Reducing Diagnostic Errors: The Next Frontier in Patient Safety

March 8, 2016 | Bryan LeBude, MD

Each year approximately 5 percent of U.S. outpatients experience a diagnostic error in their care. The numbers are staggering. Diagnostic errors contribute to about 10 percent of patient deaths and account for up to 17 percent of adverse events in hospitals. Diagnostic errors lead to patient harm by delaying necessary treatment, propagating unnecessary and sometimes detrimental therapies, and culminating in both psychological damage and financial turmoil. They are the leading category of paid medical malpractice claim, and compared with other types, are twice as likely to have contributed to the patient’s death.

Despite their significant impact, diagnostic errors have historically been overlooked in efforts to improve the quality and safety of health care. In September 2015, the Institute of Medicine (IOM) released a report entitled “Improving Diagnosis in Health Care” thereby laying the framework for a revolution. Cardiovascular imaging specialists are uniquely positioned to improve diagnosis in health care by practicing at the interface between diagnosis and therapeutics. Three areas for improving diagnosis readily come to mind:

Clinical Reasoning

Post hoc analyses of errors in diagnosis have shown that shortcomings in clinical reasoning rather than knowledge deficiencies are the more prevalent cause of diagnostic error. Advances in clinical psychology have informed our understanding of the clinical decision-making process, which is largely composed of intuitive versus analytical thought processes. Intuitive reasoning is the predominate process at play; it is fast, associative, and susceptible to bias. A variety of cognitive biases are known to potentially lead to diagnostic errors. Premature closure is the failure to consider other possibilities after reaching an initial diagnosis. Availability bias refers to the tendency to judge a diagnosis as being more likely if it readily comes to mind, and thus recent experiences with a disease increase the probability of it being diagnosed. Confirmation bias is the tendency to search for confirming evidence to support an already considered diagnosis rather than consider the more compelling evidence that refutes it.  

Debiasing strategies hold promise for reducing errors in clinical reasoning. Insight and awareness of cognitive biases is an important first step. Forced consideration of alternative diagnoses can aid in avoidance of biases such as premature closure. Metacognition is the reflective process of stepping back from the specific problem to analyze and reflect on the thought processes at hand. Making relevant data easier to obtain, and limiting time pressures on diagnosis would promote analytical as opposed to intuitive reasoning.


The IOM defines diagnostic error as “the failure to (a) establish an accurate and timely explanation of the patient’s health problem(s) or (b) communicate that explanation to the patient.” Many diagnostic errors involve disease processes that were entertained at one point in the patient’s course but lost in translation. Good medicine is more commonly guided by simple mechanisms to ensure adequate follow-up and communication of diagnostic information than the exotic diagnosis missed by a handful of physicians before eventually being nailed by a brilliant specialist. This entails timely and thoughtful communication of the available diagnostic information to both the patient and other providers involved in their care. For cardiovascular imagers, effective communication most times involves more than simply faxing a dictated report to the ordering provider.

Patients must also be empowered with portable, up-to-date records of their medical care. The U.S. health care system remains fragmented, and patients are often the one unifying link across levels of care. The IOM has recognized the importance of patient engagement in the diagnostic process by providing a “Checklist For Getting The Right Diagnosis.” This includes useful tips for patient engagement in areas such as telling your story well, keeping good records, knowing your test results, and recording your health information.


In general, physicians lack systematic means for calibrating diagnostic decisions based on outcomes feedback. Cardiovascular imagers typically become aware of their diagnostic successes or failures through ad hoc means such as bumping into a colleague who is taking care of a mutual patient. This system is not sufficient as physicians are robbed of valuable information on countless diagnostic decisions made each day. The current landscape of increasingly consolidated health care organizations with larger networks and growing usage of electronic health records should form the groundwork for a closed-loop system based on the thousands of collective diagnoses that are made each day within an organization. Feedback is necessary to allow physicians to engage in the learning process, and correct individual or systematic errors going forward. Institutions that are serious about improving patient safety need to promote a non-punitive culture that places emphasis on continually providing it’s team members feedback on their diagnostic performance.  

The majority of people will experience at least one diagnostic error in their lifetime. The time has come for us to focus our efforts on improving this pillar of clinical medicine.

This article was authored by Bryan LeBude, MD, a fellow in training at MedStar Washington Hospital Center.