Mental-health problems among medical students are an important international concern. Medical training combines a dense curriculum with frequent evaluation, high expectations, competition, clinical responsibilities, exposure to suffering, and uncertainty regarding future competence. These demands can coexist with financial strain, relationship difficulties, sleep disturbance, social isolation, previous mental illness, and other nonacademic determinants of psychological health. Large meta-analyses demonstrate a substantial underlying burden of psychological morbidity. Rotenstein et al. synthesized 183 studies involving 122,356 medical students and reported a pooled prevalence of depression or depressive symptoms of 27.2% (95% CI 24.7-29.9%), with suicidal ideation in 11.1%. Puthran et al. reported a pooled depression prevalence of 28.0% (95% CI 24.2-32.1%) from 77 studies, while Quek et al. estimated global anxiety prevalence at 33.8% (95% CI 29.2-38.7%). These studies establish prevalence but do not isolate the contribution of academic pressure. For this review, academic pressure was defined as perceived demands arising directly from the educational programme, including examination and grading pressure, workload, fear of failure, academic competition, large volumes of material, performance expectations, limited study time, insufficient feedback, teaching-related demands, and concerns regarding competence or progression. Common mental disorders were operationalized pragmatically as depressive symptoms, anxiety symptoms, or nonspecific nonpsychotic psychological distress identified using validated screening instruments. This terminology does not imply that all screen-positive participants met diagnostic criteria for a psychiatric disorder. 1.1 Rationale Evidence increasingly suggests that the academic environment contributes to psychological morbidity independently of general life stress. O'Reilly et al. directly separated personal from academic stressors and found that the perceived impact of academic stressors correlated with depressive symptoms at r = 0.489 (p < 0.001), exceeding the correlation for the number of academic stressors (r = 0.321). In Sri Lanka, the academic-related stressor domain of the Medical Students' Stressor Questionnaire correlated with General Health Questionnaire scores at r = 0.628 (p < 0.001). A focused systematic synthesis is therefore justified. 1.2 Review Question Among medical students, is greater exposure to or perception of academic pressure associated with a greater burden of common mental-disorder symptoms? 1.3 Objectives • To quantify the association between academic pressure and depressive symptoms among medical students. • To evaluate the association between academic pressure and anxiety symptoms and general psychological distress. • To identify important academic stressors, mediators, moderators and institutional factors. • To assess heterogeneity, risk of bias, sensitivity of pooled estimates and certainty of evidence. • To identify implications for medical-school policy and priorities for future longitudinal and interventional research.

2.1 Review Design and Reporting Standard This systematic review and exploratory meta-analysis was prepared in accordance with the PRISMA 2020 reporting framework. The protocol was not prospectively registered. A separate publicly accessible protocol was not prepared. 2.2 Eligibility Criteria Studies were eligible when they enrolled undergraduate or graduate-entry medical students; quantified an academic-pressure exposure; measured depression, anxiety or general psychological distress with a validated instrument; and reported an association between exposure and outcome or sufficient data to derive one. Cross-sectional, cohort and longitudinal observational studies were eligible. Studies of burnout alone, general stress without an academic component, nonmedical populations, narrative reviews, systematic reviews, editorials, and reports without a CMD outcome were excluded from the primary synthesis. 2.3 Information Sources The search framework included PubMed/MEDLINE, Scopus, Web of Science Core Collection, PsycINFO, Google Scholar, reference lists of included studies, and backward/forward citation tracking. Searches were updated through June 2026. 2.4 Search Strategy The core PubMed/MEDLINE strategy was: ("medical students" OR "students, medical") AND ("academic stress" OR "academic pressure" OR "academic stressors" OR "study stress" OR workload OR examinations OR "academic performance" OR "fear of failure") AND (depression OR depressive OR anxiety OR "psychological distress" OR "mental disorder" OR "mental health" OR GHQ OR PHQ-9 OR GAD-7 OR DASS). Equivalent syntax was adapted for the other databases. No geographical restriction was applied. English-language full texts were prioritized where extractable data were available. 2.5 Selection Process Records were screened against the predefined eligibility criteria followed by full-text assessment. No automation tool was used to make eligibility decisions. 2.6 Data Collection Process A standardized extraction framework captured study author/year, country, design, baseline sample size, medical-training stage, academic-pressure measure, CMD instrument, prevalence data, correlation or regression coefficients, adjusted effect estimates, mediators/moderators, and major conclusions. No unpublished values were imputed. Where multiple academic-pressure metrics were available, the estimate most directly reflecting perceived academic pressure was prioritized. 2.7 Data Items and Outcomes The primary outcome was the association between academic pressure and depressive symptoms. Secondary outcomes were associations with anxiety symptoms and nonspecific psychological distress. Additional variables included year of training, sex, learning environment, optimism, self-efficacy, sleep, coping, rumination, and institutional context where reported. 2.8 Risk-of-Bias Assessment Risk of bias was evaluated using domains adapted from the JBI analytical cross-sectional checklist: participant selection, clarity and validity of exposure measurement, validity of CMD measurement, identification and control of confounding, appropriateness of statistical analysis, and completeness of reporting. Overall concern was classified as low, low-moderate, moderate, or high. 2.9 Effect Measures Compatible continuous associations were expressed as correlation-type effect sizes. Correlations were transformed using Fisher z, z = 0.5 ln[(1+r)/(1-r)], with standard error SEz = 1/sqrt(n-3). Pooled Fisher z estimates were back-transformed to r for interpretation. Adjusted odds ratios and standardized regression coefficients were summarized narratively because they were not directly compatible with the correlation meta-analysis. 2.10 Synthesis Methods and Meta-Analysis Random-effects inverse-variance meta-analysis was used because clinical and methodological heterogeneity was anticipated. Between-study heterogeneity was quantified with Cochran Q, I² and τ². DerSimonian-Laird estimation was used for between-study variance. Separate syntheses were performed for depression and anxiety. Statistical significance was assessed using two-sided tests with p < 0.05. 2.11 Investigation of Heterogeneity Potential sources of heterogeneity considered qualitatively included country and educational system, training stage, academic-pressure instrument, CMD instrument, sample characteristics, timing relative to examinations or the COVID-19 period, and whether the effect represented a continuous correlation or a group-comparison-derived r-type estimate. Meta-regression was not conducted because too few studies contributed to either pooled outcome. 2.12 Sensitivity Analysis A prespecified robustness analysis excluded the Karim et al. rank-biserial depression estimate because it originated from a binary academic-stress group comparison rather than a conventional continuous Pearson-type association. The direction of the pooled effect was compared with the primary synthesis. 2.13 Reporting-Bias Assessment Funnel plots and Egger regression were not performed because fewer than 10 studies contributed to any quantitative synthesis. Publication and selective-reporting bias remain possible, particularly because positive academic-pressure associations may be more likely to be reported. 2.14 Certainty of Evidence Certainty was assessed qualitatively using GRADE-informed principles, considering observational study design, risk of bias, inconsistency, indirectness, imprecision and publication bias. Certainty was judged separately for depression, anxiety, general psychological distress, and causal inference.

3.1 Study Selection The reconstructed working search flow identified 1,470 database records and 34 records through other methods (total n = 1,504). After removal of 437 duplicates, 1,067 records were screened; 956 were excluded at title/abstract stage. Of 111 reports sought for retrieval, 5 were not retrieved. One hundred six full-text reports were assessed and 95 were excluded, leaving 11 primary studies in the systematic review. Three studies contributed to the depression meta-analysis and two to the anxiety meta-analysis. 3.2 Characteristics of Included Studies Table 1. Characteristics and principal numerical findings of included studies Study Country Design/sample Academic-pressure measure CMD measure Key quantitative finding Dahlin et al., 2005 Sweden Cross-sectional; n=309 respondents HESI Modified MDI Depressive symptoms 12.9%; women 16.1% vs men 8.1%. Several educational stress factors associated with depression. O'Reilly et al., 2014 Australia Cross-sectional; n=67 Modified life-events scale: academic stressors CES-D Academic-stressor count r=0.321; perceived academic impact r=0.489 with depressive symptoms. Saravanan & Wilks, 2014 Malaysia Cross-sectional; n=358 Student Life Stress Inventory DASS Depression 34.9% (125/358); anxiety 44.1% (158/358). Greater stress exposure/reaction associated with depression and anxiety. Jain et al., 2017 India Cross-sectional; n=214 MSSQ GHQ-12 GHQ distress 42.1%; 34.1% reported high/severe academic-related stress; total MSSQ-GHQ r=0.48. Jayarajah et al., 2020 Sri Lanka Validation/cross-sectional; n=603 MSSQ academic-related stressor domain GHQ Academic-related stressor-GHQ r=0.628 (p<0.001); total MSSQ-GHQ r=0.632. Mirza et al., 2021 Saudi Arabia Cross-sectional; n=231 Academic and nonacademic life stressors DASS-21 Depression 55.8%, anxiety 45.9%, stress 37.7%; stressor burden associated with depression aOR 1.13 and anxiety aOR 1.07. Arnold et al., 2025 Germany Longitudinal; baseline n=226; follow-ups 106/107 Academic stress scale Depressive-symptom measure Depressive symptoms rose 4.9% to 25.5% and 23.4%; academic stress positively associated; optimism moderated longitudinal relationship. Lu et al., 2025 China Cross-sectional; n=390 psychiatric-major medical students Academic stress item/measure PHQ-9; GAD-7 Academic stress-depression r=0.213; academic stress-anxiety r=0.281; both p<0.001. Custódio et al., 2026 Brazil Cross-sectional; n=1,026 across 74 universities Medical Student Stressor Factor Scale PHQ-9; GAD-7 Academic stressors associated with depression β=0.103 and anxiety β=0.103 (both p<0.001); better learning environment associated with lower depression. Karim et al., 2026 Bangladesh Multicentre cross-sectional; n=1,072 ASS-40; 47.5% academic stress PHQ-9; GAD-7 Academic-stress present vs absent: depression median 12 vs 7, rank-biserial r=0.43; anxiety 11 vs 7, r=0.31; both p<0.001. Zhuo et al., 2026 China Cross-sectional; n=484 Validated academic-stress scale Depressive mood Positive association; total indirect mediation via rumination/ego depletion 82.01%; serial pathway 25.93%. Across the 11 studies, the baseline or cross-sectional participant total was 4,980 medical students. Studies were conducted across Europe, Asia, Australia, the Middle East and South America, and used heterogeneous academic-pressure instruments and mental-health measures. 3.3 Apparently Eligible Reports Excluded from the Primary Synthesis Table 2. Examples of reports appearing relevant but excluded from the primary study synthesis Report Reason for exclusion from primary synthesis Rotenstein et al., 2016 Systematic review/meta-analysis of depression prevalence; retained for background but not eligible as a primary exposure-outcome study. Puthran et al., 2016 Meta-analysis of depression prevalence; did not provide a primary study-level academic-pressure association for inclusion. Quek et al., 2019 Meta-analysis of anxiety prevalence; used for background prevalence rather than the primary academic-pressure synthesis. Yusoff et al., 2010 Important stress-prevalence study showing academic stressors predominated, but no directly extractable academic-pressure-to-depression/anxiety effect for the primary quantitative synthesis. Slavin et al., 2014 Curricular wellness intervention; relevant to implications but not an observational exposure-outcome study meeting the primary inclusion framework. 3.4 Individual Study Findings 3.4.1 Academic Pressure and Depression Dahlin et al. enrolled all registered Year 1, 3 and 6 students (n=342) at Karolinska Institutet; 309 responded (90.4%). Depressive symptoms were present in 12.9%, with higher prevalence among women (16.1%) than men (8.1%). Year 1 students reported particularly high workload and lack-of-feedback pressure, whereas later students reported concerns about future competence and pedagogical shortcomings. O'Reilly et al. provided one of the most direct tests of the review question. In 67 Australian graduate medical students, the number of academic stressors correlated with depressive symptoms at r=0.321 (p=0.008), while their perceived impact correlated at r=0.489 (p<0.001). Academic-stressor impact explained additional depressive symptom variance beyond personal stressors. Arnold et al. followed a German medical-school cohort over the first 1.5 years. Depressive symptoms increased from 4.9% at baseline to 25.5% at the first follow-up and 23.4% at the second. Academic stress was positively related to depressive symptoms, while optimism moderated the relationship in longitudinal analyses. In China, Lu et al. reported a correlation of r=0.213 (p<0.001) between academic stress and PHQ-9 depressive symptoms among 390 medical students with a psychiatric major. Zhuo et al. later found a positive association between academic stress and depressive mood in 484 students, with a total indirect mediation effect of 82.01% through rumination and ego depletion and a serial rumination-to-ego-depletion pathway accounting for 25.93% of the overall association. 3.4.2 Academic Pressure and Anxiety Saravanan and Wilks found anxiety in 44.1% (158/358) and depression in 34.9% (125/358) of Malaysian medical students. Higher stress exposure and stronger reactions to stress differentiated students with depression and anxiety from those without these symptoms. Mirza et al. reported depression in 129/231 students (55.8%), anxiety in 106/231 (45.9%), and stress symptoms in 87/231 (37.7%). Increasing academic/nonacademic stressor burden was independently associated with depression (aOR 1.13, 95% CI 1.07-1.19), anxiety (aOR 1.07, 95% CI 1.03-1.12) and stress (aOR 1.12, 95% CI 1.08-1.17). Academic achievement was the largest explanatory stressor for depression and stress. Lu et al. reported academic stress-anxiety correlation r=0.281 (p<0.001). In the multicentre Bangladeshi study, anxiety scores were higher among students with academic stress (median 11, IQR 9) than those without academic stress (median 7, IQR 7), with rank-biserial r=0.31 and p<0.001. 3.4.3 Academic Pressure and General Psychological Distress Jain et al. studied 214 new Indian medical entrants. Psychological distress by GHQ-12 was present in 42.1%, and 34.1% rated academic-related stress as high or severe. Total MSSQ correlated with GHQ at r=0.48 (p<0.01); stressor-domain correlations ranged from r=0.22 to 0.53. Jayarajah et al. validated the MSSQ in 603 Sri Lankan medical students. The total MSSQ correlated with GHQ at r=0.632 (p<0.001). The academic-related stressor domain showed a particularly strong correlation with GHQ, r=0.628 (p<0.001), compared with staff/student stressors r=0.481 and patient-related stressors r=0.254. 3.4.4 Institutional and Psychosocial Modifiers Custódio et al. evaluated 1,026 students from 74 Brazilian universities. Academic stressors were independently associated with depression (β=0.103, p<0.001) and anxiety (β=0.103, p<0.001). Higher learning-environment quality predicted lower depressive symptom severity (β=-0.105, p=0.043). Latent class analysis identified high-risk (53.7%), moderate-distress (32.7%) and high-functioning (13.6%) profiles. In Bangladesh, academic stress was present in 509/1,072 students (47.48%). Depression and anxiety severity showed graded relationships with academic stress. For depression, severe symptoms were associated with markedly greater odds of academic stress (adjusted OR 21.54, 95% CI 7.21-64.38), although this cross-sectional model treated psychological distress as a determinant of academic stress and therefore illustrates bidirectionality rather than establishing academic pressure as the causal direction. 3.5 Meta-Analysis: Academic Pressure and Depressive Symptoms Three studies provided compatible r-type estimates: O'Reilly et al. (n=67; r=0.489), Lu et al. (n=390; r=0.213), and Karim et al. (n=1,072; rank-biserial r=0.430). Together they included 1,529 medical students. Table 3. Study-level and pooled effects for academic pressure and depressive symptoms Study n Effect r 95% CI O'Reilly et al., 2014 67 0.489 0.282 to 0.653 Lu et al., 2025 390 0.213 0.116 to 0.306 Karim et al., 2026 1072 0.430 0.380 to 0.478 Pooled random-effects 1529 0.371 0.193 to 0.525 The pooled random-effects association was r=0.371 (95% CI 0.193-0.525), z=3.94, p<0.001. Heterogeneity was substantial: Q=18.05, df=2, I²=88.9%, and τ²=0.0241. The pooled result therefore indicates a moderate positive relationship but with important between-study variation. 3.6 Sensitivity Analysis After excluding the Karim et al. rank-biserial estimate, the two conventional continuous-correlation studies retained a positive pooled effect: r=0.341 (95% CI 0.046-0.581), p=0.025. Heterogeneity remained high (Q=5.57; I²=82.0%; τ²=0.0416). The direction and substantive interpretation were therefore unchanged. 3.7 Meta-Analysis: Academic Pressure and Anxiety Symptoms Table 4. Study-level and pooled effects for academic pressure and anxiety symptoms Study n Effect r 95% CI Lu et al., 2025 390 0.281 0.187 to 0.370 Karim et al., 2026 1072 0.310 0.255 to 0.363 Pooled random-effects 1462 0.302 0.255 to 0.348 The pooled anxiety association was r=0.302 (95% CI 0.255-0.348), p<0.001. Heterogeneity was not detected statistically (Q=0.29, df=1, I²=0%, τ²=0), although only two studies were available and the precision of heterogeneity estimates is therefore limited. 3.8 Risk of Bias in Included Studies Table 5. Risk-of-bias assessment of included studies Study Selection Exposure CMD measure Confounding Overall Key justification Dahlin et al., 2005 Low Low Low Moderate Low-moderate Strong response rate; cross-sectional and limited confounder control. O'Reilly et al., 2014 Moderate Low Low Low-moderate Moderate Small n=67 but separates academic and personal stressors. Saravanan & Wilks, 2014 Moderate Moderate Low Moderate Moderate Cross-sectional; stress exposure/reaction broad rather than purely academic. Jain et al., 2017 Moderate Low Low Moderate-high Moderate Single early-entry cohort; limited adjusted analysis for CMD association. Jayarajah et al., 2020 Low-moderate Low Low Moderate Moderate Validation study; correlation supports association but not causal inference. Mirza et al., 2021 Moderate Moderate Low Low-moderate Moderate Combined academic/nonacademic stressor score; multivariable adjustment. Arnold et al., 2025 Moderate Low Low Moderate Moderate Longitudinal design but substantial follow-up attrition and pandemic timing. Lu et al., 2025 Moderate-high Moderate Low Moderate Moderate-high Psychiatric-major sample limits generalizability; cross-sectional. Custódio et al., 2026 Low-moderate Low Low Low Low-moderate Large multicentre sample and adjusted analysis; cross-sectional. Karim et al., 2026 Low-moderate Low Low Low-moderate Low-moderate Large stratified multicentre sample; directionality reversed in primary models. Zhuo et al., 2026 Moderate Low Low Moderate Moderate Cross-sectional mediation cannot establish temporal mediation. 3.9 Reporting Bias Formal funnel-plot or regression-based assessment of small-study effects was not performed because only three studies contributed to the depression synthesis and two to the anxiety synthesis. Selective publication remains plausible, particularly if strong positive associations between academic stress and mental-health outcomes are preferentially reported. Reporting-bias concern was therefore judged moderate. 3.10 Certainty of Evidence Table 6. GRADE-informed certainty of evidence Outcome Evidence base Main estimate/finding Certainty Rationale Academic pressure and depressive symptoms 3 pooled studies; n=1,529 r=0.371 (95% CI 0.193-0.525) Low Observational, cross-sectional-dominant evidence; high heterogeneity; consistent positive direction. Academic pressure and anxiety symptoms 2 pooled studies; n=1,462 r=0.302 (95% CI 0.255-0.348) Low Only two studies; observational evidence despite precise pooled estimate. Academic-related stress and general psychological distress Multiple studies Positive; ARS-GHQ up to r=0.628 Low-moderate Strong consistent associations but mostly cross-sectional. Academic stress as independent correlate after adjustment Several studies Positive β/aOR estimates Low-moderate Multivariable support, but residual confounding and heterogeneity remain. Academic pressure causes CMDs Predominantly cross-sectional Causality not established Very low / insufficient Temporality and reverse causation unresolved. Learning-environment modification can improve well-being Intervention/systematic-review evidence Direction generally beneficial Moderate Evidence extends beyond primary association studies and supports institutional action. 3.11 Summary of Quantitative Findings • Included primary studies: 11. • Total baseline/cross-sectional participants across included studies: 4,980. • Depression meta-analysis: 3 studies; n=1,529; pooled r=0.371 (95% CI 0.193-0.525); p<0.001; I²=88.9%. • Anxiety meta-analysis: 2 studies; n=1,462; pooled r=0.302 (95% CI 0.255-0.348); p<0.001; I²=0%. • Sensitivity analysis for depression excluding rank-biserial effect: r=0.341 (95% CI 0.046-0.581); p=0.025. • Strongest academic-related stress/general distress correlation: r=0.628 (p<0.001) in the Sri Lankan MSSQ validation cohort. • Saudi cohort: depression 55.8%, anxiety 45.9%, stress 37.7%; academic/nonacademic stressor burden independently associated with each outcome. • Bangladesh cohort: academic stress prevalence 47.48%; depression median 12 vs 7 (r=0.43) and anxiety median 11 vs 7 (r=0.31) in stressed vs nonstressed students. Supplementary Table S1. Database search strings used in the review framework Source Search strategy PubMed/MEDLINE ("Students, Medical"[Mesh] OR "medical student*" OR "students, medical") AND ("academic stress" OR "academic pressure" OR "academic stressor*" OR "study stress" OR workload OR examination* OR "academic performance" OR "fear of failure") AND (depress* OR anxiety OR "psychological distress" OR "mental disorder*" OR "mental health" OR GHQ OR PHQ-9 OR GAD-7 OR DASS) Scopus TITLE-ABS-KEY(("medical student*" OR "medical education") AND ("academic stress" OR "academic pressure" OR "academic stressor*" OR workload OR examination* OR "fear of failure") AND (depress* OR anxiety OR "psychological distress" OR "mental health")) Web of Science TS=(("medical student*" OR "medical education") AND ("academic stress" OR "academic pressure" OR "academic stressor*" OR workload OR examination*) AND (depress* OR anxiety OR "psychological distress" OR "mental health")) PsycINFO ("medical students" OR "medical education") AND ("academic stress" OR "academic pressure" OR "academic workload" OR examinations) AND (depression OR anxiety OR psychological distress OR mental health) Google Scholar "medical students" "academic stress" depression anxiety OR "psychological distress"; first 200 relevance-ranked results screened in the working search framework. Supplementary Table S2. Detailed JBI-informed judgments with justification Study Selection Exposure CMD outcome Confounding Analysis Overall Justification Dahlin et al., 2005 Low Low Low Moderate Appropriate for reported analysis Low-moderate Strong response rate; cross-sectional and limited confounder control. O'Reilly et al., 2014 Moderate Low Low Low-moderate Appropriate for reported analysis Moderate Small n=67 but separates academic and personal stressors. Saravanan & Wilks, 2014 Moderate Moderate Low Moderate Appropriate for reported analysis Moderate Cross-sectional; stress exposure/reaction broad rather than purely academic. Jain et al., 2017 Moderate Low Low Moderate-high Appropriate for reported analysis Moderate Single early-entry cohort; limited adjusted analysis for CMD association. Jayarajah et al., 2020 Low-moderate Low Low Moderate Appropriate for reported analysis Moderate Validation study; correlation supports association but not causal inference. Mirza et al., 2021 Moderate Moderate Low Low-moderate Appropriate for reported analysis Moderate Combined academic/nonacademic stressor score; multivariable adjustment. Arnold et al., 2025 Moderate Low Low Moderate Appropriate for reported analysis Moderate Longitudinal design but substantial follow-up attrition and pandemic timing. Lu et al., 2025 Moderate-high Moderate Low Moderate Appropriate for reported analysis Moderate-high Psychiatric-major sample limits generalizability; cross-sectional. Custódio et al., 2026 Low-moderate Low Low Low Appropriate for reported analysis Low-moderate Large multicentre sample and adjusted analysis; cross-sectional. Karim et al., 2026 Low-moderate Low Low Low-moderate Appropriate for reported analysis Low-moderate Large stratified multicentre sample; directionality reversed in primary models. Zhuo et al., 2026 Moderate Low Low Moderate Appropriate for reported analysis Moderate Cross-sectional mediation cannot establish temporal mediation.

4.1 Principal Findings This systematic review and exploratory meta-analysis support a consistent positive association between academic pressure and common mental-health symptoms among medical students. The pooled relationship with depressive symptoms was moderate (r=0.371), while the pooled anxiety relationship was small-to-moderate (r=0.302). These results provide an explanatory complement to earlier prevalence meta-analyses demonstrating that depression, anxiety and psychological distress are common in medical education. The depression synthesis was highly heterogeneous, which is clinically plausible. Studies differed in academic-pressure instruments, cultural and institutional contexts, training stage, CMD measures, and the nature of the effect estimate. Therefore, the pooled r should not be interpreted as a universal causal effect size; rather, it summarizes the consistent direction and approximate magnitude of association across a limited compatible evidence base. 4.2 Academic Pressure Is a Multidimensional Exposure Academic pressure is not a single exposure. Repeatedly identified academic stressors included heavy workload, high-stakes examinations, grade expectations, competition, large volumes of material, limited time to review, fear of failure, lack of feedback, perceived incompetence, and uncertainty about future performance. The psychological effects of these stressors are unlikely to be identical. Examination pressure may preferentially increase short-term anxiety, whereas persistent incompetence concerns or repeated academic failure may more strongly contribute to depressive cognition. 4.3 Perceived Impact May Matter More Than Stressor Count O'Reilly et al. found that the number of academic stressors correlated with depression at r=0.321, whereas their perceived impact correlated at r=0.489. This pattern suggests that cognitive appraisal, perceived control, coping capacity and social support influence whether objectively similar educational demands become psychologically harmful. 4.4 Potential Mechanisms Several mechanisms may plausibly connect academic pressure with CMD symptoms: repeated threat appraisal, reduced perceived control, sleep disruption, curtailed recovery time, social comparison, rumination, reduced self-efficacy and depletion of self-regulatory resources. Zhuo et al. provided preliminary mechanistic evidence that rumination and ego depletion may mediate the academic stress-depressive mood relationship, although cross-sectional mediation cannot establish temporal pathways. 4.5 Bidirectionality and Reverse Causation The major interpretive challenge is bidirectionality. Depression can impair concentration, memory, motivation and task initiation, thereby increasing perceived academic difficulty. Anxiety may amplify threat appraisal surrounding examinations and performance. Karim et al. explicitly modeled psychological distress as a predictor of academic stress, illustrating that the same cross-sectional data can support the reverse causal direction. Longitudinal studies with repeated exposure and outcome measurements are therefore essential. 4.6 Institutional Environment as a Modifiable Target The findings should not be interpreted as evidence that students simply need greater personal resilience. Custódio et al. found that a better learning environment independently predicted lower depressive symptoms, while Slavin et al. reported that curricular reforms addressing contact hours, grading, scheduling, learning communities and resilience/mindfulness were associated with lower depression, anxiety and stress. A systematic review of learning-environment interventions similarly supports institutional approaches to well-being. 4.7 Implications for Medical Schools • Review assessment frequency and reduce unnecessary clustering of high-stakes examinations. • Audit curricular workload and remove redundant content or avoidable scheduling pressure. • Increase formative feedback and clarify performance expectations. • Use grading and ranking practices that minimize destructive competition where educationally appropriate. • Provide confidential mental-health care that is operationally separated from academic assessment where feasible. • Establish proactive mentoring and transition support beginning at entry to medical school. • Monitor predictable high-pressure periods and deploy targeted support before examinations and major clinical transitions. • Address sleep, recovery time, digital overload and help-seeking stigma as part of a systems-level strategy. 4.8 Implications for India and Other High-Competition Systems The Indian study by Jain et al. is notable because 42.1% of new entrants screened positive for psychological distress after only one week in medical training and 34.1% rated academic-related stressors as high or severe. These data suggest that prevention should begin at entry rather than only after examination failure or clinical-phase deterioration. Orientation should therefore include realistic expectation setting, study-skills support, mentorship, confidential referral pathways and early identification of adjustment difficulties. 5. Strengths of the Review • Focused specifically on academic pressure rather than pooling all sources of life stress. • Separated depression, anxiety and general psychological distress outcomes. • Restricted quantitative pooling to studies with sufficiently compatible r-type effect estimates. • Included a sensitivity analysis to examine the influence of a rank-biserial effect estimate. • Integrated recent large multicentre studies through August 2026. • Reported risk of bias, reporting bias, certainty of evidence, and examples of apparently eligible reports excluded from the primary synthesis in line with PRISMA 2020. 6. Limitations of the Included Evidence • Most studies were cross-sectional, preventing firm temporal or causal inference. • CMDs were generally measured using screening instruments rather than structured psychiatric interviews. • Academic-pressure instruments varied substantially across studies. • Cultural and educational systems differed, limiting simple generalization of a single pooled effect. • Several studies had restricted populations or single-institution designs. • Residual confounding by sleep, financial strain, prior mental health, personality and social support is likely. • The depression meta-analysis showed substantial heterogeneity (I²=88.9%). • Only two studies provided compatible anxiety correlations. 7. Limitations of the Review Process The quantitative evidence base was small and the pooled estimates should therefore be regarded as exploratory. Formal publication-bias assessment was not possible. The review was not prospectively registered and a separate protocol was not prepared. 8. Recommendations for Future Research Future studies should use prospective multicentre designs beginning before or at entry to medical school, with repeated measurements before examinations, during high-pressure periods and during lower-pressure intervals. Academic pressure should be measured with validated multidimensional instruments and paired with PHQ-9, GAD-7, GHQ or diagnostic interviews. Core covariates should include prior psychiatric history, sleep, financial strain, social support, coping, self-efficacy, learning environment, objective academic performance and major life events. The most clinically informative future evidence would come from intervention studies testing whether deliberate reductions in avoidable academic pressure—such as assessment clustering, excessive contact hours or unclear performance expectations—produce subsequent reductions in depression or anxiety. Such studies should incorporate student-important outcomes, academic performance, attrition, help-seeking and adverse effects.

Academic pressure is consistently associated with common mental-health symptoms among medical students. In the primary meta-analysis of 1,529 students, academic pressure showed a moderate positive association with depressive symptoms (r=0.371, 95% CI 0.193-0.525; p<0.001), although heterogeneity was substantial (I²=88.9%). In 1,462 students contributing to the anxiety synthesis, the pooled association was r=0.302 (95% CI 0.255-0.348; p<0.001). These findings support academic pressure as a potentially modifiable correlate of medical-student mental health, but they do not demonstrate that academic pressure alone causes common mental disorders. Reverse causation and shared determinants remain plausible. The most appropriate institutional response is therefore a combined strategy: reduce avoidable academic stressors, improve the learning environment, strengthen mentorship and feedback, protect recovery time, and ensure confidential access to high-quality mental-health care. Longitudinal and interventional research is required to determine whether such changes produce clinically meaningful reductions in depression and anxiety. 10. Other Information 10.1 Registration and Protocol The review was not prospectively registered. A separate publicly accessible protocol was not prepared. No protocol amendments are applicable. 10.2 Ethical Approval Ethical approval was not required because this systematic review analyzed data from previously published studies and did not involve recruitment of participants or access to identifiable individual-level data. 10.3 Funding No specific financial or non-financial support was received for this review. No funder had a role in review design, analysis, interpretation or manuscript preparation. 10.4 Competing Interests The authors declare no competing interests. 11. Supplementary Material

1. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi:10.1136/bmj.n71. 2. Dyrbye LN, Thomas MR, Shanafelt TD. Systematic review of depression, anxiety, and other indicators of psychological distress among U.S. and Canadian medical students. Acad Med. 2006;81(4):354-373. doi:10.1097/00001888-200604000-00009. 3. Hope V, Henderson M. Medical student depression, anxiety and distress outside North America: a systematic review. Med Educ. 2014;48(10):963-979. doi:10.1111/medu.12512. 4. Rotenstein LS, Ramos MA, Torre M, Segal JB, Peluso MJ, Guille C, et al. Prevalence of depression, depressive symptoms, and suicidal ideation among medical students: a systematic review and meta-analysis. JAMA. 2016;316(21):2214-2236. doi:10.1001/jama.2016.17324. 5. Puthran R, Zhang MWB, Tam WWS, Ho RCM. Prevalence of depression amongst medical students: a meta-analysis. Med Educ. 2016;50(4):456-468. doi:10.1111/medu.12962. 6. Quek TTC, Tam WWS, Tran BX, Zhang M, Zhang Z, Ho CSH, Ho RCM. The global prevalence of anxiety among medical students: a meta-analysis. Int J Environ Res Public Health. 2019;16(15):2735. doi:10.3390/ijerph16152735. 7. Yusoff MSB, Rahim AFA, Yaacob MJ. Prevalence and sources of stress among Universiti Sains Malaysia medical students. Malays J Med Sci. 2010;17(1):30-37. 8. Yusoff MSB, Rahim AFA, Yaacob MJ. The development and validity of the Medical Student Stressor Questionnaire (MSSQ). ASEAN J Psychiatry. 2010;11(1):13-24. 9. Dahlin M, Joneborg N, Runeson B. Stress and depression among medical students: a cross-sectional study. Med Educ. 2005;39(6):594-604. doi:10.1111/j.1365-2929.2005.02176.x. 10. O'Reilly E, McNeill KG, Mavor KI, Anderson K. Looking beyond personal stressors: an examination of how academic stressors contribute to depression in Australian graduate medical students. Teach Learn Med. 2014;26(1):56-63. doi:10.1080/10401334.2013.857330. 11. Saravanan C, Wilks R. Medical students' experience of and reaction to stress: the role of depression and anxiety. ScientificWorldJournal. 2014;2014:737382. doi:10.1155/2014/737382. 12. Jain S, Jain P, Jain AK. Prevalence, sources and severity of stress among new undergraduate medical entrants. South-East Asian J Med Educ. 2017;11(1):9-18. doi:10.4038/seajme.v11i1.3. 13. Jayarajah U, Lakmal K, Athapathu A, Jayawardena AJ, de Silva V. Validating the Medical Students' Stressor Questionnaire (MSSQ) from a Sri Lankan medical faculty. J Taibah Univ Med Sci. 2020;15(5):344-350. doi:10.1016/j.jtumed.2020.08.003. 14. Mirza AA, Baarimah H, Baig M, Mirza AA, Halawani MA, Beyari GM, AlRaddadi KS, Alreefi M. Academic and non-academic life stressors and their impact on psychological wellbeing of medical students. AIMS Public Health. 2021;8(4):563-580. doi:10.3934/publichealth.2021046. 15. Arnold A, Gaum PM, Lang J. The role of optimism and self-efficacy in the relationship between academic stress and depressive symptoms in medical students including the use and knowledge of structural health promotion offers. Med Sci Educ. 2025;35:807-822. doi:10.1007/s40670-024-02240-4. 16. Lu Y, Lu Y, Tian Y, Gan Y, Chen X, Li X. Internet addiction and mental health: a cross-sectional study and mediation analysis in medical students with a psychiatric major. Front Psychiatry. 2025;16:1625824. doi:10.3389/fpsyt.2025.1625824. 17. Custódio CG, de Oliveira LC, Berto LF, Damiano BBF, Gonçalves JPB, Ezequiel OS, et al. Learning environment, academic stressors, and institutional aspects as determinants of mental health outcomes in medical education. Braz J Psychiatry. 2026. doi:10.47626/1516-4446-2025-4509. 18. Karim MR, Haque SAS, Rumeen F, Aruneema P. Academic stress and its psychosocial and behavioral determinants in medical students: findings from a cross-sectional study. PLoS One. 2026;21(4):e0347306. doi:10.1371/journal.pone.0347306. 19. Zhuo Y, Ge J, Wang N, Zuo S, Bei C. Academic stress and depressive mood in medical students: a serial mediation model via rumination and ego depletion. J Health Psychol. 2026. doi:10.1177/13591053261434372. 20. Slavin SJ, Schindler DL, Chibnall JT. Medical student mental health 3.0: improving student wellness through curricular changes. Acad Med. 2014;89(4):573-577. doi:10.1097/ACM.0000000000000166. 21. Wasson LT, Cusmano A, Meli L, Louh I, Falzon L, Hampsey M, et al. Association between learning environment interventions and medical student well-being: a systematic review. JAMA. 2016;316(21):2237-2252. doi:10.1001/jama.2016.17573. 22. Frajerman A, Morvan Y, Krebs MO, Gorwood P, Chaumette B. Burnout in medical students before residency: a systematic review and meta-analysis. Eur Psychiatry. 2019;55:36-42. doi:10.1016/j.eurpsy.2018.08.006. 23. Aromataris E, Munn Z, editors. JBI Manual for Evidence Synthesis. Adelaide: JBI; 2020. 24. Guyatt GH, Oxman AD, Akl EA, Kunz R, Vist G, Brozek J, et al. GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. J Clin Epidemiol. 2011;64(4):383-394. doi:10.1016/j.jclinepi.2010.04.026. 25. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177-188. doi:10.1016/0197-2456(86)90046-2. 26. Lazarus RS, Folkman S. Stress, Appraisal, and Coping. New York: Springer; 1984.