Key Takeaway

Baloxavir is the only antiviral that probably shortens symptoms (~1 day) and may reduce hospitalizations in high-risk outpatients without increasing adverse events; oseltamivir provides little to no benefit on hospitalization or mortality and likely increases treatment-related adverse events.

Background

Antiviral therapy for nonsevere influenza is widely used, yet the optimal agent remains uncertain. Prior reviews often lacked absolute effects, certainty ratings, or recent trials. This systematic review and network meta-analysis aimed to compare all approved direct-acting antivirals on patient-important outcomes to inform guideline updates.

Patients

• Population: Outpatients with suspected or laboratory-confirmed nonsevere influenza (WHO definition; severe illness requires hospitalization)
• Trials/participants: 73 randomized clinical trials; 34,332 participants
• Risk strata: Low-risk and high-risk for severe complications (per WHO definitions)
• Timing: Most trials initiated treatment within 48 hours of symptom onset
• Demographics: Median of mean age 35 years; ~50% male

Intervention

Direct-acting influenza antivirals: baloxavir, oseltamivir, zanamivir, laninamivir, peramivir, umifenovir (arbidol), favipiravir, and amantadine.

Control

Placebo or standard care (and head-to-head comparisons among antivirals within the network).

Outcome

• Primary: Hospital admission (focus on high-risk outpatients)
• Secondary: Time to alleviation of symptoms, mortality, ICU admission, duration of hospitalization, any adverse events, adverse events related to treatment, serious adverse events, emergence of resistance

Study Design

• Systematic review and frequentist network meta-analysis of randomized clinical trials
• Search: Multiple databases from inception to September 20, 2023; PROSPERO-registered; PRISMA-NMA compliant
• Risk of bias and certainty of evidence assessed using GRADE

Level of Evidence

Level I (systematic review and network meta-analysis of RCTs).

Follow up period

5–29 days across studies.

Results

Primary outcome: Hospital admission (high-risk outpatients)

• Baloxavir: Absolute risk difference (ARD) −16 per 1,000 (95% CI −20 to +4); low certainty.
  • NNT to prevent one hospitalization ≈ 63 (CI crosses no effect).
• Oseltamivir: ARD −4 per 1,000 (95% CI −10 to +4); high certainty of little/no effect.
  • NNT ≈ 250 (CI includes no effect).
• Zanamivir: ARD +4 per 1,000 (95% CI −4 to +15); high certainty of little/no effect.
  • NNH ≈ 250 (CI includes no effect).
• Low-risk outpatients: All antivirals showed little or no effect on hospitalization (high certainty).

Secondary outcomes

• Time to alleviation of symptoms:
  • Baloxavir: Mean difference (MD) −1.02 days (95% CI −1.41 to −0.63); moderate certainty.
  • Umifenovir: MD −1.10 days (95% CI −1.57 to −0.63); low certainty.
  • Oseltamivir: MD −0.75 days (95% CI −0.93 to −0.57); moderate certainty and judged no important effect.
  • Zanamivir: MD −0.68 days (95% CI −0.93 to −0.43); moderate certainty (small, likely not clinically important).
• Adverse events related to treatment:
  • Baloxavir: ARD −32 per 1,000 (95% CI −52 to −6); high certainty.
    • NNT to prevent one AE ≈ 32.
  • Oseltamivir: ARD +28 per 1,000 (95% CI +12 to +48); moderate certainty.
    • NNH ≈ 36.
• Any adverse events:
  • Baloxavir: ARD −52 per 1,000 (95% CI −87 to −17); high certainty.
    • NNT to prevent one AE ≈ 20.
  • Zanamivir: ARD −28 per 1,000 (95% CI −48 to −7); low certainty.
    • NNT ≈ 36.
• Mortality (low- and high-risk): Little or no effect for all antivirals vs standard care/placebo (high certainty); absolute differences near zero.
• ICU admission:
  • Oseltamivir: ARD −2 per 1,000 (95% CI −7 to +3); high certainty of little/no effect.
    • NNT ≈ 500 (CI includes no effect).
  • Peramivir: Very uncertain.
• Duration of hospitalization: Evidence very uncertain (few small trials; oseltamivir, peramivir, zanamivir).
• Emergence of resistance:
  • Baloxavir: May increase resistance (~10% of treated; low certainty).
  • Zanamivir: May have no important resistance signal (low certainty).
  • Oseltamivir/peramivir: Very uncertain.

Limitations

• Sparse data for some drugs and outcomes; very low event rates for hospitalization and mortality reduce power.
• Some risk of bias (allocation concealment, blinding of outcome assessors/analysts).
• Heterogeneity in populations, dosing, timing from symptom onset, and durations.
• Minimal within-trial subgroup data; no credible subgroup effects detected.
• Assumptions about minimal important differences may be debated.
• Resistance outcomes are complex to interpret and inconsistently reported.

Funding

Supported by the World Health Organization. The funder had no role in study design, conduct, analysis, manuscript preparation, or publication decisions.

Citation

Gao Y, Zhao Y, Liu M, et al. Antiviral Medications for Treatment of Nonsevere Influenza: A Systematic Review and Network Meta-Analysis. JAMA Internal Medicine. 2025;185(3):293-301. Published online January 13, 2025. doi:10.1001/jamainternmed.2024.7193