Replace diagnostic errors with excellence
Five strategies can help hospitalists achieve diagnostic excellence, a researcher says.
Diagnostic error is an uncomfortable topic for physicians, Hardeep Singh, MD, MPH, acknowledged during his session on the subject at Internal Medicine Meeting 2023.
Yet its prevalence and potential for patient harm mean it can't be avoided. “No matter what setting you're looking at, whether it's the hospital, emergency room, or outpatient clinic, there are a variety of diagnoses that can be missed,” said Dr. Singh, co-chief of the health policy, quality, and informatics program at the Center for Innovations in Quality, Effectiveness and Safety at the Michael E. DeBakey Veterans Affairs Medical Center and professor in the department of medicine at Baylor College of Medicine, both in Houston.
He polled the audience on an example that hit home with hospitalists: “In the hospital, who has seen a missed diagnosis of spinal epidural abscess in the last five years? Wow, most hands are raised.” In an analysis of 119 new diagnoses of this condition, 55.5% involved diagnostic error, Dr. Singh said, citing results published in the American Journal of Medicine in 2017.
That's only one of many studies he and his colleagues have done on diagnostic error, and based on these findings and additional research, they've developed an approach to reducing error that focuses on the positive. “We've tried to define the concept of diagnostic excellence,” he said.
In addition to a definition, he offered physicians five strategies to pursue diagnostic excellence for themselves, as well as advice on getting their institutions to support this goal.
Improving diagnosis
Diagnostic excellence involves more than just getting the diagnosis right, Dr. Singh noted. “That's the most essential—you've got to get the accurate diagnosis in a timely fashion,” he said. “But you've got to do that using the fewest resources. If I order a CAT scan on every single patient who walks into my door, that is not a good use of resources.”
In addition, diagnostic excellence provides the best patient experience possible and helps manage the uncertainty inherent in the situation. “We deal with uncertainty all the time. We may or may not manage it correctly or even communicate it to the patients,” said Dr. Singh. Watchful waiting, “which we often don't practice,” is another component of diagnostic excellence, he said.
So how does a physician move toward these goals? Start by assessing and learning from your own performance, recommended Dr. Singh. “Seek diagnostic feedback” was the first of the five strategies in a Practice Pointer on advancing diagnostic excellence he and colleagues published in The BMJ in February 2022.
One way to do this is to review your cases on your own. “You're on a hospital service two weeks. Maybe three patients are really bothering you. You make a list and look at their diagnoses a few months later,” said Dr. Singh.
To help with this process, his team released a tool, Calibrate Dx, that is available through the Agency for Healthcare Research and Quality (AHRQ). “Select five cases. You can look at your last five readmissions or your last five diagnoses of pneumonia,” explained Dr. Singh. “Just look at those cases yourself, or preferably find a trusted colleague who can look at them with you, too.”
It can be very helpful to get others' feedback as part of this process, he noted. “Colleagues or even patients can provide data about your diagnostic performance.” Geisinger, a health system in Pennsylvania, piloted a formal attempt at this in a diagnosis improvement program described in the February 2021 The Joint Commission Journal on Quality and Patient Safety.
Instead of calling the targeted issue diagnostic error, “For this project we used a better term that resonated more with physicians: learning opportunities,” said Dr. Singh, senior author of the study. “We delivered feedback through the physician's supervisor in a very nonpunitive, confidential way—a debrief where the physicians could go through their diagnostic process with their supervisor and figure out if there's any learning involved. No one became defensive because it was a very casual conversation.”
A second strategy to further advance excellence was “byte-sized” (brief and digital) learning activities. “We called it bite-sized because we don't always like to read in six-hour chunks,” said Dr. Singh. “Try to look at maybe one or two cases online every day. It's like exercise. Try to do it for a few minutes, but try to do it regularly.”
Clinicians can obtain these small bits of education from social media, podcasts, or medical journals, he suggested. “You look at these cases, and you can get to know how to improve your diagnostic reasoning.”
The third strategy is considering your biases. “There's a pretty good vignette study that shows that difficult patients may be more prone to the risk of misdiagnosis,” said Dr. Singh. “Try to find common ground and build empathy,” he said.
In addition, check for bias in your overall care. “Identify it in your practice-level data. If you're ordering [urine toxicology] on only one type of patient population, you might want to consider if that could be a bias,” said Dr. Singh.
The fourth step is to make diagnosis a team sport, with more multidisciplinary collaboration. “This includes nurses, physical therapists, specialists, and patients,” he said. “We've got to flatten the hierarchy and hear from everybody. “ The diagnostic team could also include computers and their advice, he noted.
The final strategy is to foster critical thinking. “This is something that is not well taught in medicine and medical school,” said Dr. Singh. “How do we take a skeptical stance towards our own decisions? That's hard to do, but oftentimes, if the patients aren't getting better, we need to be able to figure out if we need to go back and ask some more questions.”
Role of systems
While the five strategies are intended for clinicians to advance their own diagnostics, “Health care organizations are going to also have to do a lot,” said Dr. Singh. “This is not just a physician problem.”
He called on health systems to become LEDE organizations, which stands for Learning and Exploration of Diagnostic Excellence, by doing five things. “The first one is to create a virtual hub of some kind or some kind of community that coordinates activities of diagnostic excellence,” he said.
Hospitals also need to generate and translate new evidence on diagnostic safety to the bedside, according to an action plan laid out by Dr. Singh and colleagues in 2020. “The third [thing] is engaging clinicians in activities to improve diagnosis, which is sort of what we're doing here,” he said.
Fourth is to foster a culture that is accountable and engages with and learns from patients, and finally, hospitals must conduct measurement for improvement and learning. Another free tool that Dr. Singh developed for AHRQ, Measure Dx, can assist with that task. “It essentially helps you implement pathways to measure and learn from diagnostic errors,” he said.
These learning pathways can gather data to improve diagnostic safety, from electronic health records, hospital staff, or patients. Patients are a particularly rich resource, Dr. Singh noted, describing a study in which his team interviewed patients whose conditions had been difficult to diagnose. “What do you think was the most common reason the patients said that their diagnosis was missed or delayed or wrong?” he asked. “‘The doctor did not listen to me.’”
