Validation of a scoring algorithm for non-overt disseminated intravascular coagulation (DIC) proposed by the International Society on Thrombosis and Haemostasis (ISTH) is still incomplete. It was the objective of this study to assess the impact of including AT to non-overt DIC scoring on the predictability for intensive care unit (ICU) death and the later development of overt-DIC defined by the Japanese Ministry of Health and Welfare (JMHW) or the ISTH. We performed a retrospective observational study conducted in 364 patients in critical care. Coagulation parameters obtained daily for DIC screening were utilised for scoring. There were 194 and 196 patients scored as positive non-overt DIC with and without AT, respectively; diagnostic agreement between the two was 78%. As compared with patients without non-overt DIC, these non-overt DIC patients had significantly higher mortality. In 37 ICU non-survivors, positive non-overt DIC scoring with AT preceded ICU death by a median of 6.8 days, which was significantly earlier as compared with a median of 5.4 days for non-overt DIC without AT (p = 0.022). In patients who developed overt-DIC after admission, the time period from positive non-overt DIC to positive overt-DIC was significantly longer when AT was utilised (overt-DIC ISTH; 1.3 days vs. 0.1 days, p = 0.004, overt-DIC JMHW; 2.5 days vs. 2.0 days, p = 0.04, with AT vs. without AT, respectively). Non-overt DIC scoring predicted a high risk of death in critically ill patients. When information on AT levels was included, non-overt DIC scoring was found to predict development of overt-DIC significantly earlier than non-overt DIC scoring without AT.