Building a Better Health Care System: Should We Be Tracking Misdiagnosis?

If you study misdiagnosis you realize how often patients get the wrong diagnosis.

But what do expert doctors think about how often it happens? And what do they think can be done to address it?

We wanted to find out so we partnered with the National Coalition on Healthcare to conduct a landmark, nationwide survey. We surveyed 400 cancer specialists from our Best Doctors database – and the findings were provocative.

The survey, “Exploring Diagnostic Accuracy in Cancer: A Nationwide Survey of 400 Leading Cancer Specialists,” focused on what doctors believe to be the most significant barriers in efforts to accurately diagnose cancers; the types of cancer they believe are most often misdiagnosed; and the tools and improvements they most need to combat misdiagnosis.

One of the most surprising findings was on how often doctors believe misdiagnosis happens. While published studies show that misdiagnosis occurs in about 15-28% of cases, the large majority of doctors we surveyed thought it happens in less than 10% of cases. At the same time, doctors recognized that the root causes of misdiagnosis were very prevalent – fragmented medical information, disparities in experience among pathologists and other factors.

So – how to explain the difference in doctors’ perceptions and the published research? I think it is because there is no systematic feedback loop for doctors letting them know of inaccuracies in their care. If you diagnose someone and they go on to get treatment someplace, and it’s later discovered that a diagnosis wasn’t exactly right, the original doctor may never find out about it. If you don’t hear about it, you can’t be blamed for thinking this problem is rare. It also means you miss out on the opportunity to improve the quality of care that these cases represent.

Another interesting point. Doctors reported that, regardless of how often they thought diagnostic inaccuracies happened, it is a problem that needed more attention from policy-makers. As NCHC President and CEO John Rother observed, “Not enough is being done on the state and federal policy end of things to acknowledge and firmly address this critical issue. Given our current health care climate and challenges, as decision-makers become more aware of the frequency of misdiagnosis and the enormous costs associated with it, they have a sizeable opportunity to make diagnostic accuracy much more of a ‘front and center’ issue in health care.”

Here’s to that promising thought.

Evan Falchuk is Vice Chairman of Best Doctors, Inc., where this post originally appeared. Prior to joining Best Doctors, Inc., in 1999, he was an attorney at the Washington, DC, office of Fried, Frank, Harris, Shriver and Jacobson, where he worked on SEC enforcement cases. This post originally appeared on Best Doctors, Inc.’s See First Blog.

10 replies »

  1. Funds could be raised in a variety of ways – per physician license and/or other license (replacing malpractice insurance) with the state medical societies or whatever other organizations still handling oversight/discipline. The advantage is getting rid of the “fault” issue and sticking with “causation” – recognizing bad things happen; even physicians sometimes do stupid things. By bringing errors into the open, more can learn from a lot more experience making everyone better. And healthcare providers wouldn’t have to be obsessed with getting sued; it would look a lot more like quality improvement activities.

  2. You will never know who the best doctors are, because the best ones ply their skills quietly and efficiently. The errors in diagnoses are extensive.

    But, do not worry. For the big ones, the hospitals and the HIT vendors alter the electronic medical record to provide the illusion that they knew what they were doing.

  3. Sandra, the vaccine program is paid for through a .75 cent sales tax on vaccines/ per disease. I’m thinking the risk is quite low.

    New Zealand’s health system is about 75% government run with the injury compensation paid for by government (taxpayer) funds.

    I’m not sure how you see a similar system being funded here and if it would be tied to doctor oversight and disipline or not?

  4. I believe New Zealand has implemented such a system – I believe there are premiums paid by everyone (regardless of risk) and compensation awards are determined by a panel of experts. I haven’t looked into it recently so I’m not certain of the details. But another example is our vaccine compensation system here in the U.S. If a vaccine injury is found by the vaccine injury court, a specialized Federal Court of Claims court, the award is given out of a fund established by taking a few cents out of each vaccine administered. The balance of that fund is quite high, I believe. There’s more information on their website.


    They pay lawyers based on amount of work, rather than as a percentage of an award, so there is less incentive to go for a large award. Plaintiffs can “opt out” of the decision (for constitutional reasons), but it’s rare because the decisions of this courts is given significant weight. I think a similar program could be developed if there’s the political will for it.

  5. Sandra, how do you envision this would work? Where would the money come from for the fund? Who would determine what was no-fault compensatable? In a single-pay system it’s easy to pay for the medical costs of errors, but the non-medical costs are still litigated.

    In our for-profit private system it’s harder to pay no questions asked for medical consequences.

  6. A no-fault approach to compensation for patient harm would go a long way to solve the problem of (1) not knowing how much misdiagnosis occurs; and (2) reducing medical malpractice litigation costs. Providers would not fear disclosing data about harm if they did not personally pay for it, and the information could be used to improve care for everyone.

  7. “consistent with the cognitive dysfunction caused by the user unfriendly HIT devices.”

    You have any scientific DATA inferring that?

    How would one even design such a study? Longitudinally sufficient, stratified by EHR product, control group of paper chart dx’s? Not that we shouldn’t try.

  8. Given that doctors think they should not be held legally responsible for anything I’m surprised they returned a less than 10% misdiagnosis rate.

    Should misdiagnosis be held to the same error standards as say surgical errors?

  9. Misdiagnosis is in the rise consistent with the cognitive dysfunction caused by the user unfriendly HIT devices. Talking about oncologists, their specialty is pushing chemo and they are so focused on that they forget the patients have other body parts and organ systems that get diseased.