As the current flu season winds down, providers understand that it will only be a matter of months before they must gear up for the next season, with flu patients typically starting to present before the end of the year. However, when the new season commences, clinicians will have a new clinical decision guideline (CDG) at their disposal. This will help them determine which of these patients should undergo testing to confirm whether they do have influenza or perhaps another respiratory condition that causes similar symptoms.

The push to develop this tool was prompted, in part, by the availability of more precise tests that can tell clinicians with a high degree of accuracy whether a patient has the flu virus.

“The tests that were used previously, up until three or four years ago ... were called antigen-based tests. Their performance was relatively poor, meaning their sensitivity for detecting influenza was moderate to low,” explains Richard Rothman, MD, PhD, vice chair of research and a professor in the department of emergency medicine at Johns Hopkins.

With a sensitivity of just 50% to 60%, the older tests did not give clinicians much assurance as to whether patients had influenza, Rothman notes.

“That, in combination with the fact that the clinical signs and symptoms are fairly nonspecific for influenza and overlap with other respiratory conditions, made it difficult for physicians in various settings, including EDs, to know if a patient in front of them actually had influenza,” he says.

However, the newer molecular tests are much better, with a sensitivity nearing 99%, Rothman shares.

“When you run a test, you really know if the person has influenza and can make a definitive diagnosis,” he says.

Given that every year emergency providers take care of patients with the flu, and there is a significant amount of morbidity and mortality related to this population, investigators were interested in developing an approach that could improve the ability of clinicians to determine which patients are most likely to have influenza. In turn, those most at risk can be tested and provided with appropriate treatment.

“There is also potential importance for emergency physicians in being able to make a diagnosis for ruling in or ruling out other conditions and potentially improving the use of antibiotics,” Rothman notes.

Not all patients suspected of having the flu necessarily require testing, Rothman acknowledges. Rather, he advises that testing is most important for cases in which influenza recognition is relevant to the patient’s treatment and disposition. This group includes patients at higher risk of complications, such as adults 65 years of age and older, pregnant women, those who are immune-suppressed, young children, those who are obese, and those with underlying comorbidities.

“These are the individuals where identification and recognition of disease has been shown to be helpful in terms of the benefits of treatment,” Rothman says. (Editor’s Note: The CDC offers more resources about diagnosing influenza online at:

Further, there is some adjunctive benefit to identifying influenza in patients who then return to the community. Healthcare providers can decrease the transmission of the virus by telling patients that they have influenza and providing guidance about basic infection control steps, Rothman explains.

“This can decrease the likelihood that these patients will transmit the virus to other people in the home or people who may be at high risk like young children or older adults they may be living with,” he says.

Deploy CDG at Triage

To develop the CDG, investigators sought to determine which signs and symptoms or combination thereof may be most helpful for predicting whether a patient has influenza based on his or her clinical presentation.

To find answers to this question, researchers conducted a study looking specifically at patients with high-risk characteristics, deeming them particularly vulnerable to flu complications. However, the researchers otherwise cast a very broad net, including any patient in this group who presented with fever or had a history of fever and/or had respiratory-related complaints such as a cough, runny nose, or sore throat.

“Then, we systematically enrolled all of those patients from triage and tested them using these new, molecular, gold-standard tests,” Rothman explains. “They had to be high risk, according to CDC criteria, so we knew they would have the greatest potential in terms of benefitting from treatment.”

After collecting these data at four participating EDs, the investigators conducted a statistical analysis to determine which factors were most predictive of a positive flu diagnosis. From this information, they developed a CDG based on four weighted questions that has a sensitivity of greater than 90% for identifying whether a patient has influenza, Rothman explains. Further, he notes that the four questions can be deployed easily at triage:

  • Does the patient have a new or increased cough? (2 points)
  • Does the patient have a headache? (1 point)
  • Does the patient report having a fever? (1 point)
  • Is the patient’s temperature at triage greater than 100.4° C? (1 point)

In the CDG, a score of 3 or higher indicates that influenza testing is warranted, according to the researchers. They further advised that incorporating the tool into ED practice during influenza season could improve diagnosis and treatment.

One curious aspect of the CDG is the fact that it asks both for the patient’s own subjective sense of fever as well as the patient’s actual temperature taken at triage. Rothman explains that this actually makes clinical sense in the ED.

“Many patients who have fevers or even high fevers at home ... may be taking antipyretics,” he explains. Consequently, that history question to the patient turned out to be an independent predictor for influenza in the study. “That is also related to the fact that there are variations in whether or not a patient actually has a fever over the course of a day when they have influenza. They may not necessarily be febrile when they present to the ED,” Rothman adds.

The CDG questions only take seconds to collect at triage. Rothman notes that the EDs in the Johns Hopkins system plan to take full advantage of the new tool.

“It has a benefit for healthcare providers, it has a benefit for the patient, and with EMR development, it can be fairly easily integrated into practice,” he says.


  1. Dugas A, Hsieh Y, Lovecchio F, et al. Derivation and validation of a clinical decision guideline for influenza testing in four U.S. emergency departments. Clin Infect Dis 2019; Mar 7. pii: ciz171. doi: 10.1093/cid/ciz171. [Epub ahead of print].


  • Richard Rothman, MD, PhD, Vice Chair, Research; Professor, Department of Emergency Medicine, School of Medicine, Johns Hopkins University, Baltimore. Email: