-point difference in group medians. Post-intervention responses were significantly higher than pre-intervention responses (Mann–Whitney U=11,739.0; Holm-adjusted p=7.93×10−36), with a rank-biserial correlation of 0.940 (95% CI 0.891–0.970).
For supraglottitis, the median self-reported confidence score was 2 (MAD 0) pre-intervention and 4 (MAD 0) post-intervention, representing a two-point difference in group medians. Post-intervention responses were significantly higher than pre-intervention responses (Mann–Whitney U=11,786.0; Holm-adjusted p=1.95×10−36), with a rank-biserial correlation of 0.948 (95% CI 0.899–0.975).
Overall, substantial differences in response distributions were observed across all three clinical scenarios, with rank-biserial correlations ranging from 0.849 to 0.948.
Discussion
Principal Findings
The study found significantly higher self-reported confidence in recognising and initiating the management of supraglottitis, epistaxis and post-tonsillectomy haemorrhage immediately following a dedicated ENT emergency simulation session. ENT emergencies, particularly those involving the airway, require rapid recognition and implementation of life-saving treatments. The higher post-intervention confidence scores were observed following a relatively brief, targeted educational intervention delivered during the students’ ENT placement.
For final year medical students approaching transition to postgraduate practice, simulation may provide an opportunity to encounter and rehearse the initial management of acute ENT presentations in a supervised environment.
The comparatively large change observed for supraglottitis should be interpreted in the context of lower baseline confidence for this scenario, while the higher baseline scores observed for some other scenarios may have limited the magnitude of measurable improvement through a ceiling effect.
Literature Supporting Our Study Findings
The present findings are consistent with previous studies reporting increased self-reported confidence following simulation-based emergency training among medical students and junior doctors.9,10 Within ENT, however, much of the existing simulation literature has focused on specialist trainees or residents rather than final-year medical students approaching the transition to postgraduate clinical practice. Previous ENT “bootcamp” interventions have used simulation to support junior doctors assuming greater specialty-specific clinical responsibility.11 There remains comparatively limited evidence evaluating multi-scenario acute ENT simulation specifically among final-year medical students before they assume postgraduate clinical responsibilities.
Current Challenges in British ENT Practice
Newly qualified doctors in ENT SHO roles may be the first responders to acute ENT emergencies while senior support is off-site.4 In a study evaluating overnight ENT cover across England, 68% of first-on-call doctors for ENT reported having no previous ENT experience and 42% reported feeling uncomfortable managing the most common ENT emergencies as the first-on-call doctor.4 Whilst precise data on the incidence of ENT emergencies in the UK is somewhat limited, audit data suggests that epistaxis accounts for 25,000 hospital presentations annually.12 This illustrates the potential volume of acute ENT presentations encountered within secondary care. Limited exposure (average placement 7.3 days) and the variability in ENT teaching across UK medical schools1 support further consideration of how acute ENT presentations are addressed within undergraduate curricula. The higher self-reported confidence observed following this study suggests that focussed simulation may have a role as an adjunct to undergraduate teaching on acute ENT presentations.
Educational Impact of ENT Simulation
The primary outcome of this study was self-reported confidence, and the findings should therefore be interpreted as changes in participants’ perceptions rather than evidence of improved clinical competence or performance. Previous literature has reported improvements in undergraduate self-reported confidence following simulation-based teaching.13,14 However, clinicians may not always accurately assess their own ability, and self-reported confidence should not be regarded as a proxy for clinical competence. Accordingly, the higher post-intervention confidence observed in this study cannot be assumed to represent an equivalent improvement in clinical knowledge, technical ability or performance.
Simulation-based assessment can generate learner anxiety, although the relationship between anxiety and objective performance appears complex.15,16 However, neither anxiety nor objective clinical performance was measured in the present study. Participants were not formally assessed using an objective clinical performance measure or standardised performance checklist. Facilitators supervised each scenario and intervened where necessary to maintain the safety of the simulated encounter; these observations were used for formative feedback rather than as study outcomes.
The intervention was designed to promote a psychologically safe learning environment in which participants could make and discuss clinical decisions without fear of adverse consequences. Psychological safety has been described as an important consideration in facilitating learner engagement during simulation.17 Being required to perform in front of peers may generate learner anxiety and potentially hinder engagement with the learning process.18 Facilitators encouraged this by setting clear expectations and pairing participants for each scenario. Whilst pairing possibly reduced the realism of being “on-call”, a priority was to help them feel more confident and empowered.
The simulated scenarios enabled students to adopt the role of the assessing clinician and practise clinical decision-making in a supervised environment without risk to patients.
The intervention also incorporated communication and human factors as explicit components of the learning experience. Students not directly managing a scenario adopted patient, nursing or observer roles, while observers focused on predefined domains including communication, delegation and resource allocation. The structured debrief subsequently enabled participants to reflect on these aspects of their performance using specific observations from the scenario.
Although this provided opportunities to practise and receive feedback on multidisciplinary communication and teamwork, these domains were not independently measured as study outcomes; consequently, no conclusions can be drawn regarding improvement in communication or human-factor performance.
Plausibility of Implementation of the ENT Simulation Programme
The structured format and use of relatively low-fidelity simulation suggest that the intervention may be adaptable to other undergraduate ENT teaching settings. However, the present single-centre study did not evaluate implementation across different institutions, and its scalability or effectiveness at a national level cannot be established from these data. Previous literature has demonstrated widespread discomfort amongst resident doctors when they are on-call as the ENT SHO.4 Further multicentre evaluation would be required to determine whether a standardised intervention could be implemented consistently across institutions with differing curricula, staffing and simulation resources.
The content of the intervention is relevant to GMC “Outcomes for Graduates” relating to the recognition of critically unwell patients and interprofessional teamwork. The same framework highlighted the importance of collaboration and inter-professional teamwork in emergencies.
The session was designed to incorporate these domains through multidisciplinary role-play, with participants not directly managing the scenario assuming the roles of other members of the multidisciplinary team. These findings support further evaluation of structured ENT emergency simulation as an adjunct to undergraduate ENT teaching, particularly during the transition toward postgraduate clinical practice.
Limitations and Future Studies
Several limitations should be considered when interpreting these findings. First, the study used an uncontrolled pre- and post-intervention design. In the absence of a comparator group, the higher post-intervention confidence scores cannot be attributed solely to the simulation intervention.
Furthermore, the teaching session incorporated several educational components, including equipment familiarisation, simulated clinical encounters, peer role-play, facilitator teaching and structured debriefing. The relative contribution of each component to the observed differences in self-reported confidence cannot therefore be determined.
Second, although the same cohort of students completed the pre- and post-intervention questionnaires, anonymous data collection prevented participant-level matching. Consequently, within-participant changes could not be evaluated, and the pre- and post-intervention response distributions were analysed as unmatched observations. This approach does not account for the correlation that may have existed between repeated responses from the same participants.
Third, the primary outcome was immediate self-reported confidence rather than objectively assessed clinical competence or performance. No standardised assessment of clinical knowledge, technical skill or clinical performance was undertaken. Self-reported responses may also have been influenced by response-shift bias, demand characteristics and social desirability.
In addition, the concentration of post-intervention responses towards the upper end of the five-point Likert scale suggests a potential ceiling effect, which may have limited the instrument’s ability to discriminate between participants reporting higher levels of confidence. The findings should therefore not be interpreted as evidence of equivalent improvements in clinical competence or performance.
The use of an identical questionnaire immediately before and after the intervention may also have introduced testing effects, while facilitator involvement and the mandatory placement context may have influenced participants’ responses despite questionnaire completion being voluntary and anonymous.
Regression to the mean cannot be excluded entirely in a pre- and post-intervention design, although participants were not selected on the basis of extreme baseline confidence scores, which reduces the likelihood that this phenomenon was a major driver of the observed differences.
The questionnaire represents a further limitation. Although its content was informed by expert consultation and its wording was refined following pilot testing with postgraduate clinical fellows, it did not undergo formal psychometric evaluation of construct validity, reliability or responsiveness. The findings should therefore be interpreted in the context of a locally developed self-report instrument.
The observed differences also cannot be separated from any concurrent learning occurring during the wider ENT placement.
The duration of each individual scenario was limited to approximately 10 minutes and therefore did not reproduce the full duration or complexity of an acute clinical encounter. However, the scenarios were designed specifically to represent the initial response to an ENT emergency, including recognition, initiation of immediate management and appropriate escalation for senior support. Future studies could evaluate longer or more complex scenarios and incorporate objective measures of clinical knowledge and performance alongside self-reported confidence, allowing the relationship between perceived confidence and demonstrated competence to be examined.
The post-intervention questionnaire was administered immediately following the simulation session, and no longitudinal follow-up was undertaken. It was therefore not possible to determine whether the higher self-reported confidence observed post-intervention was sustained over time or translated into subsequent clinical behaviour. Longitudinal follow-up after graduation would be valuable to evaluate retention of self-reported confidence and its relationship with subsequent objective measures of clinical performance.
As one author was involved in delivering the teaching intervention, facilitator-related response bias cannot be excluded, although questionnaire responses were anonymous and had no bearing on academic assessment or progression.
Finally, although the three scenarios were selected to represent important acute ENT presentations relevant to junior doctors, they do not encompass the full breadth of ENT emergencies encountered in clinical practice. The study was also conducted within a single undergraduate ENT placement, which may limit the generalisability of the findings to other institutions, curricula and learner populations. Future multicentre studies could evaluate a broader range of acute ENT presentations and determine whether the findings are reproducible across different educational settings.
Conclusion
In this prospective pre- and post-intervention evaluation involving 110 final-year medical students, self-reported confidence in recognising and initiating the management of post-tonsillectomy haemorrhage, epistaxis and supraglottitis was significantly higher immediately following a structured ENT emergency simulation session. These findings add to the existing literature on simulation-based undergraduate education by demonstrating its application to acute ENT presentations during the transition toward postgraduate clinical practice.
The intervention provided a structured and standardised environment in which students could encounter acute ENT scenarios and participate in multidisciplinary role-play, clinical decision-making and structured debriefing. However, the study assessed immediate self-reported confidence rather than clinical competence, performance or longer-term preparedness, and the absence of a comparator group limits causal attribution of the observed differences to the simulation intervention alone. Further evaluation incorporating objective performance measures, longitudinal follow-up and multicentre evaluation is required to determine whether these findings translate into sustained educational or clinical benefits.