If the inline PDF is not rendering correctly, you can download the PDF file here.
TO THE EDITOR: I read with interest the article by Michalak et al.2 (Michalak SM, Rolston JD, Lawton MT: Prospective, multidisciplinary recording of perioperative errors in cerebrovascular surgery: is error in the eye of the beholder? J Neurosurg 124:1794–1804, June 2016). Much has been written about the lessons that health care providers can learn from aviation. Indeed, many physicians have probably grown weary of hearing about how checklists and automation can save patients from thousands of potential medical errors. But checklists and automation are only part of the story when it comes to the vast improvement
TO THE EDITOR: I read with interest the article by Michalak et al.2 (Michalak SM, Rolston JD, Lawton MT: Prospective, multidisciplinary recording of perioperative errors in cerebrovascular surgery: is error in the eye of the beholder? J Neurosurg 124:1794–1804, June 2016). Much has been written about the lessons that health care providers can learn from aviation. Indeed, many physicians have probably grown weary of hearing about how checklists and automation can save patients from thousands of potential medical errors. But checklists and automation are only part of the story when it comes to the vast improvement in aviation safety that occurred in the 1980s and beyond.
In the preceding decades, there were numerous fatal air-plane crashes whose ultimate cause was found to be “pilot error.”3 Most importantly, there is evidence that copilots or other flight personnel recognized errors, dangerous situations, or potential lifesaving solutions, but either did not tell the pilot or were not able to effectively communicate the situation to the senior officer. Airlines realized that one of their substantial problems was a steep hierarchy in the cockpit. Junior officers were unwilling to question the authority of senior pilots, even in situations where it was clear that an error was occurring. The solution to this was Crew Resource Management.1 Training and standardized procedures were widely adopted throughout commercial aviation to assure a culture of psychological safety and a flattened hierarchy, such that the most junior member of the flight crew not only could, but was required to notify the pilot of dangerous situations. The parallels to the neurosurgical operating room (OR) are obvious, and the need for Crew Resource Management in neurosurgery is huge.
The paper by Michalak et al.2 is a novel and very important contribution to our field. The authors conducted a survey of OR teams, asking team members to identify errors (“deviations from optimal care”). For the first time, not only was the attending neurosurgeon surveyed, but residents, circulating nurses, and anesthesiologists were also asked to identify errors. They found that in 31 cerebrovascular cases, a total of 118 errors were reported, 94 of them unique (3.1 errors per case). The attending neurosurgeon identified 25 errors (0.8 errors per case). Most neurosurgeons would likely agree that the attending neurosurgeon is ultimately responsible for the care of the patient and the outcome of the operation. But how can the attending neurosurgeon optimize the outcome if he or she is only aware of one-third of the potential errors during a case?
For example, the authors note that there were 9 total contamination errors reported. Attending neurosurgeons reported only 2. It is possible that the attending was not in the room at the time, and thus was not aware of the other 7. Far more concerning is the thought that the attending was present, but was not notified of the contamination event.
The authors found “a paucity of duplicate errors,” and determined that team members were more likely to report errors that were directly related to their role in the case. Although this may seem intuitive, it is an important finding. They conclude that to fully understand OR errors, we must solicit information from all members of the OR team. Their results certainly support this conclusion, but there should be more to the story. As surgeons, we need to do more than understand OR errors. We must recognize that it is human nature to make errors. And we must recognize that the surest way to prevent an individual's error from leading to an adverse event is to provide every member of the operating team with the full support of all other team members. In brief, we must recognize the importance of Crew Resource Management in our field.
How can we flatten the hierarchy in the OR? How can we assure that the resident, the circulating nurse, even the medical student is comfortable speaking up if she sees an error? Taken further, how can we assure that everyone in the OR knows that he is part of the team caring for the patient on the table, and knows that he has a responsibility to notify the rest of the team if a potentially harmful situation is developing?
To consider these questions for an entire specialty, or even for a hospital, seems daunting. Perhaps this is a change that can occur on an individual level. For example, in my own OR, at the completion of the formal time out, I have added a simple statement: “We are the team caring for this patient today. There are many ways this case can go wrong. I expect everyone here to speak up and inform me if they see a dangerous situation developing. Does everyone agree?” It is a simple step, taking less than a minute, but it establishes the expectations and accountability from every team member. Is this enough? Probably not, but it is a start.
The authors of this paper should be commended for studying this problem and calling our attention to it. If our field is to make the giant steps in safety that we have seen in air travel, we must find answers to these questions. We must begin to systematically apply the principles of Crew Resource Management to the neurosurgical operating theater.
We are grateful for Dr. Rocque's insightful response to our article. As Dr. Rocque emphasizes, the vast majority of errors we documented were not found by the attending neurosurgeon, but rather by the nurses, anesthesiologists, and residents involved in each case, underscoring the importance of systems-level thinking. As neurosurgeons, we tend to focus on the technical and decision-making errors we commit. But errors can strike patients from any part of the health care system that touches them, from anesthesia to hospital bureaucracy. For example, when we reviewed other prospective studies of neurosurgical errors, we found that only 25% were technical in nature, the remaining errors often involving domains outside the neurosurgeon's direct control.3,4 To put it in perspective, even if a surgeon perfects his or her craft, patients are still subject to 75% of the errors and adverse events to which they were exposed before.
Improvement in health care systems can occur at many levels, with the OR being of critical importance for neurosurgeons. Improving coordination among the many OR players by using techniques like Crew Resource Management, as suggested by Dr. Rocque, has been shown in other surgical specialties to improve outcomes, and has been suggested for neurosurgery specifically at least as early as 2001.1 The fact that 15 years have passed and Crew Resource Management is still not widely studied in neurosurgery is disheartening. However, strides are being made to flatten the intraoperative hierarchy and improve teamwork, as suggested by Dr. Rocque. His routine of asking for all those in the OR to speak up if they see a dangerous situation developing is simple and exemplary. At the University of California, San Francisco, our work-flow includes introductions of all OR personnel before the case begins, and concludes with structured postoperative debriefings, which can be used to identify issues that occurred during the case. These are just a small subset of the many other safety protocols detailed in our neurosurgical “Culture of Perioperative Safety” video.2 In the end, it will take more people like Dr. Rocque, who are committed to improving patient safety and teamwork in the OR, to move our field forward.