Diagnostic performance of clinical symptoms and rapid test for influenza in primary care

Abstract

Primary care doctors usually rely on clinical symptoms to judge the likelihood of influenza, but its accuracy is affected by prevalence of influenza and test sensitivity and specificity of symptom(s). We conducted a study to evaluate the diagnostic performance of influenza-like illness (ILI) criteria and a rapid point-of-care test for influenza.Patients presenting with acute respiratory illness to 20 private out-patient clinics within 72 hours of illness onset were enrolled and tested for influenza A and B viruses by rapid antigen test and subsequently by RT-PCR. ILI is defined as fever of at least 37.8 degree plus cough or sore throat. The sensitivities and specificities of ILI and rapid test among enrolled patients were calculated using RT-PCR results as the gold standard.626 ARI patients were recruited over three influenza epidemics from 2014 to 2015. 255 (40.7%) subjects were positive for influenza A or B by RT-PCR. 362 (57.8%) of them fulfilling the ILI criteria. 99 (15.8%) of all ARI patients tested positive for influenza A or B by rapid antigen test. Sensitivity (SN), specificity (SP), positive predictive value (PPV) and negative predictive value (NPV) of ILI criteria were 79.2%, 56.9%, 55.8% and 79.9%, while those for rapid antigen test were 37.3%, 98.9%, 96.0% and 69.6%, respectively. Net SN, SP, PPV and NPV of their serial application were 29.5%, 99.5%, 97.8% and 67.3%, while those of their parallel application were 87.0%, 56.3%, 57.7% and 86.3%. ILI was a sensitive criteria to detect influenza infection and a negative result could allow 80% chance to rule out influenza and consider other pathogens during influenza epidemics. Rapid antigen test can provide highly specific results for influenza among the ARI patients and guide specific treatment for them. Using rapid antigen test in parallel can improve net PPV and NPV compared with using ILI criteria alone.