Key points are not available for this paper at this time.
Design
Editorial
Answers are grounded in this paper and cite their sources.
This editorial emphasizes the need to transition from weak administrative solutions to strong human factors engineering interventions to improve perioperative patient safety.
Anesthesiology’s claim as one of the safest medical specialties can largely be attributed to 3 factors: technological innovation, quality improvement initiatives borne of critical self-examination, and a commitment to patient safety represented by education, training, and professionalism. Advances such as capnometry, pulse oximetry, and smart alarms helped achieve early gains in reducing devastating events during and immediately after surgery. Building on these improvements, the past 40 years have seen increasing recognition of the role that human performance plays in patient safety and outcomes. The published literature in anesthesia reveals a gradual but steady consideration of the ways that anesthesia clinicians function, dissecting the skills needed for optimal performance in the high-stakes environment of the surgical suite. Highlights of this study include Cooper et al’s1 1978 study of human factors contributing to anesthesia mishaps, Fletcher et al’s2 definition and evaluation of anesthesia nontechnical skills, Domino and colleagues’3–5 analyses of the American Society of Anesthesiologists’ Closed Claims Project, and Weinger and Slagle’s6 exposition of “nonroutine events” in perioperative care. As our specialty’s attention to human performance has developed and matured, we have looked to other fields for insight. One such field is the engineering discipline known as human factors engineering (HFE), also called human factors and ergonomics, or more simply, ergonomics. HFE principles date back to the 19th century, but the first professional HFE organizations emerged in the mid-20th century. HFE has 3 major branches: (1) physical ergonomics (ie, design and layout of the work environment), (2) cognitive ergonomics (ie, features and limitations of human perception and cognition), and (3) organizational ergonomics (ie, policies, procedures, and roles that guide and constrain work).7 What does HFE have to teach us about improving perioperative performance? In this issue of Anesthesia & Analgesia, Neily et al8 describe their analysis of a national Veterans Health Administration (VHA) database detailing root cause analyses (RCAs) of anesthesia-attributable adverse events. The authors analyzed 3 years’ worth of RCA reports (n = 3228) from the National Center for Patient Safety database of the VHA. The focus of their article was the 1% of reports (n = 36) in which adverse events occurred intraoperatively, and in which anesthesia care appeared to be a primary contributor. The authors categorized the RCA reports by type (eg, medication error, regional block error), ranked adverse events with severity scores, and discussed corrective actions suggested by local VHA RCA teams, juxtaposing these action plans with HFE solutions designed to reduce the likelihood of an adverse event. RCAs are used widely in health care, but are problematic for reasons relating to cognitive bias. Examples of the cognitive phenomena that might influence RCAs include memory reshaping9 by witnesses, outcome bias,10 and hindsight bias.11 These cognitive biases may compromise the identification of factors influencing the event leading up to the RCA. A powerful alternative to RCAs is prospective risk assessment using failure mode and effects analysis,12 a tool developed in industry that has been adapted by some to better align with health care settings.13 Whether risks are identified prospectively or retrospectively, there is an endemic tendency of investigators to settle for administrative and perhaps “weaker solutions”14 to patient safety vulnerabilities. The work of Neily et al8 bears this out; across 36 RCAs over 3 years, the most commonly cited action plans included improvements in policy, process, or education, which are generally considered to be weak actions, which are low on the hierarchy of lists that characterize the ability of interventions to prevent error and/or effect change.15 To fully understand the local sites’ RCA action plans, one would need to understand the constitution and training of each RCA team, including whether individuals with human factors expertise or senior hospital administrators with budgetary authority routinely participated. This information was not available to the authors, but there are still potential learnings from the solutions proposed by the local RCA teams. In HFE, “weak” actions are those that require the operator to remember something or to pay attention, while “strong” actions do not. Consider, for example, a transparent glass door with handles on both the inside and outside surfaces. From looking at the door, it is not immediately apparent how to enter or exit, and many people will choose the wrong direction to move the door. In this situation, a weak solution would be posting “push” and “pull” signs, which only work when people heed them. A stronger solution would be to replace 1 handle with a push plate. A still stronger solution might be installing automatic doors. Stated differently, strong actions make it difficult to do the wrong thing. Neily et al8 suggest several HFE-informed solutions to the medication errors (10/36, 27.8%) and regional block errors (5/36, 13.9%), and airway management problems (5/36, 13.9%) revealed by the VHA RCAs.8 These solutions range from using checklists and cognitive aids to “using intelligent pumps that will use image recognition…to identify the drug/concentration being infused.”8 Importantly, some safety challenges are more amenable to strong interventions (eg, medication errors) than are others (eg, errors in communication). The authors recognize this complexity by offering a range of solutions suited to clinical practice. All their solutions draw on HFE tenets, including accounting for the fallibility of human memory and promoting resilience in clinical teams. Ultimately, the HFE-informed approach calls for the design (and redesign) “of systems so that they fit the needs, abilities, and limitations of people.”7 In their article, Neily et al8 suggest that local VHA RCA teams may not have had the resources to implement expensive structural changes such as medication barcoding. However, there are interventions with low structural costs, such as use of emergency checklists or low-fidelity simulation that hold promise in improving team performance and patient safety. As we consider the gap between HFE-informed solutions and the solutions proposed by local VHA teams, we note that strong solutions are sometimes more expensive than weaker ones. In the current US health care environment, seemingly expensive solutions may be difficult to justify in the absence of a 1:1 association between patient safety hazards and demonstrable harm. Certainly, cost savings may be realized in the form of lower insurance premiums, but associations with more traditional patient outcomes (eg, length of stay, mortality) are difficult to tie to particular safety interventions. More studies are needed to help bolster the “business case” for implementation of strong, HFE-informed solutions. The major strength of the work by Neily et al8 is their use of a national data source. This complements prior studies published using other national databases, including the American Society of Anesthesiologists Closed Claims project16 and the National Anesthesia Clinical Outcomes Registry.17 These studies hold the potential to identify trends and vulnerabilities that are difficult to detect at a local level. However, they all have blind spots. A comprehensive national safety database like the United Kingdom National Reporting and Learning System18 would have much more power to identify patient safety threats and to generate hypotheses about factors that increase or mitigate risk. Until such time as a national comprehensive database is available, multi-institutional databases represent an important tool in identifying patient safety hazards. As explained by Weinger in the Handbook of Human Factors and Ergonomics in Health Care and Patient Safety, anesthesiology was an early adopter of human factors principles and strategies, including critical event analysis and use of simulation to train individuals and teams.19 The study of Neily et al8 serves as a reminder to continue this tradition and develop a high-quality, comprehensive national patient safety database composed of various hospital types to identify more and undiscovered adverse events. Modifying systems and processes and testing interventions brings us closer to safer solutions. By applying and disseminating these principles and system changes, we will make health care as safe as the disciplines from which HFE arose. DISCLOSURES Name: Benjamin T. Cobb, MD. Contribution: This author helped write and critically revise the manuscript. Name: Meghan B. Lane-Fall, MD, MSHP. Contribution: This author helped write and critically revise the manuscript. This manuscript was handled by: Richard C. Prielipp, MD.
No takes yet. Share an insight, caveat, or question.
Cobb et al. (2018) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: