One in 14 American patients may be harmed by hospital misdiagnosis

Harmful diagnostic errors can occur in up to one in 14 hospital patients receiving medical care, a new study based at a single U.S. medical center has found. Up to 85 percent of these errors are preventable, highlighting the need for better surveillance in hospital environments.

Previously published reports from the US have suggested that existing surveillance tools underestimate the prevalence of misdiagnosis in healthcare settings. To explore these findings, a team of researchers led by Brigham and Women’s Hospital in Boston randomly selected records from 675 patients admitted to the hospital between July 2019 and September 2021.

“In most cases, the diagnostic process works well, leading to a timely and accurate diagnosis,” said Anuj Dalal, associate professor at Harvard Medical School and lead author of the study. News week. “But sometimes things fail. Intertwined systems, complex processes and human factors can contribute to a missed diagnostic opportunity.”

In their study, published in the journal BMJ Quality and SafetyDalal and colleagues concluded that, based on this sample from a single medical center, harmful misdiagnosis occurred in 7% of patients, or one in 14, who received general medical care. They added that most of these errors were avoidable.

“In our study, the main process failures identified include failures in initial assessments and diagnostic tests,” Dalal said. “It is critical to note that it is not an individual or process that is to blame.”

These findings add to previous work by Dalal and colleagues exploring existing electronic health records systems and their ability to monitor diagnostic errors in medical settings.

“We suspect that a mix of underlying issues are causing the problems with test choice and clinical assessments that we saw,” said Andrew Auerbach, professor of medicine resident at the University of California San Francisco and co-author of this earlier research. News week.

“These, in turn, likely fall under system gaps, such as problems with how transfers happen or how the electronic health record displays data, as well as cognitive and workload issues, such as being too busy with many tasks simultaneously.”

But how applicable are these new findings to hospitals across the country?

“As a single-center study at a tertiary academic medical center, one must be cautious when extrapolating these estimates to other hospitals across the country,” Dalal said. “There is likely to be variability across different hospitals for a variety of reasons. It is difficult to say without rigorously conducted multicenter research studies.”

He continued: “As part of our research efforts, we are conducting studies to measure diagnostic error rates using a similar process in other hospitals.”

Once more nationally representative data is collected, what can we do to minimize these diagnostic errors in hospital settings?

“Measuring error rates routinely as part of hospital quality and safety programs is a first step toward understanding the burden of the problem in any institution,” Dalal said.

“Addressing the problem will require multidisciplinary efforts to monitor and address the complex systems and human factors that contribute to these errors; create a culture of diagnostic safety among clinicians; learn from cases in which a timely and accurate diagnosis has been achieved.”

These solutions can also be improved with the help of AI. “Artificial intelligence approaches will certainly play a role in improving the way we detect cases and trigger interventions,” Dalal said.

“I would note that testing new AI interventions must be done in the context of safely and rigorously conducted research studies that consider the potential risks introduced by AI before it can be safely deployed in healthcare settings.”

Better diagnostic care helps both doctors and patients, and research in this field is critical to public health. “Focusing on diagnosis and diagnostic thinking is a critical part of not only improving patient care, but also improving physician performance and well-being,” Dalal said.

“We are optimistic that projects like ours will help move the field in both directions at once.”

Are there any health problems that are worrying you? Do you have any questions about sleep? Let us know via [email protected]. We can ask experts for advice and your story can be featured in News week.

References

Dalal, A.K., Plombon, S., Konieczny, K., Motta-Calderon, D., Malik, M., Garber, A., Lam, A., Piniella, N., Leeson, M., Garabedian, P. , Goyal, A., Roulier, S., Yoon, C., Fiskio, J.M., Schnock, K.O., Rozenblum, R., Griffin, J., Schnipper, J.L., Lipsitz, S., & Bates, D.W. (2024) . Adverse diagnostic events in hospitalized patients: a single-center, retrospective cohort study. BMJ Quality and Safety. https://doi.org/10.1136/bmjqs-2024-017183

Dalal, A. K., Schnipper, J. L., Raffel, K., Ranji, S., Lee, T., & Auerbach, A. (2023). Identifying and classifying diagnostic errors in acute hospital care: first lessons from the Utility of Predictive Systems in Diagnostic Errors (UPSIDE) study. Journal of Hospital Medicine, 19(2), 140–145. https://doi.org/10.1002/jhm.13136