Real-time Forecasting of Pandemic Influenza (H1N1-2009)
Author Information
Author(s): Nishiura Hiroshi
Primary Institution: The University of Hong Kong
Hypothesis
Can a simple discrete time stochastic model effectively forecast the incidence of pandemic influenza in real-time?
Conclusion
The proposed model successfully forecasts epidemic trends, especially after the peak incidence, using a simple computation of uncertainty bounds.
Supporting Evidence
- The model's forecasts improved significantly after the epidemic peak.
- Parameter estimates were derived from observed incidence data.
- The model can account for various types of heterogeneity when more data is available.
Takeaway
This study created a way to predict how many people might get sick during a flu outbreak, using just a few numbers and some math.
Methodology
A discrete time stochastic model was developed and applied to weekly incidence data of H1N1-2009 in Japan.
Potential Biases
The estimates may be biased due to reporting delays and the assumption of a homogeneously mixing population.
Limitations
The model's accuracy is limited before the epidemic peak due to reliance on parameter estimates.
Participant Demographics
The study focused on the population of Japan during the H1N1-2009 pandemic.
Statistical Information
Confidence Interval
95%
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website