Influenza vaccine effectiveness in children

[LindsayK]

I was looking at one of the CDC releases on the 2007-2008 season. As of 2006, I think, death related to influenza is a reportable condition in people under 18. Here is the data I was looking at:

Of the 63 cases aged >6 months for whom vaccination status was known, 58 (92%) had not been vaccinated against influenza according to the 2007 Advisory Committee on Immunization Practices (ACIP) recommendations

http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5725a5.htm

Earlier in the article, it states that there were 83 deaths, and there were 9 kids were under 6 months old. So, that means there were 83-9=74 kids above 6 months. Let's assume the most possible vaccine failures, and say that 74-58=16 kids were vaccinated and 58 weren't. That means that the vaccine is 58/74=78.4% effective for preventing death in the 6 months - 18 year old crowd.

Am I reading this right? Am I missing some key point? I guess I was just really surprised by this statistic, if this is the case.

[Dmitrizmom]

For strains of influenza A (H3N2) -- the most prevalent virus during the 2007-08 season, the vaccine was 58 percent effective. But it was 100 percent ineffective against influenza B infections, leaving an overall vaccine success rate of about 44 percent, according to the U.S. Centers for Disease Control and Prevention.
http://www.washingtonpost.com/wp-dyn...041702653.html

Thought this was interesting relative to your question. bolding mine.

[LindsayK]

Originally Posted by Dmitrizmom

Thought this was interesting relative to your question. bolding mine.

Hmm.... that would seem to lend credence to the claim that flu strains don't have to be matched to gain protection from death?

[LindsayK]

Originally Posted by LindsayK

Am I reading this right? Am I missing some key point? I guess I was just really surprised by this statistic, if this is the case.

Ok, I think I might have figured out the answer to my own question. I think I was comparing apples to oranges - you can't look at the percentage of vaxed/unvaxed in this data to determine effectiveness. You'd need to look at the total deaths of vaxed kids relative to how many were vaxed, and the total deaths of unvaxed kids relative to how many were vaxed.

So, according to the National Immunization Survey:

http://www.cdc.gov/flu/professionals...ragelevels.htm

32.2% of kids aged 6-23 months were vaxed
26.4% of kids aged 24-59 months were vaxed
12.4% of kids aged 5-18 years were vaxed

The U.S. Census tells us that there were 20.3 million under 5 kids, and 53.2 million 5-18 year old kids in 2005 (I can't seem to get 2006 data, but this should be a good approximation) Unfortunately, there is no breakdown for 0-2 years and 3-5 years. So, I will give the vaccine the better odds, and say that all kids aged 0-5 were vaxed 32.2%. So... (I did this in an excel sheet)

Est # of vaxed kids: 13.1, deaths (low): 5, death rate: 0.000038%
Est # of unvaxed kids: 60.4 million, deaths: 58, death rate: 0.000096%
Death rate is 2.5 times higher in unvaxed kids.

But on the other hand.....

if you assume that those "undetermined status" kids were actually vaxed kids, and the CDC is underreporting, the math goes like this...

Est # of vaxed kids: 13.1, deaths (high): 16, death rate: 0.000112%
Est # of unvaxed kids: 60.4 million, deaths: 58, death rate: 0.000096%
Death rate is 1.3 times higher in VAXED kids.

But on the other (other?) hand...

if you assume that the number of "undetermined status" kids are vaxed at the same rate as the whole population you get...

Est # of vaxed kids: 13.1, deaths (est): 7, death rate: 0.000053%
Est # of unvaxed kids: 60.4 million, deaths: 58, death rate: 0.000096%
Death rate is 1.8 times higher in unvaxed kids.

So, I guess this is just a really long way around of saying that, 1. there is only a tiny chance of any child dying from flu related complications, and that 2. there just aren't enough deaths (and accurate data) to truly determine if the influenza vaccine causes a decrease in mortality in children between the ages of 6 months and 18 years.

Am I on the right track?

[lacysmommy]

[QUOTE=LindsayK;13279474
if you assume that the number of "undetermined status" kids are vaxed at the same rate as the whole population you get...

Est # of vaxed kids: 13.1, deaths (est): 7, death rate: 0.000053%
Est # of unvaxed kids: 60.4 million, deaths: 58, death rate: 0.000096%
Death rate is 1.8 times higher in unvaxed kids.
[/QUOTE]

Except you would need to increase the # of unvaxed at the same rate.

[Deborah]

they are splitting hairs if they are claiming that vaccines make a huge difference in the death rate from flu in toddlers.

Plus the stats are crap to begin with.

[LindsayK]

Originally Posted by lacysmommy

Except you would need to increase the # of unvaxed at the same rate.

Thanks! I realized that when I woke up this morning

[LindsayK]

Originally Posted by Deborah

they are splitting hairs if they are claiming that vaccines make a huge difference in the death rate from flu in toddlers.

Plus the stats are crap to begin with.

Had you read what I posted, you would have understood that these were the rough numbers that I calculated, not the CDC or any other governmental body. You'd also see that I wasn't talking about toddlers, I was looking at the 6 months - 18 years population. There isn't any specific information on toddlers that I have been able to find.

If you think my numbers are "crap", I would truly and honestly like to know how you think I can improve them either through analysis or data. I'm just crunching numbers the best I can, and comments like this are just, at best, unhelpful.

Wow. I guess I am just really unimpressed by the lack of thoughtful response from this board. I find it really hard to get any real numbers from the CDC to justify reasoning that many doctors use to scaremonger, and in this case, there are real numbers- albeit a small sample size. I was happy to get my hands on actual numbers, and was hoping to have a little feedback fromthe smart mamas here along the lines of what lacysmommy wrote.


A sidenote - the CDC this week released the deaths from motor vehicle accidents from the years 1999-2005 in the 0-18 year age group. That's what got me started on these calculations since I was wondering, if you add up the risk of death from all the VPDs which have vaccines on the CDC schedule, how does it compare to the risk of death from motor vehicle accidents? I spent about 4 hours working on that last night, and got a surprising answer, but I'm sure my stats are crap anyway, so I won't post any of those calculations here

[mamakay]

The reason the CDC didn't use these figures to calculate the effectiveness is probably because there's no way to work around the confounding factors. Why were the unvaxed kid unvaxed? Were they immunocompromised and already on death's door in the ER with a preexisting condition? Or were the vaxed kids more likely to be vaxed if they were considered "at risk"?

Plus, a lot of the key information you'd need to calculate the very basic numbers is missing. So this just isn't enough data to work with, I think.

[LindsayK]

Originally Posted by mamakay

Plus, a lot of the key information you'd need to calculate the very basic numbers is missing. So this just isn't enough data to work with, I think.

Yeah, I totally agree with you there. I guess what I was aiming to do is for variables I didn't have the info for, to skew the numbers totally in favor of vaccines being effective. If doing that couldn't yield a result which showed vaccine effectiveness, then its a pretty easy call, using the CDC's own numbers to show that the flu vaccine isn't effective in children.

I guess we'll have more data soon, given that the pharm companies and their $$$ will hand us a larger pool of vaccinated children in future winters!

[Deborah]

Sorry! I didn't mean your numbers, I meant the general stats that are collected on influenza rates, deaths etc. I should have been clearer.

So many cases are never even correctly tested to make sure they actually are influenza and not "influenza like illnesses".

[aniT]

Originally Posted by Deborah

Sorry! I didn't mean your numbers, I meant the general stats that are collected on influenza rates, deaths etc. I should have been clearer.

So many cases are never even correctly tested to make sure they actually are influenza and not "influenza like illnesses".

And how many people never ever go to the doctor for it in the first place?? I know I wouldn't. And that was BEFORE I was a nonvaxer. There is nothing a doctor can do but treat the symptoms of a virus.

[mamakay]

Well, they do have Tamiflu now. But a lot of flu strains are already resistent, iirc.

[LindsayK]

Originally Posted by Deborah

Sorry! I didn't mean your numbers, I meant the general stats that are collected on influenza rates, deaths etc. I should have been clearer.

So many cases are never even correctly tested to make sure they actually are influenza and not "influenza like illnesses".

Sorry, was in a bad mood from family issues, and didn't look twice to see if maybe I was misinterpreting your comment

Originally Posted by Deborah

So many cases are never even correctly tested to make sure they actually are influenza and not "influenza like illnesses".

That's why I was somewhat interested in these numbers, because all of the cases had actually been tested, and the CDC had determined which strain (A or B) the deceased had. Reported flu cases are most certainly a mixed bag when it comes to self-reported flu and lab confirmed cases, but I guess they take the pediatric deaths more seriously, and test all of them.

It will be interesting to see if this level of detailed reporting disappears in later years as there is more data available for pediatric flu vaccine use.

[aniT]

Originally Posted by mamakay

Well, they do have Tamiflu now. But a lot of flu strains are already resistent, iirc.

I'll take my chances with the flu. Weren't there reports of Tamiflu killing teenagers in Japan a few years back?

[Tofu the Geek]

Originally Posted by aniT

I'll take my chances with the flu. Weren't there reports of Tamiflu killing teenagers in Japan a few years back?

http://www.timesonline.co.uk/tol/new...cle1549260.ece

[k9sarchik]

When I had the flu I got admitted into the hospital because my asthmatic lungs were having a hard time. 1/2 hr after they gave me an anti-viral, I'm assuming it was Tamiflu, I had to be rushed into the ICU. I swear that SH!T made it WAY worse. Oh yea.....and I was hallucinating too. No way it was from the flu.
I am not surprised that some people died on it.

[Deborah]

Wow, it sounds like that drug did wonders for your flu and your lungs and your mind. Great stuff. But very profitable for someone.

[aniT]

Donald Rumsfeld?

[Rosehip]

Originally Posted by LindsayK

I was looking at one of the CDC releases on the 2007-2008 season. As of 2006, I think, death related to influenza is a reportable condition in people under 18. Here is the data I was looking at:



http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5725a5.htm

Earlier in the article, it states that there were 83 deaths, and there were 9 kids were under 6 months old. So, that means there were 83-9=74 kids above 6 months. Let's assume the most possible vaccine failures, and say that 74-58=16 kids were vaccinated and 58 weren't. That means that the vaccine is 58/74=78.4% effective for preventing death in the 6 months - 18 year old crowd.

Am I reading this right? Am I missing some key point? I guess I was just really surprised by this statistic, if this is the case.

I didn't click through to the article, but I think I'm answering your question here. There were 83 deaths. Out of the 83 children who died, there was documentation as to whether or not the child had received the flu vaccine for 63 children (so for 20 of the children who died, it is unknown whether or not they received the flu vaccine, therefore, we cannot consider them if we are discussing the relationship between vaccine status and death from flu). Of the 63 for whom we know vaccine status, 58, or 92% had not received the flu vaccine.

You cannot infer overall effectiveness at death prevention from this. What you can say, is that OF children who died from flu for whom vaccine status is know, 92% were not vaccinated.

To even begin to make hypotheses about efficacy from this, you would need lots more information about what % of children, in general, are vaccinated, whether this group of children was unusually ill (as in, were there children who were already immune compromised, etc.) which makes them not comparable to the general population of children. Also important - were there reasons that these children were or were not vaccinated that might have made them more or less vulnerable than the average child? For instance, if a child has an allergy that precludes flu vaccine, one might consider if children with said allergy are more vulnerable to flu to begin with.

These are epidemiological data that are very different from a study or trial designed to assess efficacy.