The researchers concluded the "inattention" and "neglect" paid to these mistakes have resulted in unacceptable harm to patients, and they projected the errors will probably worsen as health care becomes more complex.
"It's probably one of the, if not the, most under-recognized issues in patient safety," said Dr. Peter Pronovost, director of the Armstrong Institute for Patient Safety and Quality at Johns Hopkins. "Much of the harm that we once labeled as inevitable we're now seeing as preventable."
The report listed several examples of devastating diagnostic mistakes.
• A 51-year-old woman with a family history of heart disease repeatedly asked her doctor's office to refer her to a cardiologist for a stress test. Three months after her initial request, on the day of her appointment, she died because of significant coronary artery disease.
• A doctor mistook a blood clot in the lungs of a 33-year-old woman for an asthma attack, leading her to her death.
• An urgent care clinician misread an X-ray and diagnosed a 55-year-old man with an upper respiratory infection instead of pneumonia. He died as a result.
• Doctors at a trauma center decided not to perform a CT scan on a 21-year-old stabbing victim and missed a knife wound penetrating several inches into his skull and brain.
• A newborn baby suffered preventable brain damage when doctors failed to test for high levels of a chemical in his blood that had turned his skin yellow from head to toe.
"It's just incredulous to the public," said Pronovost, who advocated for the study but wasn't involved in it. "We just too often accepted bad outcomes as the norm."
Inadequate openness, limited data
Diagnosing patients' health problems is at the core of what doctors and clinicians do. So what leads them to sometimes get things wrong?
The report identified a number of factors, from inadequate communication and collaboration to a culture that discourages disclosure of mistakes, impeding attempts to learn from them.
The researchers acknowledged that data on diagnostic errors is limited. Good measurements are hard to come by, and there's no real consensus on what even constitutes such a mistake. Studying medication or surgical errors or infections that patients acquire inside hospitals is less of a challenge.
"Frankly, this is not low-hanging fruit," said Dr. Albert Wu, director of the Center for Health Services and Outcomes at Johns Hopkins Bloomberg School of Public Health. "We tend to focus on things that we can measure, and this is hard to measure."
While Tuesday's report says urgent change is needed, Wu says there's no fast and easy way to figure it out.
"We're asking people to manage without data," he said. "So, for example, I do not know what proportion of my diagnoses are correct. There's no feedback mechanism that I get at any point in the system about whether or not I got the most appropriate test that would allow me to make the diagnosis, and there's no one keeping track of whether or not I did make the right diagnosis."
Wu, who also wasn't involved in Tuesday's study, said it sets a course toward reducing the mistakes, even though the solutions are not all clear yet.
One significant recommendation in the report: Medicare and organizations that accredit health care organizations should require facilities to monitor how they're diagnosing patients. Right now, there is no requirement to monitor diagnostic mishaps.
"You'd need to be specific about the diagnosis you're interested in," Wu said. "There's a lot of places where this could be messy."
Another recommendation: The federal government should conduct autopsy studies that essentially ask whether you died from the same thing you were being treated for.
While these kinds of postmortem exams can generate useful data on diagnostic errors, they've declined substantially since the 1960s, in part because of cost. The average autopsy costs about $1,275, according to the report.
The report also says that doctors and nurses should make patients feel comfortable about asking questions. Fear about being a complainer or being seen as difficult can silence patients and potentially lead to a bad outcome.
The researchers cite a study in which 87% of cancer patients didn't report concerns that their care had been compromised.
"We need to encourage patients to speak up and ensure that when they do speak up, it's well received," Pronovost said.
The National Patient Safety Foundation and the Society to Improve Diagnosis in Medicine have a checklist
to help patients get the right diagnosis.
"It is OK to go over the doctor's head or get another opinion if you're really convinced that something is being missed," Wu added.