Background

Recent reviews into how medical students cope with caring for the dying and attending to patient deaths have raised questions into how medical schools support students across their training trajectories [1]. Pivotally, reports reveal that rising anxiety, distress and compromises to mental and general wellbeing amongst medical student predispose them to medical errors, and jeopardize patient communication and care [2,3,4,5,6,7,8]. This does not only underscore the need to review current curricula and support services, but also the need to improve and innovate education into self-care. However, self-care education escapes the focus of most medical school curricula and remains rudimentary.

The call for robust self-care education is further underlined by increasing evidence that medical students require individualized approaches to cope with their competing academic, research, clinical, administrative, social, relational, familial and individual commitments and existential and ethical dilemmas [9,10,11,12,13]. Here, empowering medical students to devise their own means of supporting themselves is key.

Acknowledging these gaps in the medical curricula, we undertook a review to map self-care education amongst medical students guided by our primary research question, “What is known about self-care education interventions amongst medical students?”. For the purposes of this review, self-care education is characterized as “a spectrum of knowledge, skills and attitudes including self-reflection and self-awareness in identifying and preventing burnout with professional boundaries and handling grief and bereavement appropriately” (p. 77) [8].

Methods

A Systematic Evidenced Based Approach guided systematic scoping review (henceforth SSR in SEBA) was adopted to map prevailing literature on self-care education amongst medical students [14,15,16,17]. This SSR in SEBA was overseen by an expert team comprising of medical librarians from the Yong Loo Lin School of Medicine (YLLSoM), and local educational experts and clinicians at NCCS, the Palliative Care Institute Liverpool, YLLSoM and Duke-NUS Medical School who guided, oversaw and supported all stages of SEBA to enhance the reproducibility and accountability of the process [14, 15, 17, 18] (Fig. 1). This SSR in SEBA is also shaped by SEBA’s constructivist ontological perspective and relativist lens, as well as the principles of interpretivist analysis to enhance reflexivity of the research analysis and discussions [19,20,21,22].

Fig. 1
figure 1

The SSR in SEBA process [23]

Stage 1 of SEBA: systematic approach

  1. i.

    Determining the title and research question and inclusion criteria

The PICOs format and the PRISMA-P 2015 checklist (see Additional file 1) were employed to guide the primary research question, “What is known about self-care education interventions amongst medical students?”. The secondary research questions were, “How are self-care education interventions structured in medical school curriculum?, “What topics are included in self-care education curriculum in medical schools?”, “How is self-care in medical schools assessed?” and “What factors support and hinder self-care education interventions in medical schools?” (Table 1).

Table 1 PICOs, inclusion criteria and exclusion criteria applied to database search
  1. ii.

    Searching

Searches were conducted on PubMed, Embase, PsycINFO, ERIC, Google Scholar and Scopus databases and key medical education journals, including BMC Medical Education, Academic Medicine, Medical Education, Medical Teacher, Medical Education Online and Canadian Medical Education Journal published between 1st January 2000 and 30th June 2023. It was conducted independently by authors DWWJ, LSHG, GLGP, CKRL, JAL, NAH, OEK, NS and LKRK. Variations of the terms “self-care education”, “medical students” and “medical education” were applied. This timeframe was selected to facilitate a viable and sustainable research process and to account for prevailing manpower and time constraints [23]. These searches were also accompanied by ‘snowballing’ of the references of included articles to ensure a more comprehensive review [24]. The full search strategy can be found in Additional file 2.

Each of the nine members of the research team consolidated their own lists of articles to be included. To reach an agreement on the final set of articles to be reviewed, the team then adopted Sandelowski and Barroso [25]‘s ‘negotiated consensual validation’ that saw “research team members articulate, defend, and persuade others of the ‘cogency’ or ‘incisiveness’ of their points of view”. The accepted list of articles was then consolidated into a master list for further sieving to determine their suitability.

  1. iii.

    Extracting and charting

The titles and abstracts were subsequently independently reviewed by GLGP, LYY, DWWJ, LSHG, CKRL, JAL, NAH, OEK, NS and LKRK using an abstract screening tool. The team then discussed their findings for the deconflicting process, similarly applying ‘negotiated consensual validation’ to finalize the list of articles to be included [25]. This process involved the screening of the abstracts and titles of the articles, followed by a deeper in-depth sieve of the full text of each article. Articles that did not fit in the inclusion criteria in any of these two stages were removed whilst articles that met the inclusion criteria proceeded to the data extraction and quality assessment stages.

  1. iv.

    Assessing quality of articles

NDAR, MC, DWWJ, LSHG, GHKY, JJHL and CJL individually appraised the quality of the quantitative and qualitative studies using the Medical Education Research Study Quality Instrument (MERSQI) [26] and Consolidated Criteria for Reporting Qualitative Studies (COREQ) [27] (see Additional file 3).

Stage 2 of SEBA: split approach

  1. a.

    Summary and tabulation of full-text articles

The Split Approach [28] was carried out by three teams. The first team (LSHG, MYKT, NR, CJLG, NS, YLL) summarized and tabulated the included full-text articles in keeping with recommendations drawn from Wong, Greenhalgh [29]‘s RAMESES publication standards and Popay, Roberts [30]‘s “Guidance on the conduct of narrative synthesis in systematic reviews”. A tabulated summary of the included articles is enclosed in Additional file 3.

  1. b.

    Braun and Clarke’s thematic analysis

    Guided by Braun and Clarke [31]‘s approach to thematic analysis, the second team of researchers (DWJW, GHKY, GLGP, OEK, LKR) independently reviewed the included articles to extract relevant findings. They subsequently crafted a code book from the extracted data categorized according to the emerging themes. In an iterative step-by-step analysis process [32], the team combined each new emerging code with previous codes. This formed fresh themes that were derived from the raw data with no prior classification [33]. Thereafter, the team organized meetings to discuss their independent findings, shortlisting the final list of themes through ‘negotiated consensual validation’ [25].

  2. c.

    Hsieh and Shannon’s directed content analysis

Concurrently, the third team of researchers (NR, JJHL, MC, NDAR) employed Hsieh and Shannon [34]‘s approach to directed content analysis. This entailed the identification and operationalizing of a priori coding categories [34,35,36,37,38,39]. Here, codes and categories were drawn from Drolet and Rodger’s study entitled, “A Comprehensive Medical Student Wellness Program—Design and Implementation at Vanderbilt School of Medicine” [40]. Known as the ‘coding agenda’ [41, 42], the research team adopted these codes as a template for coding the included articles. This served to diminish concerns on the inconsistency, incoherence and omission of negative results seen in thematic analysis [18, 43,44,45,46,47,48,49,50]. The team also prescribed new codes to any data uncaptured by the priori codes [41]. ‘Negotiated consensual validation’ was similarly practiced by the team to attain consensus on the final categories [25, 37].

Stage 3 of SEBA: jigsaw perspective

The Jigsaw Perspective employed Phases 4 to 6 of France et al. [51]‘s adaptation of Noblit et al. [52]‘s seven phases of meta-ethnographic approach. This stage entailed DWJW, LSHG, MYKT, CJLL, GHKY, NR, OEK, GLGP, NS and LKRK contrasting themes and subthemes with the categories and subcategories identified. Upon verifying the similarities by comparing the codes contained within each group of data, the researchers then merged complementary categories and themes, as well as complementary subthemes and subcategories, to form larger ‘themes/categories’.

Stage 4 of SEBA: funnelling

DWJW, LSHG, MYKT, CJLL, GHKY, NR, OEK, GLGP, NS and LKRK compared the ‘themes/categories’ with the tabulated summaries [51, 52] and included quality appraisals using MERSQI and COREQ [26, 27]. This led to domains that formed the basis of the discussion’s ‘line of argument’ in Stage 5 of SEBA.

A total of 6128 abstracts were reviewed, 429 full text articles were evaluated, and 147 articles were included (Fig. 2).

Fig. 2
figure 2

PRISMA flowchart

Results

The Funnelling process revealed six domains: definition, topics, pedagogy, influencers, outcomes and assessments. Many of these domains were merely listed in the included articles without any accompanying descriptions or clarifications. Thus, to enhance clarity and facilitate the review, we have summarized and presented the domains in tables.

Domain 1: conceptualization of self-care education

Self-care education in medical schools can be characterized as interventions that seek to promote positive coping strategies and reflective practice to boost psychological, emotional, and physical well-being [5, 53,54,55,56] whilst fostering competent, caring, and resilient physicians [57]. This ‘catch-all’ characterization allows the inclusion of an expanding array of interventions and acknowledges the notion that a variety of options are being tried, adapted, or used on their own or in combination to meet the needs, goals, individual preferences, working styles, experience, attitudes and skills of medical students. This wide conceptualization of self-care education accounts for a mix of options to cater to a medical student’s individual historical, socio-cultural and contextual narratives, as well as psycho-emotional states, in different settings, stages of training, specialities, cultures and curricula.

Domain 2: topics and intervention

A wide range of interventions are used to introduce self-care. Table 2 details the various contents espoused within the included publications for ease of review.

Table 2 Topics and interventions

The most prevalent intervention topic is mindfulness-based interventions perhaps due to growing interest in this field [58,59,60,61,62]; increasing social acceptability of this form of meditation [63,64,65]; its ease of use [4, 5, 66]; its proven efficacy in alleviating anxiety and depression amongst medical students [5, 67]; and its ability to promote attention, relaxation, and emotional intelligence [63,64,65]. Mindfulness-based interventions promote non-judgmental awareness and acceptance of internal and external events, thoughts, and emotions that foster the capacity to respond to situations with equanimity [63,64,65]. The versatile nature of this form of meditation sees it used in a variety of self-care interventions, including mindful breathing, mindful eating, mindful yoga [4, 5, 66], managing reflexivity, reflective listening, and journaling [4, 5, 66].

However, mindfulness is not a ‘one-size-fits-all’ solution and may not be uniformly accepted nor applicable to all users [68]. When poorly supported or inculcated within appropriate settings, it may precipitate negative effects [58]. Engaging in mindfulness may also pose a challenge for acutely stressed or anxious students [5, 67]. Reviews on mindfulness are also divided on its overall efficacy [58,59,60,61].

Other stress management interventions [59, 62] are also proffered. These interventions tend to inculcate elements of mindfulness and focus on instilling more effective coping mechanisms, recognizing the symptoms of stress, and capturing the negative effects of stress on their learning, personal health and patient care [69,70,71]. Additional interventions include lifestyle interventions, such as increasing physical activity and improving eating habits and sleep quality [72,73,74,75]. More recently, psychoeducation or the use of activities, such as mask making to promote self-reflection and development of personal identity, have been adopted [76,77,78].

Domain 3: pedagogy

Current self-care interventions vary in duration, place in the curriculum, group size, facilitator, and delivery methods. The features of pedagogy utilized are illustrated in Table 3.

Table 3 Pedagogy used

Much of the debate on self-care pedagogy is premised on whether it should be voluntary or mandatory. Whilst mandatory self-care interventions maximize audience reach, such actions may render them counterproductive [113, 131]. When made mandatory, medical students may perceive it as a violation of their autonomy. This could reduce engagement and precipitate stress and feelings of resentment and coercion [4, 5, 64, 65, 87, 90, 111, 131, 137, 139]. Proponents of voluntary participation in self-care programs also argue that the effectiveness of such interventions far outweighs greater audience reach—in turn boosting active participation and enhanced engagement and better outcomes [113, 131].

Domain 4: influences

Factors facilitating or hindering the success of self-care interventions occur at the student or program level. These are summarized in Table 4. Both options rely on the choice of program delivery, contextual considerations, approach, and the presence of a conducive environment that facilitates active and open sharing [154, 155]. A conducive environment is also inclusive of protected time to attend and actively engage in these interventions [68, 137, 161]. Indeed, when poorly supported, these programs become an additional source of stress [91, 145, 173].

Table 4 Influences upon self-care education

At the student level, interest and adherence are pivotal facilitators to effective self-care interventions. Conversely, poor understanding of the interventions, mandatory participation and low adherence hinder success.

At a program level, smaller peer or clinician-led sessions are more successful in facilitating open, safe and collaborative discussions [154, 155]. Contrarily, large group sizes and the lack of time and resources impede engagement in these programs [68, 137, 161] and in some cases, become an additional curricular demand [91, 145, 173].

Domain 5: outcomes

The positive impacts of self-care education are illustrated in Table 5. However, some reviews reveal equivocal or even negative outcomes [58]. The effects are categorized into student and patient levels.

Table 5 Benefits of self-care education

At the student level, self-care programs enhance student wellbeing, reduce psychological distress and effects, inculcate positive values and skills, and increase academic performance [67, 88, 93, 111, 129, 132, 138, 140]. At the patient level, there is an improvement in patient safety, quality of care and patient satisfaction [2, 4,5,6, 93, 94, 96, 130, 137, 167].

Domain 6: outcome assessment method

Current assessment methods are listed in Table 6. Most studies employ validated questionnaires, wherein the Perceived Stress Scale presents the most common tool used, as observed in 16 studies [66, 80, 83,84,85, 88, 91, 94, 95, 110, 130, 137, 159, 161, 169, 174]. Four studies utilize interviews [82, 91, 96, 111] whilst two studies employ laboratory tests, such as measuring salivary cortisol [3, 140], as their methods of assessment. The remaining studies adopt non-validated questionnaires and surveys [5, 54, 56, 75, 76, 84, 86, 87, 113, 131, 139, 145, 149].

Table 6 Methods of assessment

Stage 5 of SEBA: analysis of evidence-based and non-data driven literature

Evidenced-based data from bibliographic databases (henceforth evidence-based publications) were separated from grey literature and opinion, perspectives, editorial, letters and non-data-based articles drawn from bibliographic databases (henceforth non-data driven literature). The two groups of data were thematically analysed separately. The themes/categories from both groups were then compared against each other to determine if there were additional themes in the non-data driven group that could influence the narrative.

There was consensus that themes from the non-data driven and peer-reviewed evidence-based publications were similar and did not bias the analysis untowardly.

Discussion

Stage 6: synthesis of discussion

In answering its primary and secondary research questions, “What is known about self-care education interventions amongst medical students?”, this SSR in SEBA provides a sketch of the current state of self-care education in medical school curricula. Each key aspect is considered by its secondary research questions. Here, the secondary research questions, “What topics are included in self-care education in medical schools?”, “How are self-care education interventions structured in medical schools?”, “How is self-care in medical school assessed?” and “What factors support and hinder self-care education interventions in medical schools?” highlight the topics and interventions used in Domain 2 (Table 2), the pedagogy employed in Domain 3 (Table 3), the influences upon the training processes in Domain 4 (Table 4) and the outcomes and outcome assessment methods in Domains 5 and 6 (Tables 5 and 6) respectively.

In answering its secondary research question, “How are self-care education interventions structured in medical schools?”, current data suggests that such programs should be provided a formal place within the curriculum, accompanied by the provision of trained tutors, protected time for engagement, an appropriate setting, and opportunities for debriefs [5,6,7, 67]. It is likely that mandatory sessions will lack the desired effects but greater education on the matter would be useful to allow students to make an informed decision on participating [113, 131]. Programs should also provide general and personalized information on self-care. General education ought to cater to the goals of the program, the group size [64, 138, 139, 167], and the setting [6, 91, 131, 173] whilst individualized advice must consider the specific needs [64, 138, 139, 167], motivations [4, 5, 64, 65, 87, 131, 137, 139] and abilities of individual medical students.

Similarly, available resources should also be accounted for where considerations are made with regards to the training environment [68, 100, 138], structure [90, 111, 131], assessment methods and outcome measures [122, 123], as well as tutor support available. Critically, at a program-level, self-care education sessions must be supplemented with role modelling, mentoring, supervision and coaching to provide timely, personalized, appropriate, holistic guidance, support and remediation [5,6,7, 56, 67, 68, 75, 90, 91, 96, 131, 137, 143, 145, 155, 161, 173]. Faculty development and the presence of dedicated facilitators must also be a key consideration.

Returning to the context of medical students who are frequently exposed to patient death and suffering where psycho-existential distress has been recognized, awareness about issues on mental and emotional health should be raised. This then necessitates the availability and access to self-care interventions for those who choose to engage in these programs. We also underscore the importance of ensuring that there is sufficient time and support allocated to these programs, as well as effective means of providing longitudinal support post-medical school.

Limitations

Focus upon guidelines published in English may have restricted the search results whilst data drawn from North America and the European countries may not be necessarily transferable beyond these regions where education, healthcare programs and healthcare financing differ.

Conclusions

Whilst awareness of mental health issues ought to be underscored, as should its role in professionalism, and access to self-care education and interventions be made easy for those who choose to engage in these practices, we believe that one area of urgent concern is tutor training. Tutors who are expected to access and support students should be provided training and longitudinal support. Similar ties and access to psychological and psychiatric medical services, formal debriefs, coaching, remediation, and supervision programs should be made clear. Further study in changing the culture and perspectives of self-care and mental and psycho-emotional well-being in medicine should be the focus of future studies, as should the design of effective assessment tools.