Comparing Methods of Identifying Outlying Nurses in Audits of Low-Risk Cesarean Delivery Rates.

BACKGROUND: The low-risk cesarean delivery (CD) rate is an established performance indicator for providers in maternity care for quality improvement purposes. PURPOSE: Our objectives were to assess nurse performance using adjusted nurse-level CD rates and to compare methods of identifying nurse outliers. METHODS: We conducted a retrospective, cohort study of 6970 births attended by 181 registered nurses in one hospital's maternity unit. Adjusted and unadjusted nurse-level CD rates were compared and agreement between 3 definitions (statistical, top decile, over a benchmark) of outliers calculated. RESULTS: Adjusted nurse-level CD rates varied from 5.5% to 53.2%, and the unadjusted rates varied from 5% to 50%. Risk adjustment had little impact on the ranking of nurses, and outliers were consistently identified by 3 definitions. CONCLUSIONS: Trade-offs between statistical certainty and feasibility need to be considered when classifying nurse outliers. Findings can help target interventions to improve nurse performance.