Background: Due to high heterogeneity and risk of bias (RoB) in previously published meta-analysis, a concrete conclusion on the efficacy of baricitinib in reducing mortality in COVID-19 patients was unable to form. Methods: Search engines PubMed/MEDLINE, ScienceDirect and other sources like preprints and reference lists were searched with appropriate keywords. The included evidence was graded with GRADEpro. The RoB, heterogeneity and meta-analysis were studied through RevMan 5.4.1 software. The heterogeneity was evaluated based on the generated p-value or I2 test. Results: Eight (8) RCTs were included in current analysis. Five studies had low RoB. Based on grading the evidence, the inclusion and exclusion of high RoB articles led to moderate and high certainty of evidence, respectively. Based on 8 RCTs (with high RoB), baricitinib statistically significantly reduced mortality where the risk ratio (RR) = 0.84 [95% CI: 0.76 to 0.92; p = 0.0002; I2 = 23%; p = 0.25]. The heterogeneity was insignificant but the RoB was high. We did subgroup analysis of low and high RoB articles and found out baricitinib statistically significantly reduced mortality with the RR = 0.68 [95% CI: 0.56 to 0.82; p < 0.0001; I2 = 0%; p = 0.85] and RR = 0.89 [95% CI: 0.80 to 0.99; p = 0.04; I2 = 0%; p = 0.43], respectively. The heterogeneity was 0% with insignificant p-values in both subgroup analyses. The percentage of mortality reduction was 31.31% and 7.79%, respectively whereas it was 13.95% in main group analysis. Conclusion: With the presence of optimal sample size of 3944 from 5 low RoB studies which represents a minimum of 300 million population of people and with 0% of heterogeneity, the effectiveness of baricitinib in reducing the mortality in COVID-19 patients is concretely proven.