MedPageToday.com
Published: Aug 15, 2014 | Updated: Aug 17, 2014
By Joyce Frieden, News Editor, MedPage Today
Story Source

---

Action Points

• Be aware that misdiagnosis continues to be a relevant and concerning problem for physicians.
• Autopsy studies suggest that the rate of misdiagnosis has been falling over the decades, but remains surprisingly high.

---

Misdiagnosis — diagnosing a patient with the wrong disease, or with the correct one too late — continues to be a big issue for healthcare providers. And despite continuing efforts to reduce it, the problem isn’t going to go away any time soon.

The major areas of misdiagnosis — commonly defined as “a diagnosis that is missed, wrong, or delayed, as detected by some subsequent definitive test or finding” — have not changed much over the years, according to Mark Graber, MD, founder and president of the Society to Improve Diagnosis in Medicine (SIDM), who coined the definition in 2005.

For example, when looking at malpractice cases filed over the years involving diagnostic errors, “there is consistency: it’s … cancers; cardiovascular disease — including strokes, heart attacks, and aortic dissections; and infections, things like sepsis and meningitis,” he told MedPage Today.

Graber, a senior fellow at RTI International in Rockville, Md., and his colleagues who are interested in reducing misdiagnosis founded SIDM 3 years ago to draw more attention to the issue. They are hoping that the society’s work — plus an upcoming report on misdiagnosis from the Institute of Medicine — will bring more interest from the medical community.

Autopsy Findings

In trying to better define the problem, autopsies are one place researchers turn to as a source of misdiagnosis data. In 2002, the Agency for Healthcare Research and Quality published an evidence report by Kaveh G. Shojania, MD, of the University of California San Francisco, and colleagues entitled “The Autopsy as an Outcome and Performance Measure.”

“At the level of the individual clinician, the chance that autopsy will reveal important unsuspected diagnoses in a given case remains significant,” the authors noted. “Moreover, clinicians do not seem able to predict the cases in which such findings are likely to occur.”

Furthermore, they said, “While ‘newer diseases’ such as opportunistic infections have undoubtedly increased in recent decades and account for some of the misdiagnoses detected at autopsy, clinically missed diagnoses continue to include common diagnoses such as myocardial infarction, pulmonary embolism, bowel perforation, and other common conditions.”

Another study, published in 2000 by Franco Salomon, MD, of the University of Zurich in Switzerland, and colleagues, looked at diagnosis errors found via autopsy in 1972, 1982, and 1992. The researchers looked at results from 100 autopsies done in each of the 3 years studied.

The authors found that the rate of diagnostic errors was halved over the course of the study (1972, 30%; 1982, 18%; 1992, 14%; P=0.007); however, the types of diagnoses missed continued to fall into the same categories: cardiovascular diseases, cancers, and infectious diseases, with the drop in error rate mainly due to improved detection of cardiovascular disease.

A third autopsy study, published in 2008 by Fabio Tavora, MD, of the University of Maryland in Baltimore, and colleagues, looked at autopsy results from three different institutions from 1999 through 2006. The investigators found that “the largest single category of unsuspected diagnoses was pulmonary embolism followed by undiagnosed infections, including cases of tuberculosis; cardiovascular disease, including undiagnosed coronary artery disease and ruptured aneurysms; and … undiagnosed neoplasms.”

That study found an overall rate of major discrepancy involving the cause of death was 17.2%. Although misdiagnosis rates in general are hard to quantify, “we think in general practice it’s at least 10%,” Graber said.

Many of the most common diagnosis errors haven’t changed much in frequency over the years because “the symptoms are nonspecific, and too often the presentation is atypical,” he continued. Besides, he added, “Diagnosis is hard!”

Physician-Reported Errors

Studies of physician-reported diagnosis errors are less common, though there are a few. A2009 study by Gordon Schiff, MD, then of Cook County Hospital in Chicago, and colleagues, analyzed 583 physician-reported errors from hospitals nationwide and found that the five most common misdiagnoses were:

• Pulmonary embolism (4.5% of all misdiagnoses)
• Drug reaction or overdose (4.5%)
• Lung cancer (3.9%)
• Colorectal cancer (3.3%)
• Acute coronary syndrome (3.1%)

 

Breast cancer, stroke (including hemorrhage), congestive heart failure, fracture (various types), and abscess (various locations) rounded out the top 10 misdiagnoses.

Types of misdiagnosis are very specialty-dependent, noted Paul Epner, executive vice-president of SIDM. “If you ask emergency physicians for their top 10 list, it’s different than if you ask pediatricians [about] their top 10 list,” he said, adding that “many of the most common diagnoses are where we find the highest number of diagnostic errors.”

In the emergency department (ED), “the common diagnoses we miss are heart attacks, acute coronary syndrome … ruptured abdominal aneurysms, and appendicitis,” said David Meyers, MD, an emergency physician at Sinai Hospital in Baltimore. “We also miss strokes.”

For the most part, the most common ED misdiagnoses haven’t changed much since the first monographs were published on them 30 years ago, Meyers told MedPage Today.

There are a few exceptions, however. “We’re not missing as many ectopic pregnancies, because the ease of getting pregnancy tests has improved over the years,” said Meyers. “Also, childhood meningitis is very rare now — [malpractice] claims in that area have gone down significantly because kids get vaccinated against the bugs that cause that disease.”

And some newer entries have crept up as well. “Sepsis and necrotizing fasciitis were very rare in previous eras, and, now, it’s not that they’re common, but they’re much more common than they used to be,” Meyers added. “Maybe by overprescribing antibiotics, we’re allowing stronger bacteria to take over and become more prevalent. We’re also able to keep sicker people alive longer with drugs that compromise their immune system.”

Error Types Vary

In addition to specific diseases, misdiagnosis errors also vary by the type of error. For example, in the radiology department, there are two types of errors, Leonard Berlin, MD, of Skokie Hospital in Illinois, explained at an Aug. 7 meeting in Washington on diagnostic error in healthcare.

First, there are perceptual errors (not seeing what is on the film), which account for 70% of errors; and then there are cognitive errors (seeing what’s on the film but attaching minimal significance to it), which make up the other 30%, Berlin said.

Perceptual errors come in several different types, he continued. One is called “satisfaction of search,” meaning that once an abnormality is found on a radiology film, physicians become less likely to find additional abnormalities.

In one study involving 25 residents and staff members at an Ohio hospital, participants had a 75% chance of seeing one abnormality on a film that only contained one, and a 75% chance of seeing at least one abnormality on a film that contained multiple abnormalities, but the chance of seeing a second or third one dropped to around 40%.

Another type of error is known as an “alliterative” error, in which physicians who see a particular finding listed on previous radiology reports for a particular patient tend to make the same finding themselves, even when it is in error.

The overall rate of radiology errors is around 4% and has not changed greatly over the years, according to Berlin.

One barrier to improving misdiagnosis rates is that generally speaking, “physicians are lousy at reporting errors,” Robert Trowbridge, MD, of the Maine Medical Center in Portland, said at the meeting, which was sponsored by the Institute of Medicine.

Part of the reason for that is the “blaming” culture of medicine, Michael Kanter, MD, medical director for quality and clinical analysis at Southern California Permanente Medical Group in Pasadena, Calif., said at the meeting.

Kanter is running a program to try to reduce missed follow-up opportunities at Kaiser facilities; his philosophy is that instead of focusing on who is to blame for a particular error, “you need to get the physicians engaged to fix that problem, so they need to feel comfortable in the process and cooperate.”

Ideally, reducing misdiagnosis would start early in physicians’ careers, Graber said. “We’d like to do things in medical education that will make doctors better diagnosticians. If they realize they make mistakes, they are less likely to make them.”

To help achieve that goal the SIDM has established an education committee, and one of its major projects is to develop a consensus curriculum for medical students that will help them spot and reduce errors, he said.