Diagnostic errors, or “the failure to establish an accurate and timely explanation of the patient’s health problem(s) or communicate that explanation to the patient,1” are a leading cause of patient harm. To Err Is Human,2 a report published by the Institute of Medicine in 1999, was one of the first publications to bring the issue of medical error, patient harm, and the need for safer systems to a national stage.
Improving Diagnosis in Health Care,1 published in 2015, continued the patient safety discussion with a focus on the impact of diagnostic errors on medical harm. This report suggested that diagnostic errors may contribute to 10 percent of all patient deaths. In addition, diagnostic error may result in serious harm to more than 500,000 Americans each year across ambulatory, emergency, and inpatient care settings.3
In a more recent study of almost 2,500 medicine patients who died or were transferred to the intensive care unit, diagnostic errors were found in 23 percent of cases.4 While the cause of these delayed, missed, or wrong diagnoses are almost always multifactorial, most cases have some contribution from inaccurate clinician diagnostic reasoning.5,6 Diagnostic reasoning is the process by which a clinician uses medical knowledge, critical thinking, and experience to gather, integrate, and interpret clinical information. Then, the clinician generates a working diagnosis, creates a testing plan, formulates an accurate diagnosis, and communicates this diagnosis to a patient.1,7
Unfortunately, identification of inaccurate diagnostic reasoning and diagnostic error often takes place after the error has occurred. Retrospective investigation into the contributing factors that led to the diagnostic error are subsequently at risk of biases. These include hindsight bias and outcome bias. Hindsight bias is overestimation of one’s ability to predict a correct diagnosis had one been asked to do so before knowing the diagnosis. Outcome bias is difficulty accurately judging a decision after the outcome of the decision is known, rather than based on the information available at the time of the decision.8
Diagnostic reasoning has historically been treated as intrinsic to the individual clinician, likely due to difficulty visualizing and measuring cognition based on the inherent complexity and internal nature of mental processes. However, there is now growing recognition that cognition is complex and affected by context, cognitive biases, resources, and physical, social, and technological environments.9 Increasing attention is also being given to cognitive load theory (CLT), an educational model for how human memory and processing occur, with a focus on how to optimize the limited resources of cognitive capacity.10
Studies have already found that cognitive overload reduces decision-making flexibility,11 is associated with task errors,12 and can promote default thinking rather than conscious and analytical thinking.13 However, CLT as it applies to diagnostic accuracy is a relatively new area of study, with most of the research occurring in the last 5 to 10 years.
This Ƶbrief will provide a broad overview of fundamental CLT concepts, discuss what is and is not yet known about the interplay between CLT and diagnostic accuracy, and review ways to measure CLT to understand its causal impact on diagnostic accuracy. The brief will conclude with recommendations for future research and improvement efforts, with the goal of better understanding how cognitive load affects diagnostic accuracy and how to optimize cognition to decrease diagnostic error-related morbidity and mortality. A clinical vignette will be woven throughout the paper to better illustrate the concepts and provide clinical context.
