Gender disparity in prevalence of mental health issues in Kerala: a systematic review and meta-analysis

Background: Kerala is a state in south India, appreciated for its education, better health indicators and quality of life. However, there is a proportionately high prevalence of mental health illnesses and suicides reported in the state. It is unclear if there is any disparity in the gender categories in this. With this systematic review, we aim to systematically study the gender disparity in the prevalence of mental health (MH) issues among adolescents, younger and older adults in Kerala. Methods: A search strategy was built and several databases like Pubmed, Cochrane, Scopus, EMBASE, EBSCOhost, Web of Science, and ProQuest were used alongside grey literature to identify relevant articles. The study was conducted according to the PRISMA guidelines following a prespecified protocol. After relevant data extraction, the estimates were pooled using random effects model due to the high heterogeneity assessed by tau-squared, Cochran Q, and prediction interval. Subgroup analyses, and meta-regression were used to reduce heterogeneity. We also identified the influence and heterogeneity contributed by individual studies using influence plots, Baujat plot, clustering, and performed several sensitivity analyses. Results: Twenty articles were included in the review and meta-analysis. The pooled odds ratio of mental health illnesses amongst females compared to males in Kerala was 1.31 (95% CI: 1.0 - 1.73) and falls within a prediction interval of 0.38 to 4.53. The individual studies showed high heterogeneity (I2 = 92%, p = p < 0.01) and hence, subgroup analysis was done for several prespecified subgroups based on etiology, geography, demography, study settings, and age groups. Heterogeneity was significantly reduced by subgrouping based on etiology, study setting and age (p, 0.001, p < 0.001, p = 0.001). In etiologic subgrouping the highest pooled odds was in comorbidities associated MH issues (2.54) and least in non-specific (0.97). In age subgrouping, the highest pooled odds was in elderly (2.53) and least in adolescents (0.63). The odds was highest in health care facility (2.21) and least in educational institution (0.78) based studies. Meta-regression based on the size of the study failed to reduce heterogeneity. Interpretations: A gender disparity was evident in the prevalence of mental health issues, with a higher Odds in females especially among the elderly and adults. A Gender transformative approach in legislative, health systems and policy frameworks will be the answer to this.