Background

Patient and public involvement (PPI) in medical education includes teaching, assessment, feedback, and curriculum development [1]. Global consensus is that medical school stakeholders should be partners in the education of the future medical workforce [2] and this is a requirement of professional regulatory bodies. For example, in the UK, PPI is a requirement of the General Medical Council (GMC) [1]. In early 2020, the Covid-19 pandemic forced medical centres globally to reduce face-to-face contact, and remote teaching became the standard approach for medical schools. This reduced opportunities for interactions between patients and students. With the benefits of PPI in medical teaching wide-ranging, including developing students’ person-centred skills [3], improving students’ professional attitudes and clinical performance, and offering professional, personal, and emotional benefits for healthcare users [3, 4], continued PPI in the era of online learning is crucial. Involving patients and carers in remote medical education is also timely and will help students respond to the evolving needs of patient groups as remote healthcare delivery continues [5]. Developing remote consultation skills will undoubtedly be a key requirement for the future medical workforce [6, 7].

A number of challenges arise from this new way of working. Firstly, patient and carer “involvement” varies widely, with implications for remote teaching practice. A recent taxonomy of involvement defines active PPI in medical teaching as a spectrum, from featuring in case studies (“Level 1”) to involvement at an institutional level, e.g. in decision-making (“Level 6”) [8]. Historically, patients and carers have held relatively passive roles in the education of medical students, but examples of good practice have increased over recent years [9]. However, these examples are from in-person teaching and other than electronic case studies and pre-recorded patient videos [8,9,10], the variety of potential uses of digital technology in medical training when patients and carers are also involved has not been explored. A recent scoping review of PPI in rural healthcare education settings found patients had been involved over telephone and in online materials, in consultations about new curricula and evaluating programmes [11]. This initial insight however needs to be expanded to all medical education contexts (including other geographical areas) to inform the future strategy of medical schools globally.

There may be barriers for patients and carers invited to join medical teaching sessions and research from traditional (face-to-face) medical training has found these include a lack of knowledge about medical education [12] having a sensitive clinical problem and concerns about privacy or confidentiality [12, 13]. However, findings from digital healthcare allude to new barriers introduced when joining remote healthcare consultations. For example, according to one study, patients report being unable to access the necessary technology, and may find connecting remotely more difficult due to their symptoms [14]. For marginalised patient groups remote healthcare may exacerbate language barriers and reduce opportunities for practical support from reception staff such as registering and signposting [15]. We must first understand the specific barriers that may limit PPI in remote medical education, to ensure medical training is inclusive of diverse voices, and representative of local populations.

Two further recent systematic reviews have described active PPI in medical education broadly [3, 16], however these reviews were not focused on the use of technology for learning. The aims of the present review therefore were to present the variety of digital technologies that have been used in medical teaching when patients and/or carers are also involved, and what has been the experience of patients, students and educators alike.

Methods

Rapid systematic review methods were employed. Rapid reviews follow standard systematic review procedures, whilst providing timely evidence and maintaining rigour [17]. Rapid methods were chosen to provide teaching teams with timely evidence for the uses of technology to support continued PPI in undergraduate medical education after the rapid shift to remote working during the COVID-19 pandemic.

Protocol

The protocol has been registered on PROSPERO Ref. CRD42021243279. The review protocol is available on this PROSPERO web page.

Search strategy and selection criteria

Searches for published and unpublished studies were performed from database inception to 27th October 2020 using MEDLINE (OvidSP), EMBASE (OvidSP) and medRXiv Preprints (https://www.medrxiv.org/) by a university librarian (NR). The search strategy is available as a supplementary file (Supplementary File 1). Boolean and proximity operators were used, for example digital*. Searches were not limited by language or publication date. Retrieved references were initially de-duplicated in Endnote before being exported into Rayyan [18] and titles and abstracts were screened by seven authors (SLW, AA, JHH, NY, CJ, NC & SP). Ten percent of titles and abstracts were screened independently by two authors (AA & NC) and any disagreements were discussed with a third author (SLW) until consensus was reached. This was limited to 10% of articles due to time restrictions.

Primary studies evaluating undergraduate medical education activities were eligible. Eligible studies also described any type of digital technology, including remote technology (e.g. telephone, video-conferencing software), or technology used in-person that could be adapted for remote use (e.g. video). Studies involving patients and/or carers at any level [8], and employing any study design, were eligible. Eligible studies also reported student-, educator- and/or patient-related outcome data. Studies explicitly describing the use of actors (without experience of the medical problem they were presenting with) or other persons not presenting as authentic patients or carers were excluded. Non-English language articles were excluded at screening stage, due to the rapid nature of this review and a lack of resources to translate studies. Attempts were made to retrieve articles from the authors’ institutions but if unsuccessful the article was excluded, due to time and funding restrictions. Reviews were excluded, but reference lists were hand searched for additional studies.

Data extraction and analysis

A data extraction form was developed by the authors based on the Sample, Phenomenon of Interest, Design, Evaluation, Research type (SPIDER) criteria, developed for reviewing qualitative and mixed methods studies [19]. Data extraction was completed by 7 authors independently (SLW, AA, NC, JHH, NY, EL, CJ), all extracted data was reviewed by SLW for completeness. All student-, educator- and/or patient-related outcome data was extracted, as well as the type of technology, demographics of involved patients and/or carers, types and levels of PPI, and study design. Due to the heterogeneity of study designs, a narrative synthesis was performed. A taxonomy of active PPI in healthcare education [8] was used to categorise the level of patient and/or carer involvement in the educational activity described by study authors. Categories range from patients being involved in developing a case study/ scenario—but had no overall influence on the theme of the content, nor on curriculum development (Level 1)—to patients being involved at the institutional level (Level 6) [8].

Quality assessment

The Mixed Methods Appraisal Tool (MMAT [20]) was used to assess study quality. The MMAT has been used for most common study methodologies and in a variety of contexts including health sciences, education, information sciences and psychology [20]. Two authors were independently involved in the appraisal process (EL & NC); double assessment was not performed due to time limitations. MMAT scores were categorised as low, moderate or high-quality using criteria employed for two recent rapid systematic reviews of public health interventions [21, 22]; a score of 0–1 was categorised as low quality, 2–3 moderate quality, and 4–5 high quality.

Patient and Public Involvement (PPI) in the research team

The review team included two public contributors (JHH, NY), who joined the team at the stage of planning the review (after the research question had already been defined). Both public contributors had specific experiences as patients, of literature reviewing, and as PPI representatives on research teams at Oxford University (JHH) and University College London (NY), as well as experiences contributing to medical education. One public contributor also had lived experience as a carer. JHH and NY were members of the research team, joined research meetings, and supported the review processes including literature screening, data extraction and interpretation, and preparing the manuscript for publication. PPI contributors informed decisions about our inclusion criteria, ensuring the review considered the carer viewpoint.

Results

Study selection

The full texts of 216 potentially relevant articles were screened for eligibility. A total of 20 studies were identified as eligible and included in the review (Fig. 1).

Fig. 1
figure 1

PRISMA Flow Chart

Full size image

Study characteristics

Characteristics of the articles included in the review and the types of digital technology used to support educational interventions involving patients and carers are presented in Table 1.

Table 1 Characteristics of included studies
Full size table

Types of technology used

Six of the final 20 studies (30%) used remote healthcare and remote learning technologies, including telehealth platforms and video-conferencing software, to engage live with patients. Telehealth platforms were used for live remote clinical consultations [27, 28, 34, 41]. One class was delivered via video-conferencing software where “simulated” patients were featured in role plays with students [36], and one class involved an online blended learning module [32]. The remaining 14 studies (70%) used pre-recordings and existing online materials featuring patients, where there were no live interactions between students and real-time patients. Three studies used existing online patient materials to facilitate in-person teaching sessions [26, 33, 38], e.g. an online medical e-forum with clinical questions previously submitted by patients [33]. Eleven studies used pre-recorded videos to provide the patient’s perspective on their illness or demonstrate doctor-patient consultations [23,24,25, 29,30,31, 35, 37, 39, 40, 42] during in person teaching sessions.

Description of patients and carers involved in medical education

Two studies described using “simulated” or “virtual” patients, [36, 42] and two studies used patient-focused videos [32, 35], but none of these three studies clarified these terms, or whether patients were authentic. Sixteen studies involved authentic patients [23,24,25,26,27,28,29,30,31, 33, 34, 38, 40, 41] and two studies included the perspectives of family members [30, 32]. One of these two studies did not provide enough detail to determine if caregivers were authentic carers of patients, or whether they were actors [32].

Levels of patient and public involvement (PPI) in medical education

In the majority of studies (n = 14), students viewed a pre-recorded video or completed online material that involved no interaction with patients and were thus categorised at Level 1 of Towle and colleagues’ [8] taxonomy [23,24,25,26, 29,30,31,32,33, 35, 37,38,39,40, 42]. Six studies involved patients in real-time clinical encounters led or observed by students [27, 28, 34, 41], reflecting Level 2 of Towle et al.’s taxonomy, although in two of these studies the authenticity of the patients is unclear [23,24,25,26, 29,30,31,32,33, 35, 36, 38, 42]. No study reported patient and/or carer involvement above Level 2.

Quality assessment

The MMAT score distribution for the included studies were summarised as follows: low quality n = 3 papers [25, 31, 33], moderate quality n = 2 [22, 40] and high quality n = 15 papers [23, 24, 26,27,28,29,30, 32, 34,35,36,37,38, 41, 42]. The two mixed-methods studies scored 5/5 and 3/5 respectively in the quality criteria for their qualitative and quantitative components respectively [23, 32].

Synthesis of results

A summary of the main results is reported in Table 2. The results have been synthesised below in relation to the impact of digital educational activities involving patients and/or carers, on medical students, educators, and patients/ carers themselves.

Table 2 Results of the included studies
Full size table

Impact on medical students’ learning and attitudes

Nineteen of the 20 articles reported the impact of PPI via digital tools on students’ learning and attitudes. Two of these 19 studies included mixed samples of medical students, residents [30, 32] and nursing and pharmacy students [30] where the outcomes for medical students from other healthcare students could not be extracted, so their findings have not been reported below. The remaining 17 studies measured student-reported outcomes of PPI on their learning when this was combined with the use of technology. These included acceptability, attitudes towards the activity, attitudes towards patients and/or carers, clinical knowledge and communication skills. Two of these 17 studies included objective measures of students’ learning, e.g. interpersonal skills (scored by a blinded simulated patient) [39].

Acceptability and general attitudes towards educational activity

Six studies found digital activities involving patients and/or carers to be educationally valuable [25, 26, 28, 31, 40, 41]. Two high quality studies reported that students found the educational activity acceptable [31, 34]. One high quality qualitative study investigating student-led remote consultations reported mixed student perceptions about the educational value and acceptability of these remote interactions with patients, with some reporting a preference for in-person consultations (e.g. due to being unable to perform a physical examination), while others found the experience valuable [27].

In one high quality study students reported positive attitudes towards video libraries featuring authentic patient cases [29]. Another high quality study found 79% of students reported that a 13 min video of a patient’s perspective of fibromyalgia was superior to a traditional in-person lecture [35].

Attitudes towards patients

Three out of 20 articles reported positive students attitudes towards patients after digital activities involving patients or carers. Two of these studies found improvement in students’ patient-centred attitudes after watching videos of patients discussing their condition or hospital experiences [23, 35]. Yoon and colleagues, however, reported traditional problem-based learning led to significantly improved attitude towards patients, compared to videos of patient cases [42]. Although, notably, it is unclear if patients were authentic in either the standardised or video-delivered approach in this study.

Knowledge of condition or treatment featuring in educational activity

Out of six articles reporting students’ clinical knowledge or knowledge about the patient group featuring in digital activities, five reported gains in students’ knowledge [31, 34, 35, 37, 38]. One high quality study reported no differences in self-reported knowledge about cervical screening when students viewed a video involving patients, versus a video featuring a clinician [39].

Clinical and communication skills

Six studies reported improvements in students’ communication skills after a digital activity involving patients and/or carers. One qualitative study reported a remote class with “simulated” patients helped students develop skills in exploring patient’s perceptions, sharing information with patients, and checking their understanding [36]. An online educational tool featuring a Muslim woman was found to improve students’ self-efficacy in communication with Arab American patients than participants in the control condition [38]. Dow and colleagues [28] reported videos helped students understand how to adapt their history taking skills and vary their approach to meet patients’ needs. Coret and colleagues [25] reported higher communication scores after a blended learning activity (with online elements) versus a standard lecture. Students reported introducing themselves more often, and taking measures to make patients feel more at ease, after watching videos of patients discussing their hospital experiences [40]. Snow et al. [39] reported higher OSCE scores, more confidence communicating with patients, and students feeling more comfortable responding to patients’ emotional needs, after watching a video of patients sharing their experiences of colposcopy, compared to a video featuring a clinician only. One low quality study found a student-led clinical hotline for patients with COVID-19 increased students’ remote clinical skills in screening, assessment, and triaging patients [23].

The traditional patient simulation was found significantly more beneficial to students in their collaborative learning, reflective thinking, and patient-doctor communication, than a video-delivered simulation in the study by Yoon and colleagues [42]. Further, student-led remote consultations were reported by some students to inhibit rapport-building with patients versus in-person consultations [27].

Perspective of medical educators

Only four of the 20 included articles reported the perspective of medical educators of digital educational activities involving patients and/or carers.

Acceptability and value of educational activity

Video-recorded GP consultations featuring patients were reported to facilitate discussions with students [28]. Tutors found student-made films about the impact of living with chronic conditions (with PPI) to be compelling and informative [37]. While GP supervisors were satisfied with some aspects of student-led remote consultations, including how students set up and maintained appropriate environments for consultations, the physical distance made it difficult to build rapport with students, with fewer opportunities to offer students feedback [27].

Perceptions of students’ skills

One high quality study reported that an e-forum for patients was a suitable learning tool for tutors to assess students’ clinical decision-making skills [33].

Perspective of patients and/ or carers

Acceptability of educational activity

Darnton and colleagues [27] reported that student-led remote consultations were acceptable to patients, but this was from the perspective of students and educators. No studies measured the acceptability of PPI in students’ learning via digital technology from the perspective of patients or carers involved.

Barriers to participating in educational activity

Weber and colleagues [41] reported difficulties for patients attempting to participate in telehealth consultations led by students. Out of 222 encounters, 46.5% of patients requested a traditional telephone call (over the telehealth consultation), 32.7% reported not having access to a compatible smartphone and/or computer and 18.4% had difficulty with the technology and were unable to join the virtual waiting room [41].

Discussion

Main findings

The aim of this rapid systematic review was to identify the uses of, and evaluate, digital technology in undergraduate medical teaching when patients and/or carers have been involved, encompassing all educational settings, technologies and geographical locations. Twenty articles met the eligibility criteria and demonstrated a variety of potential uses of digital technology in undergraduate medical education when patients and carers are involved.

The review found that PPI was perceived to be educationally valuable to students and educators, acceptable to students, and increased students’ knowledge of patient groups, as well as communication and clinical skills. Limited evidence also demonstrates enhanced student engagement, and improved patient-centred attitudes. Although it is important to note study designs were heterogenous and it is difficult to draw firm conclusions about the outcomes of digital medical education when patients and carers have been involved, particularly where it is unclear whether participants had lived experience or were scripted. Furthermore, patient and carer involvement was generally at a low level where there was no interaction with students, suggesting that digital technology has not yet driven the involvement of patients and carers much further beyond simulation. “Live” encounters with patients offered an opportunity to enhance students’ clinical and communication skills, although introduced additional barriers related to building rapport (between students and patients, and students and their supervisors) and issues with technology. This review does however demonstrate the potential benefits of involving patients and carers in medical education when teaching is delivered remotely.

The research in this area was limited in scope, with no studies directly capturing the perspective of the patients or carers involved in remote teaching. Thus, a balanced view of patient or carer participation, including any benefits and negative impacts for participants, and how educators might address these when organising teaching sessions, has not been obtained. In contrast, evidence from in-person medical education has identified a number of barriers to participation [12, 13]. Studies in this review identified potential challenges, including difficulty building rapport with patients, and between GP supervisors and students [27] and patients lacking access to connect with students [41]. However, no study has directly captured these issues from the viewpoint of patients’ and carers’ themselves. Without these key stakeholder perspectives, it remains unclear what additional barriers using remote teaching tools may introduce for those wishing to be involved.

Links to previous research

Patients and carers have not been meaningfully involved in medical education when digital technologies have been used in teaching. This finding is not replicated by the growing body of literature reporting good examples of PPI in medical education [3]. The majority of studies in a recent systematic review described patients as educators and assessors, reflecting Level 4 of Towle’s taxonomy [3, 8]. One study included in our review involved patients with lived experience of the medical problems they portrayed, but their involvement was a scripted role, supposedly with the aim of standardising students’ learning experience [31]. This suggests there is still progress to be made to ensure patients and carers are equal partners in remote medical learning ensuring spaces for authentic interaction between students and patient about their lived experiences of illness and disease. Research from in-person teaching contexts has helpfully identified ways patients and carers wish to be involved, including wanting clear information before student encounters and a desire for their consent to be taken at each stage (e.g. may consent to student being present, but not taking a clinical examination) [12, 13]. With technology in educational activities significantly increasing in use since the beginning of the COVID-19 pandemic [7, 43, 44], the use of more interactive technologies (e.g. video-conferencing software) can provide students with valuable experiences interacting with and learning from patients and carers in real-time [7, 43]. Furthermore, without identifying the barriers associated with remote participation we risk further marginalising people already excluded [45], for example people with disabilities or who are homeless. Medical students would benefit from these viewpoints to better understand how to improve future healthcare service access in the era of digital health.

There was poor consistency in the use of terminology to describe patients and carers, including in studies where authentic patients or carers had been involved and where there had been no genuine patient or carer involvement (e.g. when actors were employed). Previous authors have highlighted the inconsistencies in meaning within and between common terms such as “virtual patient” or “simulated patient” [46, 47]. The diversity of meaning in these terms (and poor reporting of study methods) has implications for the replicability of medical education research evaluating the involvement of patients and/or carers. Going forward, researchers and educationalists may benefit from a new, standardised approach to terminology to ensure study replicability. For instance, Towle and colleagues clearly differentiate “patients” (who have a medical problem), from “simulated/ standardised” patients who role play symptoms and signs they do not actually have [8].

Limitations

This was a rapid systematic review, conducted under time constraints and we acknowledge the potential to have excluded some relevant research. For example, articles published in foreign language and unpublished ongoing trials. We also acknowledge the inclusion of four studies where it remains unclear whether authentic patients or carers were involved, due to poor describing of methods. We decided to retain these studies as there was also no indication that patients or carers were not authentic. This raises an important issue whereby a lack of description inhibits a thorough assessment and replication of the study methods. We acknowledge that our PPI contributors were not involved in defining the research question, however their contribution to the review processes, and to our understanding of issues related to whether “authentic” patients were involved in educational activities or not, as described by study authors, was invaluable.

Conclusions

Medical schools should ensure students’ learning is reflective of everyday healthcare practice during the COVID-19 pandemic and beyond, by incorporating PPI in remote learning. We have identified a variety of digital technologies used in medical teaching where patients and carers are involved. With the majority of studies in this review describing low levels of involvement, there is a need for medical schools to embrace recommendations to involve patients and carers as equal partners in the design, delivery and evaluation of medical curricular. Digital teaching sessions involving patients or carers were beneficial and found educationally valuable by students and educators, acceptable to students, and increased their engagement, patient-centred attitudes, clinical knowledge, and communication skills. Overall, quality of the studies included in this review was moderate to high; the results of studies of poor quality and those lacking clear descriptions of patients and carers should be viewed with caution. Future research should capture patients’ and carers’ views about their involvement in remote medical education (including any barriers and facilitators) to ensure future medical training is representative of local populations and to avoid digitally excluding marginalised groups.