BACKGROUND: Adverse drug events related to patient-controlled analgesia (PCA) place patients at risk. METHODS: We reviewed all critical incident reports at three tertiary care hospitals dated January 1, 2002, to February 28, 2009. In this longitudinal cohort study, critical incidents attributable to PCA errors were identified, and each incident was investigated. A safety intervention was implemented in February 2006 and involved new PCA pumps, new preprinted physician orders, nursing and patient education, a manual independent double-check, and a formal nursing transfer of accountability. RESULTS: A total of 25,198 patients were treated with PCA during this study, and 62 errors were found (0.25%), with 21 (0.08%) involving pump programming. All errors occurred before the safety interventions were put in place. Compared with the preintervention period, the odds ratio of a PCA error postintervention was 0.28 (95% CI = 0.14, 0.53; P < 0.001) whereas the odds ratio of a pump-programming error postintervention was 0.05 (95% CI = 0.001, 0.30; P < 0.001). Programming the wrong drug concentration was the most common programming error (10 of 21). Improper setup of intravenous tubing was also common (8 of 62), with one incident leading to respiratory arrest. Most PCA errors resulted in no harm, but there was negative impact to patients 34% of the time. CONCLUSION: At less than 1%, the incidence of PCA errors is relatively low. Most errors occur during PCA administration. Safety can be improved by addressing equipment, education, and process issues.
BACKGROUND: Adverse drug events related to patient-controlled analgesia (PCA) place patients at risk. METHODS: We reviewed all critical incident reports at three tertiary care hospitals dated January 1, 2002, to February 28, 2009. In this longitudinal cohort study, critical incidents attributable to PCA errors were identified, and each incident was investigated. A safety intervention was implemented in February 2006 and involved new PCA pumps, new preprinted physician orders, nursing and patient education, a manual independent double-check, and a formal nursing transfer of accountability. RESULTS: A total of 25,198 patients were treated with PCA during this study, and 62 errors were found (0.25%), with 21 (0.08%) involving pump programming. All errors occurred before the safety interventions were put in place. Compared with the preintervention period, the odds ratio of a PCA error postintervention was 0.28 (95% CI = 0.14, 0.53; P < 0.001) whereas the odds ratio of a pump-programming error postintervention was 0.05 (95% CI = 0.001, 0.30; P < 0.001). Programming the wrong drug concentration was the most common programming error (10 of 21). Improper setup of intravenous tubing was also common (8 of 62), with one incident leading to respiratory arrest. Most PCA errors resulted in no harm, but there was negative impact to patients 34% of the time. CONCLUSION: At less than 1%, the incidence of PCA errors is relatively low. Most errors occur during PCA administration. Safety can be improved by addressing equipment, education, and process issues.