Bird flu data: how good?

H5N1 bird flu is now in 55 countries, in each of which it has severe economic consequences on the poultry industry and carries with it an unknown but potentially catastrophic public health threat. Since it is s a disease of animals (primarily birds), much of the work has been done by veterinarians and ornithologists. One would expect us to have a great deal of information, given the attention this nasty virus has received, in the laboratory, the field and among the general public and press.

And we do. But a recent paper in the journal BioScience (published by the American Institute of Biological Sciences) asks how accurate it is:

The article's authors, Maï Yasué, Chris J. Feare, Leon Bennun, and Wolfgang Fiedler, made use of the Aiwatch (avian influenza watch) e-mail forum to gather information for their article from sources worldwide. They describe several instances in which the species of an infected wild bird was incorrectly or inadequately recorded--sometimes just as "wild duck," for example--and others in which the bird's sex and age were misidentified.

Likewise, reported details of the location and time of discovery of an infected bird often lack specificity, yet they are crucial for a good understanding of the virus's spread. Information about capture and sampling methods and other species in the vicinity of an infected bird has also often been inadequately described.

The authors end their article with a plea for greater involvement by ornithologists and ecologists in H5N1 research and monitoring. (Science Daily)

In a number of previous posts we have deplored the sorry state of clinical data collection (see here, here and here).

It now appears we may not be much better off on the animal side.

Tags

More like this

Revere, I think the same would apply for the people that get it and are misdiagnosed or just ignored in my opinion. Stuff that appears to be flamingly evident for H5N1 turns out later on to be dengue or vice versa. I think that when it decides to become infectious in a High Path way, we wont know it when we see it, we will go into denial that its here, and then if its High Path we will go into the night by the body bag loads.

DEFRA hiring the cats to be in charge of the canaries in the UK likely allowed for H5 to enter the country undetected. Not only did they not find H5, they didnt find even the baseline for Europe for LPAI. So we might just take a long look to see if the same things are happening here. I for one dont want the Audubon Society checking up on this one. Too many sympathetic hearts that would object to the use of turgitol on large areas of waterfowl and migrating birds if it came down to that and they might intentionally fail to report a case of HPAI if it was found.

All in all humanity is doing a bad job of policing this stuff and combined with crappy bird vaccine that this stuff continues to evade, we are I think going to get popped and hard.

By M. Randolph Kruger (not verified) on 06 Nov 2006 #permalink

It is important to get accurate test results back to the communities involved in the sampling. I have seen some news media results but I prefer the actual test results to understand something of the meaning. ("no bird flu" or do you mean no HPAI H5N1? Which birds and how many had what kind of specific results?)

Keep in mind the hunters who collected samples (and ate the birds) and sent their children out to gather eggs, did so with some apprehension. A generic response of "no bird flu in North America" when the newsmedia reports various strains in wild birds harvested afterwards (or in previous years) in other states-- doesn't instill confidence.
http://ykalaska.wordpress.com/2006/06/09/test-results-maybe/

So far, the actual test results are not available to the general public. And the HEDDS table of testing results, "LOW PATHOGENIC "NORTH AMERICAN" H5N1 AVIAN INFLUENZA STRAIN IN WILD BIRDS PRESUMPTIVE AND CONFIRMED TEST RESULTS" doesn't include the 16,150 or so Alaska samples. Does this mean Alaska isn't part of the US or North America or that none of our birds has any form of H5N1?

I don't understand the secrecy; if it was done to keep the public from getting confused, it's not successful. If it was done to keep the public from "panic", Fish & Wildlife has already claimed the public was "freaking out" even before the birds had arrived to be swabbed.

If the data are confirmed, is there any way to get a hold of them? bird species, #, locale, year, what found

A couple of weeks back i described the results of a recently published scientific study in which researchers infected a number of different species of wild waterfowl with HPAI H5N1, and how only a couple of the test species actually became sick while the rest, although definitely infected with the virus, did not show any signs of illness. This study, IMHO, didn't get nearly the attention it deserved. To me that result is a huge red flag: What it says is that there may be many migratory birds out there carrying the High Path H5N1 virus asymptomatically. The prior assumption that we will get a "canary in the coal mine" type warning from birds dropping dead around us is thus a dangerous misconception if this study is to be believed.

And Randy makes a very good point about the potential for misdiagnosis or overlooking the first wave of human cases should they come. Just this past week we had a young student out for 10 days with a very high fever, and I started thinking 'bird flu'? Well, it wasn't - I presume - as she finally got well. But how can we be sure: no tests were run, the parents didn't even take her to the doctor until yesterday, after the fever finally broke. Maybe they would have run some tests if she had become critically ill or died, but probably even then not for H5N1, not here. No one is expecting it here. Not until it began taking out people by the tens, at least, would someone begin to suspect a bird flu epidemic, and by then it would be too late.

By mary in hawaii (not verified) on 06 Nov 2006 #permalink

MiH: The story of asymptomatic wildfowl has been with us from the outset. It is normal for high path virus to be mild or symptomatic in wild quatic birds but lethal for terrestrial birds (poultry). This is the basis for the migratory bird spread scenario, since sick birds don't migrate long distances.

Regarding accurate diagnosis of inlfuenza-like illness (L:I), this, too, has been with us for decades. It is the problem faced by surveillance programs. It is too expensive and impractical to apply invasive and/or difficult diagnostic procedures to every ILI, since there are millions of them daily. Devising a system that does this is a difficult task that CDC approaches with at least three different systems we have posted about at EM at one time or another. If you look at the CDC site under surveillance you will get an idea of what is involved. It isn't that you and Randy are wrong, it's that the problem has been long recognized and it is difficult to solve. Sometimes the World doesn't want to cooperate.

Please read this quote carefully, in order to understand what may be wrong with bird flu data:
In talking about the Spanish Flu Pandemic of 1918, the author says:
"Influenza A could have been the innocent passenger virus in 1918,as Barr-Epstein and cytomegalovirus were in AIDS."
Indonesia has one of the highest TB rates in the world.
It is possible many patients being diagnosed as H5N1 positive, be co-infected with both H5N1 and TB.
There are now more people infected with XDR-TB than with bird flu. XDR-TB is epidemic in the entire world, and highly contagious.
To diagnose TB, they still depend on microscopic examination of sputum. This method allows for the detection of 45 to 65% of those infected. And the rate of detection is even lower for those with HIV and TB.
So in Indonesia, where there are clusters of H5N1, are there also clusters of MDR-TB and XDR-TB. If you are a doctor is Indonesia, are you going to check the sputum of the patient of TB, if the patient has already tested positive for H5N1. And even if you test, you may only detect 45 to 65% of those infected.
Many TB patients have develop pneumonia, and bleed from the mouth, just like H5N1 patients do. And TB can drill holes in your lungs.
And HIV is causing TB to increase 6 times faster than it would normally.
In a comment of ARDS, Acute Respiratory Distress Syndrome, one researcher states: "Roger and others (lung specialists) favor suspecting TB in all cases of acute respiratory failure of unknown origin.

Might the U.S. Consulate General have had this in mind:

Henry Niman November 7, 2006 09:10 AM: "Sequences released today show S227N in Egyptian patient from the spring infections. To date only four sequences have been released from patients in Turkey (two had S227N) and one in Azerbaijan. --- a 2006 H5N1 sequence from a goose in Shantou has THREE changes in its receptor binding domain"

when releasing this:
http://hongkong.usconsulate.gov/ci_avian_2006110301.html

Also thinking of this:
http://tinyurl.com/w3z2p ?

Just speculating...

There are at least THREE 2006 H5N1 isolates in Shantou with all three changes in the receptor binding domain (K222R, V223I, S227R) as well as at least two others with K222R or V223I.

This is recombination in action, and the action is in the middle of the receptor binding domain (Shantou isoloates also have a novel cleavage site, QRERRKKR).

H5N1 doesn't read press reelases.

Howdy Henry,

You have been working under extraordinarily difficult conditions vis a vis restricted access to H5 sequence samples which would, if globally released, analyzed and published, chart the homologous, and not random, trajectory of recombination within various regional H5 strains... Frankly, I'm just a novice and am yet to understand the genetic lexicon to a degree which confers "competence." But, I do understand the fear in your writing which underlies the phrase, "changes in receptor binding domain" -- the implications of H5 strains recombining further along this path, becoming increasingly efficient in H2H transmission without a decrease in case fatality rate!

Now, I read the Nov '06 JVI.org article "The NS1 Gene Contributes to the Virulence of H5N1 Avian Influenza Viruses". In this article, Bu Zhigao from China's Harbin Veterinary Research Institute states the NS1 gene is key to blocking H5 viral evolution -- get a human vaccine halting the amino-acid sequence of the NS1 gene and, hey presto, end of H5 pandemic!

The question I ask is simple -- can the H5 virus do a recombination dance around this "NS1 magic bullet"!?!

Cheers:*) and Aloha pumehana -- Jon

By Jon Singleton (not verified) on 07 Nov 2006 #permalink

Hi Revere. Sorry to keep on with something that must seem like pretty dumb questions, but....The source of my confusion is regarding the difference between High Path H5N1 and Low Path H5N1. I read reports that such and such mallards or teals or whatever were tested and found to have H5N1, but - not to worry, folks - it's low path and harmless to humans. My question is how do they know? Do the HP and LP varieties share H5 and N1 sites, but their other genes are different which leads to the difference in their lethality? I may be misinterpreting Henry Niman's posts - but it sounds as if genetic sequences on at least some of these - such as the Prince Edwards Swans - have not been released. My overall point: if the difference between HP and LP H5N1 is determined by evaluating the differences in their genetic sequences, antibodies or some other definitive test, fine. If it is determined by how sick the infected birds get, not so fine.

By mary in hawaii (not verified) on 07 Nov 2006 #permalink