Medical Mistakes

Doctors are not infallible.

They often make diagnostic errors. Though the incidence of such errors can be hard to measure, autopsy studies provide one metric that is hard to dispute: “major diagnostic discrepancies” were identified in 10–20% of cases (Graber 2013). Other types of studies find similar results (see Graber 2013).

In some cases, doctors are systematically mistaken about important medical facts. In one study, gynecologists were asked about the likelihood that a woman who has tested positive on a mammogram actually has breast cancer. They were presented with four alternative answers, one of which was correct, and they were given the statistical facts needed to calculate their way to the correct answer, so the task should have been easy.

Only 21% chose the correct answer, which means that the doctors did slightly worse than we would expect them to do if they chose the answer at random (Gigerenzer et al. 2008).

Should We Be Worried?

These facts are troubling. When doctors are wrong, the consequences may be severe. It is tempting, therefore, to react with a scathing criticism of doctors and medical education.

In part, this is warranted. The human tendency to crash and burn when faced with problems that require Bayesian reasoning, which is what foiled the gynecologists in the study above, can be corrected with proper teaching (Gigerenzer et al. 2008). Diagnostic errors that result from cognitive biases could be removed using formalized procedures such as checklists (Ely et al. 2011).

However, as long as doctors remain human, errors will occur. Moreover, since medicine is a field characterized by risk and uncertainty, focusing on individual blame for mistakes runs the risk of focusing on outcomes rather than the procedure leading to those outcomes.

Malpractice

Malpractice suits, which are the legal manifestation of such a focus on individual blame, are more likely to be filed when outcomes are bad, such as when someone dies because of a delayed diagnosis of cancer. The likelihood of the filing (in the case of diagnostic errors) increases with the severity of the outcome (Tehrani 2013). But a bad outcome does not automatically entail any error of medical judgment.

Any positive diagnosis involves a risk of overdiagnosing a healthy patient. Any negative diagnosis involves a risk of underdiagnosing a patient with a serious ailment. As both overdiagnosis and underdiagnosis can lead to serious harm, the trick is to balance the risks according to their costs and benefits, but there is no way to completely avoid the risk.

The Costs of Blame

One serious cost of blaming doctors for mistakes is the phenomenon known as defensive medicine. The harms resulting from underdiagnosis and undertreatment are usually much more spectacular and easy to understand than the harms resulting from overdiagnosis and overtreatment. This means that doctors can minimize the risk of being sued for malpractice by erring on the side of the latter. According to one estimate, defensive medicine costs the US between $650 billion and $850 billion annually (jacksonhealthcare.com).

Another significant cost of the focus on blame is the harm that befalls doctors. Being a physician is stressful. Depression and burnout are common, and the suicide rate among doctors is frighteningly high—41% higher than average for men and 127% higher than average for women (Schernhammer 2004). A likely contributor to this is the blame and guilt associated with making mistakes, or even with making completely justified decisions that, because they involve risk, happen to result in bad outcomes.

Less obviously, focusing too much on the responsibility of the physician obscures the fact that the institution of modern medicine tends to marginalize and overlook a significant healthcare resource: the patient.

The Doctor as Authority

Modern healthcare is still very much an authoritarian institution, where patients come in and are told what to do by the Olympians in white coats. Even title of “patient”, which you automatically gain once you enter the system, denotes passivity, someone “to which something is done” (oed.com). Doctors have access to a special set of skills and knowledge, which is demarcated by high social status and pay and often romanticized in popular culture. To a patient, the doctor is an unapproachable expert, one to which you listen, sometimes literally, on pain of death.

It is no wonder, then, that most of us are afflicted by what Wegwarth and Gigerenzer call the trust-your-doctor heuristic, which is the decision-making rule most of us follow in matters regarding our medical needs: consult your doctor and simply follow her commands (2013).

Because the gap in relevant knowledge between physician and patient is assumed to be astronomical, the responsibility for arriving at the right conclusions is placed squarely on the shoulders of the physician. Though the 20th century has given us the doctrine of informed consent, an institution intended to protect patient autonomy; the underlying picture is still that of a commanding doctor and consenting patient. By being bound by this framework, we risk losing out on the resources patients could bring to bear on solving their own medical problems.

Bridging The Gap With Google

As Andreas Eriksen discussed in his excellent post a couple of months ago, the advent of the internet and Google has increased the information easily available to the average person by several orders of magnitude. This means that the knowledge-gap between doctor and patient is less absolute.

No doubt it’s true that a doctor with Google is better suited to diagnose and propose treatments than most patients with Google. However, it is also true that most patients spend a lot more time thinking about their medical condition, their symptoms and how it affects their life than their doctors do. A doctor can’t spend hours researching on Google every consultation, and they cannot routinely monitor their patients as they go about their daily lives.

Every patient should be considered an expert on her circumstances of life. More and more, the medical knowledge they can muster through the use of Google and other resources should be taken seriously. When combined, these two insights make a good argument that a healthcare model based on the idea that responsibility and authority in medical matters should belong solely to the physician is obsolete.

Taking The Patient Seriously

The involvement of patients in medical decisions should not be regarded merely as matters concerning the protection of their autonomy but as an important part of improving the medical decisions themselves. Through the last couple of centuries, medicine has seen a gradual shift towards a focus on the patient in several ways, through informed consent and, more recently, the ideal of shared decision-making. This is a trend that should continue.

Doctors are sometimes wrong. Patients are sometimes right. On an authoritarian model, the instances where these situations overlap will result in doctors overriding their patients’ correct judgments with their own mistaken ones. In an ideal situation, a patient’s correct judgment should correct the doctor’s mistake. Taking the patient’s resources to make medical decisions seriously should be a step towards achieving this ideal.

Litterature

Ely, John W., Graber, Mark L. & Croskerry, Pat. 2011. “Checklists to Reduce Diagnostic Errors”. Academic Medicine. 86 (3).

Gigerenzer, Gerd, Gaissmaier, Wolfgang, Kurz-Milcke, Elke, Schwartz, Lisa M. & Woloshin, Steven. 2008. “Helping Doctors and Patients Make Sense of Health Statistics”. Psychological Science in the Public Interest. 8 (2). 53–96.

Graber, Mark L. 2013. “The incidence of diagnostic error in medicine”. BMJ Quality & Safety. Online First.

Schernhammer, Eva S. & Colditz, Graham A. 2004. “Suicide Rates Among Physicians: A Quantitative and Gender Assessment (Meta-Analysis)”. The American Journal of Psychiatry. 161. 2295–2302.

Tehrani, Ali S. Saber, Lee, Hee Won, Mathews, Simon C. Shore, Andrew, Makary, Martin A., Pronovost, Peter J. & Newman-Toker, David E. 2013 “25-year summary of US malpractice claims for diagnostic errors 1986–2010: An analysis from the National Practitioner Data Bank.” BMJ Quality & Safety. 22. 672–680.

Wegwarth, Odette & Gigerenzer, Gerd. 2013. “Trust Your Doctor: A Simple Heuristic in Need of a Proper Social Environment”. In Simple Heuristics in the Social World. Hertwig, Ralph, Hoffrage, Ulrich & The ABC Research Group. Oxford University Press.

Authors comment: This post was written after binging a season of “Doctors vs. Google” (originally: “Hva feiler det deg”) the Norwegian TV series that pits a team of people without medical education, but with access to google, against a team of doctors without google. The task: to correctly guess the diagnosis of people based on a brief anamnesis and some rounds of questioning. Andreas mentions the show in his post, which is where I found out about it, and it is worth watching, as it’s both entertaining and a fair showcase of the potential (and the limits) of what patients can achieve with the help of google. Though the doctors often come out on top, this is probably in part because in the weightiest task point-wise, a time constraint means that there is almost no time to use google.

Ainar Miyata-Sturm is a PhD student at the Centre for the Study of Professions (SPS), and part of the project Autonomy and Manipulation: Enhancing Consent in the Health Care Context. He is also the editor of Professional Ethics.

Photo: Sonja Balci