Advancing Diagnostic Safety: An AHRQ Leadership Conversation

Editor’s Note: Recently, Acting Director David Meyers, M.D., spoke with AHRQ’s chief patient safety official, Jeff Brady, M.D., M.P.H., to discuss diagnostic safety, which Dr. Brady has called “the newest frontier in patient safety.” Following is a summary of that conversation.

Dr. Meyers: Jeff, AHRQ is increasingly spending energy and resources on diagnostic safety research. But I’m not sure that many of us understand this field of study fully. So, let’s clear up any misconceptions. What do we mean by “diagnostic safety?”

Dr. Brady: As a physician yourself, David, you know that a critical early step that’s on the pathway to high-quality healthcare is an accurate and timely diagnosis—an informed opinion of what’s wrong with a patient. But we know that in many instances this just doesn’t happen. People get the wrong diagnosis, or they get the right diagnosis, but it’s made too late. And these failures can set off a chain of very harmful actions. In some cases, not only are opportunities missed to help a patient, but in other cases, a patient may receive potentially harmful care that carries risk without potential benefits because the wrong diagnosis was made. 

So, at its most basic, diagnostic safety is the field of study that aims to ensure that every patient who is ill receives a diagnosis that is accurate and timely.

DM: Do diagnostic errors happen often?

JB: They’re more prevalent than one might think. They affect an estimated 5 percent of U.S. adults every year. Four million people a year in the United States suffer serious harm as a result. Missed, inaccurate, or delayed diagnoses contribute to about 10 percent of patient deaths, and they’re the single greatest source of medical malpractice claims. The estimated cost of diagnostic errors is more $100 billion a year, but the human cost is incalculable.

DM: How do we know?

JB: These are estimates, but they’re based on reliable studies where data are available that can shed light on the problem. For instance, as part of their work, malpractice insurers collect a lot of data, and these indicate that diagnostic errors are one of the most frequent sources of poor-quality care that leads to a malpractice claim. That’s just a subset of the data that are out there, and we have enough to confirm that diagnostic errors occur much too frequently.

DM: Am I right that patient safety starts with diagnostic safety?

JB: Yes. Patient safety threats are present in all aspects of care, and diagnostic safety research is the branch of patient safety research that focuses specifically on the diagnostic process.

If you think broadly about patient safety research, we understand that it’s often instructive to compare the things we intend to happen (or to avoid) in a healthcare encounter with what actually happens, and then closing the gap between these two situations. This basic comparison represents much of healthcare safety and quality writ large. In this instance, we’re applying that same concept to the process of diagnosis.

DM: Are all wrong diagnoses considered diagnostic errors?

JB: No. Diagnosis is often an iterative process. For example, if a patient has a rare illness, early on a clinician may think the person has a more common condition with similar symptoms. Only after the patient does not respond to treatment or when new symptoms emerge will the correct diagnosis be made. The initial diagnosis, while not correct, was not necessarily due to an error.

The field of diagnostic safety focuses on helping teams make the best diagnostic decisions given the information available and also avoid unnecessary delay in collecting, interpreting, and using additional information that may be needed. By this I mean if a screening test result is abnormal, then acting quickly to perform the appropriate follow-up diagnostic testing that will ultimately lead to the right diagnosis.

As with other areas of patient safety research, we focus on the errors in order to understand the diagnostic process as completely as possible. These insights help us think more effectively about solutions that can help prevent errors and improve the overall performance of the system. Talking about error can lead to perceptions that inadvertently suggest blame, but it’s important to remember that the purpose isn’t to find fault with one person. Rather, we are examining the whole system and process of diagnosis and how to improve it.

DM: Why do diagnostic errors occur in the first place?

JB: Research suggests that many diagnostic errors have their roots in two major types of challenges: how people think and issues with systems and processes. 

Errors related to the way we think are called cognitive errors. Doctors and nurses have years of training and are devoted to caring for their patients. But they’re people, and people don’t always perform perfectly. An example of a common type of cognitive error that can lead to a diagnostic error is called recency bias. People tend to remember things that happened recently. So when a busy emergency room doctor is thinking about a patient they are seeing who has stomach pain they may focus on appendicitis as the likely diagnosis, because they recently cared for a patient with a ruptured appendix. This focus can occur even when their current patient’s history or symptoms suggest a different diagnosis.

Healthcare professionals work hard to avoid these problems, but sometimes despite how well-trained and conscientious they are, they make cognitive mistakes. It is critical that we learn more about how healthcare professionals think so that we can help reduce the risk of cognitive errors. The use of artificial intelligence and clinical decision support are some of the supportive strategies that can help us think through complex situations and more consistently make the right diagnosis.

Systemic errors, in contrast, can occur when processes within organizations aren’t designed well or don’t work as designed. Say a biopsy is performed to determine if a suspicious lesion is cancerous. But because of a workflow backlog, the result is not entered into the system fast enough to be interpreted and acted on by a clinician. Or, the same biopsy result could be lost and never entered, so that a once manageable problem could become untreatable. In other situations, a busy, rushed environment in an emergency department during peak hours can cause a patient who needs immediate care to be missed by the system for triage, because the emergency department’s capacity  is not sufficiently augmented when the demand increases beyond capacity.  

DM: It sounds like well-designed systems can make cognitive errors less likely and that poorly designed systems might make them more likely. Is that true?

JB: Absolutely. Oftentimes a diagnostic error is the result of a combination of cognitive and systemic issues. For example, in a busy emergency department, a clinician working quickly to keep up with patient care when the ED is short-staffed might be more apt to make a thinking error. A clinician might not completely and deliberately consider all the different types of diagnoses that could be affecting a patient, and therefore might not do all the necessary testing. We often take thinking shortcuts whenever we’re busy, and we fall back on the explanations for problems that seem most obvious. But the most obvious explanation isn’t always the correct one.

Ultimately, what we really care about is how we can prevent these errors from happening in the first place so that diagnoses are correct and timely.

DM:  Jeff, you’ve taught me that a key part of both causing and preventing diagnostic errors is communication. Could you share some of your thinking?

JB:  The need for effective communication is present throughout the diagnostic process.  It is part of how clinicians begin the diagnostic process as they listen carefully to patient’s stories.  It is how teams share information that is needed to make a correct diagnosis.  It is how clinicians inform patients about a management or treatment plan and steps they can take to more actively participate in their care. Good communication leads to better outcomes and breakdowns in communication between patients and healthcare teams or among healthcare professionals can lead to errors. 

DM: What is AHRQ’s role here?

JB: AHRQ is the Nation’s patient safety research agency. Safe, high-quality healthcare starts with an accurate and timely diagnosis. So diagnostic safety is right in our patient safety wheelhouse, and we’re busy learning more about how to improve the safety and reliability of the diagnostic process.

Beyond that, there are other aspects of diagnostic safety that make it particularly interesting to AHRQ. I think specifically of the systems issues that are related to diagnosis. AHRQ has a long history of funding research on how to help healthcare delivery systems improve care, and diagnostic safety aligns with this experience. And, AHRQ has already produced some research and tools that are designed to improve diagnosis. For instance, we know that patient safety culture has a direct impact on diagnostic safety, and AHRQ created a set of supplemental items for use with the Medical Office Survey of Patient Safety Culture to inform organizations about key aspects of their culture that support better diagnosis.

The most exciting thing about more activity and resources for improving diagnostic safety is that they are on the pathway to actually preventing harm. They will make it possible to avoid the high costs and human toll that diagnostic errors exact. 

DM: Can you share some specific examples of how researchers are approaching not only understanding diagnostic errors but also designing better systems and processes?

JB: One of the reasons I think the field is poised to make real change is the potential of digital health technologies—which of course is another reason why it makes sense for AHRQ to take on this challenge. I mentioned clinical decision support systems, and these can provide checks and balances to assist doctors and nurses to help them avoid cognitive errors when making a diagnosis. Additionally, improved electronic health records can ensure the right information, about the right patient, gets to the right team members, at the right time, reducing communication errors and delayed diagnosis. Other innovators are working to develop patient- and family-centered ways to improve how healthcare teams communicate diagnosis and management plans to their patients. The National Academy of Medicine affirmed the importance of a patient’s understanding of their diagnosis by including the failure to communicate a diagnostic explanation to him/her as a fundamental criterion for a diagnostic error.  

DM: Given the importance and size of this issue, who else are we working with to tackle it?

JB: It’s important to note that we aren’t the only game in town. Our colleagues at the NIH have funded significant research about the diagnostic process, we’ve worked closely with the professional community via the Society to Improve Diagnosis in Medicine, and we’re coordinating a Federal Interagency Workgroup on Improving Diagnostic Safety and Quality in Healthcare. So we’re waving the flag to raise awareness about our work, and we’re also happy to have plenty of important partners working in this area.

DM: Thank you, Jeff. I think all of us at AHRQ are excited about the shared progress we’re making in this area. It’s clear that patients will benefit from this research and the tools and resources the Agency develops for decades to come.

JB: David, thank you. I really feel like we’re just getting started. If we continue to fund the research and work with our partners in the field, we can use what we learn about diagnostic safety to make sure we are treating the whole patient and better meeting their needs, which is the objective of high-quality care.

Dr. Meyers is Acting Director of AHRQ. Dr. Brady is Director of AHRQ’s Center for Quality Improvement and Patient Safety.

Get more AHRQ news, sign up for AHRQ news via email and follow AHRQ on Twitter, LinkedIn, Facebook, and YouTube.