Current strategies for minimizing non-urgent visits to emergency departments are questionable, study finds

Emergency departments in the United States have more than 140 million visits each year—a rate of four visits for every 10 people—that cost nearly $80 billion. Each interaction is carefully documented, including the reasons the patient gives for the visit upon arrival and the diagnosis for the illness or injury the doctor reports when the patient is discharged.

But how often do doctors and patients agree about how serious the situation is based on what the patient says when they arrive?

Not as often as you might think. A new, cross-sectional study found that emergency department doctors and patients agree on the urgency level only about 38 percent to 57 percent of the time. The research, by Benjamin Ukert, PhD, with the Texas A&M University School of Public Health, and colleagues at the University of Alabama at Birmingham and University of South Carolina, was published in the journal of the American Medical Association.

“This is important because nearly 40 percent of emergency department visits are not medical emergencies, which is very costly financially and in terms of staffing and other hospital resources,” Ukert said. “As a result, state legislatures and health insurers have implemented policies to transfer less-urgent cases to doctors’ offices and urgent care centers, but clinicians face profound challenges in making this decision based on what patients tell them about their condition.”

This legal process—retrospective review and adjudication—is based on medical claims and algorithms related to discharge diagnoses and can be used to decide whether insurance pays for emergency care.

“Our findings fundamentally challenge this plan design because if patients and doctors provide different evaluations of the urgency of the condition, then incentives to reduce emergency room visits may not be effective,” Ukert said. “For example, if patients could go to a primary care doctor but payment policies rely on reviewing the patient’s diagnosis and treatment after the visit to determine whether the physician assessed the condition correctly, then this would require patients to know that their condition could be treated in a doctor’s office instead of an emergency department.”

To shed light on concerns about the use of retrospective review for emergency departments, the researchers characterized visits to high-level groups based on the medical urgency of the presenting reasons for visit and to explore the concordance between discharge diagnoses and reasons for visit. They mapped all possible discharge diagnoses to the same reasons for visit for 190.7 million emergency department visits among adults aged 18 years or older for 2018 and 2019 using data from the National Hospital Ambulatory Medical Care Survey.

Most emergency room patients were women (57 percent) and had public health insurance, including Medicare (24.9 percent) and Medicaid (25.1 percent). Visits resulted in hospitalization for 13.2 percent of visits.

The researchers found that 38.5 percent of emergency department visits were classified with 100 percent certainty as involving injuries, needing emergency care, being treatable by primary care, not urgent, or related to mental health or substance use disorders, based on discharge diagnoses. In comparison, only 0.4 percent were classified the same way based on the reason patients gave for their visit.

“In sum, we found no association between the reasons patients gave for their visit at the time of arrival at an emergency department, their need for emergency department care and their final discharge diagnosis,” Ukert said.

For example, the team found that even among discharge diagnoses defined and classified as very emergent, such as strokes or heart attacks, the initial reasons given for the visit for these conditions were likewise classified as emergent only 47 percent of the time.

“This underscores the difficulty physicians face in making definitive assessments at the triage level without first evaluating patients, given that a single reason for seeking care could have multiple possible underlying causes,” Ukert said. “Alternatives to discharge diagnoses are needed.”

He said these could include getting additional information from patients upon their arrival at the emergency department, such as their main concern, symptoms and other information like mode of arrival.

“This information could lead to the development of objective tools that could more accurately assess the complexity of these visits,” Ukert said.