Thursday, May 28, 2015

Parsing The USDA's Avian Flu Numbers

# 10,106

Yesterday an article appeared on NPR’s that examines some of the confusion surrounding the numbers on the USDA’ s APHIS Avian Flu list.

The author points out – correctly – that the main chart only depicts commercial flocks that have tested positive, and often runs several days behind what individual states have already report on their websites.  

Understandable as states often publish their `presumed positive’ outbreaks, but the USDA waits for confirmation.

Additionally, other types of detections (wild birds & captive wild birds) are listed separately. A small table of captive wild birds is posted below the main list, and the wild bird list (n=72 as of 5/14)  is maintained on a separate PDF file.

Both can be accessed from the APHIS Avian flu page, but they are easy to miss. 

As a result some news reports have been based solely on the main chart’s tallies, which is really just a subset of the total number.  While confusing, it would be even convoluted if all three types were combined.

First a link to the NPR report - well worth reading in its entirety - after which I’ll return with more.

Are the USDA's bird flu numbers accurate?

Peggy Lowe
The avian influenza virus is moving fast and it’s already the largest outbreak ever in the U.S.

Since the beginning of May, we’ve watched and Tweeted what the U.S. Department of Agriculture has titled “ALL Findings” of bird flu. The updates come daily, about 3 p.m. Central, from a page run by the USDA’s Animal and Plant Health Inspection Service.

The latest update, posted Tuesday, looked like this:

The numbers are alarming. But are they accurate?

Yes. And no.

(Continue . . . ) 

The author also points out additional discrepancies between the number of outbreaks reported by the USDA and the OIE, and differences in the `flock size’.

Most of which can be attributed to the fact that the chain of transmission of information between agencies isn’t always as quick or efficient as the chain of transmission of the virus.

As to the question posed by the NPR article; Are the USDA's bird flu numbers accurate?

The USDA’s numbers are at best a `snapshot in time’ of lab confirmed cases, based on specific criteria (commercial flocks, wild bird, captive wild bird) - and due to delays in reporting and human error – (like all such numbers) are best consumed with a grain of salt.

That said, I don’t know of any place with 100% accurate and up-to-date figures. 

While far from perfect, the USDA’s numbers do provide us with a sense of the size and trajectory of the avian flu outbreak in this country.  

And while I don’t tend to quote their tallies very often (for many of the reasons stated above), when dealing with an extremely fluid situation, sometimes that’s the best you can hope for.

No comments: