Prevalence of Depression Among Undergraduate Medical Students in India: A Systemic Review and Meta-Analysis

Background: Systematic reviews have proved that there is a high rate of depression among medical students when compared with their age-matched peers. Very few studies have evaluated the pooled prevalence of depression among medical students in India. Objectives: To determine the pooled prevalence of depression among medical students in India. Materials and methods: This review was done by searching databases like PubMed, Google Scholar, and Scopus for available original articles published between 2019 and 2022 on depression among Indian medical (MBBS) undergraduate students using PRISMA guidelines. Results: A total of 19 original research articles were included in this review, involving students at different medical colleges from various regions of India. The pooled prevalence of depression among 5944 medical students was 50.0% (95% CI: (31%-70%)) based on the random effect model. This meta-analysis also found that the pooled prevalence of depression among females (pooled prevalence: 38.0%, 95% CI: 20.0 to 58.0) was slightly higher than among males (pooled prevalence: 34.0%, 95% CI: 15.0 to 55.0). Conclusion: The high prevalence of depression among medical students demands regular screening for depression along with counselling services. It shows that there is a need to raise awareness among students and other stakeholders, such as parents and medical educators, concerning symptoms and signs of depression among medical students.


Introduction And Background
The World Health Organization (WHO) defined health as "a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity", a definition relevant today [1]. Further, it is stated that "not depressed" is not the end goal as there is a spectrum of well-being, with the disease at one end and optimal well-being at the other. The World Health Organization recognizes mental health disorders as important causes of morbidity and disability, with depression as one of the leading causes of mental health disorders.
The COVID-19 pandemic has taken a toll on the mental health of different individuals worldwide for various containment measures and the disease itself [9]. During the COVID-19 pandemic, college students are more prone to mental health disorders along with COVID-19 patients and medical personnel. Compared to other training programmes, medical education has the highest academic and emotional requirements. A systematic review and meta-analysis with research works published globally show depression prevalence of 37.9% among medical students [10].
The pooled prevalence of depression based on standard screening instruments among medical students in India was 40% before the COVID pandemic [11]. However, no studies have been conducted to date evaluating the effect of this healthcare crisis on medical trainees in India during the pandemic. This metaanalysis includes cross-sectional studies on depression among medical students in India to determine the pooled prevalence of depression among medical students in India.

Materials and methods
This study was completed following the PRISMA checklist [12], and it was registered in PROSPERO (CRD42022331012) [13]. Before beginning the literature search, the study's framework was created using PRISMA criteria; after that, no changes were made.

Literature search
A systematic search was undertaken in three databases, namely PubMed, Google Scholar and Scopus, for all the available articles published in the English language during 2019-2021 on the prevalence of depression amongst undergraduate medical students in India by two independent investigators. (Appendices 1) Based on the "PICOS" principle for formulating the search strategy, the search terms include "Depression", "Undergraduate students", "medical students", "undergraduate medical students", and "MBBS students". We have added relevant articles identified by several databases to the search using Rayyan software. It was also attempted to contact the authors of articles whose full texts were unavailable in the databases. Appendices 1 presents a comprehensive search strategy.

Inclusion and exclusion criteria
The inclusion criteria were 1) Cross-sectional study design, 2) The study population was undergraduate medical students from India, 3) Assessing the prevalence of depression using a standardized instrument, and 4) The study period was from January 2019 to April 2022. The articles which did not state the information on the prevalence of depression or outcome data among the undergraduate medical students were excluded.

Study selection and data extraction
The studies were retrieved based on the search strategy discussed a priori. Multiple publications of the same study were identified and collated. Two independent reviewers screened titles and abstracts to identify the studies that meet the inclusion and exclusion criteria. Any disagreements were resolved by discussion or by a third reviewer. Data were extracted using a pre-designed spreadsheet from the studies which included the first author name, year of publication, study period, study setting, sample size, the instrument used for assessing depression, the total number of medical students, number of females, number of males, age (mean) of the student, prevalence of depression overall, and among male students and female students.

Quality assessment/risk of bias (ROB)
Using the Joanna Briggs Institute's (JBI) Critical Appraisal tool for systematic reviews of prevalence studies, we evaluated the study's methodology which was shown in Table 1 [14,15]. The reviewers independently assessed the ROB using the Modified JBI criteria. In case of a mismatch of results, the discrepancies or disputes have been checked, and the reviewers came to a common opinion by discussing it with a third investigator. The evaluated articles were divided into three categories: high ROB (JBI score 49%), moderate ROB (JBI score between 50% and 69%), and low ROB (JBI score >70%) [16].

Data synthesis and analysis
In this meta-analysis, to determine the prevalence of depression among medical students, the pooled estimates with 95% confidence intervals (95% CI) were calculated using the random-effects model (DerSimonian-Laird method) due to high heterogeneity [17]. The forest plots have been used to determine the prevalence of pooled estimates. The estimation was calculated using MetaXL version 5.3 and an Excel spreadsheet. A P-value less than 0.05 was considered significant. The Q and I 2 statistics were used as tests for heterogeneity. Q test with P < 0.10 was considered statistically significant heterogeneity and I 2 >75% was labelled as high heterogeneity [18]. This review used the Doi plot and the LFK index for publication bias. The double arcsine prevalence was taken as the x-axis for the funnel plot. Arcsine transformation was required with data with extreme values like 0 or 1. Otherwise, the confidence interval for the proportion could include values outside the range of 0 and 1, representing 0% and 100%. The y-axis of the graphic was set to the precision obtained from the inverse of Standard Error (SE). Doi plot and the LFK index for publication bias were used to validate the funnel plot's asymmetries. Values describe the publication bias over ±1 of the LFK index [19]. Sensitivity analysis was done to indicate the major determinant for the pooled prevalence of depression.

Study Characteristics
The initial search from PubMed, Scopus, and Google Scholar yielded 3297 studies. Only 56 pertinent papers were reviewed for eligibility after removing duplicates based on the screening of titles and abstracts. Out of 56 papers, 19 articles were included in the quantitative synthesis after various publications were excluded for multiple reasons ( Figure 1).

Sensitivity Analysis
The effect of each study (i.e. nineteen studies)  on the pooled prevalence of depression has been analysed by excluding each study step by step using sensitivity analysis (

Discussion
Although it is of global public health concern, depression among medical students is still under-recognized and the associated discrimination and stigma deter most of the students from seeking help. Research done among medical students even before the pandemic were showing higher rates of depression and physical distress in comparison to the general population [3][4][5][6][7][8][9]. COVID pandemic may have further declined the mental health of all individuals, especially medical students. This is one of the few systematic reviews and meta-analyses on medical trainees which has ventured into exploring the prevalence of depression amidst the COVID pandemic. It has incorporated 19 original articles published from 2019 to 2022 from medical colleges situated in different regions of India. These articles used various standard screening instruments such as Patient Health Questionnaire (PHQ-9), Depression Anxiety Stress Scale (DASS 21), Beck Depression Inventory (BDI) Scale, Centre for Epidemiologic Studies Depression Scale and Hospital Anxiety and Depression Scale (HADS) for screening depression. The pooled prevalence of depression among medical students (n=5944) was found to be 50.0% [95% CI: (31%-70%] based on a random effect model pertaining to high heterogeneity [Q test (p <0.001) and I2 = 99.6%]; the depression calculated in the various studies ranged from 16.0% to 100.0%. This pooled prevalence is higher than the pooled prevalence value of 40%, which was done before the COVID pandemic, as mentioned in a meta-analysis conducted by Dwivedi N et al. [11] In Jia Q et al. study, a meta-analysis done in 41 studies among medical students from different countries showed a high pooled prevalence of depression of 37.9% among the students. Comparatively, this meta-analysis has shown a higher prevalence among Indian medical undergraduates, slightly higher in female students [10]. This variation of the depression may reflect the trend of higher rates of depression in females in the general population [42]. The severity of depression has also been measured in various studies, though it is not uniform throughout the studies as different screening tools had different cut off for measuring depression. The pooled prevalence of depression in this review varied from 27.0% to 99.0% when stratified based on the study tools. Depression Anxiety Stress Scale was the most commonly used scale among the studies, followed by Patient Health Questionnaire and Beck Depression Inventory.
According to this meta-analysis, one out of two students was found to have some degree of depression which is a concerning number. The prevalence has increased during the COVID pandemic compared to the prevalence before the pandemic since the pandemic itself may have a negative impact on the mental health of the students [11]. Further, it was also stated that it would increase the burden of depression in South Asian populations and healthcare systems [2]. There have been multiple explanations behind depression among medical students such as the challenging medical curriculum, lack of sleep and recreational activities due to vast academic tasks given, transition from a familiar to a non-familiar situation, lack of communication skills, etc. Several strategies can be implemented at different levels like changing the grading system, setting clear learning objectives to decrease academic burden, incorporating team-based learning and group activities rather than following traditional didactic lectures, self-directed learning so that students are aware of their shortcomings and promoting professionalism and communication skills for dealing patients properly. This pandemic had led to online teaching of medical students, [43] further, the effects of social distancing and self-isolation could have made the students more vulnerable to depression and anxiety [44]. This global health concern can also be taken care of using evidence-based structured programmes for mental upliftment like mindfulness therapy, life skill training, and counselling sessions can be kept for the students [45,46].

Strengths and limitations
The strengths of this study are that it has no asymmetry in the doi plot and is one of the few meta-analyses which has investigated the pooled prevalence of depression among the undergraduate students in India.
Despite the strengths, this study has a few limitations like high heterogeneity and there might be a possibility of overestimation of the prevalence of depression because pooled prevalence has been calculated based on screening instruments and in various studies which are self-reporting and does not involve any clinical diagnosis.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.