Near-peer Tutoring Compared With a Standard Faculty-led Course for Undergraduate Training in Abdominal Ultrasound.

Background:

Over the last decades, ultrasound (US) examination has become commonly used in many medical specialties, including internal medicine and primary care. Young doctors often perform simple ultrasound investigations as residents in internal medicine or primary care especially in emergency care. Currently, 30% of Swiss primary care practices are equipped with an ultrasound machine.

The Swiss Society for Ultrasound in Medicine (SGUM/SSUM) is the accreditation body for ultrasound education in Switzerland. Doctors must obtain a certificate from the SGUM before they can bill ultrasound examinations for insurance payments. However, ultrasound courses are time-intensive (63 hours total, plus a final examination), and are offered in the postgraduate phase, when doctors often have few resources and little time. This may explain why only about half of General Practitioners who use ultrasound have the SGUM certificate. For the half without the certificate, not only can they not bill their ultrasound examination to an insurance company, but also the quality of the examination is less assured, as they have not demonstrated their competence in US skills in the SGUM examination. This may lead to insecurity for the patient and for the performing doctor. There is also a risk that this results in unnecessary examinations.

The SGUM certificate is achieved by passing three courses (basic/advanced/final; 21 hours each) as well as a summative final examination. Between courses, trainees must perform a certain number of US examinations on real patients, under direct and distant supervision. These courses are now being opened to undergraduates, and the 2017 revision of the Swiss learning objectives for medical studies (PROFILES) added basic US skills as mandatory learning goal.

In the past years US training courses at Swiss medical universities were mostly optional, and were quickly overbooked. This shortage of courses could be addressed by using near-peer tutors, which are increasingly used in faculty-organized US classes. Near-peer tutors are "trainees one or more years senior to trainees on the same level of medical education spectrum". Five randomized-controlled trials of ultrasound training have compared near-peer tutoring with faculty-led teaching and assessed students' practical ultrasound skills as a primary outcome measure. Four of these found peer-tutoring to be non-inferior and one found that students in the near-peer-tutoring group slightly underperformed.

Near-peer tutoring in undergraduate US training offers many potential advantages over traditional faculty-based teaching:

• Since mostly reciprocal US is being done by the participating medical students, perfect conditions for learning basic sonographic anatomy can be found. Medical students, in contrast to patients, are mostly slender, healthy and mobile.
• In near-peer tutoring settings, students are more likely to ask questions and participate actively. In addition, having ultrasound skills can help students feel more secure in their basic clinical examination skills.
• Near-peer tutoring relieves pressure on the university by alleviating the increasing teaching demand and by being cost-effective.
• There is evidence that students who work as tutors constantly improve their own skills.

In Autumn 2017, the Institute of Primary Health Care (BIHAM) at the University of Bern set up a working group of 24 US experts from all Swiss universities and of representatives of SGUM sections. The group decided on the "Swiss standard" for basic abdominal ultrasound examination after discussing it at two consensus-building meetings. In Spring 2018, videotapes of SGUM experts demonstrating the ultrasound examination of the various abdominal organs, based on that consensus, were recorded. These videos were integrated into interactive e-learning modules. They have been available to all Swiss medical students since the beginning of 2019.

In parallel, a formal system of training and continuous education for near-peer tutors was put in place by a collaboration of motivated students and faculty staff. Near-peer tutors receive at least 6 days of US training, including teaching on didactic techniques, and afterwards be registered as so-called 'junior-tutors' by SGUM. They must offer at least four hours of US class per semester, but can choose the times and dates. Tutors will be reimbursed for their teaching through course fees paid by course participants.

Methods:

The standard "Basic course in abdominal ultrasound", delivered over 2.5 consecutive days, contains 5 hours of theory and 16 hours of practical training. In contrast, our blended-learning course is more student-centred, in that it expects students to take the initiative to learn the theory from (A) the e-learning developed by the working group, and (B) 16 hours of near-peer tutoring in groups of two to four, at times that are mutually agreeable for tutors and students, without faculty involvement.

The e-learning is made up of five modules starting by an introductory e-learning module on technical basics in ultrasound, followed by four practical modules comprising the abdominal organs. For each of the four practical modules, students start by working through the e-learning module, followed by four hours of individual near-peer tutoring sessions.

Students decide the pace of their 16 hours of training, typically over a four-month period. When a student completes all five e-learning modules and has gathered 16 signatures from the peer-tutored classes, they can apply to take a summative final examination. If they pass the Objective Structured Clinical Examination (OSCE) validated by Hofer et al. (2011) and adapted to Swiss needs, they are accredited by SGUM, according to Swiss national accreditation guidelines, for the "Basic course in abdominal ultrasound" - the first of the three ultrasound courses within the SGUM curriculum.

The OSCE will be assessed twice for every participant. Once at the end of the course and once 6 months after the end of the course. The second OSCE will be the primary outcome, since it evaluates long-term retention of acquired US knowledge. Every student will be tested by six different examiners in each OSCE. Differences in difficulty of the four series of OSCE stations as well as difficulty of every station will be taken into account in the statistical analysis.

The blended-learning course offers several other advantages by implementing modern didactics: by distributing teaching sessions over 16 weeks instead of just 2.5 consecutive days, students benefit from "spaced repetition" which improves long term retention. In our programme, students can decide which session to choose next, and tutors and students arrange the teaching sessions according to their availability and needs, so learning is student-centred. These factors differentiate our study from the near-peer tutoring in ultrasound education reported in some studies, in which tutors engage in scheduled courses organized by university faculty.

Between randomization and the end of the training (t = 16 weeks), students will be instructed not to undergo any additional training (e-learning, practical sessions or other) other than those provided within the intervention. Additional training will be recorded directly after the end of the training by questionnaire and by going through e-learning logs. During the 6-month follow-up period, students in both groups will be equally encouraged to continue self-organised training and both will be granted access to ultrasound machines.

Adjusting to local curricula we will primarily recruit students in semester 7 in Bern (total medical student intake per year = 220) and in semester 6 in Zurich (240) and Fribourg (120). At each site, we will present the programme during a plenary session, assisted by our local co-investigators.

To avoid systematic differences in additional training during the follow-up period, the following rules will put in place:

• Students will only be told their test results from the first test after the end of the study
• Test performance in the follow-up test will have no effect on the accreditation

To reduce loss to follow-up, we will:

• Require all students who apply for the programme to sign a separate form confirming their availability and intention to complete the course and the examinations
• Reimburse participants 50.- CHF after they have taken the second OSCE.
• Reimburse travel costs for the second OSCE
• Hand out the test result and the course certificate only after the second OSCE

Students' motivation for ultrasound training is very high. The optional courses at Swiss universities are typically overbooked within minutes and many students are disappointed that they are unable to take part. Our study will give students the opportunity to get a certificate accredited by the SGUM for a course fee of 200.- CHF, regardless of the intervention they are assigned to. Postgraduate training for the same degree costs around 1200.- CHF.

We will use centralized computed randomization in blocks of 4 to allocate each of the participants to one of the two study arms, with allocation in a 1:1 ratio, stratified by study site. The reason for block randomisation is the requirement of the control group to comprise of exactly 24 or 28 students due to the maximum group size of 4 students per tutor. The randomization will be performed by a blinded external partner (Clinical Trials Unit in Bern).

Outcome assessors will be experienced sonographers who have not taught course participants of either group.

To ensure high interrater-reliability, faculty training for all assessors will be held immediately before the OSCE. In the first part of this training, the assessors discuss key aspects of the rating using videotaped examples of OSCE stations to achieve a common standard. At the end of the training, assessors will rate another video example, where the interrater-reliability will be assessed.

OSCE assessors will be blinded to group allocation. The assessments will be at the same time for each group. We will instruct students not to reveal their group allocation to the OSCE assessors.

Sample size calculation has been done as follows: Based on a pilot with 20 students we expect a mean score of 33/50 points per station with a standard deviation (SD) of 4 points. The study was powered for non-inferiority using a margin of a -1.5 points for the difference in mean OSCE scores between the intervention and control group in the practical exam at 6 months (primary outcome). 1.5 points difference would correspond to an effect size of 0.375 standard deviations (SD), which we would still consider as non-inferior. We calculated the sample size with the assumption that the intervention group will perform somewhat better (+0.5 in mean score) than the control group based on test results from previous pilots. We chose a 1-sided alpha-error of 0.05 and a power of 90% (138 participants). Allowing for a 9% dropout rate we decided to include 152 students.

Discussion:

If the new blended-learning format is shown to be at least as good as the existing standard course, we could quickly expand our capacity to offer such courses to undergraduate medical students, replacing resource-intensive faculty-led courses with peer-assisted learning programmes. This will eventually allow all undergraduate students to acquire ultrasound skills that will help them in their residencies. The national accreditation body fully supports this near-peer-tutoring-based course both financially and through full accreditation of a pilot project and has committed to continue its support if we find that this teaching format produces results that are just as good as, or better than, the usual faculty-led courses. Learning more about the advantages conferred by student-centred learning through near-peer teaching will enable us to design better peer-assisted learning programmes and to design new, similar programmes in other areas.