New US study supports Wakefield's findings

[bubbleymummy]

Here I have limited Internet access at the moment so I can't really read up much more about it but I'd be interested to see what you all think.

Riveninside: That research has already happened. It does not support Wakefield's claims.

silverfrog: "the best 12 examples"? Research isn't about cherry-picking the bits of data that best support your hypothesis. It's about statistically significant divergences from results predicted by the null hypothesis taking into account prior probabilities.

Any observed divergence from the null hypothesis would be statistically insignificant: 12 is a tiny sample size (it lacks statistical power), and considering only those that are autistic means your sample is not representative of the population (which skew your priors completely). The methodology was flawed, the analysis was flawed, and the conclusions were flawed.

"Gooseberrybushes and bubbleymummy: Are you actually defending the claim that MMR causes autism? Or just seeking to spread fear, uncertainty and doubt about the safety of MMR?"

Oh dear. I logged in thinking, at last someone who can look at the studies and is ready for an informed discussion.

Instead it's just same old same old.

"the claim that MMR causes autism"

Which I DO hope you're not attributing to AW. I hoped you knew something about this.

There never was a "golden age" of MMR uptake.

Until 1997 MMR, the single jab and a double measles jab were all available. If you combined the uptake of all three, you could conclude a good level of coverage.

But in 1997, jabs other than MMR were withdrawn, and rates of uptake fell. It was the wrong policy. Parents (especially those with children both sides of the abolition) felt uneasy eg because of problems with overseas batches, and also because of a general uneasiness (which persists) of giving so many jabs so young.

The better policy decision would have been to do nothing and continue to provide the range of jabs, plus public information and opinion forming on the preferred option. That is, of course, if your aim is to have children immunised (not to have them have one particular regime). The unyielding and patriarchal approach of the policy makers in 1997 exacerbated the whole issue.

G33k: You give one paper: there are others which have been published which have backed up the research. They've been linked to elsewhere and if I have the time and energy later I will copy and paste, but have just come off another long long thread like this and am rather depressed to find the same old "baby-killing doctor" nonsense going on here as well.

Considering the thousands of parents who report autistic regression, and considering the explosion in autism diagnoses, I don't think one paper is very convincing. I can't analyse the paper, I'm better at the epidemiological studies.

singles were withdrawn in late 1998, after the paper was published, and well after the press conference.

g33k - as already pointed out, the 1998 paper was not research. it never was, and never will be. it was a presentation of some cases that Wakefield had come across in his daily practice. he called for research.

A quick look at the abstract and I don't find any mention of the number of children. Also these were blood samples: from the gut? It doesn't say. Perhaps you can explain?

Silver did you look at the abstract that G33 posted? Were they looking in the right place? It doesn't seem that way to me.

"Any observed divergence from the null hypothesis would be statistically insignificant: 12 is a tiny sample size (it lacks statistical power), and considering only those that are autistic means your sample is not representative of the population (which skew your priors completely). The methodology was flawed, the analysis was flawed, and the conclusions were flawed."

But it wasn't a sample size designed to have statistical power. The methodology was flawed? It was a case study. The results were presented as evidence that more research was needed.

"Research isn't about cherry-picking the bits of data that best support your hypothesis."

But he wasn't doing this : he wasn't doing anything to do with this kind of research. Either you've been misled or you are trying to mislead.

I'm not sure that people realise that Wakefield didn't start something. Something was happening. Parents were reporting reactions.

Everybody talks as if Wakefield presented this information out of the blue and then wham, people jumped on the autism-mmr reaction reporting. They were being reported before he published. Not just this kind of regression either: auto-immune problems, asthma and so on. They were reported in the Guardian before it went all "parents are stupid" on the issue.

have the windows opened, gooseberry, but am in the middle of tea time crowd control . will have a read once feeding time is over!

I stand corrected.

But I think even with the different time line, the policy in itself was wrong for the reasons I outlines above. (Please not I am talking here about the policy - especially in light of parents' behaviour - not about the safety evidence on any of the jabs).

sorry, meditrina - I tend to go into overdrive on correcting misinformation

I do understand the point oyu were making

right, the d'souza study seems to be the same hoary old chestnut of "examining children with ASD" not actully trying ot look at the group as identified by Wakefield et al. this is important, of course, as it has never been suggested that all autistic children have been affected by mmr/vaccinations.

there was no indication that the children with ASD used in the study were in any way selected for GI issues (a key factor).

it also has a dodgy conclusion in "epidemiological burden of evidence against such an [causal] association between MMR and autism is overwhelming" - no reference to Cochrane at all, which of course rubbishes most epi stuff, so not sure where the overwhelming bit comes in....

the Afzal study appears to be a case of not looking in the rigth place

he looked in the large intestine, whereas what he was trying to disprove was found in the ilium. a small, but important difference.

I have to do bedtime now, so cant read any further on that one for now, but will read more later.

a bit more as dd2 is faffing about...

the Afzal study was based on 15 children (where are the cries of not statistically important now?!) who all had mmr and were autistic.
again, not clarified whether early-onset ASD or regressive ASD - a pretty important point, really, given what is is trying to disprove...