Diagnosing flu still not straightforward

A flu diagnosis is not always as simple as a cough plus a fever. But just what combination of signs and symptoms may prove more accurate remains unclear, suggests a new review of 12 studies on the topic.

According to lead researcher Dr Mark Ebell, of The University of Georgia in Athens, most doctors use some variation of an informal rule: "If it's flu season in my community, and someone comes in with fever and cough, then they have the flu."

"That rule, while simple, is only correct about half the time," he said.

Better formula

A better formula, Ebell added, could help doctors identify patients who might benefit from further tests or treatment with drugs such as Tamiflu. Additional likely benefits would be fewer healthcare dollars spent on testing and treating people who don't need it, and fewer patients missed who do.

In the new review, Ebell and his colleague Anna Afonso identified 12 studies that had looked at patients with a respiratory tract infection to see if certain combinations of factors - including sore throat, headache, cough, fever and fatigue - might predict the presence of influenza virus.

As they report in the Annals of Family Medicine, the rate of flu detected ranged widely among the studies: from 7% to 79%.

The findings

Such great variability in the study results, small study group sizes, and lack of rigor in the study designs hindered evaluation of diagnosis guidelines in the current review. The best Ebell and Afonso could conclude was that common rules-of-thumb based simply on fever and cough - or fever, cough and an acute onset of those symptoms - offered "modest" accuracy.

Overall, they found, a slight majority of the patients initially receiving a flu diagnosis based on such rules actually had the virus and more than half of patients in whom flu was ruled out were truly flu-free.

While the popular guidelines for diagnosing flu and the studies of those guidelines both leave room for improvement, the new finding could at least inform the design of future studies, according to the researchers.

Ebell suggested that data from large groups of patients with symptoms of respiratory tract infections are needed to determine the best set of signs that predict flu.

Prevalence of flu

"That can be used in combination with information about the overall prevalence of flu in the community to identify high-risk patients who might benefit from treatment with Tamiflu, low-risk patients who need neither testing nor treatment, and moderate-risk patients who might benefit from point-of-care testing," he said.

Ebell and Afonso recently completed just such a rigorous study of a rule that incorporates four symptoms. The results have not yet been published.

In the meantime, he said, the public can continue to do its part in limiting the number of people that get sick with the flu. "The best thing you can do is to get your flu shot," advised Ebell, noting that the Centres for Disease Control and Prevention website provides a map of recent flu activity (http://bit.ly/1KBVsR).

"If you have fever, cough, and body aches during flu season, there is a good chance that it is the flu," he added. "For most folks, staying home so you don't spread the infection, resting, taking acetaminophen for fever, and getting adequate fluids are the best things you can do."(Reuters Health/ January 2011)

Read more:   
Influenza myths and facts

How flu viruses changeHow flu viruses change  How flu viruses change


We live in a world where facts and fiction get blurred
In times of uncertainty you need journalism you can trust. For only R75 per month, you have access to a world of in-depth analyses, investigative journalism, top opinions and a range of features. Journalism strengthens democracy. Invest in the future today.
Subscribe to News24