Abstract Interpretation of coagulation testing in neonates currently relies on reference intervals defined from adult cohorts. These likely do not reflect the normative values for neonates, in whom immature synthesis of vitamin K-dependent factors may contribute to this divergence. Direct reference interval estimation studies are difficult, particularly in neonatal populations; however, advances in indirect estimation may allow us to infer normative distributions from routinely collected clinical data. To accomplish this, initial coagulation testing results from all patients admitted to a level IV neonatal intensive care unit between 1/1/2018 - 1/1/2023 were extracted from the electronic health record and analyzed. Results prior to June 2021 were performed on the viscosity-based STA Compact Max (Diagnostica Stago. Parsippany, NJ, USA). Afterward, results were performed on a nephelometry-based ACL Top (Instrumentation Laboratory. Bedford, MA, USA). Specimens collected after transfusion of blood products were excluded. Prothrombin times (PT) and international normalized ratios (INR) were available for 1,117 neonates, while activated partial thromboplastin times (aPTT) were available for 790 neonates. The 95% confidence intervals (CI) of the median PT for extremely preterm, very preterm, preterm and term neonates were [19 - 21s], [18 - 22s], [17 - 19s] and [17 - 18s]. For INR, the CI were [1.6 - 1.8], [1.5 - 1.8], [1.4 - 1.6], and [1.4 - 1.5]. For aPTT, the CI were [51 - 61s], [47 - 58s], [45 - 50s], and [42 - 46s]. Indirect reference intervals (IRIs) were then estimated across gestational age groups (binning all patients delivered at less than 37 weeks into one preterm category) using refineR. The IRI for PT was 10 - 24s for preterm,10 - 22s for term, and 10 - 24s for all neonates. The IRI for INR was 0.7 - 2.1 for preterm, 0.8 - 1.8 for term, and 0.8 - 1.9 for all neonates. The IRI for aPTT was 24.5 - 68.4s for preterm, 24.7 - 58.2s for term, and 25 - 62.3s for all neonates. When compared to currently reported reference intervals, the lower limits of the IRI were similar to currently reported reference intervals; however, the upper limits of the IRIs were 1.3 - 1.8 times higher than current reference intervals. Theoretically, implementation of these IRIs would lead to a 3-fold decrease in the number of results flagged as abnormal. In conclusion, IRIs for coagulation assays estimated from neonates using routine clinical data show substantial divergence from current reference intervals in their upper limits, possibly due to the decreased factor concentrations seen prior to liver maturation. If implemented, these IRIs would substantially reduce the proportion of abnormal results. Future work will explore whether switching to these reference intervals could lead to a reduction in unnecessary transfusions or improved clinical outcomes.