Synopsis: Studies estimate that about five percent of diagnoses are wrong, leading treatment down the wrong road. Experts discuss why misdiagnoses occur, and a new Institute of Medicine report on how they might be prevented.

Host: Reed Pence. Guests: Dr. Mark L. Graber, President, Society to Improve Diagnosis in Medicine and Senior Fellow, RTI International; Dr. Lewis Levy, Senior Vice President of Medical Affairs and Chief Quality Officer, Best Doctors; Helen Haskell, President, Mothers Against Medical Error

Links for more information:


Reed Pence: Getting the right diagnosis is the first step in treating any illness. But it’s not as easy as many of us assume. We figure that the big question is choosing a treatment, when in reality, figuring out what’s wrong can be more difficult…and more important.

Graber: Treatment begins with the correct diagnosis so if you have the wrong diagnosis then all the treatment you get is completely wasted.

Pence: Yet, Dr. Mark l. Graber, President of the Society to Improve Diagnosis in Medicine and a Senior Fellow at RTI International, says nobody really knows how often doctors misdiagnose illnesses.  

Graber: The sad fact is that there isn’t any hospital or healthcare organization that’s actually measuring the rate of diagnostic error right now. So I don’t know my own; I don’t know what it is for my hospital. Nobody really is good at measuring or knows how to measure. So all we have are estimates from research studies, which suggest it’s probably one in ten diagnoses are wrong. Another study from a friend of mine, Hardip Singh, found that one in twenty primary care patients will experience a diagnostic error every year. So, if you do the math on that if your odds are one in twenty every year and you live to be eighty that’s pretty close to a hundred percent.

Levy: Because of the prevalent nature of misdiagnosis it is likely that most of us will experience at least one diagnostic error in our lifetime, sometimes with devastating consequences.

Pence: That’s Dr. Lewis Levy, Senior Vice President of Medical Affairs and Chief Quality Officer for Best Doctors. He notes by a new report released by the Institute of Medicine, an authoritative adviser on healthcare to the nation. The report finds that misdiagnosis is a neglected problem despite its importance.

Levy: I believe that the IOM report has clearly shown a very bright spotlight on this issue. Misdiagnosis is prevalent and there is little work being done right now at addressing this widespread problem.

Graber: The Institute of Medicine got us all thinking about patient safety in 1999 with a landmark report titled To Err is Human and that really woke up the hospitals and healthcare organizations and physicians. We’ve all been focused on patient safety since then, working on all sorts of things — falls, medication errors and wrong side surgery and hospital acquired infections. But essentially all of them were related to treatment. There was virtually nothing in that report on diagnostic error. There was no discussion, no recommendations, so this report breaks new ground and really is the missing chapter.

Pence: But it’s a chapter Helen Haskell has been living with ever since the IOM put out that report a decade and a half ago.

Haskell: Fifteen years ago my fifteen-year-old son died in the hospital from an undiagnosed perforated ulcer, which was caused by pain medication following elective surgery. He had really extreme symptoms, but his caregivers were so fixed on the idea that he had constipation that they really couldn’t put it together. So no attending physician was ever called in to see him. The symptoms were getting worse. It was really a classic presentation of a patient going into shock from blood loss and infection. He died of blood loss and peritonitis.

Pence: Today Haskell is President of Mothers Against Medical Error, a patient safety advocacy group. She was also a reviewer of the New Institute of Medicine report. Haskell says each kind of provider has unique problems that may prompt misdiagnosis. Primary care providers, for example, are very different from major medical centers, which she says can sometimes be simply too big.

Haskell: The care is so fragmented and there’s so many people working on it and so they do tend to stick with the first thing that goes into the chart. To change a diagnosis means a major change in what you’re going to do. When people have a lot of patients that’s very hard to just turn the ship around and decide that you’re going to drop everything and change the plan for this patient. So, there’s a lot of, I think, wishful thinking that goes into sometimes sticking with the initial diagnosis just trying to make the symptoms fit the diagnosis.

Pence: However, if estimates are correct and doctors misdiagnose five percent of the time, that means they’re correct 95 percent of the time. And given today’s medical world, some experts are amazed they do that well.

Levy: Too often individuals view diagnosis and misdiagnosis as issues of individual physician competency. I think that one of the key findings of the report is that a lot of the root cause of misdiagnosis has nothing to do with individual physician competency. In fact, it makes the point that doctors in general are actually very good at making diagnosis. So what causes the misdiagnosis? A lot of this has to do with the way in which healthcare is being arranged and being practiced. There are very practical steps that one can take around the structure of medical care that I believe will help to dramatically reduce the rate of misdiagnosis in the US.

Graber: Doctors are human and we make the same mistakes in medicine that we make in our every day lives. We kind of jump to conclusions and we assume we know what’s going on when we may not. We think we’re in one context and we’re really in another. It’s kind of like when the pilot thinks there’s no ice on the wings, but there is and then there’s an airline accident. So, the cognitive errors are very important. The other big category are system related things; our health care systems are immensely complex and becoming more so everyday –communication problems, breakdowns in coordinating care, getting the right test to the right person, having expertise available when you need it, supervising the trainees — the list goes on and on. There are all sorts of ways that system problems can produce a diagnostic error.

Pence: In fact, the first reason for misdiagnosis cited by all our guests is the fragmentation of information in the system.

Levy: Even in the day of the electronic medical record, the reality is that for many patients their medical information is still contained in several different areas that are not connected, so an individual may have medical information that is in the hospital computer system and another set of information in a variety of outpatient medical record systems. So, there is this fragmentation and we do not have this universal superhighway, if you will, of medical information. I can give you an example: my wife had some sort of meningitis that she acquired after a mosquito bite on a trip a couple years ago. We saw our primary care doctor, our emergency room, a local neurologist, she had some tests done, and some imaging. Nobody could see any of the results from any of these other sites even though we’re in a medically sophisticated area of the country. So the biggest problem is we’re just not connected and if you get your care at more than one site it’s gonna be difficult or impossible for that information to be shared appropriately to give you high quality care.

Pence: Doctors order all kinds of tests in an effort to reduce the chance of misdiagnosis. And Graber says the tests themselves do a good job.

Graber: The laboratories actually the one place in the hospital that’s achieved the highest levels of quality so the error rate in the analytical phase of testing, you know, just measuring my potassium, for example, is like one in ten thousand or one in a hundred thousand so somewhere in that range because of the instrumentation is so sophisticated they have that down pat. The problems arise in all the steps before the actual test is done and afterwards so the doctors don’t always order the right tests or the specimen is incorrectly obtained or something happens to it on the way to the lab and then there’s problems getting the results. Fifteen, twenty percent of laboratory tests don’t make it to the provider or aren’t acted upon or appreciated in a timely manner.

Pence: But tests go only so far, and doctors may be tempted to rely on them too much. Sometimes, good old-fashioned doctoring works better. Levy recalls the advice of Sir William Osler, often considered the father of modern medicine.

Levy: Listen to your patient. He is telling you the diagnosis. This really emphasized the absolute importance of taking a good history. Fundamental to that is really having the time to spend. Doctors do need to take the time to establish a great rapport with their patients, so that patients feel comfortable in sharing detailed medical information, that they feel as though the doctor is focused on them.

Haskell: I think a good history and physical examination of a patient can probably tell a doctor or another healthcare professional more than a test. Tests are great for confirming suspicions and sometimes they are definitive, I mean, you want to know if the arm is broken or not.

Pence: However, Haskell admits if avoiding misdiagnoses requires a good history and a good physical, that puts a lot of responsibility on the patient.

Haskell: That’s one of the things that we really have to do is to be very clear and concise when talking to a doctor. Particularly if you’re in an outpatient setting you’ve got seven minutes. If you’re in a hospital you probably have even less. You really have to have your thoughts organized when you talk to a doctor so that you can explain to him what your main concern is and what your worried about, how problems came about and what you’ve done about it, what you think. It might be, I mean, patients really have to take it upon themselves. We all wish they didn’t but communication is a two-way street.

Graber: The report from the Institute of Medicine envisions patients becoming much more active partners in a diagnostic process, which we think would be very valuable and it would help provide a safety net for catching diagnostic errors. So, if patients had a better understanding that we’re always just playing the odds and [if they] knew how to get back if their symptoms didn’t resolve or didn’t respond to treatment, we think that would be very powerful in terms of preventing diagnostic error, but there’s a little tension there because we don’t want to be a burden to patients. For some patients that would be perceived as a burden and it’s absolutely correct. I think the medical profession, it’s their responsibility, it’s our responsibility as physicians and leaders of healthcare organizations to improve the system. We shouldn’t put that burden on patients, but we’re just not there yet. So, until we get there I think it’s very important for patients to be as involved as they’re willing or interested to be.

Pence: To do that, Haskell says patients need to stay skeptical.

Haskell: One of the things that I tell patients is that you have to diagnose not just the doctor but yourself. You have to always think what else could this be? Have we taken every thing into account? Keep pushing yourself  because its very easy to be persuaded and to come to a sort of a consensus with your doctor and yourself so that none of you are looking for what might be obvious red flags if you stepped back and took a second look.

Pence: Second looks are important. All of our experts say patients should never hesitate to seek a second opinion, and to find a doctor to give it who has the most expertise possible. But while second opinions are a good check on care, Levy says the biggest changes that could help avoid misdiagnoses are systemic. Care needs to be coordinated. And the rapid pace of medicine needs to be slowed down.

Levy: Some of the most important steps that can be done is, 1) a restoration of time that patients have to spend with their doctor in discussing their history so that the physician can take a prudent approach to that patient ordering only the required laboratory tests and then 2) making sure that there are good systems in place for the follow-up of those laboratory tests so that there are no laboratory tests that are going unread and uninterpreted.

Pence: You can find out more about the Institute of Medicine Report on Misdiagnoses at More information on the Society to Improve Diagnosis in Medicine is at For more on best doctors, see And for more on Mothers Against Medical Error, visit mame-moms-online-dot-org. You can find out more about all our guests on our website, I’m Reed Pence.

Join the discussion

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.