Vaccine statistics...

[leerypolyp]

So I have to do a project for a class in which I compare a numerical result under two conditions (and the result should be expected to vary somewhat). And I'll need at least 40 observations. We can use existing data for the project, and my first thought was -- Vaccines! I totally want to use some of that CDC data. But 1) I can't decide how to set up the project, and 2) I can't find that CDC page I once had bookmarked that's just loaded with raw data going back to the 1960s.

Any suggestions? Help me do my homework (and maybe give a few classmates some education they weren't expecting)!

[mamakay]

Is this the page you were looking for?

http://www.cdc.gov/vaccines/pubs/pin...ses&deaths.pdf

[leerypolyp]

Yep! That's what I'm looking for. Now I just have to figure out something interesting to do with the data -- compare to another country? Hmm...

Either way, very useful link, thank you.

[mamakay]

Finding the incidence and complication rates of VPDs in the prevax era would be interesting.
The way the CDC comes up with their "one in 250 people with measles dies" figure is...a...creative way of interpreting the data, for example.

[mamakay]

For example, here's what they say about CP:

http://www.cdc.gov/MMWR/preview/mmwrhtml/mm5237a2.htm

Varicella (chickenpox) is a common, highly infectious, and vaccine-preventable disease. Before the introduction of the live attenuated varicella vaccine in 1995, approximately 4 million cases of varicella occurred annually in the United States, resulting in approximately 11,000 hospitalizations and 100 deaths (1--3).

http://www.journals.uchicago.edu/JID...157513580Guest

Varicella is a highly infectious disease that is preventable by vaccine. Before vaccine licensure in 1995, 4 million cases per year resulted in 9300 hospitalizations [1] and 100 deaths each year

4 million is the US birth cohort. You can look at the blood of people born before the vaccine and see that basically everyone caught CP in childhood.
There have never been 4 million cases reported, though. But they don't need to use reported cases, because the serology tells the tale.

But what would the complication and death rate be for CP if you *did* use reported cases? It would make CP look really, really deadly.

[mamakay]

Haha...I'm on a roll...this is fun...

Ok, here's the reported cases of CP from the late 70's.
http://findarticles.com/p/articles/m...0/ai_101956293

It hung out around 160-200 thousand reported cases a year.
So, if we wanted to make CP sound really deadly, we could find out the incidence of hospitalization for CP and combine that with the reported incidence of plain old CP.

What's 9,300 hospitalizations into 167,000 cases?
(I suck at math)
How scary can you make CP look by doing the math like that?
What's 100 deaths in 167,000 cases? Is that one death per 1670 cases of CP?

[mamakay]

Now let's pop over to the NIP's WWHIWSV page and switch to measles:

http://www.cdc.gov/vaccines/vac-gen/whatifstop.htm

Before measles immunization was available, nearly everyone in the U.S. got measles. An average of 450 measles-associated deaths were reported each year between 1953 and 1963.

In the U.S., up to 20 percent of persons with measles are hospitalized. Seventeen percent of measles cases have had one or more complications, such as ear infections, pneumonia, or diarrhea. Pneumonia is present in about six percent of cases and accounts for most of the measles deaths. Although less common, some persons with measles develop encephalitis (swelling of the lining of the brain), resulting in brain damage.

As many as three of every 1,000 persons with measles will die in the U.S.

Where are they getting that last bolded part from?

We would have had 4 million cases of measles a year before the vax. 400 deaths.
How do they come up with three in every thousand measles cases resulting in death?

[leerypolyp]

Right, exactly! That's why I wanted an excuse to run it all through some stats programs. Because there are many ways to manipulate statistics, and it looks like they're using every last one.

The before and after vax incidences (and complications) were what I was hunting for. I guess I shouldn't expect a neat data table, huh?

Hmm. I need forty observations, meaning at least 20 vpds with before and after data. Except that several of the vpds weren't even reportable before the vaccines...hmm.

Yeah. So far beyond the project there. But so fascinating!