Predicting the Epidemic Sizes of Influenza A/H1N1, A/H3N2, and B: A Statistical Method
2011

Predicting Influenza Epidemics

Sample size: 12 publication 10 minutes Evidence: moderate

Author Information

Author(s): Edward Goldstein, Sarah Cobey, Saki Takahashi, Joel C. Miller, Marc Lipsitch

Primary Institution: Harvard School of Public Health

Hypothesis

Can routine surveillance data be used to predict the sizes of influenza epidemics caused by different strains?

Conclusion

The study developed a method to predict the sizes of influenza epidemics, showing that early circulation of one strain can reduce the incidence of others.

Supporting Evidence

  • High infection rates with one strain can interfere with the transmission of other strains.
  • The method accurately predicted the whole-season cumulative incidence for each strain.
  • Predictions were generally made several weeks before the peak incidence of the strains.

Takeaway

This study helps us understand how to predict flu outbreaks by looking at how different flu strains affect each other.

Methodology

The researchers used CDC influenza surveillance data from 1997 to 2009 to analyze the incidence of different strains and develop a prediction algorithm.

Potential Biases

The quality of the incidence proxy may vary, affecting the accuracy of predictions.

Limitations

The model is based on only 12 seasons of data and may not account for all factors affecting influenza transmission.

Participant Demographics

Data from the US population were used, but specific demographics were not detailed.

Statistical Information

P-Value

7·10−10

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1371/journal.pmed.1001051

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication