Abstract
OBJECTIVE: Academic stress has been increasingly identified as a possible risk factor for suicidal behavior among students at all levels of education and in international settings. This systematic review and meta-analysis was conducted to synthesize the existing evidence for the association between academic stress and suicidal ideation and behavior. METHOD: A systematic search was performed across six databases, viz., PubMed, Scopus, PsycINFO, Web of Science, CINAHL, and the Cochrane Library, from inception until May 2025. The criteria for study inclusion necessitated them to quantitatively measure academic stress and suicide-related outcomes, viz., suicidal ideation and suicide attempts, among adolescent or young adult student groups. Data extraction involved demographic details, measurement tools for stress and suicidality, statistical findings, and covariate adjustments. The AXIS tool was applied to evaluate the risk of bias. Meta-analysis was carried out using RevMan 5.4.1, employing a random-effects model to estimate pooled mean differences (MD) for continuous outcomes and odds ratios (OR) for dichotomous outcomes. RESULTS: 10 cross-sectional studies were included in the review. Academic stress was measured with a range of instruments, mostly self-report questionnaires. Most studies reported suicidal ideation as the main outcome, and fewer used suicide attempts or probability estimates. There were distinct methodological variations in stress severity classification, psychologic or academic confounder adjustment, and psychometric properties reporting. Pooled mean difference between females and males in academic stress was nonsignificant (MD=10.38; 95% CI: -6.34 to 27.09; p=0.22), as was between suicidal ideation (MD=14.43; 95% CI: -22.16 to 51.02; p=0.44). Overall continuous outcomes were also nonsignificant (MD=12.38; 95% CI: -3.70 to 28.47; p=0.13). The pooled odds ratios for suicidal ideation and suicide attempts, however, were statistically significant (OR=0.01; 95% CI: 0.01 to 0.02 and OR=0.00; 95% CI: 0.00 to 0.02 respectively; both p<0.0001) indicating a correlation with increased academic stress. Additionally, there was extremely large heterogeneity in most of the pooled analyses (I(2)> 98%). CONCLUSIONS: The prevalence of academic stress had a strong association with suicidal ideation and suicide attempts in dichotomous outcome models. However, continuous outcomes did not yield significant differences in the two genders. Associations that were found were confounded by measurement instrument variability, not making proper adjustments for psychiatric and academic confounders, and methodological differences in studies.