Discussion

In this study, only 42% of children hospitalized with laboratory-confirmed influenza received antivirals, consistent with post COVID-19 pandemic trends. ^{Reference Frutos, Ahmad and Ujamaa4,Reference White, OHalloran and Sundaresan7} Clinical presentation, illness severity, and underlying comorbidities influenced antiviral prescribing. Documentation in the EHR suggested the following additional factors impacted treatment: (1) parental declination, (2) provider emphasis on risks over benefits, (3) guideline adherence variation, and (4) oral medication tolerance concerns.

Gaps in clinical decision-making documentation were prevalent; half of untreated children lacked EHR documentation regarding antiviral non-prescription. A study evaluating oseltamivir use among infants in inpatient and outpatient settings found that 40% of untreated infants lacked non-prescription documentation. ^{Reference Zaheer, Moehling Geffel and Chamseddine8} Poor documentation of antibiotic prescribing is also prevalent, highlighting the role of tools such as clinical decision-supports (eg, EHR smart forms) to prompt better recording of treatment decisions. ^{Reference Moccia, Arvinte, Lee and Langford9} Improved documentation could identify knowledge gaps or misconceptions to inform targeted influenza antiviral education.

Treated children were more likely to have comorbidities or severe illness, suggesting clinicians prioritize treatment for higher risk children. This aligns with prior studies showing that providers often rely on perceived illness severity, rather than guideline-based criteria alone, to influence treatment. ^{Reference Frutos, Ahmad and Ujamaa4,Reference Walsh, Schnadower, Zhang, Ramgopal, Shah and Wilson10,Reference Bassett, Rao and Beck11} Children admitted earlier in their illness were more likely to receive antivirals despite evidence that later antiviral initiation can still reduce risk of severe outcomes in hospitalized children. ^{Reference Adlhoch, Delgado-Sanz and Carnahan12} Providers often documented patients as “outside the [48 hour] window” for antiviral treatment, suggesting the application of outpatient treatment guidelines ³ to hospitalized inpatients. In a recent national survey, one-third of clinicians were unaware of treatment recommendations and practices varied among those familiar with the guidelines, reflecting differing interpretations of the limited pediatric inpatient evidence. ^{Reference Bassett, Rao and Beck11} Decision-support tools that provide guideline rationale may improve adherence and address provider uncertainty.

GI manifestations are common in children with influenza. ^{Reference Wolf and Antoon2} They were highly prevalent in our cohort and associated with lower antiviral use due to documented concerns for symptom exacerbation. Inpatient antimicrobial stewardship efforts could include clearer guidance on managing antiviral therapy when GI symptoms are present and oral intake is limited. Antibacterials were prescribed more often than antivirals, despite similar GI side effects and low bacterial co-infection rates, making this a high-yield stewardship opportunity.

Parental concerns about side effects contributed to antiviral non-prescription and were often noted following provider discussions of treatment risk. These findings underscore the importance of providers offering unbiased recommendations that outline both potential benefits and risks of treatment during shared decision-making. ^{Reference Fisher, Nguyen, Fouayzi, Singh, Crawford and Mazor13}

This study is subject to several limitations. Data are from a single influenza season and children hospitalized at 2 hospitals in Western NY, limiting generalizability. Influenza testing was at the discretion of the provider, and our data reflects only what was documented in the EHR, which may not capture the full clinical decision-making process. Despite these limitations, our findings identify patient, clinician, and family factors influencing antiviral treatment decisions among hospitalized children with influenza.

Antiviral prescribing in our cohort was low despite national recommendations for treatment of hospitalized children; treatment decisions were influenced by illness duration and severity. Documentation suggests that concerns about side effects and varied application of treatment guidelines also influenced prescribing. Better understanding of the drivers of these decisions may help inform strategies to improve guideline-concordant antiviral use.