Being on the receiving end of reports about imagined misconduct has sensitized me to the misuse of anonymous event reporting systems. Health care organizations need reporting systems to identify events and behaviors that reduce safety and harm patients. Frank and honest reporting is encouraged by making such reports anonymous.

Anonymization invites abuse. A disgruntled colleague, nurse, or administrator can file an anonymous report against anyone, anytime. Since the report is anonymous, all that the presumed offender may learn is that someone was upset about something. Without facts or resources, the institution takes the path of least resistance. The reported facts are presumed to be true. The targeted physician is told to apologize and accept whatever punishment the institution doles out. Usually there is no overt punishment, other than a black mark against the physician buried in the institutional records.

Last February, I sent a note to the ASA Community Open Forum (asamonitor.pub/3Okyc5s) as well as several hundred professional colleagues:

I am considering a future editorial for the ASA Monitor about how patient safety and workplace behavior reporting systems are weaponized against physicians. I believe this is a common occurrence. By and large physicians are too busy to file reports against other health care professionals. However, if other health care providers (e.g., nurses, therapists, administrators) feel slighted, in my view they all too readily file complaints. As a consequence, well-intentioned systems designed to increase patient safety and reduce workplace hostility are weaponized to exact retribution for perceived slights.

I would welcome anesthesiologists sharing such experiences (steven.shafer@stanford.edu). Emails will remain confidential. I will work with authors to be sure that any examples used in my editorial are fully anonymized to protect confidentiality.

Wow, did that hit a raw nerve!

Within a few days, I received hundreds of jaw-dropping reports. Several anesthesiologists described how anonymized reporting had been orchestrated against them, ultimately leading to termination of employment. One anesthesiologist had evidence that CRNAs at his former hospital used incident reporting to systematically rid the hospital of pesky competitors (us).

Those are the extreme examples. It’s hard to know what to make of them. It’s hard to even know if the reports are true.

The vast majority of first-hand accounts reflect the different perspectives of physician and non-physician responsibilities. Most of the first-hand accounts described nurses reporting physicians.

Physicians are expected to make decisions and execute decisive action, particularly in anesthesia and critical care. Nurses are taught to follow orders and enforce hospital policy. Nurses exercise judgment in their work, but within constraints imposed by the institution. Conflict is guaranteed.

For example, many of the reports I received involved anesthesiologists responding to emergencies. The anesthesiologist was focused on keeping the patient alive. Protocols may be breached in the process, but who cares as long as the patient survived? It turned out the nurses cared. The nurses didn’t appreciate the non-sterile I.V. insertion or the order barked out to grab a suction catheter.

I’ve included a small fraction of vignettes in the article “You Can’t Make This Stuff Up” in this issue of the ASA Monitor. I rewrote each vignette to preserve the anonymity of the correspondent.

My query also produced a vigorous defense of incident reporting systems. Several colleagues pointed out that my claiming “reporting systems are weaponized against physicians” reflected my own bias and experience and is not a statement of fact.

Dr. Alyssa Burgart, a Stanford colleague, wrote that “a more interesting question is WHY do nurses often feel reporting is the only safe avenue for addressing behavior they feel is harmful? What cultural factors, structures, and systems have contributed to the phenomenon you are investigating? I don’t think it’s a fair assessment to suggest that these reports are being ‘weaponized’ against physicians without more fully exploring the milieu in which reporting takes place.”

Dr. Burgart is right. What I see as weaponized reporting may simply reflect a nursing belief that anonymous reporting is their only avenue of raising issues that involve physicians. I invited Dr. Burgart and Dr. Steven Bradley to pursue this. Their article “Write-Ups, Retribution, and ‘DARVO’” appears in this issue.

DARVO stands for “deny, attack, reverse victim and offender.” That stood me on my head. Could the vignettes I summarized, as well as my own personal experiences, simply be examples of DARVO? Maybe we are the offenders after all. It gives me pause.

Dr. James Lamberg pointed out that incident reports can only be weaponized against physicians (or anyone else) if the institution permits it. He offered two suggestions: “1) Encourage physicians to become part of the peer-review process/committees, and 2) Improve education for these committees to focus on standard safety principles and Just Culture.” I invited Dr. Lamberg to contribute “Clinical Peer Review and Just Culture,” which appears in this issue.

Another Stanford colleague, Dr. Anita Honkanen, pointed out the fundamentally different perspectives between nurses and physicians. Dr. Honkanen wrote, “it’s a complex issue but one facet I believe is that there is something of a culture clash between the professional approach taught for nursing care and that taught for physician care. One is fairly rigid on following protocols and rules. The other encourages free thinking and action. I see this as a bit of a recipe for conflict.” Dr. Honkanen, Ms. Noelle Marton (a nurse), and Dr. Edward Damrose (a surgeon) ask the question “Can’t We All Just Get Along?” in this issue.

Dr. Emily Methangkool is the Vice Chair of Patient Safety at UCLA. She wrote, “We need these systems to improve patient safety. At the same time, how can we prevent them from targeting individuals for retribution?” In her article “Redesigned Incident Reporting Systems for Patient Safety,” Dr. Methangkool reviews common pitfalls of incident reporting systems. She offers detailed suggestions for how systems can be designed to truly enhance patient safety rather than arming disgruntled colleagues with weapons to use against each other.

One correspondent introduced me to Goodhart’s law: “when a measure becomes a target, it ceases to be a good measure” (asamonitor.pub/3zkjis0). Goodhart’s law proves to be a major issue in using metrics like incident reports to assess quality. An extreme example is financial penalty imposed on hospitals for readmission within 30 days for patients with heart failure, acute myocardial infarction, and pneumonia. That was considered to be a quality measure in the 2010 Affordable Care Act. Once this quality measure became a target, hospitals aggressively denied readmission within 30 days for these seriously ill patients, increasing overall mortality (JAMA 2018;320:2542-52). A recent viewpoint noted that Goodhart’s law invariably compromises any efforts to use quality metrics to gauge the quality of patient care (Emerg Med 2021;28:176-7). Incident reports will go up, or down, depending on whether the institution thinks that an increase or decrease in incident reports is a measure of either the quality of patient care or the robustness of their reporting systems. It likely has nothing to do with quality or robustness at all.

Another correspondent sent a 2018 psychology paper in Science (Science 2018;360:1465-7). The authors demonstrated that research subjects altered their perception to maintain an expected prevalence of events. “When blue dots became rare, participants began to see purple dots as blue. When threatening faces became rare, participants began to see neutral faces as threatening. When unethical requests became rare, participants began to see innocuous requests as unethical.”

Perhaps this explains why we have trouble assessing what constitutes bad behavior. We keep moving the goal posts to maintain a nominal incidence of bad behavior. As behavior improves, something as innocuous as remaining calm and collected during a code becomes a reportable incident (see “You Can’t Make This Stuff Up” on page 29).

It is my impression that critical incident reports are weaponized against physicians. However, there are two (or more) sides to every story. I recognized that nurses, therapists, and administrators have their own stories to tell.

My impressions may be wrong. My impressions may be “DARVO”: deny, accuse, reverse victim and offender. My impressions may reflect an unconscious change in goal posts, as suggested by the paper in Science.

I conclude with the words of Rodney King, spoken 30 years ago last May: “Can we all get along?” (asamonitor.pub/3PmGQBN). Before we enter the daily fray of clinical medicine, where patient lives, production pressure, and big egos mix and collide, perhaps we should start each day with a deep breath, a glance in the mirror, and a personal commitment to get along.