OBJECTIVE: Our goal is to describe the association between total quantitative blood loss (QBL) and risk of obstetric hemorrhage-related morbidity (OBH-M) in order to assess utility of the current definition of obstetric hemorrhage (OBH). DESIGN: Retrospective cohort study. SETTING: Urban safety-net hospital in Boston, Massachusetts. POPULATION or SAMPLE: Deliveries at our institution over a two-year period. METHODS: We categorized deliveries into ten equally sized deciles based on QBL and compared the proportion with OBH-M in each. Among the two deciles with the highest proportions of OBH-M, we stratified deliveries into seven groups of ascending intervals of 250cc QBL. Finally, we compared the positive predictive value (PPV) of the standard definition of OBH (QBL ≥ 1000cc) to a definition extrapolated from our stratified analysis. MAIN OUTCOME MEASURES: The primary outcome was proportion of deliveries within each QBL decile affected by OBH-M. The secondary outcome was PPV. RESULTS: We found a significant increase in OBH-M from decile 9 (895-1201cc QBL) to decile 10 (1205-8325cc QBL) (p<0.001). In our stratified analysis, we found QBL of 1500cc to be an inflection point for an increased proportion of OBH-M. Our secondary analysis showed an increased PPV for OBH-M using QBL of 1500cc (20.5%) compared to that of QBL 1000cc (9.8%). CONCLUSIONS: Our findings suggest that a higher QBL threshold than the currently accepted definition of OBH is more predictive of OBH-M.