Suicide is a leading cause of death globally for people aged 15 to 29 years old (WHO, 2014). University students studying medicine, the majority of whom are within this age range, appear to show higher rates of suicidal ideation and attempted suicide compared with members of the general population who are the same age (Tyssen et al., 2001). One large meta-analysis reported that 27.2% of medical students reported symptoms of depression, and 11.1% reported suicidal ideation (Rotenstein et al., 2016).

Indeed, medical students face unique stressors compared with the general population and other university students, such as emotionally and intellectually demanding learning, frequent examinations, and exposure to severe illness and death including suicide. Medical students are also less likely to seek help despite better knowledge of treatment options and services, with fear of negative consequences with respect to career progression and the right to practice cited as common barriers (Beyondblue, 2013).

Universal interventions are programmes or support systems provided to a whole population, aiming to improve outcomes in the group overall. In this case, this would be an intervention designed to be delivered to all medical students to address mental ill-health irrespective of risk and/or current levels of distress. Previous research has shown that several aspects of medical school curricula have been identified as being associated with poor mental health outcomes, indicating these as a potential avenue for universal interventions (Radcliffe et al., 2003).

There is therefore a need to develop effective universal interventions that can address mental ill-health, and in particular suicidal ideation and suicidal behaviour in medical students. This study conducted a systematic review and meta-analysis to draw together the existing evidence in this area (Witt et al, 2019).

Methods

Studies were eligible to be included in this review if they were published in English and met the following criteria:

• They were a randomised controlled trial (RCT), a non-randomised controlled trial (non-RCT), or they used a historically controlled design or uncontrolled longitudinal design
• They evaluated a university intervention to address suicidal ideation, suicidal behaviour, depression, anxiety, stress and/or burnout
• The participants were medical students.

The authors searched three major databases (Embase, MEDLINE and PsycINFO) for papers published before 1^st December 2017, and also hand-searched for missed studies. Retrieved papers were double screened independently and disagreements resolved through team discussions. The quality of each included study was evaluated using the appropriate tool; the Cochrane Risk of Bias tool for RCTs, the Risk of Bias in Non-randomised Studies of Interventions tool for non-RCTs, and the Newcastle-Ottawa Scale for longitudinal studies.

Quantitative data were synthesised using meta-analysis, calculating standardised mean differences for continuous data and pooled odds ratios for dichotomous data. Data were not pooled across individually-focused and curriculum-based interventions.

Results

Characteristics of included studies

1,505 papers were identified from the electronic search, and 116 from hand-searching. From these, 39 studies were included in the final review. 24 of these were conducted in the USA and 13 were RCTs. The majority of interventions described were based on mindfulness principles (18 studies), with the next most common being stress management training (5 studies). Five studies also investigated the impact on stress of changing from a lecture-based curriculum format to a problem-based format. Across the 39 studies, 7,387 medical students were included, aged between 18 and 32 years. 25 studies reported participants’ gender, and in 21 of these the majority of participants were women.

Suicidal ideation

• Only one study investigated an intervention specifically designed to address suicidal ideation, and none reported on suicidal behaviour outcomes.
• Only one study included information on the proportion of participants reporting suicidal ideation, and this was measured one year after the psychoeducation intervention. This study found some evidence of a moderate reduction in the proportion of participants reporting suicidal ideation.

Depression

• In terms of depression, the majority of the included studies on individually-focused interventions found no evidence of an effect of the interventions studied, including: mindfulness programmes (6 RCTs), stress management programmes (2 RCTs and 1 non-RCT), guided meditation (1 non-RCT), omega-3 fatty acid supplementation (1 non-RCT), and a yoga programme (1 uncontrolled longitudinal study)
• Six uncontrolled longitudinal studies did report a benefit of a range of interventions, including yoga, psychoeducation and relaxation training, but this was no longer significant when studies recruiting graduate-entry medical students were removed
• Similarly, the three studies which investigated curriculum-based interventions did not find evidence of a beneficial effect on depression.

Anxiety

• In terms of anxiety, the majority of the included studies on individually-focused interventions found no evidence of an effect of the interventions studied, including mindfulness-based interventions (5 RCTs), a multicomponent stress management programme (1 RCT), and a guided meditation and group-based stress management programme (3 non-RCTs)
• Two non-RCTs reported an increase in anxiety scores at final follow-up, after guided meditation and psychoeducation (respectively)
• One RCT found a small beneficial effect of omega-3 fatty acid supplementation on anxiety scores
• Six uncontrolled longitudinal studies did report a benefit of a range of interventions, as for depression, and this remained significant (though reduced) when studies recruiting graduate-entry medical students were removed
• The one study which investigated a curriculum-based intervention did not find evidence of a beneficial effect on anxiety.

Stress and burnout

• One RCT reported no impact of mindfulness on stress post-intervention, but two RCTs found a positive impact on stress and burnout at later follow-up
• Three uncontrolled longitudinal studies found a modest benefit of stress management programmes on stress, though one non-RCT failed to replicate this finding. No studies which investigated a curriculum-based intervention reported data on stress or burnout.

Conclusions

In summary, the findings are mixed but overall there is little evidence that university interventions, whether individually-focused or curriculum-based, are effective at improving suicidal ideation, depression, anxiety and stress in medical students.

The authors note that there are too few studies at present, and what does exist is of poor quality. More evidence would be needed firstly to establish whether any of the interventions included are indeed effective at improving medical students’ mental health, and secondly to ascertain their impact specifically on suicidal ideation and behaviour.

Strengths and limitations

Overall, this was a solid review of preliminary evidence in an under-researched area. Nevertheless, there are some important limitations to consider:

• Medical education structure and culture is likely to differ between institutions and in particular between countries. Many of the included studies in this review were conducted in the USA, and therefore the findings, particularly relating to curriculum-based changes, may not be applicable in the UK
• A variety of different interventions were considered alongside one another here, and when more research is available it will be important to evaluate which specific elements of each intervention may be more promising than others, rather than considering all ‘stress management programmes’ (for example) as one group despite the heterogeneity
• Only one study included suicidal ideation as an outcome. The authors reported on other outcomes such as stress and burnout which may be related but are qualitatively different, and therefore may be expected to respond differently to interventions. This is likely due to the lack of evidence available, but should be kept in mind when interpreting the findings together.

Implications

Future research should focus on the development, evaluation and implementation of mental health interventions specifically designed to be delivered to medical students at a university level. The authors noted that the interventions tested here are likely to have been of too low intensity, so it will be important to consider what exactly is being targeted (for example, anxiety and suicidal ideation are very different experiences) and what the appropriate ‘active ingredients’ of interventions might be.

Most studies included in this review focus on individual-level mindfulness-based interventions and largely ignore the systemic pressures students face. Further research is needed regarding curriculum-based interventions.

In the meantime, universities should consider tackling the organisational and structural stressors associated with medical education and training, such as competitive cultures, frequent examinations and high workloads, in order to improve students’ mental health. For students who do experience mental health problems, it is imperative that universities commit to reducing stigma around help-seeking and ensuring that a range of support is available.

Statement of interests

None.

Links

Primary paper

Witt K, Boland A, Lamblin M, McGorry PD, Veness B, Cipriani A, Hawton K, Harvey S, Christensen H, Robinson J. (2019) Effectiveness of universal programmes for the prevention of suicidal ideation, behaviour and mental ill health in medical students: a systematic review and meta-analysis. Evidence-based mental health. 2019 May 1;22(2):84-90.

Other references

World Health Organization. Preventing suicide: a global imperative. Geneva, Switzerland: World Health Organization, 2014

Tyssen R , Vaglum P , Grønvold NT , et al . Suicidal ideation among medical students and young physicians: a nationwide and prospective study of prevalence and predictors. J Affect Disord 2001;64:69–79.

Rotenstein LS , Ramos MA , Torre M , et al . Prevalence of depression, depressive symptoms, and suicidal ideation among medical students: a systematic review and meta-analysis. JAMA 2016;316:2214–36.

beyondblue. National Mental Health Survey of Doctors and Medical Students. Hawthorne, VIC: beyondblue, 2013.

Radcliffe C, Lester H. Perceived stress during undergraduate medical training: a qualitative study. Medical education. 2003 Jan;37(1):32-8.

Photo credits

• Photo by Jaime Lopes on Unsplash
• Photo by Paweł Czerwiński on Unsplash
• Photo by David Travis on Unsplash