This study employed a retrospective cohort analysis using New York State's Statewide Planning and Research Cooperative System (SPARCS) to improve the Patient Safety Indicator (PSI) definition of postoperative hemorrhage/hematoma (POHH) and to identify patient risk factors associated with POHH. Study participants were nonobstetric, inpatient surgical admissions in SPARCS and readmissions within 30 days with a principal diagnosis of POHH. The main outcome measures were mortality rate, length of stay, and readmissions. The mortality rates of events identified by a secondary diagnosis only and by the PSI were not significantly different. The number of POHH events increased by 9.3% when readmissions were captured. The PSI definition of POHH may need modification to capture events with no secondary procedure. The PSI misses events identified on readmission, but the consequences of these events are not as severe as those currently captured. A variety of patient and hospital characteristics are predictive of a higher risk of POHH.