Ruling Out the Wrong Diagnosis

Although misdiagnosis may kill up to 80,000 annually—more people each year than firearms and motor vehicle accidents combined—you won’t find it on the list of the country’s leading causes of death.

Most Americans don’t realize how frequently well-meaning medical providers get it wrong. Just last year Johns Hopkins researchers found that one in 12 ICU patients die from something other than what they were being treated for. Aside from a handful of instances covered by the national media, misdiagnosis hasn’t received much attention from the public or the medical community. One such tragedy is the death of Rory Staunton, a 12-year-old boy who was treated for an upset stomach and dehydration instead of sepsis, a severe response to infection that requires immediate treatment with antibiotics. To make a complex diagnosis like sepsis, a doctor may need to assess a couple dozen different factors.

One solution is to arm clinicians with better problem-solving tools and improved IT systems to help them identify possible diagnoses faster and more accurately, especially for conditions that are commonly confused or missed altogether. This week at Johns Hopkins, a team of researchers shared some promising results about a new way for emergency medicine doctors to accurately detect stroke in patients with dizziness.

That team, led by Armstrong Institute faculty member David Newman-Toker, found that a portable bedside device that measures eye movements could accurately predict whether a patient’s dizziness was caused by a life-threatening stroke or a less time-critical issue like a balance disorder. The machine could be a time saver for busy emergency department doctors who often rely on brain imaging, typically a CT scan—an expensive and inaccurate technology for this diagnosis.

“We’re spending hundreds of millions of dollars a year on expensive stroke work-ups that are unnecessary, and probably missing the chance to save tens of thousands of lives because we aren’t properly diagnosing their dizziness or vertigo as stroke symptoms.”

-David Newman-Toker, lead author

If their results can be replicated on a larger scale, the device could one day be the equivalent of an electrocardiogram, the go-to tool to rule out heart attack in patients with chest pain.

For more details, read a press release about the study or watch David’s interview with CBS This Morning.

Among other efforts on the diagnostic errors front at Johns Hopkins, an international conference was hosted on our campus this past fall.

Congratulations to David and the whole team on their exciting and important work to save or improve the quality of life for the 20,000 to 30,000 patients who suffer an overlooked stroke each year.

Director of the Johns Hopkins Medicine Armstrong Institute for Patient Safety and Quality, Peter Pronovost, MD, PhD is a practicing anesthesiologist and critical care physician who is dedicated to making hospitals and health care safer for patients. Pronovost has chronicled his work in his book, Safe Patients, Smart Hospitals: How One Doctor’s Checklist Can Help Us Change Health Care from the Inside Out. His posts appear occasionally on THCB and on his own blog, Points from Pronovost.

Spread the love

11 replies »

  1. It has been my experience that when a doctor can’t figure out what’s really wrong with your arms, legs, shoulders or back, s/he sends you to physical therapy. Buff and turf…

  2. Given that the technology is evolving, it is also important to educate the patients carefully to build knowledge and self awareness about diseases and risk factors to avoid wrong diagnosis. Diagnosis is one of the most important phase. Patients pay a steep price for this medical mistake.

  3. The following are two review articles describing new electronic tools that utilize electronic-nose (e-nose) devices to check and help validate diagnoses in order to greatly reduce the incidence of misdiagnoses of diseases and disorders for which biomarkers and/or complex metabolite mixtures have been identified and used to build electronic aroma detection (EAD) reference databases. Many human diseases and disorders now can be effectively identified from samples of the human breath using trained, portable e-nose devices. These new tools will become part of standard clinical methods used to confirm many types of diagnoses in the future.



  4. I really liked the article – and think it’s good that Dr. Pronovost is doing something about the problem of misdiagnosis and suggesting things to do to follow up on it. I look forward to hearing his next steps. I have written some blogs on patients having a voice and being a decision maker in their treatment and I also think this is an important aspect in care that has been presented in a couple of other comments.

  5. Dr. Pronovost is dead on…but if anything the 80,000 number is low. It does not take into account nursing home deaths due to misdiagnosis. When you include the new research just published in JAMA Internal Medicine, you come away with 150,000 people harmed or killed every year because of misdiagnosis.

    Is there any other profession that could make that many serious preventable mistakes and not be accountable? Blame is not going to solve anything, but ignoring the problem won’t either.

    Misdiagnosis is a third rail issue in medical schools, hospitals, OP clinics, the CDC, NIH, the media and Congress. A leading cause of death in the U.S. (and the world) has been pretty much ignored by everyone.

    In the recent paper by Hardeep Singh, MD, MPH, about this very thing he and his colleagues suggest letting patients help by including them and their families in the process. Engaged patients have a lot of interest in getting the diagnosis right.


  6. Looking for a blame line is not the answer. I once worked for an organization that had a problem solving culture that insisted that all possible causes needed to be identified prior to moving to the next step in solving the problem. IT can be used to accomplish this task. The physician needs to be the instrument to gather the data, systems need to provide the possible causes, and prioritize the possibilities and then the physician needs to be the final decision maker. That bright kid who could memorize and figure things out is just not enough and never has been. In a perfect world the final step in the problem solving process would be an after action analysis that creates in some instances will correct an error and document the outcomes and rate the physician.

  7. The problem sometimes is due to doctors miscalculations or might be misdiagnosis. But we cannot blame them all just much knowledge to acquire more skills and learn more vigorously from experts. Seminars and other workshop must be develop too not only that some factors affecting similar to this are stress among doctors… I don’t know if you agree with me but that’s what happens sometimes stress and being too busy handling more two patient at the same time.

  8. Scott, I like your comment. We are often looking for an exciting new innovation, but we could do a lot of good if we just learned to implement the basics more consistently.
    Follow-up the test result. Contact the family.
    It sounds simple but it’s hard to do well…but that’s not because we need a tech breakthrough, it’s because we need an implementation breakthrough.

  9. In the case of Rory Stauton, if the NYTimes article is accurate, Rory was discharged from NYU Langone before the his last test result indicated a substantial infection and that sepsis was a threat. Then no one was notified of this last test result, neither his parents nor his family physician.

    While the technology you describe may have helped prevent Rory’s suffering and death if present and used properly, it seems that there is something more fundamentally wrong here that goes beyond the infatuation with technology.

    The sepsis tool indicated Rory had two symptoms of sepsis; three symptoms would have indicated he needed critical attention. In this case, why not just keep him until the last test results came back and then decide how to proceed given the possibility of sepsis?

    The root cause is not a technology issue, although in all fairness, technology can enable better care if created and used properly. The root cause here is that the culture and care processes at NYU Langone did not support the comprehensive care Rory needed at that time.

  10. While I agree that this is exciting and important work, we shouldn’t lose sight of the power of listening to the patient as well. Often the diagnostic error would be avoided if (we) physicians were better listeners and realized that patients are the “experts” about their own symptoms.

    • I will not be surprised,if wrong dignosis incidence increases as years go bye.Because ,the young generation of medical doctors depend on laboratory findings more and more.Yes,they do not listen their patients’s story.The pres-cription is ready in their brains before diagnosis and the diagnossis is written in the last sentence of a report. Treatment?That is another and more important subject!

Leave a Reply

Your email address will not be published. Required fields are marked *