Nurse staffing and medication errors: cross-sectional or longitudinal relationships?

We used autoregressive latent trajectory (ALT) modeling to examine the relationship between change in nurse staffing and change in medication errors over 6 months in 284 general medical-surgical nursing units. We also investigated the impact of select hospital and nursing unit characteristics on the baseline level and rate of change in medication errors. We found essentially no support for a nurse staffing-medication error relationship either cross-sectionally or longitudinally. Few hospital or nursing unit characteristics had significant relationships to either the baseline level or rate of change in medication errors. However, ALT modeling is a promising technique that can promote a deeper understanding of the theoretically complex relationships that may underlie the nurse staffing-medication error relationship. 2008 Wiley Periodicals, Inc.