EXECUTIVE SUMMARY

Data demonstrating reliability of e-triage allow the defense to refute arguments that using the tool breached the standard of care.

  • Various triage tools have been attacked as unreliable yet remain widely used.
  • Some EDs continue using outdated triage systems.
  • Plaintiffs would have to show another triage tool could have prevented a bad outcome.

New e-triage tools have produced some solid data demonstrating their validity.1 But what are the liability implications for EDs who are early adopters?

“None of the tools are perfect. E-triage is new but in my opinion has a reasonable solid foundation for implementation,” says Daniel J. Sullivan, MD, JD, FACEP, president and CEO of the Sullivan Group, a Denver-based medical risk management and patient safety organization. Given the evidence of the tool’s reliability, it is unlikely one could argue convincingly that e-triage is below the standard of care. “I don’t think there is much, if any, room for criticism by a plaintiff on that count,” Sullivan predicts.

The plaintiff would have to show that another, more reliable triage tool should have been used instead, a daunting task. “The other triage methods are definitely imperfect and increasingly under attack,” Sullivan explains.

This does not mean those tools are below the standard of care, either; these tools are used in many EDs. Thus, it is hard to argue that using these other tools is so unreasonable that such use constitutes a breach of the standard of care. “Looking at how many hospitals use them, it would be difficult to effectively attack anyone or a hospital for using them,” Sullivan says. “Currently, there is little or no basis either way for a legal challenge.”

It is possible a plaintiff could criticize e-triage if the bad outcome occurred because of mistriage. “But with so many other variables at play in emergency room decision-making, I just don’t see a case turning on this issue,” Sullivan offers.

Sullivan is unaware of any malpractice cases in which the use of a particular triage tool was the primary focus, even if EDs used outdated three-level systems. However, as evidence accumulates, that could change. Under one scenario, e-triage could continue to demonstrate superiority to other tools, leading to higher adoption rates in a particular geographic region or nationally. “Down the road, I could see calling the use of an outdated or inadequate triage tool substandard,” Sullivan says.

Although triage tools take somewhat different approaches, they suffer from the same issue, says Scott Levin, PhD: “There is a lot of variability in how they are applied.” With the Emergency Severity Index (ESI) tool, most ED patients are assigned to level 3 triage, some with significant illness. “That is a mixed bag of patients. Some are probably safe waiting, and some are not,” Levin says. “It is a safety issue.”

Levin led a team of researchers who developed an e-triage tool to risk-stratify patients and sort out those who should not be waiting. “Triage is an important decision done with little information,” says Levin, an associate professor of emergency medicine at the Johns Hopkins University School of Medicine.

E-triage showed equal or better identification of patient outcomes compared to ESI, according to Levin and colleagues’ work on a multisite retrospective study of nearly 172,726 ED visits in urban and community EDs.1 Of the more than 65% of visits triaged to ESI level 3, e-triage identified about 10% who may have benefited from triage to a level 1 or 2. This group was at least five times more likely to experience a critical outcome, such as death, emergency surgery, or ICU admission.

The researchers’ goal is to glean information from millions of ED visits and put all that information to use when the triage decision is made. “We really feel that there is all this data in the EMR and it’s not being used like it should,” Levin says. “Other industries leverage their data to make decisions, but healthcare is new to this.”

Better predictive models could determine the risk of a patient dying, needing an emergent procedure, or being hospitalized, with triage decisions informed by that data. “We’ve also found that you need the nurse no matter what,” Levin notes. “The nurse, in harmony with this technology, can do better than the tool alone or the nurse alone.”

Of course, more accurate triage mitigates legal risks. “Any case where the patient with a time-dependent condition is waiting too long and there’s a delayed diagnosis, triage plays a part in that,” Levin warns.

REFERENCE

  1. Levin S, Toerper M, Hamrock E, et al. Machine learning-based electronic triage more accurately differentiates patients with respect to clinical outcomes compared with the Emergency Severity Index. Ann Emerg Med 2018;71:565-574.e2.