In Response

We were pleased to read the letter from Dr. Webster,1 who was a pioneer in anesthetic drug administration safety with his colleagues in New Zealand, especially Dr. Merry. We said in our paper2 that It is important to understand that there is no single universally accepted standard for syringe label content and color because several regulatory agencies and standards-setting bodies, including TJC, Centers for Medicare and Medicaid Services, US Pharmacopeial Convention, ISO, American Society for Testing and Materials, American Society of Anesthesiologists, and Australian Commission on Safety and Quality in Health Care (Table 1) have published guidelines. In the United States, TJC label content requirements are the most relevant because member hospitals can be sanctioned if syringe labels do not meet TJC specifications. We did not mean to imply that there are no international color standards. Although we agree that universally accepted label colors may be useful in avoiding drug administration errors, the evidence is not extensive. In the study cited by Dr. Webster, there was indirect evidence for the efficacy of color labeling.3 Dr. Webster et al. simultaneously applied a variety of drug safety measures, including label colors, which were associated with a reduction in drug errors. The specific evidence for the efficacy of label colors was the fact that drug swap errors tended to occur within drug classes, rather than between drug classes when the drug safety measures were applied. We also said in our paper that Merry et al.4 conducted a prospective randomized trial of a comprehensive anesthesia drug safety system that included scanning the barcodes of syringes before drug administration but did not find a statistically significant difference between the rate of drug administration error when the drug safety system was used compared with that of conventional practice. However, they did note that the rate of error was lower when anesthesia providers consistently scanned the drug barcode before administration and kept the voice prompt active (6.0% vs 9.7%; P = 0.004). Unfortunately, the compliance with these 2 features was only 18%. We understand that the drug safety system used in this study had a variety of parts that can be used together or separately. Our point was that the system was only effective in reducing actual drug errors when the syringes were scanned before administration and the voice prompt was active. This result is not surprising, because these are key features of the “New Zealand” drug safety system. As noted in the Results section, “The rate of errors in drug administration per 100 administrations was 0.16 with the new system (one in 625) and 0.33 with conventional methods (one in 303; P = 0.120).” Thus, the reduction in actual drug errors in the study, overall, was not statistically significant. However, in a subset analysis, there was a significant reduction of errors when the system was used as intended. “The mean error rate per 100 administrations was 6.0 (3.1–8.8) when the anesthetist complied with the 2 key principles of scanning the drug barcode before administering each drug and keeping the voice prompt active, and 9.7 (8.4–11.1) when the anesthetist did not comply with them (P = 0.004).” Thus, we believe that the “New Zealand” drug safety system is effective when used as intended, as shown by this study. We thank Dr. Webster and his colleagues for their pioneering work in this field. They have been the primary inspiration for our own interest in drug administration safety. Srdjan Jelacic, MD Andrew Bowdle, MD, PhD Department of Anesthesiology and Pain Medicine Division of Cardiothoracic Anesthesiology University of Washington Seattle, Washington [email protected]