Health care providers are employing a number of innovative techniques to address the preventable errors, biases, and oversights that can cause harmful and costly misdiagnoses, the Wall Street Journal's Laura Landro reports.
According to researchers at Johns Hopkins University, missed diagnoses cause permanent damage or death in as many as 160,000 patients annually. In addition to being more common than other medical errors, misdiagnoses are also the leading cause of malpractice lawsuits, accounting for 35% of nearly $39 billion in payouts in the United States between 1986 and 2010, according to Johns Hopkins.
At the same time, diagnostic errors are more likely to be preventable than other medical mishaps, and health care providers across the nation are seeking to take advantage of the potential for improvement through computer automation, new tests and devices, and changes to medical culture.
Mining EHR data to catch missed signs of trouble
The Southern California Permanente Medical Group is one of the many large health care systems that have turned to data mining to identify opportunities for improvement. Under its "Safety Net" program, the system periodically surveys its database of 3.6 million members for missed lab results or testing.
For example, a review of 8,076 patients with abnormal PSA results encouraged more than 2,200 patients to receive follow-up biopsies. Between 2006 and 2009, 745 cancers were diagnosed among those patients, Landro writes.
The program also is used to find patients with undiagnosed kidney disease, which can be diagnosed with an abnormal test result for creatinine that needs to be repeated within 90 days. From 2007 to 2012, the system found 7,218 lab orders for patients with abnormal tests that were not repeated. Of those, 3,465 were repeated within 90 days of notifying the patient, leading to 1,624 confirmed cases of the disease.
Michael Kanter, the medical group's regional medical director of quality and clinical analysis, says the system allows Kaiser to go back "as far as is feasible to find all of the errors that we can and fix them."
Using EHRs to 'double check' test results
Some researchers warn that EHR data mining can flood doctors with more information than they can effectively manage. Instead, the data could be better used as a "second set of eyes" to detect things that may have been overlooked, thereby preventing dangerous delays in follow ups.
Delivering value with 'big data'
In a new pilot program at VA facilities in Houston, researchers developed "trigger" queries to electronically flag medical records of patients with potential delays in prostate and colorectal cancer evaluation and diagnosis. The queries were run on 600,000 patient records, identifying about 1,048 cases of delayed or missed follow ups after abnormal findings annually and 47 high-grade cancers.
Hardeep Singh, chief of health policy and quality at the Michael E. DeBakey VA Medical Center, says the program is "like finding needles in a haystack, and we use the information technology to make the haystack smaller and smaller so it's easier to find the needles."
Hospitals are also using electronic decision-support programs that help physicians rank possible diagnoses by their likelihood based on symptoms and chart notes. One such system, called Isabel, was able to provide the correct diagnosis 96% of the time when data from 50 challenging medical cases were entered into the system.
Meanwhile, a number of companies are manufacturing new devices—including portable diagnostic equipment and lab tests—that hold promise for detecting disease. For example, MelaFind allows dermatologists to noninvasively examine moles as deep as 2.5 millimeters beneath the skin's surface for signs of melanoma.
Other efforts to reduce diagnostic errors have centered on improving communication. Patients are being encouraged to get more involved in the diagnostic process, and physicians-in-training are being taught to be more receptive to patient input and avoid common processes that often lead to incorrect diagnoses.
For example, the Critical Thinking program at Dalhousie University in Halifax, Nova Scotia, aims to help medical residents understand how different cognitive biases may affect their thinking. "If we can teach physicians how to think more critically, they would be more effective in delivering good care and arriving at the right diagnosis," says program developer Pat Croskerry (Landro, Wall Street Journal, 11/ 17).