Nurse staffing has been linked to hospital patient outcomes; however, previous results were inconsistent because of variations in measures of staffing and were only rarely specific to types of patient care units. To determine the relationship between nurse staffing in general and intensive care units and patient outcomes and determine whether safety net status affects this relationship. A cross-sectional design used data from hospitals belonging to the University HealthSystem Consortium. Data were available for approximately 1.1 million adult patient discharges and staffing for 872 patient care units from 54 hospitals. Total hours of nursing care [Registered Nurses (RNs), Licensed Practical Nurses, and assistants] determined per inpatient day (TotHPD) and RN skill mix were the measures of staffing; Agency for Healthcare Research and Quality risk-adjusted safety and quality indicators were the outcome measures. TotHPD in general units was associated with lower rates of congestive heart failure mortality (P<0.05), failure to rescue (P<0.10), infections (P<0.01), and prolonged length of stay (P<0.01). RN skill mix in general units was associated with reduced failure to rescue (P<0.01) and infections (P<0.05). TotHPD in intensive care units was associated with fewer infections (P<0.05) and decubitus ulcers (P<0.10). RN skill mix was associated with fewer cases of sepsis (P<0.01) and failure to rescue (P<0.05). Safety-net status was associated with higher rates of congestive heart failure mortality, decubitus ulcers, and failure to rescue. Higher nurse staffing protected patients from poor outcomes; however, hospital safety-net status introduced complexities in this relationship.