A failed attempt to improve quality of out-of-hospital CPR through performance evaluation

Quality of CPR performed by professionals has been reported to be substandard even with automated corrective feedback. Our hypothesis was that providing CPR performance evaluation (CPR-PE) to three ambulance services would facilitate local education and implementation of CPR guidelines and, consequently, improve CPR quality.

Quality of CPR in 85 consecutive cases of adult out-of-hospital cardiac arrests after CPR-PE was compared to 39 cases prior to CPR-PE. Real-time automated verbal and visual feedback on CPR performance was given in all cases. No general implementation strategy was provided because the sites were expected to use the CPR-PEs in development of local strategies. Because the strategies were expected to vary, the sites were analyzed separately.

No significant improvement was seen in quality of CPR after CPR-PE. No chest compressions were given 40% of the time before versus 41% after CPR-PE. The median (95% confidence interval) percentage of chest compressions within the recommended depth range (38-51 mm) was 35% (27-57) before versus 51% (42-60) after CPR-PE (p = 0.12). In site-specific analysis, chest compressions within guideline depth increased from 31% to 61% after CPR-PE (p = 0.05) in one site.

Overall our attempt to improve CPR-quality was unsuccessful. Quality improvement likely requires a full range of implementation strategies to change current attitudes and practices.