Objective The value of plasma fibrinogen (FIB), platelet count (PLT) and D-dimer (D-D) test in predicting postpartum hemorrhage was analyzed. Methods A total of 70 parturients admitted to our hospital from August 2022 to August 2024 were selected for this study. According to whether postpartum hemorrhage occurred, they were divided into non-bleeding group (n=42) and bleeding group (n=28). FIB, PLT and D-D tests were given to analyze their general data, and Pearson analysis was performed on their related data. Results Age (≥35 years), delivery times (≥2 times), delivery mode (cesarean section), pregnancy complications (yes), gestational week (<37 weeks), placenta previa (yes), fetal body weight (≥4 000 g), placental adhesion (yes) and placental abruption (yes) in hemorrhage group were higher than those in non-hemorrhage group (P<0.05). There was no significant difference in body weight (≥23 kg/m², <23 kg/m²) and conception method (natural conception, assisted reproductive technology) between the two groups (P>0.05). The levels of FIB and PLT in bleeding group were lower than those in non-bleeding group (P<0.05). The level of D-D was higher than that of non-bleeding group (P<0.05). The AUC of FIB, PLT and D-D were 0.882, 0.884 and 0.834, respectively. The AUC of combined test in predicting postpartum hemorrhage was 0.969. The levels of FIB, PLT and D-D of combined test were higher than those of single test (P<0.05). Logistic analysis of results: Age, delivery time, delivery mode, pregnancy complications, gestational weeks, placenta previa, fetal body mass, placental adhesion, placental abruption, FIB, PLT, D-D were independent variables, and postpartum hemorrhage was dependent variables. Delivery mode, pregnancy complications, placenta previa, fetal body mass, placental adhesion, placental abruption, FIB and PLT were all risk factors for postpartum hemorrhage (P<0.05). D-D was the protective factor (P<0.05). Conclusions FIB, PLT and D-D have important value in the prediction of postpartum hemorrhage, and the combined detection can significantly improve the prediction accuracy.