Stability of risk-adjusted primary cesarean delivery rates over time

The purpose of this study was to compare how hospital rankings, based on risk-adjusted primary cesarean rates, change over time Study design Washington State Birth Events records data from 1995 through 1998 were used. Risk-adjustment models were built for each year. Risk-adjusted hospital rankings and outlier status (either “above” or “below” expected cesarean delivery rates) were calculated for each year. Hospitals that were identified as outliers for each year were compared with the use of κ statistics. The top and bottom 25% of the hospital rank list from each year were also compared. Results The overall κ statistic that compared outlier status over the 4 years was 0.44. The κ statistic for the upper 25% of rankings over the 4 years was 0.42. The κ statistic for the lowest 25% of rankings over the 4 years was 0.57. Conclusion Although exact ranking may not be meaningful, outlier status, especially if repeated in multiple years may be a good marker for the identification of hospitals for further scrutiny of their quality of care.