Patients assigned a “nonurgent” status on arrival in the emergency room might still be sick enough to be hospitalized, a new study shows.

Patients deemed by triage nurses to be “nonurgent” often receive diagnostic services and procedures, and some are even admitted to critical care units, researchers found. Triage was never intended to completely rule out severe illness, only to give patients an estimate of how long they may have to wait to see a doctor, the researchers note.

Dr. Renee Y. Hsia of the University of California, San Francisco, and colleagues used a national survey of emergency department (ED) visits between 2009 and 2011, with triage scores assigned by a nurse when the patient arrived. The scores range from one to five, with one through three including immediate, emergency, and urgent patients, and four and five being semi-urgent and nonurgent.

They used data on almost 60,000 observations of patients age 18 to 64 collected between 2009 and 2011, which represented 240 million E.D. visits. More than 90% had a score of one to four and were deemed “urgent,” while about 8% had a score of five and were “nonurgent.”

Almost half of nonurgent patients received diagnostic scans, imaging, or blood tests, and a third underwent procedures, e.g., splinting or IV fluid replacement. For urgent visits, about three-quarters involved diagnostics and half involved procedures. About 4% of nonurgent visits resulted in hospital admission, as reported online April 18 in JAMA Internal Medicine.

Dr. Hsia told Reuters Health by email, “Triaging patients is an extremely difficult task, since . . . there has not been time to do a full history and physical, and nurses have limited information upon which to base their determination.”

“We should not expect triage categorization to be perfect, and one of the goals of this paper is to show that, indeed, triage is not,” Dr. Hsia said.

Many states have policies that patients with Medicaid insurance who present to the ED for nonurgent visits will be charged a copayment, which may keep people from seeking needed care, even though the triage system is not perfect, she said.

“It is important that we do not blame the patient for going to the ED if there were no alternatives that were available in a timely manner,” Dr. Hsia said. “Our study cannot distinguish the reasons behind why we found such a high proportion of visits that received diagnostic services or procedures.”

Some of the procedures may not have needed to happen in an ED setting, but since the patients presented there, they were treated, Dr. Hsia added.

In an accompanying Editor’s Note, Dr. Joseph S. Ross notes the results “suggest that there is some uncertainty during ED triage assessment of visit urgency and policies that are based on this assessment must take this uncertainty into account or risk unfairly, and inappropriately, imposing cost-sharing penalties.”