What is the number needed to treat (NNT) to prevent one hospitalization due to influenza?

Number Needed to Treat for Influenza Hospitalization Prevention

The number needed to treat (NNT) to prevent one influenza-related hospitalization varies substantially by population risk group, ranging from approximately 40-50 for high-risk adults to several hundred for healthy younger adults, with antiviral treatment (oseltamivir) showing an NNT of approximately 63 to prevent one hospitalization when started early.

Understanding the Context

The NNT for preventing influenza hospitalization cannot be expressed as a single number because it depends critically on:

• Baseline hospitalization risk in the specific population being treated 1
• The intervention being used (vaccination vs. antiviral treatment) 2
• Timing of intervention (prophylaxis vs. early treatment) 3, 2

Baseline Hospitalization Rates by Population

Understanding baseline risk is essential for calculating meaningful NNT values:

Highest Risk Groups 1

• Infants <6 months: 1,040 per 100,000 population (approximately 1% hospitalization rate)
• Children 0-4 years with high-risk conditions: 500 per 100,000 population
• Adults ≥65 years: 200 to >1,000 per 100,000 population
• Adults 45-64 years with high-risk conditions: 80-400 per 100,000 population

Lower Risk Groups 1

• Healthy children 0-4 years: 100 per 100,000 population
• Healthy children 2-4 years: 8-136 per 100,000 population
• Healthy adults 15-44 years: 20-30 per 100,000 population
• Healthy adults 45-64 years: 20-40 per 100,000 population

Treatment-Specific NNT

Antiviral Therapy (Oseltamivir/Zanamivir)

Meta-analyses demonstrate that early oseltamivir treatment reduces hospitalization risk by 63% when started within 48 hours of symptom onset 2. This translates to:

• NNT ≈ 63 to prevent one hospitalization in mixed populations when treatment is initiated early 2
• Greatest benefit occurs within 24 hours of symptom onset, reducing illness duration by approximately 24 hours 3
• Zanamivir shows equivalent effectiveness to oseltamivir in preventing hospitalization (adjusted hazard ratio 1.01,95% CI 0.96-1.06) 2

High-Risk Subgroups

For patients at elevated baseline risk, the NNT is lower (more favorable):

• Adults ≥65 years: The 63% relative risk reduction applies to a higher baseline rate, yielding an NNT of approximately 40-50 2
• Patients with chronic lung disease: Similar NNT range of 40-50 based on higher baseline hospitalization rates 2

Vaccination

While the provided evidence does not give specific NNT values for vaccination, guidelines emphasize that vaccination is the most effective means of reducing influenza impact 1. The NNT for vaccination would be calculated based on vaccine effectiveness (typically 40-60% in matched seasons) applied to population-specific baseline hospitalization rates.

Clinical Application Algorithm

When treating an influenza patient, calculate the likely benefit:

1. Identify baseline risk category using age and comorbidities 1
2. If high-risk (elderly, very young, comorbid conditions): NNT approximately 40-63 for preventing hospitalization with early antiviral treatment 2
3. If low-risk (healthy adults 15-64 years): NNT likely >200 based on low baseline hospitalization rates of 20-40 per 100,000 1
4. Timing matters critically: Treatment must start within 48 hours, ideally within 24 hours 3, 2

Important Caveats

The NNT concept has inherent limitations when applied to influenza hospitalization prevention 4:

• NNT values are population-specific and cannot be generalized across different risk groups 4
• The measure reflects both the exposure effect and the specific intervention deployed 4
• During the 2009 H1N1 pandemic, hospitalization rates were tenfold higher than typical seasons, dramatically altering NNT calculations 1

Mortality considerations: While hospitalization is important, >90% of influenza deaths occur in adults ≥65 years, with 20,000-40,000 deaths occurring during severe epidemic years 1. This mortality burden reinforces treating high-risk patients regardless of NNT calculations.

Resistance patterns: Increasing resistance to antivirals, particularly in Japanese children, may affect future NNT values 5. Methicillin-resistant Staphylococcus aureus secondary pneumonia carries high mortality and should be covered empirically during influenza season 6.