How hospitalists can avoid diagnostic errors


Rates of diagnostic errors range from 10% to 15% in internal medicine, and hospitalists are particularly positioned to mitigate these mistakes, according to nephrologist Mark L. Graber, MD, FACP, who has been involved in raising awareness of patient safety for the last 15 years.

In 2008, he convened the Diagnostic Error in Medicine conference series and then in 2011 founded the nonprofit Society to Improve Diagnosis in Medicine. Most recently, in 2014, Dr. Graber founded Diagnosis, an international journal for research studies on the issue. He is also the co-creator of Patient Safety Awareness Week (PSAW), an annual initiative that runs from March 13 to 19 this year.

Dr. Graber, a senior fellow of RTI International's Health Care Quality and Outcomes Program, recently spoke with ACP Hospitalist about his work and how hospitalists can improve their diagnostic skills.

Q: What led you to focus on the issues of patient safety and diagnostic error?

A: The patient safety movement began in earnest in 1999 with the first report from the Institute of Medicine (IOM), “To Err Is Human.” I was the chief of medicine at a VA hospital in Northport, N.Y., at that time, and...1 of the things we did in my own medical service is start peer review meetings about the quality of care we were providing. We addressed many of the issues that were in the IOM report, like medication errors, falls, and hospital-acquired infections, but we also noticed there were a fair number of diagnostic errors. As internists, that's supposed to be the thing we do well, so I was concerned that that wasn't on the patient safety radar screen and started to look into it. There were delays that could have been prevented—and diagnoses that we missed completely.

Q: What is Patient Safety Awareness Week, and why is it important?

A: Patient Safety Awareness Week began in 2002. This was the brainchild of Ilene Corina, who is a Long Island mother and patient safety advocate whose son suffered a serious medical adverse event. So Ilene and I thought it would be appropriate for there to be at least 1 week a year where providers and health care organizations focus on patient safety. It gives us a time to talk about what we're doing and what still needs to be done and the importance of addressing the issue. With support from the National Patient Safety Foundation and the VA, almost all hospitals now recognize PSAW.

Q: What is the role of hospitalists in improving diagnosis?

A: Hospitalists are extremely important in establishing the correct diagnosis. They're the captains of the ship once the patient is in the hospital. They have the opportunity to reconsider what's been decided up to that point, so they are in an ideal position to catch any diagnostic errors that have been made in the clinic or in private settings or in the emergency room.

Q: What causes diagnostic errors?

A: The causes are almost always complex because diagnosis itself is complex. This begins with our health care organizations, which are also incredibly complicated and becoming more so by the day. We have problems communicating effectively from one provider to the next and coordinating care, or the specialist we need isn't available to us right at that moment, or we have problems supervising trainees. There are dozens of ways in which our health care organizations can contribute to diagnostic error. And with equal frequency, there are also cognitive problems that contribute to diagnostic error. We just don't think of the right diagnosis, for example: We're too biased by a previous diagnosis or some bit of information that we trust more than we should. The most common reason for a missed diagnosis is simply that “I just didn't think of it.”

Q: How can hospitalists improve their diagnostic skills and reduce errors?

A: Fresh eyes are very important in catching mistakes, so the hospitalists have a very important role to play in rethinking the diagnosis of patients that are being admitted to their service. And for patients who don't have a diagnosis yet, they'll be the primary diagnosticians. So they need to excel in all the skills necessary to reach the correct diagnosis: Taking a thorough history, conducting an appropriate physical examination, considering all the past medical history that's relevant, all the tests that have already been done, and to really think things through in a comprehensive manner.

Hospitalists should be aware of both the system-related and cognitive causes of diagnostic error so that they can avoid them. Steering clear of cognitive errors involves practicing reflectively and consciously reflecting on any intuitive biases that might have been in play. Hospitalists typically have substantial workloads and pressures of time, but the advice to just “stop and think” is very useful, if only to consider the universal antidote: “What else could this be?” Possibly the most successful strategy to avoid missing a key diagnosis is to always derive a differential diagnosis.

Q: Who should read your journal, Diagnosis?

A: There are many, many stakeholders interested in diagnosis; it's not just doctors and patients, although those are probably the 2 primary audiences. Nurses have an incredibly important role to play in making and monitoring diagnoses. It's a wide audience that we're hoping to reach, including insurers, educators, cognitive scientists and psychologists, payers, accrediting organizations.

Q: What kind of progress have you seen since you first started studying these issues, and what still needs improvement?

A: I think over the last 15 years, what we've seen is a remarkable uptake of electronic medical records (EMR), which are right in the middle of all the issues regarding diagnosis and diagnostic quality and diagnostic error. There are so many ways that the EMR can help me in terms of making a diagnosis: All the data I need is there at my fingertips, it's typically organized more effectively than in the paper records, I can communicate with other providers, I can read all the notes, I can see all the data, and some of it is organized in graphs, which make it easier for me to see trends.

There's so many ways it can improve diagnosis, but at the same time, most of us agree that it hasn't reached its full potential, and many providers are very frustrated with their EMR and think that it degrades the process, so there are certain aspects of it that are negative. For example, copy/paste notes are extremely common and very deleterious to diagnostic quality. Whenever a provider sees a patient, we're obliged to write a note and document what we found and what we think, but with the EMR, it's far too easy to copy an old note and paste it in as today's note and then try to edit it appropriately. But too often, providers don't do the editing appropriately, so what ends up in there is information that's either outdated or has changed or is confusing or just, frankly, wrong. And once you see a note like that, you've lost your trust and faith that medical care relies on, so it's extremely toxic.

The most encouraging sign of progress is that awareness of diagnostic errors is growing, thanks in large part to the [2015] IOM report, “Improving Diagnosis in Health Care.” Every doctor and every health care organization should be aware of this major report.