It's now ten years later. Next month marks the ten year anniversary of "To Err Is Human", the study which suggested almost 100K people a year die in hospitals due to medical errors. It was meant to shock us all. The summary from the Institute of Medicine runs through the types of errors occurring.
Diagnostic
- Error or delay in diagnosis
- Failure to employ indicated tests
- Use of outmoded tests or therapy
- Failure to act on results of monitoring or testing
Treatment
- Error in performance of an operation, procedure, or test
- Error in administering the treatment
- Error in the dose or method of using a drug
- Avoidable delay in treatment or in responding to an abnormal test
- Inappropriate (not indicated) care
Preventive
- Failure to provide prophylactic treatment
- Inadequate monitoring or follow-up of treatment
Other
- Failure of communication
- Equipment failure
- Other system failure
One main conclusion of the report that the majority of medical errors did not result from bad apple characters, but rather systems failures. In other words, we are forced to operate in conditions that create errors and increase the likelihood of having a bad outcome. In this case, death. However, I find the list above in contradiction to that conclusion.
As a hospitalist, I am keenly in tune with many of the possible errors that can happen to my patients. I look at the list above and have to wonder how one can possibly differentiate some of these deaths as complications of illness from a preventable death. Hearst media is bringing attention to the ten year anniversary over at DeadByMistake. Their suggestion is that 200,000 people will die this year due to mistakes and hospital infections. They go on to profile thirty people who died or were left with serious disability after contacts with hospital systems. Their conclusion is that powerful forces have buried the quality movement.
I look at some of the errors, called preventable, and have to wonder what expectation we have as a nation. Do we want perfect care, or do we want quality care. One example given by Hearst was a young surfer who went undiagnosed with a skull fracture after being hit in the head with a surf board. The patient died hours later of a brain hemorrhage.
Was the failure to diagnose negligence? Or was it an unfortunate bad accident with an atypical presentation. Should every patient hit in the head with a surf board who comes to the ED get a head CT? If I tripped and fell and hit my head against a wall, should I go the the ED and get a head CT?
Rarely is there a field of medicine, a disease process, a presentation or a complication that presents itself as a diagnostic certainty. Even the gold standard therapies we order come with uncertainty. The question is, how much uncertainty are we as a nation willing to accept, and if we are not willing to accept uncertainty, how much money are we willing to pay for perfection, or at least the pursuit of perfection. And if we, personally are not willing to pay for that pursuit, how much in tax money are we willing to pay or have others pay for that pursuit of perfection. And how much in premiums are we willing to pay in that impossible pursuit of perfection.
In every day situations, failure to diagnose is a part of the evaluation process. Not a day goes by where I will admit a patient with a diagnosis of weakness, and have multiple differentials, but no definitive diagnosis. Perhaps that patient suddenly collapses of an MI. Perhaps they have a massive PE. Maybe they have a stroke. Perhaps early sepsis. Maybe adrenal insufficiency. Is my failure to make the diagnosis on presentation a sign of negligence on my part?
Failure to diagnosis is not a failure on the physician. It is a failure of the patient to present themselves within a defined box of illness. Perhaps 99 people who hit their heads on a surf board will have no complication. Perhaps 999 people with a head ache will not have an aneurysm. Are we as a nation willing to pay for a head CT on all of them, every time to exclude disease that is possible in one of them, even if that possibility is unlikely, and even if the clinical data suggests otherwise?
The expectation of perfection as a driver of medical errors is irrational. It creates irrational standards of care that perpetuate a perception of negligence when bad outcomes occur. Defining failure to diagnose as an error, equivalent to negligence, sets physicians up for perpetual failure. It is not possible to prevent delays in diagnosis. These errors will always exist.
I look at the list of errors above and I have to wonder what we expect as a nation. Many of these errors could occur through the normal course of illness and are part of the evaluation process. Because the data does not come with an answer, my job is to put the puzzle together. And that puzzle can take hours, days, or weeks or even months to put together.
What I find, often enough, is that time is one of the best diagnostic tools I have. Often times I will have conversations with daughters of patients who say they will never go back to Dr. Primary ever again because "he missed it". I stand up for them. Just the other day I had a daughter tell me a community internist, someone whom I trained with in residency was a terrible doctor because they "missed the diagnosis." I told them they were crazy. I told them they were lucky to be treated by that doctor. That he is one of the best doctors they could have. A delay in diagnosis is not negligence. A delay in diagnosis is often how the diagnosis is made. It's called watchful waiting that often ends with a declaration of the disease process. Sometimes that declaration, unfortunately, is death. It does not imply negligence, only a natural progression of disease.
Many bad outcomes occur because we are working with bad protoplasm. Patients do not present with defined data boxes. Patients do not respond with defined end points. Patient's livers do not metabolize medications the same.
When I think of medical errors, I think more broadly at systems processes. How do we get physicians to order VTE prophylaxis? How do we get everyone to wash their hands? Not how do we get physicians to make the right diagnosis on first contact. That expectation will never happen.
I have accepted the unpredictability of disease. On any given day I have patients who crash and burn. Who get transferred to the ICU for rapidly progressive organ failure. Does that constitute an error on my part or a failure on the patient's body to respond? Perhaps they developed acute heart failure as a response to their acute renal failure. Is their decompensation negligence on my part? Of course not. Is failure to prevent the CHF negligence? Of course not.
I am all for quality in medicine. I also understand the best way to generate improvements are on system wide hospital platforms. Perhaps national protocols for best practices. I agree that there is too much variation between what we know works and what we no doesn't. Hospital culture is strongly dictated by the culture generated by the people within. Start with an excellent transparent culture, and you get excellent support to improve things that don't work. Start with a combative, secretive, arrogant and obstructive culture of administrators, nurses and physicians, and you get a culture of distrust, anger, and discontent.
I think the key to quality within hospitals lies in generating national best practices for things that are
- Easy to implement
- Have a high yield to little effort
- Have low cost solutions
- Generate excitement with immediate improvement in patient care parameters.
Calling a physician negligent and in error for not ordering a head CT when one should have been ordered is not the road to quality. That is the road to irrational bankrupting of our country. Bad medicine does not kill patients. Bad policies do.
