OBJECTIVE The aim of the study was to evaluate the association of patient factors, amount of in-hospital postoperative narcotics, and pain scores on postdischarge narcotic use (PDNU). METHODS This is a secondary analysis of a randomized controlled trial comparing a postoperative usual-care regimen with multimodal pain regimen after pelvic reconstructive surgery. This analysis evaluated patients in the multimodal arm. Postdischarge narcotic use (as mg oral morphine equivalents, MME, calculated from narcotic tablets remaining) was assessed postoperative days 7 to 10. Brief Pain Inventory (BPI) surveys were collected preoperatively and at postoperative day 1. Patient factors were evaluated using univariate and multivariate analysis. Correlations examined the relationships between PDNU and postoperative in-hospital narcotic use and BPI scores. RESULTS Sixty-eight patients randomized to the multimodal pain regimen arm had median (interquartile range) PDNU of 22.5 (0-159.4) MME. After excluding postdischarge narcotic nonusers (34.8%), the median PDNU was 127.5 (22.5-180.0) MME. The median PDNU was 172.5 (150.0-180.0) MME after abdominal reconstructive surgery (n = 7), 82.5 (28.1-180.0) MME after laparoscopic reconstructive surgery (n = 22), and 37.5 (13.1-181.2) MME after vaginal reconstructive surgery (n = 14). A linear correlation was noted between the amount of postoperative narcotics used in-hospital and the amount needed postdischarge after abdominal (r = 0.588, P = 0.057), laparoscopic (r = 0.439, P = 0.019), and vaginal (r = 0.455, P = 0.017) reconstructive surgeries. The BPI scores on postoperative day 1 for "average" pain (r = 0.388, P = 0.002) and "now" pain (r = 0.490, P < 0.001), and on postoperative week 1 for "average" pain (r = 0.383, P = 0.002) and "now" pain (r = 0.389, P = 0.002) correlated with PDNU. CONCLUSIONS Amount of postoperative in-hospital use of narcotic medications and BPI scores can be valuable predictors of PDNU in patients undergoing pelvic reconstructive surgery.