Provably Clean: A Formal Analysis of Hand Hygiene During Anesthesiology Induction

Healthcare-acquired infections (HAIs) are a serious problem in operating rooms. Anesthesia providers’ hand hygiene during induction is often cited as a major contributor to this problem. HAIs can lead to grave consequences such as sepsis and death. Proper infection control practices are expected to be in place. However, multiple people working concurrently results in more hands with less clearly defined roles. This makes the process more complex and error-prone. Moreover, anesthesia induction is a rapid procedure that may make it difficult for practitioners to remember to clean their hands. Previous work sought to address this problem using a novel computational method based on model checking. This allowed an analyst to model an induction process and prove whether the procedure would ever allow things that should remain clean to become dirty. In this work, we extended this method’s capabilities and used it to evaluate the induction procedure used at a large, mid-Atlantic, academic medical center. To accomplish this, we conducted observations and used this information to construct a model of a real anesthesiology induction procedure. We then used the method to evaluate this normative procedure, but we also introduced modifications that included nondeterministic factors such as skipping steps and accidentally touching hands together. Finally, we accounted for the possibility of improperly cleaning equipment between patients. By analyzing these models, we discovered that the evaluated induction procedure was robust both normatively and to practitioners skipping hand hygiene steps. However, we found two ways improperly cleaned equipment could allow serious patient cross-contamination. We discuss the importance of these findings and make recommendations for improving anesthesiology induction protocols and the analysis method we used.