To evaluate the effect of cesarean section (CS) and vaginal delivery (VD) on postpartum stress urinary incontinence (SUI) and pelvic floor muscles strength and to find out the correlated obstetric factors and prevention for postpartum SUI. Totally, 788 women, who visited the antenatal clinics, delivered and had the follow-up at 6-8 weeks after delivery in Beijing Obstetrics and Gynecology Hospital in the year of 2008, were enrolled in this study and were divided into 3 groups: CS group (n = 212); normal vaginal delivery (NVD) group (n = 534) and forceps delivery (FD) group (n = 42). Women in the NVD and FD group were merged into one VD group and then divided into SUI and non-SUI group. Information of delivery mode and the correlated obstetric factors were obtained through questionnaires and medical records. Femiscan pelvic floor muscle examine system was applied to measure the pelvic floor muscle strength to understand the relationship between postpartum SUI and pelvic floor muscle strength. (1) Incidence of SUI: the overall proportion of women who complained of urinary incontinence (UI) during pregnancy was 15.4% (121/788), and it was 15.9% (85/534), 11.9% (5/42) and 14.6% (31/212) in the NVD, FD and CS group, respectively (P > 0.05). The overall incidence of postpartum SUI was 17.1% (135/788), and it was 19.1% (102/534), 26.2% (11/42) and 10.4% (22/212) in the NVD, FD and CS group, respectively, with significant difference between the NVD and FD group, and between the CS and NVD group (all P < 0.01). (2) The associated obstetric factors of postpartum SUI: among the VD group, 113 women were in the postpartum SUI group and 463 in the non-SUI group. Univariate analysis and logistic multivariate analysis showed that delivery mode, neonatal birth weight and UI during pregnancy were risk factors of postpartum SUI. CS decreased and higher neonatal birth weight and UI during pregnancy increased the risk of postpartum SUI. In the VD group, neonatal birth weight, forceps delivery and UI during pregnancy increased the incidence of postpartum SUI (P < 0.01), but no correlation was found with labor analgesia, duration of labor, episiotomy, breast feeding, volume of postpartum bleeding, gestational weeks at delivery, induction and pre-pregnant BMI, etc (all P > 0.05). (3) Pelvic floor electromyogram: pelvic floor muscles strength in the CS group was significantly higher than that of the VD group [activity value: (19.7 +/- 9.9) microv vs (14.8 +/- 8.4) microv; work value: (84.5 +/- 37.2) microv vs (78.8 +/- 28.2) microv; peak value: (25.5 +/- 12.5) microv vs (19.7 +/- 11.8) microv, all P < 0.01]. Among women in the VD group, the relaxation value and the ratio of relaxation value over activity value (r/a) in the postpartum SUI group were significantly lower than those in the non-SUI group [relaxation value: (1.7 +/- 1.8) microv vs (3.0 +/- 3.9) microv; r/a ratio: 0.2 +/- 0.2 vs 0.3 +/- 0.5, all P < 0.01]. The r/a ratio in the VD group showed no difference compared to that in the CS group (0.2 +/- 3.5 vs 0.2 +/- 0.2, P > 0.05). Women experienced vaginal delivery, either NVD or FD, have a higher incidence of postpartum SUI than those delivered through CS. UI during pregnancy, forceps delivery and neonatal birth weight are risk factors of postpartum SUI.