Common mental health problems in medical students and junior doctors – an overview of systematic reviews

Background: common mental health problems (cMHP) are prevalent among junior doctors and medical students, and the cOviD-19 pandemic has brought challenging situations with education disruptions, early graduations, and front-line work. cMHPs can have detrimental consequences on clinical safety and healthcare colleagues; thus, it is vital to assess the overall prevalence and available interventions to provide institutional-level support. Aims: this overview summarises the prevalence of cMHPs from existing published systematic reviews and informs public health prevention and early intervention practice. Methods: Four electronic databases were searched from 2012 to identify systematic reviews on the prevalence of cMHPs and/or interventions to tackle them. Results: thirty-six reviews were included: 25 assessing prevalence and 11 assessing interventions. Across systematic reviews, the prevalence of anxiety ranged from 7.04 to 88.30%, burnout from 7.0 to 86.0%, depression from 11.0 to 66.5%, stress from 29.6 to 49.9%, suicidal ideation from 3.0 to 53.9% and one obsessive-compulsive disorder review reported a prevalence of 3.8%. Mindfulness-based interventions were included in all reviews, with mixed findings for each cMHP. Conclusions: the prevalence of cMHPs is high among junior doctors and medical students, with anxiety remaining relatively stable and depression slightly increasing during the cOviD-19 pandemic. Future research on mindfulness-based interventions is required for a resilient and healthy future workforce. PRISMA/PROSPERO: the researchers have followed PRiSMA guidance. this overview was not registered with PROSPeRO as it was conducted as part of an MSc research project.


Introduction
Medical careers are renowned for their demanding nature and challenging environments (Sekhar et al., 2021).Due to laborious working conditions and exposure to occupational hazards (e.g.occupational infections and radiation exposure) (Naithani et al., 2021), all healthcare professionals are susceptible to illness (Søvold et al., 2021).Mental difficulties are a growing public health concern (Jacob et al., 2020), and evidence suggests that the prevalence may be higher among healthcare professionals (Weibelzahl et al., 2021).Junior doctors and medical students are recognised for experiencing stressors during their rigorous training programmes; and, consequently, increased strain on their mental health (Sekhar et al., 2021).This risk is further compounded by lengthy work/study times and an increased financial burden (Tam et al., 2019), along with an increased likelihood of possessing personality characteristics like maladaptive perfectionism (Prabhu & Rashad, 2021).
Common mental health problems are characterised by changes to behaviour and emotional control; they are typically linked to functional impairment, and the most prevalent are anxiety disorders, depression, and obsessive-compulsive disorder (NICE, 2022a).Common mental health problems can have detrimental effects on working ability and performance (Khan et al., 2006).A 2016 systematic review and meta-analysis estimated the prevalence of depression in medical students to be 27.2% and 11.1% for suicidal ideation (Rotenstein et al., 2016).A further systematic review found a higher prevalence of depression and anxiety among medical students during the COVID-19 pandemic than that of the general population (37.9% and 33.7% vs 24.2% and 21.3%) (Castaldelli-Maia et al., 2021;Jia et al., 2022).According to the NHS sickness absence rates from January 2022, anxiety/stress/depression/other psychiatric illnesses are the most frequently represented sickness absence reasons among all NHS workers (19.9%) (Statistics -NHS Digital, 2022), which, correspondingly, increases the pressure on healthcare services and affects patient care.This reinforces the significance of assessing the prevalence and available interventions during medical training to inform efforts to prevent and treat causes of mental distress and ensure that there is institutional-level support for their mental health, especially after the COVID-19 pandemic.
There have been previous attempts to summarise the body of evidence from published systematic reviews on the prevalence of common mental health problems and interventions to tackle them, albeit with a discrete focus on individual common mental health problems.Overviews of systematic reviews have the benefit of summarising different prevalence rates and effects of interventions for the same disorder or target group where many systematic reviews are already available (Pollock et al., 2022).Given the plethora of existing systematic reviews, it was decided to conduct an overview of systematic reviews to gather and assess the evidence on the prevalence of common mental health problems and interventions to inform clinical practice, understand their consequences and avoid the need for additional systematic reviews.
Appendix 1 Provides definitions of the abbreviations.

Aim and objectives
The overall aim is to summarise the prevalence of common mental health problems and interventions for junior doctors and medical students from existing published systematic reviews; this information can then be used to inform public health prevention and early intervention practice.The objectives are: 1. To collate the overall prevalence of common mental health problems in junior doctors and medical students.2. To highlight the change in prevalence rates during the COVID-19 pandemic.3. To assess the evidence on the effects of interventions to tackle common mental health problems.

Study design and research protocol
This overview was conducted in adherence with the recommendations of the Cochrane Handbook of Systematic Reviews of Interventions (Higgins et al., 2019) and the Preferred Reporting Items for Systematic Reviews and Meta-Analysis-(PRISMA) guidelines (Moher et al., 2009).

Eligibility criteria
Eligibility criteria were divided into two parts, and all published systematic reviews met the following inclusion criteria.
1. Prevalence of common mental health problem(s): • Study designs: a systematic review and/or meta-analysis • Participants: Junior Doctors and/or Medical Students • Common mental health problems: • Common mental health problems are characterised by clinically significant impairment to behaviour, emotional control, and individual intellect; it is typically linked to functional impairment and the most prevalent and included in the inclusion criteria are anxiety disorders, stress, depression, obsessive-compulsive disorder, burnout, suicidal ideation, and post-traumatic stress disorder (NICE, 2022a).
• Outcome measure: Prevalence of common mental health problem(s) 2. Interventions to tackle common mental health problem(s): • Study designs: a systematic review and/or meta-analysis

Literature search
A search of major electronic databases (MEDLINE, EMBASE, ERIC, PsycINFO) was conducted on 16 May 2022 to identify eligible systematic reviews published in the English language from 2012, allowing for ten years' worth of data to be collected that is most relevant to today's junior doctors and medical students.
The search strategies were designed to include an appropriate combination of MeSH and text words for the population of interest (e.g.medical students, student doctors, student physicians), types of common mental health problems (e.g.depression, anxiety, burnout, stress) and interventions to tackle them (e.g.counselling, cognitive behavioural therapy, mindfulness).Different facets of the search were combined using the Boolean operators "AND" and "OR" when appropriate.An example of the search strategies is included in Appendix 2. Additionally, the citation lists of identified systematic reviews were perused to identify further eligible reviews.

Data extraction and quality assessment of included reviews
Citations' titles and abstracts were reviewed by the primary author (SA), who then retrieved potentially relevant articles for full-text screening; any uncertainty or doubt about eligibility was discussed among all review authors to ensure that no potentially relevant papers were discarded.The full-text assessment was conducted according to the pre-specified inclusion criteria.Data extraction was performed by the primary review author (SA); however, 20% of the included reviews were randomly checked by co-authors, three experts of this overview.Details recorded for systematic reviews assessing the prevalence of common mental health problems included: publication dates and the scope of the review, target population and the number of participants, and included studies, type of common mental health problem, and review outcomes.For systematic reviews assessing interventions to tackle common mental health problems in addition to the review characteristics, information on the characteristics of interventions, comparators and outcome measures were recorded.Appendix 3 provides the data extraction forms.Measures of effect and accompanied confidence intervals (CI) were noted for systematic reviews with a meta-analysis component.
The quality of the eligible systematic reviews was assessed using the validated AMSTAR-2 tool (A Measurement Tool to Assess Systematic Reviews-2), consisting of 16 items (Shea et al., 2017), where each review received an overall quality rating.Reviews were not excluded based on AMSTAR-2 ratings.

Data synthesis
The main characteristics of each identified review and the relevant outcomes (prevalence rates and the effects of interventions) were summarised using summary tables with no attempt to standardise results across reviews.

Literature search
The literature searches retrieved 624 records.After eliminating duplicates, 395 citations were screened for eligibility, and 47 citations were selected for full-text screening, of which 45 were accessible.Twenty reviews were subsequently excluded as they failed to meet the eligibility criteria.Eleven reviews were identified by citation searching, and ultimately, 36 systematic reviews (with or without meta-analysis) were deemed suitable for inclusion.
The PRISMA diagram in Figure 1 summarises the study selection process.
All reviews included both male and female participants.Reviews that reported information on the gender of participants were predominantly female.
The basic characteristics of the 36 included systematic reviews assessing prevalence are shown in Table 1, and the systematic reviews assessing interventions in Table 2.

Prevalence
Table 3 shows the prevalence rates of common mental health problems among junior doctors and medical students as reported by the identified systematic reviews, and Appendix 8 provides definitions of the included common mental health problems.

(d) Obsessive-compulsive disorder.
One review looked at the prevalence of obsessive-compulsive disorder in medical students and reported a prevalence rate of 3.8% (Pacheco et al., 2017), using the Obsessive-Compulsive Inventory-Revised-(OCI-R) (Pacheco et al., 2017).
(e) Stress/psychological distress.Two reviews investigated the prevalence of stress (Pacheco et al., 2017) and psychological distress (Hope & Henderson, 2014) among medical students and one review looked at stress in trainee physicians (Zhou et al., 2020).Prevalence rates from the three reviews were 49.9% (Pacheco et al., 2017), 29.6% (Hope & Henderson, 2014)-and 44.6% (Zhou et al., 2020), respectively.The differences in prevalence rates can be explained by year of study as psychological distress is known to be more common as medical students progress through their course (Hope & Henderson, 2014).All reviews used the General-Health-Questionnaire-(GHQ)-instrument.

Discussion
This overview aimed to investigate the worldwide prevalence of common mental health problems among junior doctors and medical students.Twenty-five published systematic reviews from 2012 with a pooled population of 531,556 medical students and 173,030 junior doctors were assessed.Across systematic reviews, the prevalence of anxiety ranged from 7.04 to 88.30%, burnout from 7.0 to 86.0%, depression from 11.0 to 66.5%, stress from 29.6 to 49.9%, suicidal ideation from 3.0 to 53.9% and one obsessive-compulsive disorder review reported a prevalence of 3.8%.Mindfulness-based interventions were the most used interventions, with mixed findings for each common mental health problem.
Findings show high prevalence rates of common mental health problems, with the peak age for onset occurring between adolescence and early adulthood (Jurewicz, 2015) owing to establishing independence, risk-taking behaviours and neurodevelopmental changes rendering it a vulnerable period for the development of common mental health problems (Colizzi et al., 2020).This, coupled with general university stresses, heavy workloads, challenging clinical environments, and career planning uncertainties, further increases the risk of common mental health problems (Card, 2018;Jafari et al., 2012).
Clear links between financial pressures over repaying debt whilst at medical school and the development of common mental health problems have been described (Dossey, 2007;Pisaniello et al., 2019), especially in rural students (Kwong et al., 2005).Medical students from disadvantaged socioeconomic backgrounds have reported higher risks of suicidal ideation (Seo et al., 2021).
The prevalence of anxiety, depression and stress is higher in junior doctors and medical students (32.9%, 30.6%, −49.9%) (Pacheco et al., 2017) compared to the general population (26.9%, 28.0%, and 36.5%)(Nochaiwong et al., 2021).The prevalence of depression among undergraduate university students was 25% (Sheldon et al., 2021), which is relatively similar to that of non-medical students reported in two of the included reviews (22.4%-and 28.7%) (Lei et al., 2016;Puthran et al., 2016).Heavier academic pressures, tense junior-senior relationships, high-intensity internships, and consequent sleep deprivation may explain this discrepancy.Common mental health problems can lead to undesirable consequences, including suicidal ideation, poor academic performance, and compromised patient safety (Zeng et al., 2019).
When understanding common mental health problems, it is vital to be aware of the wider social context and cultural aspects that may influence treatment-seeking behaviours (Doll et al., 2021).This overview found that when comparing the prevalence of anxiety in medical students across different continents, the Middle East and Asia had the highest prevalence of anxiety (Quek et al., 2019).Cultural differences across countries may explain this; in the Middle East, emotions are often concealed due to the associated stigma (Abdullah & Brown, 2011), and in Asia, common mental health problems are considered a shame to one's family (Littlewood et al., 2007).In contrast, Caucasians may experience less stigmatisation than other sociocultural groups (Rao et al., 2007).Therefore, it is important to consider these unique socio-cultural contexts when dealing with common mental health problems, which can be considered when treating and designing public health interventions.
The COVID-19 pandemic is thought to have increased the prevalence of anxiety and depression in the general population (Sousa et al., 2021), with medical students particularly reported to have higher prevalence rates (33.7% (Jia et al., 2022)-and 37.9% (Jia et al., 2022)) than the general population-(27.7%-and26.9) (Sousa et al., 2021).Reasons for this could include the change to inherent training models of medical schools, limited clinical experience, and online learning (Jia et al., 2022;Natalia & Syakurah, 2021).
This overview highlighted that the most common forms of public mental health interventions for common mental health problems are mindfulness-based interventions and MBSR.Results show that mindfulness-based interventions significantly reduced stress, which is consistent across reviews despite the interventions being undertaken in different countries with varying lengths and components (e.g.meditation, cognitive restructuring) (Hathaisaard et al., 2022).These results are comparable to those of an earlier review that assessed interventions to reduce burnout in doctors (SMD 0.38, 95%-CI 0.26-to-0.46)(Regehr et al., 2014).Other approaches, namely mental practice and assistantship training, have also reported promising findings (Krishnan et al., 2022).
It is evident from the included reviews that there is a lack of consensus on the definitions of common mental health problems.For example, "burnout" has been defined differently over time with several validated measuring tools being used.The Copenhagen Burnout Inventory highlights fatigue and exhaustion as the main characteristics of the problem (Kristensen et al., 2005), whilst other tools describe it as a depressive disorder (Bianchi et al., 2014).Even when using the MBI, there is no consensus on the definition of burnout, which limits the interpretation and generalisability of findings (Galaiya et al., 2020).The way forward may be to separate depersonalisation and emotional exhaustion to determine those at most risk for each component and design targeted interventions (Eckleberry-Hunt et al., 2018).Duty hour restrictions have shown mixed results on burnout; however, debriefing curriculums that foster professional development and enhance the quality of care seem to be beneficial (DeChant et al., 2019).

Strengths
This overview included the use of comprehensive literature searches, the assessment of a large body of evidence (1,143 primary studies across systematic reviews) worldwide and of several common mental health problems.All included systematic reviews were published from 2012, making the observed findings more applicable to current junior doctors and medical students.

Limitations
Some limitations must be acknowledged.The information gathered from existing systematic reviews varied in terms of study designs, screening tools, demographics of the assessed populations and eligibility criteria.The prevalence of the common mental health problems cannot be precisely determined due to the heterogenous screening tools, student/ trainee populations and low methodological rigour of some reviews.It is worth noting that the nature of psychiatric screening tools relies on subjective and self-reported measures with varying cut-off values and, therefore, the reliability and validity of these measures must be interpreted with caution.• two studies reported no significant reduction in burnout: • (p = 0.204) for physical and mental exercises and (p value not reported) • one study reported the burnout subscale to be significantly reduced (p = 0.001) for body scan, sitting meditation etc.

Depression:
Mindfulness based courses: • Meditation, guided imagery, journal writing vs control • MBSr: body scan, raisin exercise, awareness of breath, yoga, walking with waitlist control

Stress:
Mindfulness based stress reduction: • Yoga postures, breathing techniques, meditation, stress management techniques) vs no control

Stress:
• four studies reported an improvement in stress levels, 2 reported no change and 1 reported an increase in stress.
• (p = 0.019) for MBSr: body scan, raisin exercise, awareness breath • Wellness curriculum -no significant differences in the MBI categories.
• Duty hours changes: significant worsening in DP (p = 0.03) and ee (p = 0.02) and significant improvement in overall burnout (p = 0.007), and ee (p = 0.056) • Professional development coaching programme: no significant differences in MBI categories.
• exercise program: no significant differences in MBI categories.
• Complementary and alternative medicine course: significant improvement p < 0.01 • respiratory one method program: significant improvement in ee (p < 0.05) • Stress management program: no significant differences in MBI scales • Debriefing curriculum: no significant differences in MBI scales Full diagnostic interviews are a more reliable assessment method.The heterogeneous study populations varied widely in locations, course structure and cultures, which may introduce various confounding variables that may influence the participation in studies and reporting of common mental health problems.Similarly, the intervention reviews had varying lengths of follow-up times and often, the absence of comparators made it challenging to draw reliable conclusions.Due to time constraints, only English language publications were included, therefore, one non-English review with no translation was not considered; this is acknowledged as a limitation of this overview.Only two reviews compared prevalence rates across groups (i.e.students from other faculties).
The degree of overlap between reviews in terms of included studies was not quantified.Finally, the definitions of mindfulness-based interventions and MBSR varied considerably across reviews and between studies included in the same reviews rendering it challenging to make proper comparisons and draw conclusions.Apart from mindfulness-based interventions and MBSR, few attempts have been made to explore other possible interventions.

Implications for practice
Medical training establishments should recognise that junior doctors and medical students are susceptible to high prevalence rates of common mental health problems.Interventions to mitigate common mental health problems should be offered to junior doctors and medical students, incorporated into medical curricula, and tailored to cultural contexts.

Implications for research
For future research, a consensus on common mental health problem definitions, valid cut-off scores, and validated outcome measures should be consistent across the field; this can be achieved by following the WHO and ICD-10 standardised definitions.Additionally, screening for co-morbid psychiatric disorders (e.g.eating disorders), and accurately reporting the data by following the Strengthening the Reporting of Observational Studies in Epidemiological guidelines (von Elm et al., 2008).
There is a need to standardise the definitions of mindfulness-based interventions, MBSR and for improved comparators to allow for fair comparisons between interventions and controls.There is a suggested benefit of extending intervention durations to up to 8-weeks (Yusoff, 2014) along with implementing longer follow-up periods to understand the long-term effects (Rotenstein et al., 2016).Comprehensive baseline diagnostics i.e. clinical interviews may be beneficial to enhance the reporting of pre-existing common mental health problems.Enhanced reporting of intervention studies is required, by following the international CONSORT statement (Schulz et al., 2010).Evidence suggests that mindfulness-based interventions may play a role in addressing common mental health problems and, therefore, future research should focus on this theoretical-based intervention and provide more robust evidence on its effects.
Finally, future studies should recruit more males and be conducted in low-middle-income countries to reach robust conclusions.Research on different formats of interventions (mobile-based, online) (Haberer et al., 2013) is also desirable.
Unsurprisingly, junior doctors and medical students have high prevalence rates of common mental health problems, which reinforces the need for more focused mental health strategies.Mental health improvements could be achieved by appropriate interventions that would result in a stronger and healthier medical workforce.

Table 1 .
Basic characteristics of the systematic reviews and meta-analysis included in the overview -prevalence.
*Global -is defined as reviews that include studies from 4 or more continents.** Chinese Databases.
Common mental health problems are frequently targeted by different forms of public mental health interventions.Given the abundance of available interventions, it is essential that effective interventions are identified and implemented (Das et al., 2016).Table 4 provides details of the included reviews, which assessed the effects of interventions to tackle common mental health problems.The definitions of mindfulness-based interventions, mindfulness-based stress-reduction-(MBSR) and other interventions are detailed in Appendix 9.

Table 2 .
Basic characteristics of systematic reviews and meta-analysis included in the overview -interventions.

Table 2 .
Global -is defined as reviews that include studies from 4 or more continents.rCt -randomised controlled trials, non-rCt -non-randomised controlled trial, Ma -Meta-analysis.Continued. *

Table 3 .
Summary of results on the prevalence of common mental health problems across systematic reviews.

Table 4 .
Summary of the findings of systematic reviews assessing interventions to tackle common mental health problems.

Table of acronyms. Appendix 2. Search strategy MEDLINE (ovid). Prevalence Ovid MEDLINE(R) and Epub Ahead of Print, In-Process, In-Data-Review & Other Non-Indexed Citations, Daily and Versions <1946 to May 16, 2022>
InterventionsOvid MEDLINE(R) and Epub Ahead of Print, In-Process, In-Data-Review & Other Non-Indexed Citations, Daily and Versions <1946 to May 16, 2022> Yes, n = no, PY = Partial Yes, nMC = no meta-analysis conducted.Yes, n = no, PY = Partial Yes, nMC = no meta-analysis conducted.