The Tip Of The Pyramid

 

 

# 3355

 

 

 

 

As I watch, and try to keep track of case counts and the number of fatalities attributed to the H1N1 virus, I’m reminded of the old story about the poker player who played at a local cash game each week, and always lost.

 

He complained bitterly to his best friend that the game was crooked.   

 

His friend asks, “If it’s crooked, why do you keep going back?”

 

He looks at him like he’s an idiot and says, “Cause it’s the only game in town!”

 

 

And so it is with the numbers we get regarding infections and deaths from the H1N1 virus.   It is not so much that they are`crooked’ –but they are incomplete, inaccurate, and misleading.   

 

But we keep going back to them because, right now, they are the only game in town’.

 

It’s not that there’s a conspiracy to hide the truth, I think most agencies are doing the best they can with the tools and systems at their disposal. 

 

But do I cringe a little bit when I see people quoting these numbers as if they somehow represented reality.

 

They don’t, and I’ll explain why.

 

The CDC openly acknowledged - weeks ago - that the true number of infections in the US was probably 20 times higher than reported.  Since then, testing in most places has been reserved for those hospitalized and in serious condition.

 

While the last numbers released by the CDC indicate fewer than 20,000 confirmed cases, the real number is 20 . . .30 . . . maybe even 100 times greater.  

 

Nobody really knows.

 

And in truth, this is how it is with most infectious illnesses.   We don’t know, with any great confidence, how many people contract West Nile, Lyme disease, or seasonal influenza every year. 

 

Not everyone is sickened enough to seek treatment, and not everyone that does is tested and diagnosed.   And so we `make do’ with estimates, like the controversial-but-oft-quoted 36,000 influenza deaths each year in the United States.

 

 

The CDC routinely refers to `laboratory confirmed’ cases as being the `tip of the iceberg’ of any outbreak.

 

 

Or in this case, the tip of the pyramid.

 

So, here in the US, while the case counts may be interesting, they’ve become pretty much irrelevant.

 

According to the WHO, as of Monday, June 15th there have been 35928 confirmed cases and 163 confirmed fatalities.  And 45 of those deaths have been confirmed from the US.

 

While this works out to about 4 deaths in 1,000 (about 4 x’s greater than seasonal flu), this obviously doesn’t reflect reality.  We already know the number of infected – but unconfirmed – cases is far higher.

 

But in all likelihood, so are the number of deaths.  

 

Overnight, New York City announced another 7 deaths attributed to pandemic flu.   That’s a total of 23 reported to date.  The operative word being `reported’.

 

7 More Swine Flu Deaths Reported in NYC

Wednesday, June 17, 2009

NEW YORK —  Health officials on Tuesday reported seven more deaths from swine flu, bringing the city's total to 23, the most in the nation.

The Department of Health didn't identify the victims or say when they died in an update posted on its Web site. It said the victims were between the ages of 25 and 64 and most had been hospitalized late last month.

 

In a `normal’ flu year, New York City sees 1,000 to 2,000 flu related deaths (again an estimate. nobody really knows).  

 

So 23 deaths in just over 6 weeks doesn’t sound so bad.

 

Except, these deaths are almost all comprised of young adults and children, a major shift from what we expect from seasonal flu, which exacts its greatest toll on the elderly.

 

And we have no great confidence that we are counting all of the H1N1 related deaths.  

 

In a city the size of New York City, somewhere in excess of 150 people die each day – mostly from `natural’ causes.  Across the entire US, there are more than 6,500 such deaths daily.

 

Those that died prematurely (children, young adults), and those who died from accidents, or violence, will just about always have a post-mortem (autopsy) exam.  

 

But those who died at home, or in a nursing home, or a hospital as the end result of a long, chronic illness will probably bypass the medical examiner system and end up released to a local funeral home. 

 

The patient’s doctor will sign the death certificate as an `attended’ death, and put down something `generic’ – like coronary artery disease -  for the cause of death.

 

If influenza was a factor, it is almost never recorded.  This is how the `system’ works in most places. 

 

Some municipalities take a closer look at `natural’ deaths than others, but even so, the likelihood of the medical examiner picking up on influenza as a contributing cause of death is pretty remote.

 

The cases we tend to identify are either young adults or children, or those who were hospitalized and being treated for influenza or pneumonia at the time of their deaths.

 

Which is why the CDC uses statistical models to calculate the number of seasonal influenza deaths each year, not an case-by-case count. 

 

Counting individual cases – as we are trying to do with H1N1 fatalities right now – simply doesn’t work.

 

So 23 H1N1 related deaths in New York City may be a serious undercount.

 

For the record, I’ve no idea how many deaths in the US have been caused by this virus.   I believe, based on the way our surveillance systems work, that it is considerably higher than we see reported.

 

Which is why I don’t put a lot of stock in the numbers we are getting.   

 

None of this is to suggest that the H1N1 pandemic virus is an aggressive killer like the 1918 virus.    While we aren’t able to identify all of the deaths related to this flu, I do believe we would have picked up on any obvious spike in mortality rates.

 

For now, this virus appears to be in the mild-to-moderate range of severity, and most people seem to recover without complications. 

 

Which is good news.

 

That doesn’t make it benign, and `mild' is a relative term.  This influenza, like any influenza, can produce serious illness and sometimes even kill its host.

 

Despite its inherent flaws, reporting from the US is probably among the best on the planet. The numbers we are getting from developing nations – where testing and surveillance capability is far less diligent – is even more suspect.

 

Keeping track of the numbers can be useful, and can often provide us with important clues as to how this pandemic is progressing. 

 

Even incomplete data has its uses.

 

But the daily totals don’t really tell us much about the true scope of this pandemic.