Sunday, June 21, 2009

Calculating The Attack Rate Of Editorialists

 

 

# 3372

 

 

I’ll admit that, of late, I’m feeling a bit inadequate.  

 

As a blogger, that is.

 

You see, at nearly every turn I find myself issuing caveats about how little we know about pandemics in general - and the H1N1 Swine flu in particular -and saying that we just don’t know what this virus will bring in the months to come.

 

Then I go and read the newspapers, and it seems that just about every editorialist has it all figured out!

 

They have no problem declaring the H1N1 virus as being `no worse than seasonal flu’.    They seem quite comfortable lambasting the WHO for declaring a pandemic, and accusing scientists of `fear mongering’.

 

Many seem unburdened by the facts, and cheerfully quote partial or even misinformation.   A prime example of this occurred yesterday in the Bangkok Post, in a piece entitled: WHO flu alerts 'lead to panic', experts say

 

Many virologists say the H1N1 flu is not as virulent as previously feared. Studies show the fatality rate of the H1N1 infection stands at only 0.4%, which is lower than the 1% toll of the average seasonal influenza each year.

 

Which is very reassuring. Too bad it’s 100% false.

Yes, early studies have pegged the H1N1 virus’s fatality rate at about 4 per 1,000 cases, or 0.4%  (although it may actually be lower).  But seasonal flu is thought to kill only about 1 person in 1,000, or. 0.1% . . . not 1% as this author claims.  

 

Of course, only Flubies and scientists are likely to pick up on this wee bit of misinformation; a ten fold misrepresentation of the seasonal fatality rate.  For 99% of the public, it’s a very convincing argument.

 

Right now, I don’t think anyone (except some commentators, of course) have any idea of the true virulence of the H1N1 virus.  

 

I don’t accept the .4% figure as being necessarily accurate, but I do suspect the number is somewhere between that and the mortality rate of seasonal flu.


The problem is, this early in a pandemic, it is awfully hard to calculate small differences in these CFRs (Case Fatality Ratios).

 

And small differences in mortality can mean big differences in impact.

 

Our surveillance systems are not designed to pick up every flu-related death.  And in many countries, their surveillance efforts are even less exacting.

 

The CDC reminds us in every press conference that their numbers represent `just the tip of the iceberg’, yet some writers continue to use them to `prove’ how innocuous this virus is.   

 

The above mentioned editorial used the figure .4% for the fatality rate of the H1N1 virus.   Let’s cut that by 75%.   That’s right, let’s just assume for a moment that this virus is `no deadlier than seasonal flu’.   That it only kills 1 person in 1,000.

 

That would make this a non-event.  Right?

 

Well, that would be true only if the attack rate (which a lot of writers conveniently forget to mention) remains the same as during a normal flu season.

 

The Case Attack Rate (CAR)for influenza each year runs between 7% and 10% of the population. In a bad year, that means that roughly 30 million Americans get the flu. 

 

As many as 200,000 – mostly elderly – flu victims are hospitalized, and the CDC estimates 36,000 Americans die from complications of influenza. 

 

But novel influenza strains – the type that cause pandemics – have a much higher attack rate.   Usually 25% to 35% of the population.  The UK last night announced they were preparing for a 50% attack rate.


To keep from piling on, we’ll take a middle number – 30% – which seems to be the most commonly used number in pandemic planning. 

 

So . . .if 30% of the population are stricken, we could expect at least 100 million Americans sickened (or more than 2 billion globally). 

 

That would likely create triple the number of normal influenza hospitalizations, and 3 times the normal number of seasonal flu deaths

 

And that’s if the H1N1 virus is `no worse’ than seasonal flu.

 

Since our hospital system gets slammed each winter with 200,000 extra admissions due to seasonal flu, you really have to wonder how well it will stand up to triple that number.  


And of course, since this flu seems to prefer people under the age of 60, a high percentage of the 100,000 or so American deaths we might expect to see would be of children and young adults.

 

And that’s something that we aren’t used to seeing.

 

None of the above is a prediction on my part.  

 

I’m not smart enough to know what the attack rate will be, or what the eventual CFR will turn out to be.  I’m still hoping this virus turns out to be a bust.

 

But scientists are still arguing over how many people died from the last pandemic – with the global number listed anywhere from 1 million to 4 million. If they haven’t come any closer than that over the last 50 years, I’m not inclined to put a lot of faith in predictions about the future.

 

In fact, the more I learn about influenza, the less I find out I know.

 

Editorialists don’t have to worry about getting it wrong.  

 

If they rail against the WHO, or the CDC, for `fear mongering’ and this virus turns out to be a major problem . . . well, few are going to remember or really care what they wrote months before.

 

Worst case, they can issue a retraction or an apology.  Most won’t bother, I suspect.


But if the WHO or the CDC were to adopt the same Laissez-faire attitude as these op-ed writers, and failed to warn the public or to adopt a proactive stance, then hundreds of thousands or even millions of lives could be put at risk.

 

They don’t have the luxury of saying, in their very best Emily Litella voice, `Nevermind.’