Background: Influenza A H5N1 has killed millions of birds and raises serious public health concern because of its potential to spread to humans and cause a global pandemic. While the early focus was in Asia, recent evidence suggests that Egypt is a new epicenter for the disease. This includes characterization of a variant clade 2.2.1.1, which has been found almost exclusively in Egypt.
We analyzed 226 HA and 92 NA sequences with an emphasis on the H5N1 2.2.1.1 strains in Egypt using a Bayesian discrete phylogeography approach. This allowed modeling of virus dispersion between Egyptian governorates including the most likely origin.

Results: Phylogeography models of hemagglutinin (HA) and neuraminidase (NA) suggest Ash Sharqiyah as the origin of virus spread, however the support is weak based on Kullback–Leibler values of 0.09 for HA and 0.01 for NA. Association Index (AI) values and Parsimony Scores (PS) were significant (p-value < 0.05), indicating that dispersion of H5N1 in Egypt was geographically structured. In addition, the Ash Sharqiyah to Al Gharbiyah and Al Fayyum to Al Qalyubiyah routes had the strongest statistical support.

Conclusion: We found that the majority of routes with strong statistical support were in the heavily populated Delta region. In particular, the Al Qalyubiyah governorate appears to represent a popular location for virus transition as it represented a large portion of branches in both trees. However, there remains uncertainty about virus dispersion to and from this location and thus more research needs to be conducted in order to examine this. Phylogeography can highlight the drivers of H5N1 emergence and spread. This knowledge can be used to target public health efforts to reduce morbidity and mortality. For Egypt, future work should focus on using data about vaccination and live bird markets in phylogeography models to study their impact on H5N1 diffusion within the country.

Background: Time series models can play an important role in disease prediction. Incidence data can be used to predict the future occurrence of disease events. Developments in modeling approaches provide an opportunity to compare different time series models for predictive power.

Results: We applied ARIMA and Random Forest time series models to incidence data of outbreaks of highly pathogenic avian influenza (H5N1) in Egypt, available through the online EMPRES-I system. We found that the Random Forest model outperformed the ARIMA model in predictive ability. Furthermore, we found that the Random Forest model is effective for predicting outbreaks of H5N1 in Egypt.

Conclusions: Random Forest time series modeling provides enhanced predictive ability over existing time series models for the prediction of infectious disease outbreaks. This result, along with those showing the concordance between bird and human outbreaks (Rabinowitz et al. 2012), provides a new approach to predicting these dangerous outbreaks in bird populations based on existing, freely available data. Our analysis uncovers the time-series structure of outbreak severity for highly pathogenic avian influenza (H5N1) in Egypt.