Technical Solutions and the Safety Big Picture

    loading  Checking for direct PDF access through Ovid


I read with interest the recent article by Eagle et al,1 which outlines 2 technically ingenious approaches to the automatic detection and recording of drug doses administered during anesthesia using lasers and radio frequency identification devices.1 However, I wonder whether this approach for tracking drug doses could in fact increase the risk to patients rather than reduce it, due to the fact that it requires a workflow in which the syringe is scanned after the drug has been given. We know from large-scale studies of drug administration error in anesthesia that drug substitutions and dose errors make up approximately 40% of drug errors overall.2 More worrying is the fact that approximately 69% of substitution errors occur between different pharmacological classes of drug, making substitutions potentially a very dangerous form of drug error.
For this reason, other drug administration systems do offer a drug identity check before administration, thus protecting the patient from substitution errors (as Eagle et al1 point out), but these systems do not offer automatic dose tracking. Should the safety of an identity check before administration be sacrificed for the convenience of automatic dose recording, given the known dangers? Eagle et al1 do state that their devices are intended to be record keepers rather than drug checkers, but it seems like poor system redesign to develop a device which solves 1 problem only to create another with potentially worse consequences. The overriding consideration for change in clinical practice should be the improvement of the quality and safety of patient care. One of the key concepts in implementing such change is that solutions to safety problems in health care are typically not purely technical—that is, we also need to consider the human factors of how clinicians work, and the potential unintended consequences of changing established workflows.3 Eagle et al’s1 approach currently takes 5–10 seconds to read each syringe. Even if we accept that this is due to the current devices being prototypes and that this read time will reduce to something workable for the clinician in the future, there remains the problem that reading the syringe after the contents have been administered will result in dose and substitutions errors being duly recorded rather than prevented. I would have liked to see some effort to make the new dose tracking system compatible with existing drug administration procedures in anesthesia in a way that considers how to preserve existing safeguards and also attempts to avoid potential unintended consequences of system change.
    loading  Loading Related Articles