Antimicrobial prescribing in adults with acute respiratory infections following rapid six-virus antigen testing: a prospective cohort study in Italian primary care

The overuse of antibiotics for acute respiratory infections (ARIs) is a global challenge. While multiplex rapid antigen detection tests (RADTs) have been proposed to reduce clinical uncertainty, evidence regarding their impact on antimicrobial prescribing in outpatient settings is limited. We aimed to evaluate the association between multiplex RADT results and antimicrobial prescription rates in adult outpatients with ARIs. This prospective cohort study was conducted in Genoa, Italy, during the 2025/2026 respiratory season. Community-dwelling adults (≥ 18 years) presenting to 27 general practitioners with an ARI were enrolled and followed for up to 30 days. At the index visit, patients underwent two nasopharyngeal swabs. The first was tested using a multiplex RADT targeting influenza A and B, respiratory syncytial virus (RSV), SARS-CoV-2, metapneumovirus, and rhinovirus. The second swab was sent to a laboratory for testing via standard-of-care real-time polymerase chain reaction (RT-PCR). The relative risk of antimicrobial prescribing between RADT-positive and RADT-negative patients was estimated using modified Poisson regression, adjusting for potential confounders. Of 682 patients tested by RADT, 18.6% tested positive for ≥ 1 virus; most cases were due to influenza A (7.8%) and RSV (5.9%). Compared to RT-PCR, the RADT showed low-to-moderate sensitivity (ranging from 0.7% for rhinovirus to 77.8% for SARS-CoV-2) and high specificity (> 99%). Sensitivity was higher for samples collected ≤ 3 days of symptom onset. Antibiotics were prescribed to 26.0% of RADT-positive and 33.9% of RADT-negative patients, with an adjusted relative risk of 0.73 (95% CI: 0.54–0.99). Antibiotic use was driven more by RADT results than RT-PCR: rates for true and false negatives (34.4% and 33.5%) were comparable, exceeding those for true and false positives (26.2% and 20.0%). All six antiviral prescriptions occurred among RADT-positive individuals. A positive multiplex RADT result was associated with slightly reduced antibiotic prescribing and increased antiviral use. However, low-to-moderate sensitivity limits its utility as a universal rule-out tool. Multiplex RADTs could be best deployed as rule-in diagnostics for early presenters and high-risk patients who would benefit from prompt antiviral therapy, rather than as a general strategy to reduce overall antibiotic consumption.