Spew #  [017]  --   The CDC is Punking America on H1N1 

As my many faithful followers, or follower, long ago realized, the Gutter Grunt temporarily suspended his persistent, whining, ineffectual anti-BS diatribe months ago and has turned his attention to the H1N1 pandemic, now endemic.  Following this story has been even more frustrating than the sludge epidemic.  I began faithfully recording data from the US CDC and the European CDC on a daily basis in order to see first hand how this pandemic is progressing.  But soon I began seeing anomalies in the data, and then suspicious discrepancies, and, most recently, absolute, documented proof of the CDC manipulating data and lying to the public.  I was tuning into SwineFlu.org multiple times a day and folks over there, like  RoRo, Selusha, and MercuryMom were also picking up on the "funny" data.

I'm not going to inventory all the problems here.  But I do want you to see the documentary evidence that the CDC is lying and manipulating the data it presents to the public.  By "documentary" I mean the CDC's own web pages.  I'll unfold this story in chronological order.   

The CDC weekly reports and the August 30th fatalities "re-set" 

These epidemiology  folks think linearly as far as the calendar goes.  Week 1 starts on the first of January, and Week 52 ends on December 31, except that sometimes there isn't a Week 52, depending on what day of the week Week 1 began on.  But never mind the details, the important thing is that you don't normally talk dates in this game; you talk weeks.  And that's because most of the data are bundled in weeks.  You talk about the number of hospitalizations in Week 17 or the number of deaths in Week 30.  

The Influenza Division of the CDC puts out a weekly report called "FluView", which can be found here.  If you hit that link today, Oct10.09, it will take you to the Weekly report for Week 39, ending Oct03.09; i.e., FluView is always at least a week behind, which is a very long time in a pandemic that is moving as fast as this one is right now.  The European CDC reports daily.  

If you hit the same CDC FluView link next week it will take you to the weekly report for Week 40.  The weekly report for Week 39 will then be archived at this link, but unless you know that the data are archived, it would likely take you a long time to find them.  In fact, weekly reports going all the way back to 1999 are archived there.  The weekly reports started out pretty simple in 1999, but they have evolved significantly to include a number of graphs and tables presenting the data in different ways, and the CDC folks don't seem to be able to settle on any fixed format.  The nature and content of the reports jump around a lot. 

The H1N1 situation from mid-April through Week 34 (Aug23-Aug29.09) was summarized in the Week 34 report as follows:

This paragraph is followed by some details, but it is the only mention of cumulative fatalities -- 593.  A fairly important number, one would think.  The increase in deaths for the week was 593 - 556 = 37. 

In the next weekly report, Week 35, the synopsis was much shorter -- one sentence:

Maybe you noticed no mention of fatalities.  But way down into the report they do give some numbers and they explain that they have decided to reset the fatality counter:

So as of August 30th, the CDC has "dropped" 593 deaths reported before the change in reporting protocol.  For instance there should be 593 plus the 196 reported during Week 35 for a total of 789 on Aug30.09.  But as for the 593 deaths prior to Week 35, if you weren't following, it would be impossible to know where that number went -- there's a shell-game here.

Before leaving the topic of how H1N1 fatalities are reported, let me raise the pediatric fatalities.  The data are much better for pediatric mortality because influenza related deaths in children must be reported through the Nationally Notifiable Disease Surveillance System.  We'll come back to pediatric mortality numbers, but for now I'm referring only to CDC's reporting of total mortality.      

 Weeks 36 and 37

Of course, one still has to add the "missing" 593 deaths from April to September to get the real total -- 1529 up to the end of Week 37.

Week 38 and the Green Bay Bar Graph 

In Week 38 the CDC published what I call the Green Bay bar graph, but before I get to that, let me run the new numbers.  The CDC says that at the end of Week 38 there were 1,379 deaths.  That represents an increase of 443 over Week 38, and a total "real" body count of 1,972.   It is obvious from these numbers that things are accelerating.  

The CDC did make an attempt to represent the incremental weekly fatality data graphically, as shown here:

If you're not a pro football fan, you'll miss the allusion to Green Bay, but you can't miss the data because you gotta' admit that's a pretty bright graph.  I am just interested in the bottom graph here, the fatality data.  The orange part of each bar shows the number of deaths for that week that were confirmed by a lab.  The green part shows the number of deaths in which the patient had flu-pneumonia symptoms -- or "syndromic-based" diagnosis, to use the CDC's term.  

It's important to note that by about Week 30 the CDC and health officials had pretty much stopped testing for H1N1.  More than 90% of the cases they tested turned out to be H1N1, and since "regular" H2N1 flu at this time of the year is extremely rare, the official presumption is that anyone who died with flu-pneumonia like symptoms died of H1N1.  So the number of lab-confirmed cases is bugger all compared to the syndromic cases because nobody is testing.

Now we come to the main point I want to make: Today, if you go to the archive link for Week 38 where the Green Bay bar graphs should be, you won't find them.  All you'll find are these pumpkin graphs:

After publishing the Week 38 Green Bay bar graphs, the CDC then went back and erased the green portions of the bars, which showed the great bulk of the fatalities.  So now, when the public looks at the Week 38 report they see just 40 or so new deaths for Week 37 and just 25 or so for Week 38, when, in fact, the numbers are 572 new deaths for Week 37 and 443 for Week 38.  What's almost as bad is the fact that by going back and cooking these graphs, the fatalities look like they are falling from Week 36 - Week 38, when, in fact, they doubled.  

You may be wondering how I got the Green Bay bar graph if the CDC went back to the archives and erased the green bits.  Well, the CDC also archives .pdf files of the weekly reports.  When you pull up an archived report, say Week 38, you will find a link to the .pdf.  The CDC b'crat who went back and removed the syndromic component from changed the Green Bay graph neglected to change the .pdf version.  When you falsify your data but don't cover your tracks, you are not just being dishonest, you are being stupid.

The Week 40 Pumpkin Graph

First of all, the numbers.  CDC reports cumulative deaths of 1,544, which is only an increase of 165 over Week 38.  That is very unlikely given that the distribution of H1N1 has increased by 30% over the week and is now reported as widespread in all but 13 states.  But it is even more unlikely when you look at the Week 40 pumpkin graph.

Like the revised and falsified Week 38 graph, this pumpkin graph also deletes the numbers of syndromic-based death reports.  But what is totally unbelievable is that this graph shows about 75 lab-confirmed deaths during Week 39, and yet the total of new deaths for the week is just 165, calculated above.  That means that there were only 95 syndromic-based deaths and it also means that lab-confirmed deaths accounted for 45% of the total.  45%!!!!  Look at that Green Bay bar graph -- lab-confirmed deaths were running only about 5% - 10% of the total.  And now they're doing very little lab testing so the proportion of lab confirmed deaths should be even smaller.  

Furthermore, compare the Week 38 pumpkin bars with Week 39.  In the Week 38 report the deaths were declining by 50% from Week 36 - 38.  Now those data have changed -- they are actually increasing.  The numbers for Weeks 35 and 36 are now well below 50 deaths each for those weeks; in the Week 38 report, those numbers were near or above 50.  How do the numbers of deaths reported for a week go down after the week has ended???  Or even before?

Summary 

What is happening is that the CDC has begun hiding the bulk of the fatality data -- the syndromic-based data.  Because so few cases are tested for H1N1, the lab-confirmed deaths that they do report are a very minor part of the whole picture.  They don't want us to see the whole picture.  Why?

It is possible --  by ignoring the bright graphs and reading the small print text -- to get numbers to calculate the weekly and cumulative body counts.  But if the CDC is falsifying the graphs -- and it is -- then why would anyone suppose that they are not also falsifying the naked numbers?   A close look at the numbers for Week 39, above, indicates that something funny is going on with the numbers as well as the graphs.  Besides, the reason you present data in graphs is because it's easier for people to comprehend.  If you are presenting fake data in the graphs -- and the CDC is -- but "real" data in the text, you are being deceitful.  

The CDC is punking America.

 -- Gutter Grunt
denis@thepatentguy.net