Flu testing is usually not cost-effective

When Michael Bornstein, MD, a pediatrician in Sugar Land, Texas, has a child in his office whom he suspects might have influenza, he performs a rapid test for presence of the virus. In about 10 minutes, the test gives him guidance on whether to prescribe antivirals, antibiotics or neither. The information also gives parents a clearer picture of why their child is sick and convinces them to be more vigilant in complying with the treatment.

"It is impossible to give children these medications," said Dr. Bornstein. Many are more likely to spit a medicine out than swallow it. "If I run the test and it turns out positive, parents will get that medicine in one way or another."

Dr. Bornstein is part of a growing cadre of physicians who have incorporated rapid flu testing into their practices. The cost of this trend, however, is coming under scrutiny with a growing number of researchers asking if the test -- with its price tag of approximately $20-- is worth the money.

Most recently, a study in the January/February Annals of Family Medicine suggested that, in most instances, a physician's clinical judgment combined with antiviral treatment is the most cost-effective strategy.

Researchers at the Medical University of South Carolina analyzed the costs of no treatment, testing and treatment, and empirical treatment for a hypothetical group of adults at high risk for influenza complications. Testing was only beneficial when flu prevalence was between 30% and 40% and the most expensive antivirals were being prescribed. The authors suggest that testing may be useful for public health surveillance but not necessarily the best option for the average primary care medical practice.

"This test adds cost to patient care without a great deal of benefit, and it should not be used as a routine adjunctive test for everyone who comes in with the sniffles," said William Hueston, MD, the study's lead author and chair of the department of family medicine at Medical University of South Carolina in Charleston. "It should be used selectively in patients who are really a diagnostic dilemma and used when it will make a difference in treating the patient."

Other studies have also suggested that testing is not always the best option. Researchers at Tufts University in Boston, for example, have published several studies suggesting that testing has very little role for those at low risk of complications, but that it might play a more important role in determining treatment for those at higher risk. But this was also dependent on how much flu was about.

"The questions are: How likely is it that the patient has the flu to begin with, and how bad will it be if they have the flu and I don't treat them?" said Michael Rothberg, MD, MPH, assistant professor of medicine at Tufts who has been the lead author on several such studies. "Doctors should be using the test in selected patients, but it really depends on who the patients are and what time of year it is."

Critics of these studies charge that determining exactly whose treatment may be changed by testing and whose will not is not something that may be immediately apparent. Also, treating without testing could lead to large numbers of people receiving unnecessary medications.

"Hindsight will be 20/20," said Dr. Bornstein. "It's impossible to know when they come in what the chances are that they're going to have the flu, and you can't just blindly treat."

Testing has benefits beyond cost. Patients are more likely to take antivirals with laboratory-confirmed influenza and not as likely to insist on antibiotics, which can help stem the problem of antibiotic resistance. Patients also benefit by knowing what they do and do not have.

"The only issue in this situation isn't cost," said Marc Siegel, MD, associate professor of medicine at New York University. "Early determination of what you're treating is always helpful from a medical if not a cost-benefit point of view."