The prevaccine Hib estimates

[mamakay]

Here's where it comes from.

It's 1985, and the CDC is gearing up to introduce the old, crappy hib polysaccaride vax.

http://www.cdc.gov/mmwr/preview/mmwrhtml/00022818.htm

April 19, 1985
Several population-based studies of invasive Hib disease conducted within the last 10 years have provided estimates of the incidence of disease among children under 5 years of age, the major age group at risk. These studies have demonstrated attack rates of meningitis ranging from 51 cases per 100,000 children to 77/100,000 per year and attack rates of other invasive Hib disease varying from 24/100,000 to 75/100,000 per year (1). Thus, in the United States, approximately one of every 1,000 children under 5 years of age develops systemic Hib disease each year, and a child's cumulative risk of developing systemic Hib disease at some time during the first 5 years of life is about one in 200. Attack rates peak between 6 months and 1 year of age and decline thereafter. Approximately 35%-40% of Hib disease occurs among children 18 months of age or older, and 25% occurs above 24 months of age

So they're saying one in 1,000 kids under age 5 developed invasive Hib. Ok.

But then they multiply that by 5 to make it one in 200 kids by age 5.

Anybody see the problem there?

ETA:

You often hear "20,000 cases of Hib disease a year prevaccine!"

One in 1,000 in babies 0-12 months is 4,000 cases...times the first 5 years of life (coz that's how they get the "one in 200 by age 5" figure) is 20,000 cases a year.

That has to be where the "20,000 cases of hib disease a year prevaccine!" comes from.

[mamakay]

I'm going to give you a clue with what's wrong with multiplying the "one in 1,000" by 5, to make it "one in 200", which give us the "20,000 cases a year" estimate:

http://www.cdc.gov/vaccines/pubs/pin...ds/hib-508.pdf

Incidence is strikingly age-dependent. In the prevaccine era, up to 60% of invasive disease occurred before age 12 months, with a peak occurrence among children 6–11 months of age. Children 60 months of age and older account for less than 10% of invasive disease.

Look at the chart on page 119.

Whoever sees what's wrong with the 1985 math gets a cookie.

[mamakay]

Another clue:

There are a few cases of invasive hib that still happened till age 10 and beyond. If they had multiplied the "one in 1000" till age 10 for a cumulative incidence, how many cases a year would be happening then according to that math? 40,000? Or, one in 100?

What if they took it to age 20?

[Spectrolite]

:

...and wishing I could concentrate enough to actually do the math right now...

[runes]

Originally Posted by mamakay

Another clue:

There are a few cases of invasive hib that still happened till age 10 and beyond. If they had multiplied the "one in 1000" till age 10 for a cumulative incidence, how many cases a year would be happening then according to that math? 40,000? Or, one in 100?

What if they took it to age 20?

Yes, if they do the math that way, then at some age, 100% (or, to be consistent with the terminology, 1 case in every 1 person ) will have contracted invasive HiB.

Does not compute.

[runes]

Originally Posted by mamakay

I'm going to give you a clue with what's wrong with multiplying the "one in 1,000" by 5, to make it "one in 200", which give us the "20,000 cases a year" estimate:

http://www.cdc.gov/vaccines/pubs/pin...ds/hib-508.pdf



Look at the chart on page 119.

Whoever sees what's wrong with the 1985 math gets a cookie.

The first thing that jumps out at me is that they are citing those 1986 figures as pre-vaccine figures, but the HiB polysaccharide vaccine was licensed in 1985, so that's not technically pre-vaccine.

But let's just say, for argument's sake, that the chart is correct and it is reflective of the prevaccine numbers.

In the early 1980s, it was estimated that about 20,000 cases occurred annually in the United States, primarily among children
younger than 5 years of age (40–50 cases per 100,000
population).

So they're talking about ALL HI infections, not JUST HiB. In the next paragraph, though, they say:

From 1996 through 2000, an average of 1,247 invasive
H. influenzae infections per year were reported to CDC in
all age groups (range 1,162–1,398 per year). Of these, an
average of 272 (approximately 22%) per year were among
children younger than 5 years of age. Serotype was known
for 76% of the invasive cases in this age group.
Three-hundred forty-one (average of 68 cases per year) were
due to type b.

I had to read this paragraph several times to try to figure out what they are saying. It's so unclear. Out of the 1247 average cases per year, 272 were in kids 5 or younger. The serotype was known for 76% of this group but then they jump to say that 341 cases were due to type B. I have to assume that the 341 cases are talking about ALL cases, not in children under 5 because that is greater than 100% of all cases in children under 5, and that is a mathematical impossibility.


Clear as mud.

[mamakay]

Originally Posted by kidspiration

Yes, if they do the math that way, then at some age, 100% (or, to be consistent with the terminology, 1 case in every 1 person ) will have contracted invasive HiB.

Does not compute.

YES!

Here is your prize.
http://thehealthblogger.com/wp-conte.../07/cookie.jpg

Most of the risk was clustered around age 6 months. Most cases happened in the first year of life. So they (probably closely to accurately) found that one in 1,000 kids under age 5 developed invasive Hib.
MOST of them were babies under one, some more under 2, and very few thereafter.
But they're just arbitrarily going to multiply the incidence by 5? That's going to render an extremely exaggerated figure.

Imagine if (this is an exaggerated fake scenario, to make the concept easier to understand) a fake disease we'll call "mamakay-type A" strikes 1,000 kids under 5 years old a year. 900 (90%) of them are tiny babes under 12 months old. The remaining 100 cases (10%) are dispursed among the 2-20 year olds. But a vast majority are happening in little babies under 1 year of age.

How many cases happen a year? (answer: a bit over 900. there are a few more cases coming in from older kids, too)

What happens if we want to pretend that the risk doesn't cluster around that one year? What if we assume the risk is equal for every year of the first 5 years of life, and just builds from there? Suddenly our 1,000 cases turns into an "estimated" FIVE THOUSAND (and beyond) cases.

But that's not true (in our hypothetical world, where mamakay-type-A is a killer). Really, every year is not the same risk-wise. Really, most of the risk happens in year one. So to take that risk and multiply it by 5 is to exaggerate the incidence almost 5-fold.

That's essentially what the CDC did to come up with the "20,000 cases", "one in 200" figure for Hib prevax.

They made a really basic mathematical error by multiplying the "under 5" incidence by 5.

[ema-adama]

Math is not really my thing, but I'll give it a go.

Thus, in the United States, approximately one of every 1,000 children under 5 years of age develops systemic Hib disease each year, and a child's cumulative risk of developing systemic Hib disease at some time during the first 5 years of life is about one in 200. Attack rates peak between 6 months and 1 year of age and decline thereafter. Approximately 35%-40% of Hib disease occurs among children 18 months of age or older, and 25% occurs above 24 months of age

If they know that children are most at risk 6-12 months of age, and just multiply that number when they are most at risk by 5 to get to 5 years, it is not proportionate. I do not think I am expressing myself clearly, but it doesn't make sense to take the time period that you are most at risk and multiply that number by 5, when in each subsequent year the numbers would be less and not the same.

ETA: booooh, I missed the boat

[mamakay]

Originally Posted by ema-adama

Math is not really my thing, but I'll give it a go.



If they know that children are most at risk 6-12 months of age, and just multiply that number when they are most at risk by 5 to get to 5 years, it is not proportionate. I do not think I am expressing myself clearly, but it doesn't make sense to take the time period that you are most at risk and multiply that number by 5, when in each subsequent year the numbers would be less and not the same.

ETA: booooh, I missed the boat

You get a cookie, too!

http://thehealthblogger.com/wp-conte.../07/cookie.jpg

Yes, multiplying the risk by 5 renders an exaggerated incidence. Most of the risk happened in year one. If you multiply the incidence by 5, you're going to triple or quadruple the number in a way that far exceeds reality.

[ema-adama]

Originally Posted by mamakay

Yes, multiplying the risk by 5 renders an exaggerated incidence. Most of the risk happened in year one. If you multiply the incidence by 5, you're going to triple or quadruple the number in a way that far exceeds reality.

That sounds much more eloquent

[Marnica]

The CDC fudges numbers??? NOOOOOOOOOO that never happens.

They do it with all the other diseases why not HIB?

What I don't get from the numbers from the 1980's is that they are estimates right? I mean HIB was not a reportable disease until 1991 so how do they even come up with numbers that are relatively accurate for their estimates??

[Marnica]

I had another thought about the numbers in general and the risk of contracting a VPD.

We are bombarded with all these numbers which instills fear most of the time. Lets say that the 1 in 200 number for HIB is accurate (which we know it;s not but for arguement sake).

If a child has no history of mental illness in the family then their risk of becoming Schizophrenic is about 1 in 100. One's risk of suffering form depression at some point in their life is about 1 in 20. Yet most folks don't sit around trembling in fear that their child will become schizophrenic or that someday they may be very depressed. Although if their was a vaccine for these things I wonders if people would run out and get it??

[tmaue]

Originally Posted by Marnica

If a child has no history of mental illness in the family then their risk of becoming Schizophrenic is about 1 in 100. One's risk of suffering form depression at some point in their life is about 1 in 20. Yet most folks don't sit around trembling in fear that their child will become schizophrenic or that someday they may be very depressed. Although if their was a vaccine for these things I wonders if people would run out and get it??

Yes, and they would be added without delay to the schedule and parents would line there kids up to get them because to do otherwise would be irresponsible, right.

[bri276]

What confuses me about it is that every pediatrician I've talked to says they almost never see bacterial meningitis anymore. Pediatric journals and magazines reflect the same attitude, "Thank GOD for Hib (and Prevnar)"- moreso than for other vaccines. WHY do they say that if it's not true- brainwashing only goes so far, but it's so strange that all peds seem to have this idea that they constantly saw bacterial meningitis before the vaccine, and now they don't. And what about epiglottitis? That's gone, isn't it?

[mamakay]

Originally Posted by bri276

What confuses me about it is that every pediatrician I've talked to says they almost never see bacterial meningitis anymore. Pediatric journals and magazines reflect the same attitude, "Thank GOD for Hib (and Prevnar)"- moreso than for other vaccines. WHY do they say that if it's not true- brainwashing only goes so far, but it's so strange that all peds seem to have this idea that they constantly saw bacterial meningitis before the vaccine, and now they don't. And what about epiglottitis? That's gone, isn't it?

How old are they?

The ones I've talked to who say that were going through med school in the 80's and working in hospitals, vs now where they see a limited number of kids in a private practice.

[k9sarchik]

My mom is an RN. She worked in pediatrics for a while in the 60's. She didn't even know about Hib. She said never was ever concerned about Hib when my sister and I were born.
Interesting how diseases are not a problem until Voila! we have a vaccine for it.

[Scattershoot]

So they're saying one in 1,000 kids under age 5 developed invasive Hib. Ok.

But then they multiply that by 5 to make it one in 200 kids by age 5.

Anybody see the problem there?

Ok, I'm just getting started, and I have not read all the posts yet. Thanks Mamakay for doing this. You break down numbers well.

So the first thing is the usage of cumulative numbers for the first 5 years of life (to make the odds seem pretty bleak), and at the same time admitting that after 12 months the odds of infection go way down.

[mamakay]

Originally Posted by Scattershoot

Ok, I'm just getting started, and I have not read all the posts yet. Thanks Mamakay for doing this. You break down numbers well.

So the first thing is the usage of cumulative numbers for the first 5 years of life (to make the odds seem pretty bleak), and at the same time admitting that after 12 months the odds of infection go way down.

Yep.
It's just a very basic case of bad math.

[bri276]

Originally Posted by mamakay

How old are they?

The ones I've talked to who say that were going through med school in the 80's and working in hospitals, vs now where they see a limited number of kids in a private practice.

50's and over. My mom was in med school in the 80s when she had me and claims to have seen many cases during residency and in her early years of practicing. Her partners, around the same age, have said the same. Our pediatrician has also. I don't know how old the authors are in the mags and journals, but I've certainly read praise of the Hib vax several hundred times. Also reading the older reference books and textbooks, they definitely identified it as a not altogether uncommon occurrence, and they made epiglottitis sound pretty terrifying.

[tmaue]

Wondering how antibiotic use plays into all of this. Could the use of antibiotics have caused changes in our bacteria and be a reason for a possible rise in Hi diseases? If Hi diseases were such a problem throughout time, why wouldn't a name have been given to them like with other diseases?