Thursday, October 3, 2013

Analysis of "Mapping Spread and Risk of Avian Influenza A (H7N9) in China"

The image below is a map from an analysis which predicts the distribution of H7N9 virus in China, based on a county by county mashup of a myriad of variables correlated to the human outbreak of H7N9 in China.

We (as well as the authors) see some problems / weaknesses with that analysis, but none-the-less the research paper is worth a read because of the raw data alone. We also think the analysis might have been significantly more elucidative had it included directional supply chain and logistics information.

Our quick take on the raw data, is that Chinese poultry markets were a force multiplier for the spread of H7N9 into the Human Population AND INTO THE WILD BIRD POPULATION.  Our SWAG is that the transmission vectors went something like this:

Wild bird -> Poultry Market
Poultry Market -> Wild Bird & Human
Human -> Wild Bird, Domestic Bird, & Human
Wild Bird -> Wild Bird & Human

That transmission chain would tend to explain how H7N9 spread around China, and why it grew in urban centers first, rural areas second, and why the geographic  disease spread was inverse square'ish at a right angle to the direction of wild bird migration. It would also tend explain why Shanghai was a nexus for transmission even though it appears there were several outbreak initiation points common to wetlands. 

Our assessment indicates that preventative closing of urban poultry markets before the  wild bird return migration hits full swing, and reducing the illicit trade in wild bird meat,  both might make for effective countermeasures.

However, as the non migratory feral bird population already appears widely infected with H7N9, such efforts may quickly be for naught as urban fecal contamination from pigeons and sparrows may reignite human H7N9 infections as temperature and humidity changes make transmission more amenable during the season change.

Source Data Link:

Mapping Spread and Risk of Avian Influenza A (H7N9) in China

1 comment: