by Effect Measure
The big newswires and health agencies are relatively quiet, but word keeps leaking out of Egypt that there are a lot of suspect bird flu cases:
CAIRO: Hospitals nationwide reportedly quarantined more human cases suspected of being infected with the H5N1 bird flu virus.
According to Al-Masry Al-Youm newspaper, Damietta — where the latest Bird Flu victim Hanem Atwa Ibrahim, 50, died late on Monday Dec. 31 in a Cairo hospital — hosts the largest number of cases with five people suspected of carrying the virus, while the Upper Egyptian city of Qena came next, with two cases, followed by El-Beheira with one case. (Egypt Daily Star)
This has been noticed on the Flu Boards (see, for example, here and here) and is the source of understandable anxiety. It is not clear the extent to which the current public awareness campaign is effective in altering behavior but the publicity has also heightened anxiety in the Egyptian public and raised the diagnostic index of suspicion. All of this is appropriate, even though it will also create many new "suspect" caes. However I think there is one element that has been missing in the discussion.
Any kind of influenza infection in humans is potentially serious and life threatening. That’s true in the US, where the estimated excess mortality exceeds 35,000 deaths a year, or in Egypt. It is flu season and people will be dying of this nasty virus on a daily basis. Thus a sudden increase in serious and fatal respiratory infections is not evidence H5N1 has run amok. But because there is a lot of flu around, it doesn’t mean that finding most or all cases are negative means nothing is happening, either.
Let’s do some quick back of the envelope calculations. Egypt’s population is about a fifth of the US population, so on that basis alone we might expect 7000 excess deaths from influenza, mostly compressed into a six month flu season. These, of course, are only a fraction of the influenza cases. In the US the case fatality ratio for seasonal influenza is about 0.1%. Let’s say it is five times higher in Egypt, 0.5%. That means the number of flu cases is 200 times more, or 1.4 million. Let’s say only 20% of these are really sick and are seen by a health care worker. That’s still almost 300,000 or close to 50,000 a month coming to medical attention if spread out evenly in the 6 month period (lack of uniformity would of course make some periods look much worse). If we assume that in the same month there are ten clusters of H5N1 of 25 patients each, the probability of a suspect case of influenza being H5N1 is only about 0.5%. There are also a bunch of other viruses that look like influenza, so these would lower the probability even more.
What does this mean? First it means that even if something pretty bad is happening, almost all suspect cases will be negative. Second, that there is so much influenza-like illness at this time of year that there is a real danger true clusters can get lost in the noise. Third, this fact notwithstanding, we shouldn’t mistake the noise for a true signal. Just because there is a lot of noise doesn’t mean it is hiding a true signal. So far it hasn’t been, at least judging from the lack of a prairie fire spread of H5N1 cases anywhere, including Egypt.
The only solution to this knotty problem is to continue intensive surveillance for H5N1 in areas, like Egypt, where there is endemic infection in poultry and other birds. That’s why the large number of "suspect cases" is a good thing, not a bad one. Suspicion means there is a greater chance of finding any H5N1 needle that might be hiding in the huge influenza-like-illness haystack.
If there is anything reassuring about this crude analysis it is that a large number of suspect cases doesn’t mean something bad is happening. The worrying part is that a lot of negative suspect cases doesn’t mean that something bad isn’t happening.
This is where we separate the optimists from the pessimists. Meanwhile the realists will be working hard to establish an infrastructure that can make accurate diagnoses of serious pneumonias of unexplained origin. We should be encouraging them and helping them to do it better and faster. We can still look over their shoulders without getting in their way.
Source: http://www.agoravox.com/print_article.php3?id_article=7459&format=print 
