Rapid Influenza Diagnostic Tests Have High False Negative Rates

by Joseph Esherick, M.D., FAAFP, FHM

We are in the midst of an aggressive influenza season right now that should continue into March, so it is nice to review the testing accuracy of standard influenza point-of-care tests that are commonly used.  There are many Rapid Influenza Diagnostic Tests (RIDTs) that detect the viral antigen that are more commonly used than the reverse transcriptase PCR (RT-PCR) testing assays that detect viral RNA or nucleic acids.  The RT-PCR assays are more expensive and take longer to perform (~90 minutes), but they have an 80-90% sensitivity and 95% specificity compared with a low sensitivity of 50-70% for RIDTs.  

Other important factors to consider during interpretation of the lab results is collection technique, the prevalence of influenza in the community, and duration of infection.  Nasopharyngeal influenza swabs should be inserted at least 1 inch into the nose until you meet some resistance for a good specimen.  A swab of the anterior nasal septum will yield a poor specimen with higher false-negative test results.  In addition, there will be a higher false negative rate of RIDTs when the test is negative during the peak of influenza season (high prevalence of influenza in the community).  Finally, the rate of false negative RIDT tests increases the longer the duration of symptoms, especially if specimen collection is >4 days of symptoms.

The bottom line is that you should consider treating a person empirically with antiviral medication (e.g., oseltamivir) if they present within 72 hours of the onset of symptoms and have classic symptoms of influenza.  Classic signs and symptoms of influenza include fever, myalgias, headache, lack of energy, dry cough, sore throat, nasal congestion, and possibly runny nose. Fever is not always present in influenza patients, especially in elderly persons. The fever and body aches can last 3-5 days and the cough and lack of energy may last for 2 or more weeks, especially in the elderly.  

The other important concept is that you should only perform an influenza test if the results will influence your management.  Consider empiric therapy for patients presenting with classic symptoms within 72 hours of onset during the peak of influenza season.  Other potential reasons for testing is for epidemiologic purposes and in institutions with confined residents (e.g., skilled nursing facilities).

For more accurate influenza testing, train personelle carefully in proper specimen collection, obtain specimens within 72 hours of symptom onset and use RT-PCR tests rather than RIDT tests.  Also, remember that influenza can lead to severe ARDS and can kill so the most important message for your patients is to vaccinate them early!

Reference:

https://www.cdc.gov/flu/professionals/diagnosis/labrolesprocedures.htm