![[ Visit Client Website ]](images/banner.png)
BACKGROUND: Influenza may be difficult to distinguish from other respiratory viruses by symptoms, and rapid diagnostic tests may not be reliable. Since influenza antiviral treatments have been shown to have the greatest effect when initiated early, accurate clinical assessment of the probability of influenza at the time of patient visit is helpful in optimizing treatment.
METHODS: From October 2009 through July 2013, the Influenza Incidence Surveillance Project (IISP) was conducted through 12 public health departments, each with 4-6 participating outpatient care providers. Respiratory specimens were collected from the first 10 patients presenting with fever and cough or sore throat each week and tested for influenza by RT-PCR. We evaluated symptoms associated with laboratory-confirmed influenza as a dichotomous outcome among pediatric and adult patients using logistic regression, and further by influenza type using multinomial logistic regression. We included cough and sore throat in the models, since neither is an absolute requirement of the case definition. We adjusted for age, gender, surveillance season, geographic region, and month of the year (dichotomous model only).
RESULTS: Influenza was detected in 3,887 (28.0%) of 13,869 patients with ILI; of those, 71.0% were influenza A (60% A(H1N1), 37.4% A(H3), and 2.9% were not subtyped) and 29.0% were B. Laboratory-confirmed influenza was associated with cough, rhinorrhea, and myalgia in all patients. Sore throat was also significantly associated in pediatric patients, and chills were significantly in adult patients. Ear ache and diarrhea were significant negative predictors of influenza in both groups. In the multinomial models, cough was the only symptom significantly associated with each type and subtype of influenza. Rhinorrhea was associated with both A(H1N1) and A(H3) detection. Myalgia was associated with A(H3) among all patients, and influenza B among adults only. The constellation of significantly associated symptoms other than cough, rhinorrhea and myalgia varied both by age group and influenza virus type.
CONCLUSIONS: Among all patients with clinician-assessed ILI, cough, rhinorrhea, and myalgia were significantly associated with the detection of influenza virus, but significant associations varied according to the specific type or subtype of influenza infection. These findings may provide guidance to clinicians regarding commonly reported symptoms specific to the type of influenza predominating in a season.