Technocrat

Where do Flu Pandemic Projected Numbers come from?

zogger
Sun Jul 09 14:11:49 -0700 2006
Health+and+Medicine.
manage

This is still puzzling to me. From what we have seen so far, when a human gets avian flu, there's a 50% mortality rate, more or less. Those are the current stats. So, while doing my daily check on some avian flu news, I see articles like this one, a report from Tennessee.

Now here's the problem, and I've seen this sort of thing from WHO reports, etc. It's this chronic low balling, at least it appears this way.. In the article they "predict" upwards of 30% infection rate if the flu mutates, etc. Along with that, the normal bad news scenario. From the article:

"Tennessee's plan predicts the virus would take six to eight weeks to move through a community, sickening about 30 percent of residents. It forecasts between about 4,200 and 38,000 deaths statewide. About 900,000 Tennesseans would seek outpatient care. In a severe outbreak, an estimated 198,000 would be hospitalized."

OK, from the US census bureau : Tennessee population in 2005 = 5,962,959. 30% of that (that they claim will get infected, again, questionable, but let's use their numbers again) is 1,788,887.7. 50% of that (back to current mortality stats) is 894,443.85

See, the numbers don't come close to jiving. If you use the only *real* numbers we have to look at with a potential "pandemic", you get a lot closer to a million people *expiring*, let alone getting just ill, and that is just one small state. I picked this news article at random because they had some figures in it and it references some huge 200 page official report.

So, where are they getting their low number predictions from, just picking a small percentage out of the ether? What am I not seeing here? Any medgeeks care to comment?

Where do Flu Pandemic Projected Numbers come from?
Margot White
Sun Jul 09 15:56:47 -0700 2006
 manage

These numbers can be confusing until you realize that the mortality rates are based on just those people who got sick from H5N1 and were diagnosed in a hospital or clinic.  There are always people who are exposed to a virus, but who do not get sick, or whose symptoms are mild and maybe not even noticed.  Some people never get to a clinic.  So the mortality rate is just those who die from among those who are sick in a hospital, diagnosed and counted.

Also, the mortality rate from the AVIAN form of H5N1 is very high for humans.  However, when a virus mutates to a form that becomes transmissible person-to-person, it may lose some of its virulence and become less deadly.  This was the case for the 1918 virus in its third wave.  So, bottom line, the "guesstimates" you are reading for motality rates (estimated) from a HUMAN form of H5N1 are based on the rates from the 1918 pandemic.  If it is a milder strain that 1918, the mortality could be less.  If it is more lethal, all bets are off.

Numbers are always extrapolations and they are "best guesses."  But the 50% mortality rate is a particular number based on the cases of BIRD to HUMAN infections that showed up in hospitals to be counted. 

I hope that helps. 

yes...

zogger
Sun Jul 09 17:00:49 -0700 2006
 manage
...it helps some, thank you for the reply. Have there been (do you know of any) studies to test the entire local demographic where human cases have occurred, say all the people in a village, to see what the real infection rate is? I understand on the "admitted to hospital" part, those would be the most seriously ill, so obviously they would have the highest mortality rate. I haven't seen a reference to such testing yet in any of my readings. I think for us to get a better handle on the potentialities we would need those sorts of stats. We can *use* the spanish flu as a refernce point, and other of the larger flu outbreaks of course, but this is now, that was then, things change, and viruses change rapidly. Up to date numbers are critical.
yes...
Margot White
Sun Jul 09 18:23:50 -0700 2006
 manage

The people who are most likely to have done these studies are Dr. Robert Webster, or the "WHO factsheet" compilers. Several reliable websites are:

CIDRAP

Effect Measure

and this one:: http://www.newsnow.co.uk/newsfeed/?name=Bird%20Flu

If you want an aggregate list of some pretty good sources, you could explore whether they have the more detailed virology statistics. I would not necessarily take government figures at face value, from anywhere. It's much more likely to come from surveillance virologists and epidemiologists. Here's a list of websites compiled by "pandemic watchers", with no attempt to categorize them as to rank in quality: (I don't know why this huge space exists here; couldn't edit it out; keep scrolling!)

Bird Flu RSS A daily worldwide bird flu news and RSS resource maintained from Provence, France. Click the RSS image to get the RSS feed from Bird Flu RSS http://www.birdflurss.com/rss.xmlhttp://www.birdflurss.com/rss.xml

Effect Measure’s avian flu site blogroll (dated May 21, 2005) Infectious Diseases of America Avian/Pandemic Flu Resources Web site Flu Information Centre A bilingual news aggregator based in Hangzhou, China. Also provides links to newly published scientific research. News Target Taiwan-based news service. Usually very stale news items and less reliable (opinion of Revere, editor) Bird Flu News Headlines (op. high ads to news ratio) Euractiv.com Euro health news Flu Wire Aggregated World News, including a "Blog Watch" feature Bird Flu News Database Sequential news articles with archive Bird Flu & Homeopathy Details of Bird flu with Homeopathic Possibilities by Dr. Abdul Gafar from India

Turkey Avian Flu News Latest avian flu news, headlines and maps from Turkey http://openturkey.com/main/taxonomy/term/3/0/feedhttp://openturkey.com/main/taxonomy/term/3/0/feed-

Global Pandemic News Web-based news aggregator created by New Zealand’s Roger Smith Topix.net News: Influenza The Poultry Site UK-based website catering to the poultry industry. Provides daily roundups of poultry industry news. Avian Flu (H5N1) commercial consultant site, frequent updates Avian Flu Info, a list of links on Ryan Schultz’ blog. (dated August 9, 2005)
Ryan also maintains the Avian Flu Watch group on flickr.com. http://www.flickr.com/groups_feed.gne?id=10996930@N00&format=rss_200http://www.flickr.com/groups_feed.gne?id=10996930@N00&format=rss_200Online NewsHour with Jim Lehrer - Bird Flu: Spread of the H5N1 virus Interactive map, video and audio archives from the NewsHour Avian Influenza News - CIDRAP very authoritative source Pandemic Influenza News - CIDRAP very authoritative source Bird Flu News-Science and Development Net excellent source NewsNow Bird flu. UK-based website. Updates from ~21,000 news sources every 5 minutes Reuters AlertNet - bird flu Bird Flu Articles find informative articles(la pomme) and views on Flu and its related topics commercial site Bird flu Singapore-based website. Maintained by Denis and Anita Menezes
who previously covered the SARS outbreak in the region HEWSWeb Avian Influenza provides updated humanitarian information and maps MSNBC Bird Flu Headlines Yahoo News Bird Flu Coverage Google News Bird flu / Avian flu The Wall Street Journal’s Avian Flu News Tracker Information about the deadly avain flu [Treatment for the dangerous virus] Bird Flu News Archives NewsLIB’s Bird Flu news and article archives. The RSS feed provides the latest articles only http://bird-flu-news.newslib.com/feed.xmlhttp://bird-flu-news.newslib.com/feed.xml

yes...
Scorsbee
Mon Jul 10 11:12:38 -0700 2006
 manage

There have been studies for antibodies in other people exposed and to date they are not finding hidden cases.  The people who show up at the hospital or morgue are the only people who have the virus.  This basically means that the mortality rate is not falsely high due to unreported cases. 

 
Where do Flu Pandemic Projected Numbers come from?
rubycodez
Sun Jul 09 15:57:53 -0700 2006
 manage
Maybe they figure mortality rate lower in first world hospital than third world? Let's see, that means we'll have 250,000 to 300,000 or more in Tenneesee's 140 hospitals at a time  during the 6 to 8 week period, assuming everyone gets over the flu in  two weeks (leave on your legs or out the smokestack).  They each have beds for 2123 people and most importantly TV remotes for them all right? 

this could be...

zogger
Sun Jul 09 17:04:31 -0700 2006
 manage
..but there aren't a lot of treatment options either, first, second or third world. From the reports, tamiflu etc is only marginally effective, and then only if given very early in the infection stage and in large doseages, and the stuff just doesn't exist in large enough quantities to treat everyone who might get nailed. So, perhaps..the main difference might be comfort of bed as you are laying there.....
this could be...
pyro9
Sun Jul 09 22:19:10 -0700 2006
 manage
There really aren't many treatment options, but the third world will still see higher mortality. Malnutrition, poor sanitation, and other untreated disease all leave a person more vulnerable to infection and less able to recover from it.

What will be interesting will be to compare first world countries. For example, in the U.S. there is a tendancy for people to drag themselves to work no matter how bad they feel (and so make everyone there sick).  Interestingly, of all of the dubious control measures being discussed, I notice that manditory sick leave hasn't even come up. I would have thought that one was obvious.

In some other nations, making an effort to avoid giving others a disease is considered basic courtesy.
Where do Flu Pandemic Projected Numbers come from?
Public Health Doc
Mon Jul 10 19:02:38 -0700 2006
 manage
The numbers you usually see in state and national projections are often based on what would happen if a mild pandemic (like 1968) or a severe pandemic (like 1918) occurred in today's population. 

Although politicians and even public health experts sometimes refer to a 1918-like pandemic as the "worst cast scenario,"  in 1918, the overall case-fatality rate in the U.S was about 2.5% of the people who became ill (or 0.6% of the population, since only about 30% become ill in a pandemic).  [There were significant local variations, such as a nearly 90% mortality of the population of Brevig Mission, Alaska, an isolated primarily Inuit community.  20% of the population died in Western Samoa.]

A 50% case-fatality rate would be 20 times worse than the 1918 pandemic in the U.S.

So does the current outbreak of H5N1 in humans in Asia and elsewhere REALLY have a 50% mortality rate?  The jury is still out; seroprevalence studies have either not been done or not been published.  Unfortunately, what little we know is not encouraging.  Interviews with some of the scientists looking for mild or asymptomatic H5N1 infection in humans suggest they are finding little evidence of it. 

Some have argued that since the current H5N1 virus would need to change to be easily transmitted person-to-person, those changes might cause the virus to be less lethal.  While that might occur, it might not.  Some have argued that a virus that kills its victims quickly will be an unsuccessful virus and die out.  I personallly don't find that very persuasive, since the people dying today are certainly living long enough to expose many other people.

I think the main reason you don't see numbers that calculate a 30% infection rate and a 50+% case fatality rate is that the number of deaths becomes so catastrophic that people either deny it could happen or see no way to prepare for it .  Dealing with a 1918-like pandemic seems so very difficult to try to prepare for.  The idea of a billion deaths  worldwide seems unimaginable. 

Unfortunately it's not impossible that this could occur.  Let's all hope and pray that it doesn't.  If we continue to invest in research to develop a flu vaccine that could be made more quickly, and that would provide broader protection across many strains, and build a manufacturing infrastructure to produce it, maybe in 10, 20, or 30 years from now, we would have the means to protect ourselves and send pandemic influenza to the pages of our medical history books.
 

Do you have questions about the site? Write to bruce at perens.com . List of all articles we've ever published. Source Code Developer's Documentation RSS XML Sitemap XML News Sitemap Valid XHTML 1.0 Strict Valid CSS! [Valid RSS]