Importance: Previous studies of the association between healthcare associated infections (HAIs) and nurse staffing are inconsistent and limited by methodological errors.
Objective: To examine whether HAIs and nurse staffing 2 days prior to infection onset are associated by using unit-level staffing data, adopting precise measure of HAIs from Centers for Disease Control and Prevention (CDC)’s National Healthcare Safety Network (NHSN), and risk adjusting for individual risks.
Design: Cross-sectional analysis of data from 2007 through 2012 linking nurse payroll data using multiple electronic medical data sources.
Setting: A large urban hospital system with three campuses.
Participants: 100,264 adult patients from ICUs, medical, medical/surgical, and step-down units.
Main Outcome Measure: A composite HAI measure (urinary tract infection, bloodstream infection, pneumonia) associated with standardized unit-level nurse staffing.
Results: On approximately 15% of patient-days there was 1 shift understaffed and 6.2% patient-days in which both day and night shifts were understaffed. 4.34% patients developed HAIs during the hospitalization. After adjusting for patient characteristics, unit type and unit turnover, patients on units with both shifts understaffed were significantly more likely to develop HAIs 2 days later (RN: hazard ratio: 1.15, 95% confidence interval, 1.02, 1.30; nursing supporting staff: hazard ratio, 1.11, 95% confidence interval, 1.01, 1.21).
Conclusions: These data suggest that low nurse staffing has a negative effect on patient outcomes such as HAIs.