Can the MMR or other vac ever cause autism?

[StarlightMcKenzie]

www.ncbi.nlm.nih.gov/m/pubmed/25114790/

Beachcomber, the original paper was a case-control study. Hooker did a cohort study "In this paper, we present the results of a cohort study using the same data from the Destefano et al. [14] analysis. The focus of the current study is differences in results in specific gender and race groups.

For the differences between case control and cohort studies, see the link here:

www.statsdirect.com/help/default.htm#basics/prospective.htm

Note that for a case control study, it says 'smaller numbers required' and for cohort study it says 'requires large numbers'. Large numbers which we know weren't available.
It also says that most cohort studies are prospective, so a retrospective cohort study is unusual. Presumably this also throws up issues.

The article is available through the net, but, crucially, it has been withdrawn by the journal it was originally published in, because the author had competing interests.
Check the full info.

mrs, I'm not saying that the whole sample was too small to run a test on. I'm saying that when you run a statistical analysis and you have a small sample size, any conclusions you draw should be treated with extreme caution.

If the line was 'black boys are at a higher risk of autism if they have the MMR, we know this because we looked at a a small sample of black boys and found this was the case' then it would rightly receive the face. That's why sample size is important. Not because you need to meet a bare minimum otherwise the computer says 'no can do.'

It's just sad that such articles slip through, end up viral and are used as evidence despite having been discredited.

Sorry about your son, Debs. It must be very difficult to cope with such a regression. I hope you find some answers.

Noble, you are assuming that the group that the statistical analysis was run on was small. The author himself suggested that further investigation was required to better understand the relationship between MMR and AA males.

I was pointing out why they put the cut off point at 5 in this study, not saying that sample size wasn't important.

Lweji, is there a problem with saying there is evidence that the MMR may trigger autism in certain AA male children and this requires further investigation? If so, why?

Debs75 I'm so sorry to hear about your son :( Before reading what you wrote I would have said without doubt that the MMR doesn't have anything to do with Autism, but after reading what you wrote there must be some correlation!

There is no problem in saying there is evidence when there is actual evidence.
But there isn't in this case.
The paper has been removed by the publisher because the evidence is crap, and as others have explained here.
It's essentially made up.

Even if there was a correlation, it still didn't mean there was cause and effect.

I think you're jumping to conclusions there too Lweji. How is it made up and what exactly is wrong with it because no one here has actually managed to explain it without jumping to conclusions about sample size.

Why are people so keen to dismiss anything that raises concerns about MMR? Surely it makes sense to identify the more susceptible subgroups so that everyone else can be reassured?

mrs look at this table from the Destefano study

[[https://lbrbblog.files.wordpress.com/2014/08/destefano-table-3.png]

Or this one:

lbrbblog.files.wordpress.com/2014/08/destefano-table-5.png?w=300&h=81

Then look at this one from Hooker

lbrbblog.files.wordpress.com/2014/08/hooker-table-2.png

What's missing from Hooker's table? What, looking at the Destefano table might we expect to see? The numbers involved. Why has Hooker not included these?

This is not an attempt to smear him, this is a valid question. Why has he not included the size of his samples when this is vital to assessing the importance of his results?

lbrbblog.files.wordpress.com/2014/08/destefano-table-3.png working link.

Lweji, is there a problem with saying there is evidence that the MMR may trigger autism in certain AA male children and this requires further investigation? If so, why?

I suspect putting out a hysterical video where it's claimed that your own findings show that the CDC are worse than Hitler might muddy the scientific waters somewhat. Like I said, 'conflict of interest' doesn't quite cover it. And that's even before examining the validity of the findings.

Hooker said it was same data that had been used in the Destefano study. He quoted the same figures that were given in the Destefano study in his Methods section. Why are you assuming that his figures are different?

As an aside, I'm trying to figure out how the Destefano figures add up. Table 5 shows the number of white/other as 218 and the number of black as 137. That gives a total of 355 in the birth cert sample. Table 3 shows a total of 311 cases in the birth cert sample.

Does anyone have a pdf of the Destefano study that they could send me? Otherwise I'll have to wait until Monday to see the paper in full.

I'm not assuming his figures are different, I'm saying that I haven't a clue what his figures are. Yes I know what the population size is, but when you get down to African American boys with normal birthweight, I would expect to see some figures. Like they were put in the Destefano paper for their subgroup breakdowns. As you would expect. As any interested reader who has a basic understanding of statistics would expect. I don't have access to the Destefano paper and it would be incredibly stupid of a scientist to assume that someone reading their analysis would have a copy of a ten year old paper next to them to fill in their gaps. His paper is supposed to be readable, not missing vital information. There is an obvious column missing from his table. Why?

Incidentally, this was my first comment on this thread, after simply reading the paper in the OP, without any knowledge of who the author is or his agenda or any idea about any whistleblowing scandal. So I wasn't desperate for reasons to dismiss his work or smear him or anything like that. It wasn't personal.

noblegiraffe Tue 19-Aug-14 19:11:39
I've had a look through the paper and sample size for each group seems notable by its absence. They comment that they had to cut off analysis at a certain point because otherwise a group would contain fewer than 5 individuals doesn't fill me with confidence about the rest of the groups.

I don't understand, tbh, how this paper got published without that information. I would expect the journal editor to have said 'that's all very well, Hooker, but how many kids are we talking about here?'

Sample size is the first thing you look for.

Noblegiraffe, I think you are missing the point of what Hooker has done.

Hooker didn't design a study to examine the possibility of a link between autism and MMR vaccination in African American male children.

What Hooker has done is re-run CDC figures according to the original study's original untampered with protocol.

And the reason that he has done that is to show that what Thompson says is valid.

And what Thompson says is this;

Decisions were made regarding which findings to report after the data were collected, and I believe that the final study protocol was not followed.

The issue is not are Hooker's numbers big enough (they aren't Hooker's numbers, they are CDC numbers). The issue is that a senior CDC scientist has admitted to scientific misconduct. And that that misconduct relates to a massive public health controversy.

All this criticizing of Hooker's numbers (which aren't his numbers, they are the CDC's) feels like criticizing the quality of a video recording of someone committing a crime rather than paying attention to the actual crime one is being shown. Utterly weird.

I agree with Age of Autism's commentary.

He's (Thompson) talking about scientific fraud on the most important health issue affecting America's children, at the agency charged with protecting them, not a gentleman's disagreement over decisions on how to apply chi square. The media coverage, such as it is, has wandered aimlessly along side issues, but the point here seems pretty basic: There was a protocol directing them how to do the study. William Thompson says he and his CDC colleagues didn’t follow it. And he thinks that's a big problem. Big.

Time Magazine’s question – “Did the CDC cover up the data, as Hooker claims?” – is ridiculous and shows just the kind of misreading of the story, and Thompson's own admissions, that I'm talking about. It should be, did the CDC cover up the data, as CDC Senior Scientist William Thompson, who co-authored the study, claims in a stunning break with his colleagues?

Ok, so you are annoyed that the figures aren't there in front of you in a table. So instead of looking at the figures in the Destefano tables, you are prepared to assume the worst and say that the sample sizes are 'apparently tiny'.

Based on the data given, I don't think that is likely to be the case. Over 60% of children in the birth cohort were black and 78% of them were male.

And whether or not it actually does turn out to be the case that there is a link between timing of MMR and autism in African American boys doesn't change the fact that a senior CDC scientist is clearly saying that he and his collegues did not follow study protocol and that they changed inclusion criteria after collecting and analyzing data.

What is needed here are a few things, including the following; a bigger study (which adheres to its protocol and doesn't make things up as it goes along), an investigation into Thompson's claims of CDC misconduct on a major public health issue, and a cautious approach to use of MMR vaccines until a thorough investigation has been carried out. We also need the CDC to explain why the study protocol was disrespected and what on earth they were playing at and what the motivation was behind that. If what they say about birth certificates is true then that should have been part of the original study protocol - why count kids without valid Georgia certificates and then discount them when you stumble upon something? Kids with autism are kids with autism no matter what sort of birth certificate they do or do not have.

What is not needed is endless speculation about whether the figures (which are not Hooker's, they are the CDC's) are big enough. Nor do we need a let's pretend none of this is happening and keep jabbing kids with a vaccine that a senior CDC scientist claims has fraudulent data being used by the government to attest to its safety.

birth cert cohort

And then we have this gem from from Sharyl Attkisson

The CDC’s DeStefano acknowledges that he and his study co-authors changed their study analysis plan midstream, which resulted in reducing the statistical vaccine-autism link among black boys. But he says they did so for good scientific reason.

“[Vaccine] exposure around [three years of age] is just not biologically plausible to have a causal association with autism,” DeStefano says. “I mean autism would’ve already started by then…it probably starts in the womb. So I think from a biological argument, it’s implausible this was a causal association.”

WTAF?

Here have DeStefano also admitting that they broke their own protocol and then he justifies it by saying that they did so because the data didn't fit in with what they expected/wanted to find/have decided they know about autism (which no-one knows the cause of).

Good Lord.

I bet the CDC are a teensy bit cross about that particular dropped brick. Sharyl must have caught DeStefano before he'd had his first coffee of the day or something. I wonder what Orac et al will manage to conjure up to spin this particular piece of dangerously unscientific arrogance away...

Meanwhile CNN, Time, etc ramble on about Hooker's figures (which are not Hooker's, they are the CDC's) and seem oblivious to two senior CDC officials (one of them quite cheerfully) admitting to the world that they cooked the CDC books. Only in the world of vaccine controversy.

Agree, Beach. That quote from DeStefano is all about making autism fit into the genetics box, isn't it?

When anyone with half a brain knows that there is more than one autism, and more than one route into most autisms.

Once more all about denying the existance of any susceptible subgroups at all - which is, of course, how this has come to be discussed again - covering up the existence of a subgroup who appear to be susceptible (on first study, suggesting further investigation needed, etc, etc) to MMR. Now, where have I heard that before, I wonder?

at Orac's latest lame 'Dr Thompson has caught the anti-vaccine' blog post.

Yes, PandasRock - where have we hear that before....? Just watch, it has started. Dr Wakefield was respected until he suggested a problem with a vaccine and the need for further investigation, now the same is happening to Dr Thompson. Orac et al didn't have a problem with him when he was authoring papers attesting to the safety of MMR but now that he says something critical of the CDC vaccine studies he is a dreaded anti-vaxxxxer and Not To Be Taken Seriously because Orac can't help but wonder (and I quote) "if he has started down that slippery slope to becoming antivaccine".

I guess it must be catching. Quick someone invent a vaccine for it.