Abstract
Background:
Studies from around the world have shown higher rates of anxiety, depression, alcohol and other drug use, and burnout in medical students.
Aims:
The aim of this study was to identify the socio-demographic factors and severity of difficulties Sri Lankan medical students face regarding psychological wellbeing and burnout.
Method:
This one-off survey used a cross-sectional design, assessing substance use, psychological wellbeing, and burnout using the CAGE, GHQ-12, and OLBI. The survey was open to all medical students in six universities in Sri Lanka. Chi-square analysis was used to assess the statistical significance related to categorical dependent variables and one-way ANOVA for continuous dependent variables.
Results:
A higher prevalence of diagnosed mental health conditions was found following admission to the medical course in comparison prior to admission. Sixty-two percent of students had a score of more than 2 on the GHQ-12 indicating caseness. The OLBI identified exhaustion in 79% of students. The CAGE questionnaire was positive in 4.8% of students.
Conclusions:
Only a small proportion of students are recognizing their mental health difficulties and seeking help. Further understanding is required as to why this is, as well as re-evaluation of the demands of the curriculum. Effective ways of regularly identifying and providing practical and evidence-based support for mental health problems in medical and other undergraduates need to be identified and introduced.
Introduction
Sri Lanka is a lower-middle-income country in South Asia with a population of 21.4 million people (World Bank, 2020). Sri Lanka has eight medical faculties regulated by the University Grants Commission (UGC) and one medical faculty within the Ministry of Defence. Students are selected for medical schools following their advanced level examination, based on both merit and a district quota system (University Grants Commission, 2020). The medical undergraduate curriculum in Sri Lanka spans 5 years, with an initial 1.5 years of pre-clinical training followed by 3.5 years of clinical teaching. Following graduation, all medical graduates are offered employment by the Ministry of Health.
Stress during medical undergraduate training is increasingly being reported as a global phenomenon. Several studies across the world have reported higher levels of depression, anxiety, and psychological distress amongst medical students than their peers (Rotenstein et al., 2016). This can result in numerous detrimental effects on medical students and their training, including impairment of classroom functioning, disorders related to stress, and performance deterioration (Gupta et al., 2014). Furthermore, anxiety and depression in medical students have been associated with various self-stigmatizing attitudes, such as feeling less confident and respected by their colleagues or being prone to increased thoughts that they will harm their future patients (Schwenk et al., 2010). The risk of dropout is reported to be three-fold higher in medical students who describe personal distress and depressive symptoms, compared to those that do not (Dyrbye et al., 2010).
Mental health problems have been frequently recognized in Sri Lankan undergraduates, with a high prevalence of depressive symptoms (Amarasuriya et al., 2015). A study among undergraduates in one university in Sri Lanka reported one of the highest rates of depressive symptoms in the world, with younger female students being at higher risk (Torabi & Perera, 2006). However, only a few studies have focused on the mental health and well-being of medical students. A recent study among medical students at the University of Sri Jayewardenepura has shown significantly high psychological impairment in final year students, with academic pressure being reported as the main contributory factor (Liyanage, 2017). Another study indicated that higher emotional intelligence was associated with a higher level of self-satisfaction, lower stress, and better academic performance amongst final year medical students (Wijekoon et al., 2017).
This study aimed at assessing the sources of stress, mental health problems, burnout, and substance use among medical students of six state universities and determine the associations between socio-demographic factors, psychological stress, and substance use among these students.
Methods
All medical undergraduates attached to the Universities of Colombo, Peradeniya, Ruhuna, Sri Jayewardenepura, Kelaniya, and Rajarata were requested to complete a one-off online survey. The survey was conducted in the English language as the undergraduate medical degrees in all universities are conducted in English. This was part of an international collaboration with the participation of medical undergraduates from several countries. Ethical approval for the study was obtained from the Ethics Review Committee of the Faculty of Medicine, University of Colombo. At each of the universities, one co-investigator coordinated the process of obtaining permission from their respective Deans. The coordinators at each site then sent out the emails with the link to the survey along with a covering letter introducing the study, informing students of anonymity, and encouraging their participation. Informed consent was obtained from all participants. An email reminder was sent after 2 weeks. Students were offered the opportunity to contact the investigators regarding any further information or clarifications needed. The survey remained open during July and August 2020.
The survey consisted of basic socio-demographic information including year of study, age, gender, educational level of parents, pre-existing mental health issues, and current mental health. The CAGE questionnaire and other questions were used to identify substance misuse. CAGE is a reliable and valid screening tool in the general population with a cut-off of 2 or more (Chan et al., 1994). The short version of the general health questionnaire (GHQ-12) was used to identify minor psychiatric disorders. The General Health Questionnaire is a screening tool for current mental health problems, used widely in research with a cut-off score of >2 (Makowska et al., 2002). The Oldenburg Burnout Inventory was used to assess the level of burnout among students. Oldenburg Burnout Inventory was used to assess for disengagement and exhaustion using cut-offs of >2.1 and >2.25, respectively (Halbesleben & Demerouti, 2005). This survey matched that carried out in 2019 among medical students in 12 countries (Castaldelli-Maia et al., 2019; Volpe et al., 2019).
SPSS software was employed to compare answers from various demographic groups. Chi-square test was used to assess the statistical significance related to categorical dependent variables and one-way ANOVA was used for continuous dependent variables. The confidence interval was set at 95%.
Results
Socio-demographic characteristics
One thousand and ninety-seven (1,097) medical students completed the online survey. Females represented 68% of all respondents. Among participants, 18% were in first year and about 20% in final year. Many of the parents of students assessed (46%) had an A-level equivalent educational qualification. Sociodemographic characteristics of the participants are shown in Table 1.
Socio-demographic characteristics of participants (n = 1,097).
Mental health
Among the participants, 3% had been diagnosed with a mental health condition before entering medical school and 9% were diagnosed after entry to university. Among the participants, 5% had been taking psychotropic medications prior to university entrance, compared to a total of 4% after university entrance. Academic work was considered as a source of stress by 81.8% of students and financial stressors by 31%.
According to the GHQ-12, 62% of medical students were above the cut-off (>2). The mean reported score was 4.68 ± 3.43 (SD). The mean (±SD) scores of total Oldenburg Burnout Inventory (OLBI), disengagement, and exhaustion sub-scales were, respectively, 2.62 ± 0.38, 2.57 ± 0.37, and 2.67 ± 0.44. According to predefined cut-offs of OLBI, disengagement was identified in 93% of undergraduates, and exhaustion in 79% of participants. Mental health characteristics before and during university are reported in Table 2.
Mental health issues among medical students (n = 1,097).
As shown in Table 3, GHQ-12 and OLBI scores showed no significant differences with regard to gender, year of study, and presence of a mental health condition before entering medical school. Housing as a source of stress was significantly associated with higher GHQ-12 scores (p = .015). Further, use of a non-prescription substance or prescription medication outside of its intended use in the medical school was associated significantly with higher OLBI exhaustion scores (p = .049).
Association between sociodemographic characteristics, mental health facts, and psychometric scores in medical students (*p < .05).
Psychoactive substance use
Lifetime use of at least one form of substance was reported by 7%, while 93% denied any use of substances in their lifetime. Five percent of medical students were found to be at risk of alcohol-related health problems by the CAGE questionnaire. Forty students (4%) reported taking prescribed or non-prescribed medication outside of its indication and 28 (3%) used medications to enhance concentration. Further details of substance use by participating Sri Lankan medical students are shown in Table 4.
Substance use among Sri Lankan medical students (n = 1,097).
Associations between gender of medical students, sources of stress, and substance use
Males had significantly higher levels of financial stress than females. In contrast, academic work as a source of stress was significantly higher in females. Males had significantly more alcohol-related problems, cannabis, ecstasy, ketamine, and other substance use than females. Other associations related to gender and substance use are shown in Table 5. Mental health problems were not significantly associated with year of study or institution.
Associations between gender of medical students, sources of stress, and substance use (n = 1,097).
Discussion
This study revealed several important findings related to the psychological wellbeing of Sri Lankan medical undergraduates, most noticeably an increase in reported problems after admission. One study reported lower levels of mental health problems as medical students start their course (Brazeau et al., 2014). Another study showed that depressive symptoms and suicidal ideas increase significantly after entering medical school (Rotenstein et al., 2016).
Nine per cent of our sample reported a current mental health diagnosis but 63% screened positive for caseness on the GHQ12, suggesting that many are not seeking or getting the help they need. When the same survey was done among medical undergraduates in 2019, 73% of their sample reached caseness criteria in the GHQ 12 in a Brazilian study while only 16% had a diagnosed mental health condition (Castaldelli-Maia et al., 2019). Fifty-three percent reached caseness criteria in a New Zealand sample while only 21% had a diagnosed mental health condition (Farrell et al., 2019b). This indicate that only a small proportion of students with mental health problems are recognizing the problem and seeking help. The majority seem to be underdiagnosed and unrecognized, by the students themselves and by professionals. A recent meta-analysis found that depression alone affects over a quarter (28%) of medical students globally but that treatment rates are low (12%) (Puthran et al., 2016). Commonly reported barriers include lack of time, confidentiality, stigma, cost, fear of documentation on academic record, and fear of unwanted intervention (Givens & Tjia, 2002). Students’ awareness about symptoms of depression and other common mental illnesses may be poor (Kuzman et al., 2014). Teacher and student stigma against mental illness is quite common (Janoušková et al., 2017) and is likely to interfere with treatment-seeking.
Higher levels of mental health problems in our sample were significantly associated with housing as a source of stress but not with the other general stressors assessed.
Most students (82%) perceived ‘study’ as a source of stress, comparable to other studies where academic pressures have been associated with a decline in students’ mental health (Dyrbye et al., 2006). Medical school entry is through a highly competitive written examination. Student selection in Sri Lanka is not supported by interviewing to ensure that the prospective student has other necessary attributes (Mettananda et al., 2008). After completing finals, job prospects are determined by their rank in a common merit list formulated using final examination marks and creating a competitive environment among students and increasing stress as much is at stake.
Money was reported as a source of stress among 31%. Sri Lankan state universities offer free education, and the students are supported with hostel facilities at least for part of their course. Many depend on parents or guardians for other expenditure such as food, transport, private boarding houses, and textbooks. Students with low parental income receive a scholarship or bursary which covers only a proportion of that expenditure. Some students face financial difficulties day to day as a result. Sri Lankan culture does not provide a lot of part-time job opportunities and students have limited free time, which may make the financial pressure worse. Only 7% of our sample did part-time work. Financial constraints as a contributor to stress have been reported among university students in other studies as well (Wolf et al., 1988). Relationships were recognized as a source of stress by 509 (46%) students. For most students, this will be their first time away from home, bringing opportunities, and challenges.
Reported rates of disengagement (93%) and exhaustion (79%) are of concern, comparable to other international cohorts (Castaldelli-Maia et al., 2019; Farrell et al., 2019a). This demands our urgent attention with regards to causative mechanisms and potential interventions. Burnout is also shown to be common among medical students and doctors, and timely interventions may prevent negative consequences for the workforce and the patients (Dyrbye & Shanafelt, 2016).
Very low rates of substance misuse were reported compared to other countries (Castaldelli-Maia et al., 2019). Problematic alcohol use among Sri Lankan students was 5%, much less than in other cohorts studied; 18% in Brazil, 15% in England, 18% in New Zealand (Castaldelli-Maia et al., 2019; Farrell et al., 2019a, 2019b).
Important gender differences were noted in our sample. Male medical students reported money as a source of stress significantly more than females, whereas female students reported academic work as a source of stress more frequently. Male students had significantly more alcohol-related problems and other substance use. In Sri Lanka the number of girls completing secondary education has equalled that of boys, and young females have become the majority among the university entrants. They are likely to experience substantial distress in a patriarchal society where they would be expected to fulfill traditional gender roles as well (Chandradasa & Rathnayake, 2019). Therefore, gender-specific needs in medical undergraduate training must be taken in to account when planning interventions.
Our group of students did not show significant associations of mental health characteristics with the year of study. Our finding that students enter medical school from all social classes in Sri Lanka was reassuring and welcome. This may be a result of our country’s strong commitment to free education since 1945.
Our findings indicate a high prevalence of burnout and mental health issues among medical undergraduates. They need to be used by the Ministry of Higher Education and medical educators to make a positive change in the country’s future medical workforce. Mental health problems are associated with student distress, underachievement, and reduced quality of life (Moir et al., 2018). Physician burnout can lead to loss of job satisfaction, loss of efficiency, and suboptimal patient care as well as being associated with major depression (Chopra et al., 2004). We need to urgently understand what is happening and take measures to address the root causes of the problem to improve our future doctors’ wellbeing.
Greater emphasis is needed to make adjustments to the undergraduate curriculum and focus not just on introducing new areas of learning but consider what to take out of the curriculum to prevent inflation of workload. University teachers need support and training to increase their role as student supporters and mentors (Hughes et al., 2018). Improving mental health literacy and empowerment can improve help-seeking and challenge stigma (Givens & Tjia, 2002). Interventions to improve resilience among students by evidence-based practices such as mindfulness and training to improve crucial interpersonal skills also have the potential to help (Galante et al., 2018). Access to mental health support could be improved by appointing a fulltime mental health advisor to faculties with regular supervision from a consultant psychiatrist (Chandradasa & Champika, 2018; Gale & Thalitaya, 2017).
The study highlights some areas for future research. Effective ways of regularly identifying mental health problems in medical and other undergraduates need to be identified and introduced. Research aimed at identifying the best means of reducing student burnout is a priority. These interventions should ideally be tailormade to address specific challenges at different levels of training including the internship and beyond. Opportunities for healthy lifestyle such as fulfilling relationships, leisure time activities, and physical fitness will always be welcomed by students. A revised goal of medical undergraduate training should be to make healthy doctors who can not only heal their patients but who can also look after themselves.
Footnotes
Conflict of interest
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
