ABSTRACT-Supplemental doses of antithrombin (AT) are widely used to treat sepsis-induced disseminated intravascular coagulation (DIC) in Japan. However, evidence on the benefits of ATsupplementation for DIC is insufficient. This multicenter retrospective observational study aimed to clarify the effect of AT supplementation on sepsis-induced DIC using propensity score analyses. Data from 3,195 consecutive adult patients admitted to 42 intensive care units for severe sepsis treatment were retrospectively analyzed; 1,784 patients were diagnosed with DIC (n ¼ 715, AT group; n ¼ 1,069, control group). Inverse probability of treatment-weighted propensity score analysis indicated a statistically significant association between AT supplementation and lower in-hospital all-cause mortality (n ¼ 1,784, odds ratio [95% confidence intervals]: 0.748 [0.572-0.978], P ¼ 0.034). However, quintile-stratified propensity score analysis (n ¼ 1,784, odds ratio: 0.823 [0.646-1.050], P ¼ 0.117) and propensity score matching analysis (461 matching pairs, odds ratio: 0.855 [0.649-1.125], P ¼ 0.263) did not show this association. In the early days after intensive care unit admission, the survival rate was statistically higher in the propensity score-matched AT group than in the propensity score-matched control group (P ¼ 0.007). In DIC patients without concomitant heparin administration, similar results were observed. In conclusion, AT supplementation may be associated with reduced in-hospital all-cause mortality in patients with sepsis-induced DIC. However, the statistical robustness of this connection was not strong. In addition, although the number of transfusions needed in patients with AT supplementation increased, severe bleeding complications did not.