Out-of-hospital cardiac arrest (OOHCA) is a major public health issue worldwide, with rates of survival to hospital discharge remaining around 10% (Yu et al., 2020). Since the outbreak of the COVID-19 pandemic, the incidence of OOHCA has risen more than two-fold, while rates of bystander cardiopulmonary resuscitation (CPR) have decreased (Lim et al., 2020). This reduction is likely to be multifactorial but is thought to be predominantly due to increased fear of contracting an infectious disease, such as COVID-19 (Baldi et al., 2020; Lim et al., 2020; Uy-Evanado et al., 2021). The provision of first aid, including basic life support (BLS) by bystanders, before the arrival of emergency medical services, is a critical step in the chain of survival behaviors, and results in improved neurological function and lower all-cause mortality (Giacoppo, 2019; Kragholm et al., 2017; Yan et al., 2020). There is substantial global variation in the provision of bystander CPR (Shekhar & Narula, 2022; Yan et al., 2020; Yu et al., 2020), and countries with lower rates suffer decreased survival following OOHCA (Yan et al., 2020).
BLS skills confidence
Being able to recognize the need for intervention and having the confidence and willingness to administer BLS when required, are arguably at least as important as proficiency in delivering BLS skills (Abelsson et al., 2020). BLS training has been demonstrated to improve hypothetical willingness to perform bystander CPR (Blewer et al., 2020; Cho et al., 2010; Yu et al., 2020). While research has assessed the effect of BLS training on the general population’s willingness to perform CPR, fewer studies have examined this effect in medical students.
The General Medical Council stipulates that doctors in the United Kingdom must provide emergency care in the community (2019). However, UK medical student BLS teaching lacks standardization and is of variable quality (Phillips & Nolan, 2001), and medical students and junior doctors can lack competence in administering BLS (de Ruijter et al., 2014; Passali et al., 2011). While medical students do not have the same professional obligation as qualified doctors (Medical Defence Union, 2022), students may feel there is a public expectation to intervene in an emergency (Xie et al., 2019). Empowering and providing medical students with the skills to administer first aid, including CPR, during an emergency is therefore of benefit to both students and the wider public.
RMD Bristol: Peer-led medical student BLS training
Peer-led BLS teaching in medical schools has emerged as an effective alternative to staff-led courses (Abbas et al., 2018; Harvey et al., 2012). University of Bristol Medical School utilizes a peer-led BLS training scheme called Resuscitation Mentorship Development (RMD Bristol) for first-year medical students, based on a model first implemented by the University of Birmingham (Harvey et al., 2012). The RMD model has been previously described and shown to be highly effective in terms of teaching delivery, assessment, and supporting students through training (Gillam et al., 2023; Perkins et al., 2002). RMD teaching follows the European Resuscitation Council’s (ERC) BLS provider program, qualifying successful participants with an ERC BLS provider certificate (European Resuscitation Council, 2023). Teaching is delivered by medical student peers from years 2 to 5 of the medical program, supported by experienced qualified clinicians. All instructors complete an ERC BLS instructor course.
RMD BLS training utilizes a blended approach, with online instructional videos and a multiple-choice question quiz, followed by an in-person teaching component, with approximately five hours of total teaching time. This in-person training covers practice performing the BLS algorithm, including CPR, and automated external defibrillator (AED) use, as well as basic first aid skills, including BLS modifications for children and drowning victims, the recovery position, and treating choking. In the final session, participants are given the opportunity to practice their skills and then undertake a scenario-based summative assessment. All training is delivered using a modified version of Peyton’s four-step approach; instructor demonstration, step-by-step deconstruction by instructor, formulation through instruction by students, and finally performance by students (Giacomino et al., 2020; Peyton & Walker, 1998). The course structure is outlined further in the methods section.
Impact of COVID-19 on RMD Bristol BLS training
Due to UK government restrictions during the COVID-19 pandemic, RMD Bristol BLS teaching was stopped in March 2020 for a period of six months. During the academic year 2020–2021, a decision was taken to change the delivery mode of BLS training to virtual-only. Teaching continued to include preparatory materials based on ERC guidelines and previously in-person components were modified for online delivery using Zoom video conferencing software (Zoom Video Communications, inc.). Virtual teaching consisted of two, two-hour sessions over a two-week period, simplified to cover BLS and AED use, and the theory behind the recovery position. The virtual-only course also used a modified version of Peyton’s approach (Peyton & Walker, 1998), without the fourth ‘performance’ step, due to the challenges of providing all students with a resuscitation manikin and delivering adequate feedback on skills-based teaching remotely. There was no formal assessment during the virtual training and consequently no ERC certification.
Prior to the COVID-19 pandemic, in-person training components were delivered over four weeks and incorporated training on rescue breaths. Since COVID-19, training has included only the theory of delivering rescue breaths, and accordingly, the in-person component has been reduced to three weeks to reflect this reduced skills complexity. Students who missed their in-person training during the pandemic and were taught virtually, undertook full training the following year, in their second year of study.
The timing of the COVID-19 pandemic and the resulting restrictions also impacted this study. The original study design was to follow a single student year group and investigate any changes to their self-reported confidence and hypothetical willingness to deliver BLS for two years post their original BLS training (initial recruitment 2018). Due to low recruitment of students in the study during 2019 and the impact of the pandemic restrictions on teaching in March 2020, a decision was taken to pause the study. Ethical approval was then obtained to modify the study design to instead investigate a series of subsequent groups of students undertaking BLS training across four academic years between 2018–2022. The objective was changed to explore the effect of a change in delivery mode on students’ confidence in performing BLS skills and to consider whether the pandemic influenced students’ hypothetical willingness to administer BLS. Limitations to the study both in general and related to COVID-19 are explored further in the discussion.
Study aims
The primary aim was to investigate medical students’ self-reported confidence and hypothetical willingness to perform BLS before and after their peer-led BLS training. We developed secondary aims due to the timing of the study during the COVID-19 pandemic, to assess the impact of virtual-only BLS training on medical students’ confidence and willingness to deliver BLS and to examine the impact of the COVID-19 pandemic on the hypothetical willingness of medical students to intervene in an emergency. Through identifying any effect on confidence and theoretical willingness to perform BLS, we hoped to further evidence the value of the RMD BLS peer-led teaching model and its potential for wider application within and outside the field of medical education.
Research questions
Are both in-person and virtual-only peer-led BLS training methods effective in improving medical student confidence and hypothetical willingness to administer BLS?
Does a change in delivery mode affect medical students’ self-reported confidence in performing BLS skills?
How did the COVID-19 pandemic impact medical students’ baseline self-reported confidence and hypothetical willingness to deliver BLS?
Methods
Study design
A prospective cohort study was undertaken of medical students attending BLS training (N = 1084). The data collection period for the study was September 2018 to December 2021. A STROBE reporting guideline was used during the study write-up and the checklist is provided in the supplementary materials. Details of the study population are provided in Table 1.
The breakdown of the study population for the research study.
| ACADEMIC YEAR | STUDENT YEAR GROUP | YEAR GROUP SIZE | BLS TRAINING DELIVERY MODE |
|---|---|---|---|
| 2018–2019 | Year 1 | 271 | Full BLS training with online learning (via ERC website) and in-person component |
| 2019–2020* | Year 1 | 274 | Full BLS training with online learning (via ERC website) and in-person component |
| 2020–2021* | Year 1** | 272 | Virtual-only BLS training (live via Zoom) |
| 2021–2022* | Year 2** | 267 | Full BLS training with online learning (via ERC website) and in-person component |
-
Note: * Academic years affected by COVID-19: Courses were paused in March 2020 and restarted virtual-only in April–May 2021; in 2021–22 courses were swapped to compression-only CPR training.
** The same students taught BLS consecutively over two years due to COVID-19 restrictions (initially virtual-only BLS training and the subsequent year full training with an in-person component).
ERC- European Resuscitation Council.
Participant recruitment
Student participants were informed about the study through a short presentation during their BLS training, alongside email reminders. All students undertaking BLS training were eligible to participate (N = 1084). Written informed consent was obtained via a paper or online form, and all participants answered the same questionnaire before and after their BLS teaching (completed in person in 2018–19 and online in 2019 onwards). Follow-up emails were sent out encouraging participants to complete post-course questionnaires to reduce the risk of loss to follow-up and attrition bias. Participants were made aware they were able to drop out of the study at any time before data analysis.
Questionnaire design
A pre- and post-training questionnaire was used to assess students’ self-reported confidence and hypothetical willingness to deliver BLS, using five-point Likert scale closed-questions. The four-section structured questionnaire (provided in the supplementary materials) is summarized in Table 2.
Study questionnaire section breakdown including question topic and design.
| QUESTIONNAIRE SECTION | QUESTION TOPIC | QUESTION DESIGN |
|---|---|---|
| 1 | Student number | To enable data pairing |
| 2 | Additional BLS experience | Questions to identify prior BLS training, previous BLS teaching experience, and experience performing BLS in real-life |
| 3 | Confidence in performing BLS | Four Likert-scale questions assessing student self-reported confidence in BLS skills: Q1: confidence in performing steps of the BLS algorithm correctly and efficiently (in the correct order and in a timely manner) Q2: confidence in administering effective chest compressions Q3: confidence in performing effective rescue breaths Q4: confidence performing BLS overall |
| 4 | Willingness to perform BLS | Two Likert-scale scenario-based questions to assess student hypothetical willingness to deliver BLS: Q1: in a hospital setting Q2: in a community setting |
Confidence-based questions used a five-point Likert scale: 1-not at all confident, 2-somewhat confident, 3-confident, 4-very confident, 5-extremely confident, while willingness to perform BLS used a five-point Likert scale: 1-not at all likely, 2-somewhat likely, 3-likely, 4-very likely, 5-extremely likely. Likert scales have been extensively utilized for medical education research, including the use of parametric tests to analyze the responses (Sullivan & Artino, 2013). Note questionnaire section 2 gathered data on previous BLS experience, since this may influence baseline student-reported confidence and willingness.
Questionnaire data were split into three groups: 1) year 1 students taught pre-COVID including in-person training (academic year 2018–19), 2) year 1 students taught during COVID with virtual-only training (academic year 2020–21), and 3) year 2 students taught post-COVID including in-person training (academic year 2021–22). Note that Group 3 sampled students from the same population as Group 2.
Due to the change in the provision of in-person rescue breath training because of COVID-19, question 3 in section 3 of the questionnaire assessing confidence in performing rescue breaths was removed for Groups 2 and 3 and excluded from the overall analysis (see Table 2). A visual summary of the structure and content of the BLS teaching delivery for the three groups is provided in Figure 1.
Flow diagram outlining the structure and content of the in-person and online components of the BLS courses.
Note: In-person teaching was delivered to study Groups 1 (pre-COVID) and 3 (post-COVID) using Peyton’s (1998) four-step approach, including lectures, and small-group learning. Teaching was delivered virtually to Group 2 (during COVID) due to UK COVID-19 pandemic restrictions. This was using steps 1–3 of Peyton’s four-step approach, with online lectures, and breakout groups, but no independent practice or formal assessment.
Data analysis
After the matching process, datasets were fully anonymized. Ordinal data are expressed as mean ± standard deviation (SD), with a 95% confidence interval (95% CI) where appropriate. The Shapiro-Wilk test was used and determined the data were not normally distributed. Data were analyzed using Microsoft Excel for Mac Version 16.71. A p-value of less than 0.05 was considered significant.
To determine if there was an overall impact of BLS training on medical student confidence and hypothetical willingness to perform BLS, paired pre- and post-course responses were analyzed from both Group 2 (during COVID) and Group 3 (post-COVID) using the Wilcoxon signed-rank test. Group 1 (pre-COVID) was excluded from this analysis since no paired data were available (Group 1 was lost to follow-up due to a multitude of non-systematic factors including the COVID-19 pandemic, so only pre-course questionnaire responses were obtained- see previous section on study design changes).
To assess whether any improvements in confidence and willingness to administer BLS were sustained over time following virtual-only training, an unpaired analysis of the post-course questionnaire responses from Group 2 (during COVID) and the pre-course questionnaire results from Group 3 (post-COVID) was performed using the Wilcoxon signed-rank test. For this and the above analysis, effect sizes were calculated using z/√n.
To determine if COVID-19 had impacted initial attitudes towards performing BLS, an unpaired comparison of the responses to scenarios 1 and 2 of section 4 of the pre-course questionnaires from Groups 1 and 2 was performed using the Mann-Whitney U test. Note, pre-course responses from Group 3 were not included in this analysis, since they were from the same student year group as Group 2 and had therefore already received virtual teaching, potentially impacting their responses.
Results
A total of 658 questionnaire responses were obtained from 347 students (participants, n = 347, population, N = 1084; approximate response rate = 32%) across the three study groups: 499 pre-course and 159 post-course. The total responses for each group are detailed in Table 3.
The number of questionnaire responses obtained from pre- and post-medical student BLS courses from each of the experimental groups.
| GROUP NUMBER | PRE-COURSE QUESTIONNAIRE RESPONSES | POST-COURSE QUESTIONNAIRE RESPONSES | PAIRED RESPONSES TO BOTH QUESTIONNAIRES |
|---|---|---|---|
| Group 1 | 232 | 0 | 0 |
| Group 2 | 155 | 121 | 91 |
| Group 3 | 112 | 38 | 24 |
Note: Group 1 (pre-COVID), Group 2 (during COVID), and Group 3 (post-COVID) (n = 658 total responses).
108 questionnaire responses were excluded from the paired analysis due to a lack of paired pre- and post-course responses (n = 34 excluded from Group 2, and n = 74 excluded from Group 3). Due to the low number of paired questionnaire responses and a lack of statistical power, it was not possible to directly compare Groups 2 and 3.
Prior BLS experience
A larger proportion of Group 1 (80/232; 34.4%) had received previous BLS training in the last 2 years, compared to Group 2 (40/155; 25.8%). Similarly, more students in Group 1 had previously taught BLS (10/232; 4.3%) than in Group 2 (6/155; 3.9%), and a greater proportion of Group 1 (15/232; 6.5%) had performed CPR on a real person, compared to Group 2 (7/155; 4.5%). Group 3 was excluded due to having received prior RMD Bristol BLS training.
Impact of peer-led BLS training on self-reported confidence in delivering BLS
Section 3 of the questionnaire assessed medical students’ self-reported confidence in undertaking aspects of the BLS algorithm. Paired data from pre- and post-course questionnaires were compared using the Wilcoxon signed-rank test for both Group 2 (during COVID) and Group 3 (post-COVID) (see Table 4). As described in the methods section, participants in Groups 2 and 3 are from the same student year group who first undertook virtual-only training as Group 2 during the pandemic, and then full training including an in-person component after the pandemic. A statistically significant increase in student self-reported confidence in delivering BLS was observed in both groups (r > 0.7 and p < 0.05 across all aspects). The mean post-course confidence rating was higher for all measures for the students in Group 3 who received training with an in-person component (post-COVID) compared to those reported by students in Group 2 (during COVID), who received virtual-only training (Table 4).
Comparison of medical student pre- and post-course responses to questionnaire section assessing confidence in delivering aspects of the BLS algorithm, in Group 2 (during COVID), and Group 3 (post-COVID).
| GROUP 2 (DURING COVID): VIRTUAL-ONLY TRAINING | ||||||
|---|---|---|---|---|---|---|
| CONFIDENCE DELIVERING ASPECT OF BLS | PRE-COURSE QUESTIONNAIRE (MEAN ± SD) | POST-COURSE QUESTIONNAIRE (MEAN ± SD) | Md | z | r | p |
| Confidence- completing BLS algorithm steps correctly and efficiently | 1.95 ± 0.97 (95% CI 2.25–2.55) |
3.90 ± 0.87 (95% CI 3.72–4.08) |
–2.04 | –7.84 | 0.82 | <0.00001 |
| Confidence- performing chest compressions | 2.37 ± 1.05 (95% CI 2.15–2.59) |
3.70 ± 0.90 (95% CI 3.51–3.89) |
–1.63 | –7.01 | 0.73 | <0.00001 |
| Overall confidence performing BLS | 2.05 ± 0.99 (95% CI 1.85–2.26) |
3.66 ± 0.82 (95% CI 3.49–3.83) |
–0.95 | –7.67 | 0.80 | <0.00001 |
| GROUP 3 (POST-COVID): TRAINING WITH AN IN-PERSON COMPONENT | ||||||
| CONFIDENCE DELIVERING ASPECT OF BLS | PRE-COURSE QUESTIONNAIRE (MEAN ± SD) | POST-COURSE QUESTIONNAIRE (MEAN ± SD) | Md | z | r | p |
| Confidence- completing BLS algorithm steps correctly and efficiently | 2.29 ± 0.69 (95% CI 2.25–2.55) |
4.04 ± 0.80 (95% CI 3.70–4.38) |
–1.71 | –4.29 | 0.87 | <0.00001 |
| Confidence- performing chest compressions | 2.29 ± 0.86 (95% CI 1.93–2.65) |
3.75 ± 0.74 (95% CI 3.44–4.06) |
–2.71 | –3.89 | 0.79 | 0.0001 |
| Overall confidence performing BLS | 2.25 ± 0.74 (95% CI 1.94–2.56) |
3.75 ± 0.68 (95% CI 3.46–4.03) |
–1.75 | –4.10 | 0.84 | <0.00001 |
Note: Data were analyzed using the Wilcoxon signed-rank test. Data are presented as mean ± standard deviation (SD), 95% confidence interval (95% CI), mean difference (Md), z-value (z), and effect size (r), with p-values (p) of <0.05 considered significant.
Impact of time on confidence and hypothetical willingness to perform BLS
Unpaired data from Group 2 (during COVID) and Group 3 (post-COVID), which contained students from the same year group, were compared to assess whether any changes to self-reported confidence and hypothetical willingness to deliver BLS were sustained over time, following virtual-only BLS training. The post-course questionnaires from Group 2 (taught virtual-only BLS in year 1 in 2020–21) were compared to the pre-course questionnaires from Group 3 (taught full BLS with an in-person component in year 2 in 2021–22) using the Wilcoxon signed-rank test (timing between courses 6–9 months). A significant reduction was observed in both confidence and perceived willingness to deliver BLS, demonstrating that the improvement in these scores following virtual-only BLS training was not sustained over time (r > 0.7 and p < 0.05 for all aspects) (see Table 5).
Comparison of medical student responses between the post-course questionnaire of Group 2 (during COVID) and the pre-course questionnaire of Group 3 (post-COVID) to sections assessing confidence and willingness to deliver BLS, enabling assessment of whether the effects of virtual-only BLS teaching were sustained over time.
| QUESTIONNAIRE SECTION | POST-COURSE QUESTIONNAIRE GROUP 2 (MEAN ± SD) | PRE-COURSE QUESTIONNAIRE GROUP 3 (MEAN ± SD) | Md | z | r | p |
|---|---|---|---|---|---|---|
| Confidence- completing the BLS algorithm steps correctly and efficiently | 3.94 ± 0.92 (95% CI 3.64–4.25) |
2.22 ± 0.83 (95% CI 1.95–2.49) |
2.19 | –4.94 | 0.87 | <0.00001 |
| Confidence- performing chest compressions | 3.50 ± 0.94 (95% CI 3.19–3.81) |
2.17 ± 0.88 (95% CI 1.88–2.45) |
1.74 | –4.86 | 0.87 | <0.00001 |
| Overall confidence performing BLS | 3.50 ± 0.85 (95% CI 3.22–3.78) |
2.08 ± 0.69 (95% CI 1.86–2.31) |
1.69 | –4.94 | 0.87 | <0.00001 |
| Scenario 1: hospital setting | 3.83 ± 0.85 (95% CI 3.56–4.11) |
2.89 ± 1.14 (95% CI 2.52–3.26) |
–0.21 | –3.72 | 0.70 | 0.0002 |
| Scenario 2: community setting | 4.30 ± 0.89 (95% CI 4.01–4.60) |
3.42 ± 1.29 (95% CI 2.99–3.84) |
0.46 | –3.49 | 0.71 | 0.00048 |
Note: Data were analyzed using the Wilcoxon signed-rank test. Data are presented as mean ± standard deviation (SD), 95% confidence interval (95% CI), mean difference (Md), z-value (z), and effect size (r), with p-values (p) of <0.05 considered significant.
Impact of peer-led BLS training on hypothetical willingness to perform BLS
Section 4 of the questionnaire assessed medical student willingness to intervene and administer BLS in both a hypothetical hospital (scenario 1) and community (scenario 2) setting. Paired data were available from pre- and post-course questionnaires for both Groups 2 (during COVID) and 3 (post-COVID) and were analyzed using the Wilcoxon signed-rank test (see Table 6).
Comparison of medical student pre- and post-course responses to questionnaire section regarding hypothetical willingness to deliver BLS in Group 2 (during COVID) and Group 3 (post-COVID).
| GROUP 2 (DURING COVID): VIRTUAL-ONLY TRAINING | ||||||
|---|---|---|---|---|---|---|
| WILLINGNESS TO ADMINISTER BLS IN DIFFERENT SETTINGS | PRE-COURSE QUESTIONNAIRE (MEAN ± SD) | POST-COURSE QUESTIONNAIRE (mean ± SD) | Md | z | r | p |
| Scenario 1: hospital setting | 2.79 ± 1.09 (95% CI 2.56–3.02) |
3.95 ± 0.90 (95% CI 3.75–4.13) |
–1.21 | –6.33 | 0.66 | <0.00001 |
| Scenario 2: community setting | 3.40 ± 1.22 (95% CI 3.15–3.66) |
4.32 ± 0.79 (95% CI 4.16–4.49) |
–1.59 | –4.74 | 0.50 | <0.00001 |
| GROUP 3 (POST-COVID): TRAINING WITH AN IN-PERSON COMPONENT | ||||||
| WILLINGNESS TO ADMINISTER BLS IN DIFFERENT SETTINGS | PRE-COURSE QUESTIONNAIRE (MEAN ± SD) | POST-COURSE QUESTIONNAIRE (MEAN ± SD) | Md | z | r | p |
| Scenario 1: hospital setting | 2.88 ± 1.11 (95% CI 2.40–3.35) |
4.17 ± 0.82 (95% CI 3.82–4.51) |
–2.12 | –3.77 | 0.77 | 0.00012 |
| Scenario 2: community setting | 3.38 ± 1.38 (95% CI 2.79–3.96) |
4.46 ± 0.66 (95% CI 4.18–4.74) |
–0.62 | –2.64 | 0.54 | 0.0083 |
Note: Data were analyzed using the Wilcoxon signed-rank test. Data are presented as mean ± standard deviation (SD), 95% confidence interval (95% CI), mean difference (Md), z-value (z), and effect size (r), with p-values (p) of <0.05 considered significant.
A statistically significant increase in student-perceived willingness to perform BLS was observed in both groups (r ≥ 0.5 and p < 0.05 in both scenarios). The mean post-course willingness to intervene in both scenarios was higher for the students in Group 3 who received training with an in-person component (post-COVID) compared to that reported by students in Group 2 who received virtual-only training (during COVID). Students’ hypothetical willingness to intervene was consistently higher in the community rather than hospital scenario (see Table 6).
Impact of the COVID-19 pandemic on pre-course hypothetical willingness to perform BLS
The responses to scenarios 1 and 2 of section 4 of the pre-course questionnaire were compared for Groups 1 (pre-COVID) and 2 (during COVID) using the Mann-Whitney U test, to see if the COVID-19 pandemic had impacted medical students’ hypothetical willingness to perform BLS. For scenario 1, which assessed students’ self-reported willingness to perform BLS in a hospital setting, there was no significant difference between Groups 1 and 2 (Group 1 mean 3.02; 95% CI 2.85–3.20; Group 2 mean 2.79; 95% CI 1.70–3.02; z = 1.73; p = 0.08364). For scenario 2, which assessed students’ self-reported willingness to perform BLS in a community setting, students in Group 2 (during COVID) were significantly less willing to perform BLS (Group 1 mean 3.60, 95% CI 3.43–3.77; Group 2 mean 3.41, 95% CI 3.16–3.66; z = 2.10; p = 0.03572). The results are displayed in Figure 2.
Medical student responses to the pre-course questionnaire scenario gauging self-reported willingness to perform BLS in both a hospital and community setting, before BLS training.
Note: Comparison of students surveyed prior to and during the COVID-19 pandemic. Responses are presented as a percentage of completed questionnaires.
Discussion
This study has demonstrated that both virtual and in-person peer-led BLS teaching methods significantly increase medical students’ self-reported confidence in utilizing the BLS algorithm and delivering BLS skills. Additionally, regardless of setting, peer-led BLS training results in increased hypothetical willingness for medical students to intervene in emergency situations, in both a hospital and community setting. The increase in student confidence and willingness to administer BLS was similar following both virtual and in-person courses. This corresponds with prior studies that have established the efficacy of virtual BLS teaching methods both for training healthcare professionals (Tobase et al., 2017), and the public, through resources such as the Resuscitation Council (UK) ‘Lifesaver’ app (Resuscitation Council (UK), 2023), and the British Heart Foundation ‘RevivR’ app (British Heart Foundation & Laerdal, 2023). We were unable to directly compare our groups to assess for non-inferiority of the teaching modalities, and since no formal assessment was performed after our virtual teaching, it is harder to establish whether this increased confidence and willingness translates into the ability to deliver effective BLS. It is also unknown whether the increased self-reported confidence and willingness would translate into genuine intervention in an emergency.
A significant drop in BLS self-reported confidence and hypothetical willingness to intervene was seen between the post-course questionnaire of Group 2 and the pre-course questionnaire of Group 3, in the 6–9 months between their virtual and in-person course. These groups were from the same student year group, taught during and after the resolution of UK pandemic restrictions. This suggests that the improvements in BLS confidence and willingness seen following the virtual teaching were not sustained. We hypothesize that this relatively rapid reduction in confidence and willingness may be in part due to the lack of ‘hands-on’ practice during the virtual course, which may have impacted students’ deep learning on the topic. There remains significant support for in-person training in resuscitation skills, for example, Resuscitation Council (UK) (2020) guidance stipulates that BLS training must include the physical performance of chest compressions.
The skill degradation in our study is roughly in line with that seen in similar studies of healthcare staff (~9 months), following traditional in-person BLS teaching (Smith et al., 2008). Further research is needed to directly compare virtual only to in-person BLS training to assess its long-term efficacy. Moreover, a reduction in confidence and competence over time is not just seen following BLS training. All clinical skills training, including BLS, requires regular updates to retain the competence and confidence to perform such skills. Recognizing this, Laerdal introduced its Resuscitation Quality Improvement (RQI) program to clinical environments for healthcare staff. This supports regular interval BLS skills training, resulting in enhanced CPR skills retention, and improved clinical CPR quality (Laerdal, 2023; Panchal et al., 2020). Further study is needed to ascertain whether these pulses of increased confidence and willingness to administer BLS through programs such as RQI, or other refresher training, could be beneficial for medical students.
Our pre-course questionnaire found that when faced with a hypothetical BLS scenario, medical students were significantly less willing to act in a community setting during the COVID-19 pandemic. This corresponds with research showing an overall reduction in the rates of bystander CPR during the pandemic (Uy-Evanado et al., 2021). This study did not explore the qualitative reasons behind this finding, but it may be related to the fear of contracting COVID-19, as highlighted in the literature (Baldi et al., 2020). A further factor may have been that the students in Group 2 (during COVID) had less self-reported BLS experience prior to their training than the students in Group 1 (pre-COVID) (25.8% vs 34.4%). This lack of prior BLS experience may itself be due to COVID-19 restrictions and a potential decrease in the running of community BLS training provision. Additionally, the nature of BLS training delivery changed during the pandemic, which may have further influenced the students’ viewpoint. A potential extension to this study could be to undertake a qualitative research project to investigate the reasons behind student decisions around willingness to perform BLS in different settings.
We note that mean student-reported confidence and hypothetical willingness to deliver BLS was consistently higher in the community-based scenario than for the in-hospital scenario, regardless of teaching modality, although this was not statistically compared. We hypothesize this difference may be due to reduced student familiarity and confidence in the hospital environment since they have had limited clinical exposure during the early years of their course. Additionally, their BLS training is essentially a lay-person course, meaning the focus is on community BLS. Further study is needed to explain these findings, including the qualitative reasoning behind these disparities, as well as exploring whether the reduced confidence and willingness to perform BLS in a healthcare setting remains if training is focused on the healthcare environment.
The efficacy of the BLS training for all students who took part in this research has been demonstrated through their increased confidence and perceived willingness to act if experiencing a real-life emergency. This adds to the literature that establishes peer-led BLS teaching as an effective, economical, and reliable method of training delivery for healthcare professional students. Moreover, the effectiveness of peer-led instruction has since been widely replicated across different continents, resource settings, and learner groups (Abbas et al., 2018; Beck et al., 2015; Binkhorst et al., 2020). Peer-led teaching models may be particularly well-suited to global expansion to low-resource settings since once established they can provide a self-sustaining, low-cost teaching program, with greater availability of instructors in regions where resources may be most stretched (Harvey et al., 2012). Given the reduced rates of bystander CPR in countries with a lower GDP per capita (Shekhar & Narula, 2022), such a low-cost and easily expandable model is sorely needed to improve the substantial morbidity and mortality burden of OOHCA.
Limitations
There are limitations to this study. Group 1 was lost to follow-up due to multiple non-systematic factors, including the pandemic. This resulted in a significantly reduced number of paired responses to assess the efficacy of peer-led BLS training, impacting the statistical power of the study. This loss to follow-up prevented a true comparison of pre- and post-COVID-19 BLS teaching methods and the effect of the pandemic on medical students’ self-reported confidence and willingness to deliver BLS. The sample sizes in this study were relatively small and this was a single-center study, reducing the wider generalizability of the results. The study would benefit from expansion across multiple centers that also offer peer-led BLS training. The validity of the study could also have been improved by more strictly basing the questionnaire and scenarios on previously validated models.
As discussed in the background, the timing of the COVID-19 pandemic meant that the original study design, to follow a single student cohort over time (2-year follow-up period), was disrupted. This meant the current study was not able to assess whether the observed increase in student confidence and willingness was retained over an extended period. Moreover, confidence was self-reported and did not use a specific confidence rating scale. Lastly, the questionnaire used self-assessed student willingness to administer BLS in two hypothetical scenarios, raising the question of real-life translatability.
Conclusion
Peer-led BLS training improves medical students’ self-reported confidence and hypothetical willingness to perform BLS and intervene in an emergency, regardless of delivery method. This is important, as it demonstrates that effective BLS training can be maintained during unexpected environmental disruption, such as occurred during COVID-19. However, training including the in-person provision of BLS skills, particularly for CPR competency, remains the recommended method of instruction and continues to be used at RMD Bristol post-COVID-19.
Peer-led teaching schemes provide significant development opportunities for medical student instructors in teaching, mentoring, teamwork, and leadership. Due to the ability of peer-led schemes to deliver highly effective training at relatively reduced costs, with proven instructor reliability, and the capability to create a self-sustaining instructor pool, such models have the potential for wider expansion within and beyond medical education curricula.
Data Accessibility Statement
The data set used in this research project is restricted as at the time of the initial project participants were not asked for their permission to share data beyond the immediate project team. Access requests can be directed to the research team at the University of Bristol.
Additional File
The additional file for this article can be found as follows:
Study questionnaire and STROBE statement - checklist for cohort studies. DOI: https://doi.org/10.25894/ijfae.6.1.94.s1
Ethics and consent
This study was undertaken at the University of Bristol under ethics approval number: 66702.
Acknowledgements
William Mears and Annie Rylance contributed equally to this manuscript and should be considered joint first authors. We would like to thank all the University of Bristol students who took part in this study and all the staff and student instructors involved in the RMD Bristol scheme. We would like to acknowledge Dr Daisy Hewitt for her assistance with participant recruitment and data acquisition for this study. We would further like to thank the University of Bristol Medical School for their ongoing support of the RMD Bristol scheme.
Competing Interests
The authors have no competing interests to declare.
Author Contributions
WM was involved in the initial concept of the work, the acquisition, analysis, and interpretation of the data, and the main drafting of the manuscript.
AR was responsible for the acquisition, analysis, or interpretation of data for the work, and the initial drafting of the manuscript.
OK was involved in the initial concept of the work, the acquisition, analysis, and interpretation of the data, and has supported the drafting of the manuscript.
SA was involved in supporting the original concept of the work, supported the data collection process, and critically reviewed the manuscript for intellectual content.
MB was involved in supporting and supervising the original concept of the work and critically reviewed the manuscript for intellectual content.
All authors have approved the final version of the manuscript to be published and agree to be accountable for all aspects of the work.
Author Information
William Mears is a junior doctor currently working as a clinical teaching fellow at University Hospitals Bristol and Weston NHS Foundation Trust, Weston-Super-Mare, UK. He is a graduate of the University of Bristol and spent 2 years on the RMD Bristol committee during his undergraduate studies. He continues to work as a Lead Instructor and Deputy Director of the scheme.
Annie Rylance is a junior doctor currently working at the Royal Liverpool University Hospital, Liverpool, UK. She is a graduate of the University of Bristol and spent 3 years on the RMD Bristol committee during her undergraduate studies.
Oliver King is a junior doctor currently working at Dorset County Hospital NHS Foundation Trust, UK. He is a graduate of the University of Bristol and spent 2 years on the RMD Bristol committee during his undergraduate studies.
Sarah Allsop is a Senior Lecturer in Medical Education at the University of Bristol, UK. She has been the Director of the RMD Bristol Scheme since 2017.
Matthew Booker is a Consultant Senior Lecturer at the University of Bristol, UK, and Research Lead at RMD Bristol since 2017.
References
1 Abbas, F., Sawaf, B., Hanafi, I., Hajeer, M. Y., Zakaria, M. I., Abbas, W., Alabdeh, F., & Ibrahim, N. (2018). Peers versus professional training of basic life support in Syria: A randomized controlled trial. BMC Medical Education, 18(1). DOI: http://doi.org/10.1186/s12909-018-1241-z
2 Abelsson, A., Odestrand, P., & Nygårdh, A. (2020). To strengthen self-confidence as a step in improving prehospital youth laymen basic life support. BMC Emergency Medicine, 20(1), 1–5. DOI: http://doi.org/10.1186/s12873-020-0304-8
3 Baldi, E., Sechi, G. M., Mare, C., Canevari, F., Brancaglione, A., Primi, R., Klersy, C., Palo, A., Contri, E., Ronchi, V., Beretta, G., Reali, F., Parogni, P., Facchin, F., Rizzi, U., Bussi, D., Ruggeri, S., Oltrona Visconti, L., Savastano, S., & the Lombardia CARe researchers. (2020). COVID-19 kills at home: The close relationship between the epidemic and the increase of out-of-hospital cardiac arrests. European Heart Journal, 41(32), 3045–3054. DOI: http://doi.org/10.1093/eurheartj/ehaa508
4 Beck, S., Issleib, M., Daubmann, A., & Zöllner, C. (2015). Peer education for BLS-training in schools? Results of a randomized-controlled, noninferiority trial. Resuscitation, 94, 85–90. DOI: http://doi.org/10.1016/j.resuscitation.2015.06.026
5 Binkhorst, M., Draaisma, J. M. T., Benthem, Y., van de Pol, E. M. R., Hogeveen, M., & Tan, E. C. T. H. (2020). Peer-led pediatric resuscitation training: Effects on self-efficacy and skill performance. BMC Medical Education, 20(1), 427. DOI: http://doi.org/10.1186/s12909-020-02359-z
6 British Heart Foundation., & Laerdal. (2023). RevivR. https://revivr.bhf.org.uk/
7 Blewer, A. L., Ho, A. F. W., Shahidah, N., White, A. E., Pek, P. P., Ng, Y. Y., Mao, D. R., Tiah, L., Chia, M. Y. C., Leong, B. S. H., Cheah, S. O., Tham, L. P., Kua, J. P. H., Arulanandam, S., Østbye, T., Bosworth, H. B., & Ong, M. E. H. (2020). Impact of bystander-focused public health interventions on cardiopulmonary resuscitation and survival: A cohort study. The Lancet Public Health, 5(8), e428–e436. DOI: http://doi.org/10.1016/S2468-2667(20)30140-7
8 Cho, G. C., Sohn, Y. D., Kang, K. H., Lee, W. W., Lim, K. S., Kim, W., Oh, B. J., Choi, D. H., Yeom, S. R., & Lim, H. (2010). The effect of basic life support education on laypersons’ willingness in performing bystander hands only cardiopulmonary resuscitation. Resuscitation, 81(6), 691–694. DOI: http://doi.org/10.1016/j.resuscitation.2010.02.021
9 de Ruijter, P. A., Biersteker, H. A., Biert, J., van Goor, H., & Tan, E. C. (2014). Retention of first aid and basic life support skills in undergraduate medical students. Medical Education Online, 19(1), 24841. DOI: http://doi.org/10.3402/meo.v19.24841
10 European Resuscitation Council. (2023). Basic Life Support Course. https://www.erc.edu/courses/basic-life-support
11 General Medical Council. (2019). Good medical practice. https://www.gmc-uk.org/ethical-guidance/ethical-guidance-for-doctors/good-medical-practice
12 Giacomino, K., Caliesch, R., & Sattelmayer, K. M. (2020). The effectiveness of the peyton’s 4-step teaching approach on skill acquisition of procedures in health professions education: A systematic review and meta-analysis with integrated meta-regression. PeerJ, 8(1), e10129. DOI: http://doi.org/10.7717/peerj.10129
13 Giacoppo, D. (2019). Impact of bystander-initiated cardiopulmonary resuscitation for out-of-hospital cardiac arrest: Where would you be happy to have a cardiac arrest? European Heart Journal, 40(3), 319–321. DOI: http://doi.org/10.1093/eurheartj/ehy911
14 Gillam, L., Crawshaw, B., Booker, M. J., & Allsop, S. (2023). Prompt identification of struggling candidates in near peer-led basic life support training: Piloting an online performance scoring system. BMC Medical Education, 23(1), 303. DOI: http://doi.org/10.1186/s12909-023-04225-0
15 Harvey, P. R., Higenbottam, C. V., Owen, A., Hulme, J., & Bion, J. F. (2012). Peer-led training and assessment in basic life support for healthcare students: Synthesis of literature review and fifteen years practical experience. Resuscitation, 83(7), 894–899. DOI: http://doi.org/10.1016/j.resuscitation.2012.01.013
16 Kragholm, K., Wissenberg, M., Mortensen, R. N., Hansen, S. M., Malta Hansen, C., Thorsteinsson, K., Rajan, S., Lippert, F., Folke, F., Gislason, G., Køber, L., Fonager, K., Jensen, S. E., Gerds, T. A., Torp-Pedersen, C., & Rasmussen, B. S. (2017). Bystander Efforts and 1-Year Outcomes in Out-of-Hospital Cardiac Arrest. New England Journal of Medicine, 376(18), 1737–1747. DOI: http://doi.org/10.1056/NEJMoa1601891
17 Laerdal. (2023). RQI – Resuscitation Quality Improvement Program. https://laerdal.com/us/services-and-programs/resuscitation-quality-improvement/
18 Lim, Z. J., Ponnapa Reddy, M., Afroz, A., Billah, B., Shekar, K., & Subramaniam, A. (2020). Incidence and outcome of out-of-hospital cardiac arrests in the COVID-19 era: A systematic review and meta-analysis. Resuscitation, 157, 248–258. DOI: http://doi.org/10.1016/j.resuscitation.2020.10.025
19 Medical Defence Union. (2022). Good Samaritan Acts. https://www.themdu.com/guidance-and-advice/guides/good-samaritan-acts
20 Panchal, A. R., Norton, G., Gibbons, E., Buehler, J., & Kurz, M. C. (2020). Low dose- high frequency, case based psychomotor CPR training improves compression fraction for patients with in-hospital cardiac arrest. Resuscitation, 146, 26–31. DOI: http://doi.org/10.1016/j.resuscitation.2019.10.034
21 Passali, C., Pantazopoulos, I., Dontas, I., Patsaki, A., Barouxis, D., Troupis, G., & Xanthos, T. (2011). Evaluation of nurses’ and doctors’ knowledge of basic & advanced life support resuscitation guidelines. Nurse Education in Practice, 11(6), 365–369. DOI: http://doi.org/10.1016/j.nepr.2011.03.010
22 Perkins, G. D., Hulme, J., & Bion, J. F. (2002). Peer-led resuscitation training for healthcare students: A randomised controlled study. Intensive Care Medicine, 28(6), 698–700. DOI: http://doi.org/10.1007/s00134-002-1291-9
23 Peyton, J. W. R., & Walker, M. (1998). Teaching in theatre. J. W. R. Peyton (Ed.), Teaching and learning in medical practice (pp. 171–180). Manticore Books.
24 Phillips, P. S., & Nolan, J. P. (2001). Training in basic and advanced life support in UK medical schools: Questionnaire survey. BMJ, 323(7303), 22–23. DOI: http://doi.org/10.1136/bmj.323.7303.22
25 Resuscitation Council (UK). (2020). Quality Standards: CPR and AED training in the community. https://www.resus.org.uk/library/quality-standards-cpr/quality-standards-cpr-and-aed-training-community
26 Resuscitation Council (UK). (2023). Lifesaver. https://life-saver.org.uk/
27 Shekhar, A., & Narula, J. (2022). Globally, GDP Per Capita Correlates Strongly with Rates of Bystander CPR. Annals of Global Health, 88(1), 36. DOI: http://doi.org/10.5334/aogh.3624
28 Smith, K. K., Gilcreast, D., & Pierce, K. (2008). Evaluation of staff’s retention of ACLS and BLS skills. Resuscitation, 78(1), 59–65. DOI: http://doi.org/10.1016/j.resuscitation.2008.02.007
29 Sullivan, G. M., & Artino, A. R. (2013). Analyzing and Interpreting Data From Likert-Type Scales. Journal of Graduate Medical Education, 5(4), 541–542. DOI: http://doi.org/10.4300/JGME-5-4-18
30 Tobase, L., Peres, H. H. C., Gianotto-Oliveira, R., Smith, N., Polastri, T. F., & Timerman, S. (2017). The effects of an online basic life support course on undergraduate nursing students’ learning. International Journal of Medical Education, 8, 309–313. DOI: http://doi.org/10.5116/ijme.5985.cbce
31 Uy-Evanado, A., Chugh, H. S., Sargsyan, A., Nakamura, K., Mariani, R., Hadduck, K., Salvucci, A., Jui, J., Chugh, S. S., & Reinier, K. (2020). Out-of-Hospital Cardiac Arrest Response and Outcomes During the COVID-19 Pandemic. JACC: Clinical Electrophysiology, 7(1), 6–11. DOI: http://doi.org/10.1016/j.jacep.2020.08.010
32 Xie, J. Y.-Y., Frost, R., & Meakin, R. (2019). Not quite a doctor, but should I help? A qualitative exploration of medical students’ attitudes towards responding to medical emergencies that occur in the public domain. BMJ Open, 9(4), e028035. DOI: http://doi.org/10.1136/bmjopen-2018-028035
33 Yan, S., Gan, Y., Jiang, N., Wang, R., Chen, Y., Luo, Z., Zong, Q., Chen, S., & Lv, C. (2020). The global survival rate among adult out-of-hospital cardiac arrest patients who received cardiopulmonary resuscitation: A systematic review and meta-analysis. Critical Care, 24(1). DOI: http://doi.org/10.1186/s13054-020-2773-2
34 Yu, Y., Meng, Q., Munot, S., Nguyen, T. N., Redfern, J., & Chow, C. K. (2020). Assessment of Community Interventions for Bystander Cardiopulmonary Resuscitation in Out-of-Hospital Cardiac Arrest. JAMA Network Open, 3(7), e209256. DOI: http://doi.org/10.1001/jamanetworkopen.2020.9256