The patient safety movement has long focused on errors in acute and long-term care settings. But now investigators are pointing their research tools at the diagnostic skills of primary care physicians because too often, they're getting it wrong, causing avoidable and significant patient harm as a result.
By backtracking patients whose deterioration necessitated that they be admitted to a hospital or another care setting, researchers could see how clinicians in their office practices forgot or missed certain critical steps that could have averted emergencies that ensued days after patients left the doctors' offices.
"Our goal with this paper is to shine a light on the problems we saw in two large healthcare systems, and offer some hope to other healthcare systems that could use this methodology to understand what types of misdiagnoses they are having," says Hardeep Singh, MD, Chief, Health Policy, Quality and Informatics Program at the Houston Veterans Affairs Medical Center and Baylor College of Medicine.
Singh's paper is published in JAMA Internal Medicine.
By using an electronic health record-based "trigger tool," the researchers were able to look through selected medical records of patients whose conditions declined because they didn't get preventable care. Singh and colleagues found 190 significant errors at these two healthcare settings, which included 68 "unique diagnoses" that were missed by the patients' treating physicians.
"This is the first step in trying to establish a measure of this problem and try to understand it. And we know, it opens up a lot of new areas to figure out what's next," Singh says.
By far, the largest share of factors contributing to the errors involved elements within the patient-practitioner encounter. For example, that the physician had a problem ordering diagnostic tests for further workup contributed to the cause of 109 of the errors.
Failure to take an accurate medical history contributed to the cause of 107 of the errors. And problems with the performance of the physician's physical examination of the patient contributed to 90. Each of the 190 errors they discovered had multiple causes.
"We found that often, the most important thing was taking the history, doing the physical exam and ordering the right tests based on your assessment of the patient," Singh says.
The researchers discovered that the patient's own actions and behavior, or that of his or her family members, in the practice setting, seldom led to a contributing cause of the error. For example, failure of the patient to provide an accurate medical history contributed to only 14 of the errors.
Other less frequent contributors included physicians' inadequate test result tracking systems, the physician setting too long of a period for a follow-up visit, or errors in which the clinician erroneously considered the condition as not serious.
Interestingly, the top missed diagnoses at one healthcare system studied were not the same as the top missed diagnoses at the other, and very few errors involved rare or unusual diseases.
The missed diagnoses often involved serious illnesses, from acute renal failure, pneumonia and cancer, to angina, cellulitis, hypertension, and urinary tract infections.
"Pneumonia and decompensated congestive heart failure were most commonly missed, although they accounted for less than 13% of all errors," the researchers wrote.
Singh says that once healthcare systems start measuring diagnostic errors in the primary care setting, many problematic practices will be revealed and process errors eliminated. "Because now, there's not a lot of measurement in this area, just the fact that if you start measuring, people will start to notice. If you give feedback to the doctors, saying, 'Hey, you know, in the last three months you had three patients who returned unexpectedly,' and maybe you will see some there were missed opportunities."
In an invited commentary, David E. Newman-Toker, MD, and Martin A. Makary, MD, of the departments of neurology and surgery, respectively, at Johns Hopkins University School of Medicine, said Singh and colleagues "are to be congratulated" for developing their trigger tool, "to help overcome shortfalls in traditional approaches to diagnostic error detection."
"With more than a half a billion primary care visits annually in the United States, if these data from Singh et al are generalizable, at least 50,000 missed diagnostic opportunities occur each year at US primary care visits, most resulting in considerable harm," they wrote.
"Combining this figure with autopsy-based estimates of US hospital deaths from diagnostic errors (40,000 per year to 80,000 per year) and unaccounted non-lethal morbidity from hospital misdiagnoses and acknowledging another half billion visits annually to non-primary care physicians, more than 150,000 patients per year in the United States might have undergone misdiagnosis-related harm."
Both acknowledge, however, that finding solutions isn't easy. But one step might be in physicians' greater use of electronic decision support tools, which are coming with expanded electronic medical record systems.
Another strategy might include "mandatory, structured recording and coding of presenting symptoms, rather than simply diagnoses," in those EHRs.
At least that would help healthcare systems better track these errors.