Postanesthesia Care Handovers

Handover—the transfer of professional responsibility and accountability for some or all aspects of care for a patient, or group of patients, to another person or professional group on a temporary or permanent basis (defined by British Medical Association, and adopted by Australian Medical Association).a Handovers in all areas of health care are highly variable. Communication can be harried and unstructured, leading to omissions or errors in transmission or reception, which are then implicated in adverse events. A recent large-scale review of medical records found a strong association between the number of intraoperative anesthesia care transitions and the multifactorial-adjusted odds of experiencing any in-hospital mortality or major morbidity.1 There are a huge variety of handover situations themselves, of course, including those between same-type clinicians during shift changes and those between same-type or different-type clinicians during patient transfers. Postanesthesia care unit (PACU) handovers have recently been reviewed in this journal.2 The authors found 31 publications on the topic; 24 contained recommendations for structuring the information content and handover process and 4 studies formulated and evaluated a structured handover practice. Other studies have been published since then, indicating a growing concern that perioperative handovers adversely impact patient safety. This issue contains a new report from a preeminent university medical system that describes a multimodal intervention to improve the quality of PACU admission handovers.3 A single type of handover, from anesthesia providers to recovery room nurses, was studied at 2 teaching hospitals. The intervention included implementation of a standardized handover report form that was automatically generated by the electronic medical record, intensive physician and nurse interprofessional training in handover protocol and interdisciplinary communication (using webinars, full-scale simulation, and a refresher course), and posttraining feedback about pooled observed handover quality. The investigators found that the quality of observed PACU admission handovers improved progressively over the 1.5 years during which the training and refresher training occurred. Handover duration was unchanged in the adult PACU and only slightly longer in the pediatric PACU. Most importantly, the quality of observed PACU handovers remained much better than baseline almost 4 years after training and feedback concluded, even though many of the clinicians were different and had never received the training. This indicates a sustained effect because of a change in culture (i.e., the behavior of new and untrained clinicians was shaped by the observations of and feedback from clinicians around them). It is encouraging that a microlevel culture change could be observed: effects of the intervention irradiated beyond the personnel it directly affected. This is not inconsistent with the literature concerned with organizational culture change through action learning (e.g., Ref. 4), although demonstrating the effects empirically in health care seems rather unique because there are examples of clinician behavior reverting to baseline levels when feedback stops.5,6 The interventions used in this study were well thought out and followed recommendations of other researchers in this area.2,7,8 However, they were also time-consuming and required the support of expert panels, information technology, and simulation education. The question this raises is whether a hospital can indeed engender a safety culture change if it just throws enough resources at it. This might then separate the well-endowed health care systems from the rest. As Wildavsky9 quipped, “richer is safer.” Yet, there were some interesting limitations to the study. Only the handover discussion between the anesthesia care provider and the receiving PACU nurse was evaluated, even though an entire PACU handover consists of 4 phases: “pre-handover, arrival, handover meeting, and post-handover.”7 The interventions and evaluations were also anesthesia-centric and did not consider the additional handovers that typically occur from the surgeon and circulating nurse to the PACU nurse.10 Most importantly, there was no evaluation of the impact of the “improved” handover on patient safety; patient outcomes were not measured nor were evaluations done down the line to identify important information that had not been effectively transmitted during the handover.7 Another interesting outcome, not measured in the study, is the knowledge received and perceived by the receiving clinician. Not only would it be interesting to pursue questions about the completeness and parsimoniousness of such receiver knowledge (did they get all they needed, and not much else?) but also to ask whether the handover calibrated the receiving clinician about their knowledge. That is, did it make them realize that they did not know certain things that they should know? At a higher level, it is important to consider that communication deficiencies might not be the real problem. Shaping factors related to roles, medical hierarchy, production pressures, or other issues might express themselves in communication, but that does not make communication the problem, or necessarily the only intervention target. If conditions for miscommunication remain embedded in professional relationships and the organization of work, then interventions focused solely on communication may be misguided and surely incomplete. Error reduction initiatives that do not take into account the goal conflicts, attentional demands, and knowledge factors behind the generation of errors have historically had little or no lasting effect.11 Achieving consistency in the execution of normal, repetitive, highly routine, yet crucial activities such as handovers has been a challenge in other safety-critical industries too. Commercial aviation, for example, might claim to have succeeded in this to an extent, in part because of the following: Careful requirements analysis, long-evolved systems engineering, and judicious checklist design based on research and operational input, which aims to minimize critical checklists or standard procedures down to “killer items”; Regimes of twice-yearly proficiency checking, recurrent training, and line examinations, of which routine task consistency is an important aspect; Systems of voice recording and electronic surveillance of crew performance to set accountability standards; Standard role divisions in small (2-member) crews (e.g., between flying pilot and monitoring pilot) that create operational expectations and help frame mutual social-professional control. The study by Weinger et al.3 might leave the reader with 3 messages that may not be correct. First, that multimodal interventions, including simulator-based training, online courses, continuous feedback, and more are needed to achieve this result. Second, that structured, high-quality handovers improve patient safety. And third, that handover consistency and replicability should be the goal of such interventions. The first conclusion might discourage and disempower hospitals or health care systems that do not have that level of resources. That would be unfortunate. A number of published studies show the effectiveness of different and less ambitious interventions to improve handover quality, many focused on the PACU admission handover.10,12,13 Some interventions were straightforward, such as the simple introduction of a handover form. Others, such as the one published in this issue, were extremely involved. Interestingly, every published study demonstrated improved handover consistency and quality postintervention compared with preintervention. While this may be the result of publication bias, we should not conclude that improving handover quality need be difficult; the important thing is to do something. What cannot be determined from the amassed literature is which interventions give the best patient safety result for the investment. Every study only compared preintervention with postintervention. None compared different interventions or used a structured iterative approach to analyze the relative contributions of different multimodal elements. Starting small and using an iterative approach, as was done by Weinger et al.,3 may be the best way to start a handover quality initiative. As discussed at length, earlier, the second conclusion is not supported by the methods of this study nor are there data in the literature to prove that better-quality handovers improve patient outcomes. Regarding the third conclusion, consistency cannot be taken for granted as the goal. This reflects a long-standing debate in the quality movement, of course. However, quality does not necessarily equate to safety. Safety in complex, messy systems can sometimes be generated more by adaptation and variation than through the intolerance of deviance. For the diverse environments, interventions, and roles that play across postanesthesia care handovers, some variation is probably crucial. Variation affects the signal/noise ratio in an invariant handover (i.e., the proportion of conveyed information that is important or not). What is superfluous information (noise) to one type of clinician or in one setting might be important information (signal) to or in another. Tapping into this concept provides an opportunity for handovers (like second opinions) to improve rather than degrade outcomes.14 Developing and evaluating handover processes with the goal of improving safety is difficult, but necessary, if we want to avoid the problems and reap the benefits of a team approach to patient care. DISCLOSURES Name: Robert G. Loeb, MD. Contribution: This author helped write the manuscript. Attestation: Robert G. Loeb approved the final manuscript. Name: Sidney W. A. Dekker, MA, MSc, PhD. Contribution: This author helped write the manuscript. Attestation: Sidney W. A. Dekker approved the final manuscript. This manuscript was handled by: Sorin J. Brull, MD, FCARCSI (Hon).