When things go wrong in medicine, as they invariably do, we try to figure out what went wrong, and why. We try to learn if there’s anything we could have done better and what we should do next time.
It used to be, in the days of the giants, that the physician responsible for the patient with the bad outcome presented the case during a morbidity and mortality (M&M) conference. It was a formal presentation before the whole department, with fellow colleagues in the audience discussing the error. When I was a medical student, the surgeon got up and talked about the surgery, the complications and what happened to the patient. To some this might sound terrifying — and it can be. It isn’t easy to talk publicly about how your error harmed someone. Not how it might have, but how it actually did. And not only that, but to discuss the case, which, in retrospect, and framed the right way, makes the error seem so obvious that any medical student should have been able to prevent it.
On the other hand, it is useful to keep the perspective that, however anguished the physician may be, it is ultimately the patient who has paid the price.
The rehashing of an M&M does two important things. It causes a Pavlovian sting to the presenting physician, so the next time they’re faced with a similar situation they’re more likely to recall this case and consider the possible complications. And it also serves the audience. Errors happen to everyone. We are, after all, human, and the reaction amongst the listeners is often: There but for the grace of God go I. It is an opportunity for everyone to examine how this particular error could have occurred on their watch, what they could learn from this experience.
Even in medicine, where we are continually processing statistics and probabilities, nothing is more powerful than the emotional anecdote. Hearing about the hundreds of thousands of people who die due to medical errors isn’t as effective in changing behavior as seeing the harm you’ve done to the patient right in front of you.
When I was a resident (a few years after the days of the giants), our medical M&M conferences were still a big deal. The chief residents prepared cases, they invited discussants from other departments, they framed the discussion and made the slides. People from all departments showed up, and when the case, as is common, included errors by multiple people, each person said loudly and clearly, yes it was my fault too. I should have followed up. I should have come earlier. I should have recognized the issue. I should have looked at the image. I should have asked better questions. I was to blame.
One of the most memorable M&Ms from my residency was run by a very capable chief resident. The medicine resident who was asked to present the case started with a disclaimer that he wasn’t involved in the care himself. He wanted to be clear that he was just asked to read: “I’ve been presenting these a lot lately, and I didn’t want you to get the wrong idea.” The case was about a man who, by all guidelines, needed to have his blood thinned. But this treatment was complicated by the man’s inability to come to scheduled visits or get blood tests to monitor the medicine levels. Blood thinning and clotting is a particularly vexing problem in medicine, as we’re always walking the fine line between trying to make sure our patients don’t die from massive bleeding on the one hand or massive clotting on the other.
It was a slow motion disaster. The man would come to clinic after missing countless visits. His blood thinner levels were always too high, or too low, he had falls and was an alcoholic. He kept drinking and he kept falling, and at his infrequent clinic visits his doctor would advise that he stop the former to prevent the latter. He had other factors that were red flags, easily recognizable to the clinicians but not included in guidelines per se. Not only was he an alcoholic but he lived alone. He had had strokes in the past, so between that and his chronic alcoholism one would have guessed some degree of cognitive impairment. The language the chief resident used to describe the continued failure by the clinic provider to stop the inevitable was particularly harsh. Finally, the man died from a head bleed after a fall at home. It was obvious it was going to happen to everyone. It was obvious to the clinic provider who had tried, and failed, to prevent it from happening.
The ensuing discussion included suggestions such as only refilling the blood thinner medicine a week or a month at a time. Many acknowledged the system factors at play. There were more issues (alcoholism, inability/unwillingness to adhere to recommended therapy) that were not fixable by a single provider on the occasional clinic visit.
The chief resident, at some point, revealed that the patient was hers. She was, in fact, the clinic provider she had spoken so harshly about. She started crying as she talked about what she thought she could have done better. But she also wanted to make a point about how M&Ms were changing. How as doctors we no longer stood up to admit our failures. Instead we were developing ever better ways of blaming the system.
One lesson that stuck with me after that M&M was the observation by a senior doctor that in the face of conflicting imperatives, our imperative first and foremost is to do no harm. Thus, he observed, if this patient had died at the same time, at home, but of a stroke (a complication of his original disease that required the blood thinner), rather than of a head bleed (a complication of the medication we were prescribing to prevent a stroke), the result would have been more desirable. “I don’t know if it really makes a difference to the patient, but it makes a difference to us. I think it is better if we’re not the ones that killed him.”
M&Ms remain a requirement of medical training programs. It isn’t the disappearance of M&Ms that I am lamenting, per se, but rather the shift in focus. At the last M&M I attended, half the time was devoted to the “Swiss cheese model of error,” which holds that for any error to occur, there are multiple layers of safety that it must have slipped through. While this concept may be new to some in the lay public, addressing system errors rather than blaming the individual has been well-accepted in all industries and has become ubiquitous in medicine in the last decade. So much so, that it is hard to find someone today who will say: “I did something wrong, and that is why someone suffered.” It is so much easier to talk about the system failures that caused my error.
Yet, some of the errors that were discussed at the M&M mentioned above were cognitive errors. Anchoring bias, confirmation bias, heuristics, availability bias, which amount to the fact that we choose a diagnosis based on what we recently (or memorably) saw and then dismiss information that doesn’t fit with the diagnosis, while actively seeking only elements that confirm our initial impression. While we can talk ad nauseum about sleep deprivation, and failure of communication, and “the system,” it behooves us to recognize that the system we’re often talking about is our own brains. But is it enough to acknowledge that we, like all humans, are susceptible to cognitive errors? I submit to you that it is not.
I wonder if this shift in the culture of medicine has eroded something fundamental about the profession and our personal responsibility therein. I will be the first to admit that any medical outcome does not hinge on one person alone. But even if you consider the Swiss cheese model of error for a particular case, there were multiple people who made the wrong decision, who had failed to see. In that model, we didn’t need everyone to figure it out — we only needed one person to prevent it! And what was particularly jarring during this recent M&M was that the presenter seemed far more absolving of errors as a “system” problem than was warranted.
And maybe this is where I am particularly vexed by the “system” blaming. It seems to absolve everyone of their sense of what they could do better. We figure, it’s natural that we’ll get it wrong sometime. And in medicine, as in life, outcomes and decisions are linked, but not as tightly as one might hope. Sometimes the outcome is good despite the wrong decision, and sometimes the outcome is bad despite a good decision.
It is time to bring back M&M. Not just as an acknowledgement of life in an imperfect world, where the system fails us daily, and a cataloging of all the ways in which it does. But rather as a call to individual action. What can I do in this system? No single intervention is a panacea, no decision support tool will fix the problem, no electronic medical record, no provider order entry system, no computerized alert to acknowledge potentially dangerous drug interactions.
But we can begin to think through which system interventions will bring more benefit than harm, and which individual techniques to combat the acknowledged cognitive limitations would be most effective to implement personally. Because while we may not be able to fix the world, or medical care for all patients, can we at least say that for our patients, we did well? And while we know we will never attain perfection, we will at least continue to aim to do better.
Denitza Blagev is a pulmonary physician who blogs at mybetterdoctor.