Study shows no link between MMR vaccination and autism

Sir

In addition to my response Re. Criticisms confounding Smeeth et al
appearing on this site (and receiving no response from author Sarah Mayor
advising us of Smeeth et al or directly from Smeeth et al) I was informed
by another responder that, concerned there had been no response from
Smeeth et al to my questions, he had forwarded my response to Professor
Smeeth for comment some weeks ago.

To date there has been no response at all.

I find this rather disconcerting - either I am incorrect in my
criticism or correct; only Smeeth et al have all data and information of
methodlogy and procedures involved in the study and, as the study has such
an impact on children globally let alone our own British children, one
expects a team with integrity to make a response to alleviate concerns.

The most glaring concern for me was that Smeeth et al appears to have
treatd the main control group differently from the case subject group - I
suggest that this would destroy the validity of a 'case-controlled' study.

Perhaps other experts in the field could answer this important
question, Smeeth et al do not seem willing?

Regards

John H.

Competing interests:
None declared

Sir

There has been no response about whether Smeeth et
al did ‘validate’ their control groups in terms of non-ASD/PDDs
status. Perhaps my point is seen as irrelevant? Adam Jacobs’ prolific
Responses and his “1% hypothesis” for the Control Group have been easily
refuted – see below – through evidence of unprecedented rates of all PDDs
of perhaps 2% and more from analysis of trends of increasing disability in
childhood, and the already ‘well hidden PDDs’ in the so-called non-PDD
population.

I now offer criticism of Smeeth et al beyond the Control Group aspect
of the study, and cite additional risks that confound results. I do not
believe Smeeth et al can make the claims they do. I feel the study avoids
confounders that are unavoidable when trying to accurately portray
relationships between MMR vaccination and the development of ASD/PDDs.

1. I provided evidence (1) of the unreliability of the GPRD as a
source of data for ASD/PDD persons, their diagnoses and ‘current’ status,
and this does not appear to have been ‘confounded out’ by Smeeth et al.
Their ‘design and methodology’ paper (2) acknowledges confounders that
introduce bias.

2. In (2) Smeeth et al disclose confounders that can be reduced by
applying a “questionnaire to parents of affected children and controls”.
They are the presence in the family of a child with a major illness,
social class, birth order, family size, education of parents, and
religion. Was the questionnaire deployed because the Smeeth et al study
actually states “ We were not able to measure or control for some
potential confounding factors such as birth order within families and
social class, both of which are known to be associated with vaccination
and might be a risk for autism”.

3. Smeeth et al quote as ‘insignificant’ risk “the unexpected finding
that the odds ratio associated with MMR vaccination varied according to
the age at which a person joined the GPRD”.

4. Another risk “It is possible that parents might have noticed signs
of PDD before the diagnosis of PDD and avoided MMR vaccination because of
these signs” might also add weight to the possibility that some ASD/PDDs
were caused by non-MMR vaccinations, eg DTP, given months prior to MMR;
another area ignored by Smeeth et al.

5. “We were not able to separately identify the subgroup of cases
with regressive symptoms to investigate the hypothesis that only some
children are vulnerable to MMR-induced disease….” confounds the study yet
the excuse cites references said to deny the existence of ‘regressive-
autism’. If regressive autism exists, this omission would seriously
confound Smeeth et al.

6. “In a systematic review we identified three other studies that had
directly assessed the risk of autism or other PDDs associated with MMR
vaccination” (Madsen KM, Hvlid A, et al) (3) whose “results were very
similar” to Smeeth et al. (3) has already received significant criticism
from Goldman GS and Yazbak FE (4) so if (3) is invalid does that add risk
to Smeeth et al? I note that (3) (studying autism/MMR) achieves a rate of
autism from the Danish national database of 1 in 728, which differs
significantly from the 1 in 381 of Hvlid et al (5) (studying
autism/thimerosal) for not too dissimilar ages/periods. Can both be right
or is one or more suspect?

7. No attempt was made to examine the effects of individual vaccine
schedules on ASD/PDD status, yet this could seriously confound Smeeth et
al. DTP vaccination, and others, have been identified by parents as
causing ASDs. A small survey I performed several years ago using
‘Communication’, the Journal of the National Autistic Society found about
38% of parents believed their childs’ ASD was caused by DTP, and about 35%
blamed MMR. Their experience, if real, of DTP vaccine causing a similar
but slightly larger proportion of ASDs than MMR vaccine could hold
enormous significance and invalidate Smeeth et al and all similar studies
that ignore such confounders.

8. No attempt was made to confound for the number of MMR vaccinations
received – yet a subject/control could have received one to several
MMR/MMR1/MMR2/ MRs (let alone many other types of vaccines and dosages)
and only single antigen measles vaccine recipients were excluded.

9. Some years ago the noted UK Association JABS analysed anecdotal
data from parents of children with ASDs, and other disorders, believed to
be caused by vaccination. Results suggested that not only specific vaccine
supplier type products could injure but also specific batches of vaccines
seemed more prone to injure. These potential confounders are ignored by
Smeeth et al illustrating apathy, perhaps ignorance, amongst publicly
funded scientists of public experience collated, analysed and presented
independently – yet it is public experience which must underpin studies
funded by, and supposedly performed for, the British public.

10. Studies have found rates as high as 121/10,000 (1in 83) for ‘all
PDDs’ at a specific age published in 1999 (6)

11. DSM-IV (7) (8) has difficulty identifying higher functioning ASDs
such as Aspergers syndrome. This must confound Smeeth et al at areas of
disability that are not easily established between Controls and Case
subjects, muddying already profusely muddy waters (9) (10) (11) (12)(13)
(14).

12. Control Group sample analysis could reduce the risk inherent when
any Case and Control sample depends on raters used (psychologist LH and
psychiatrist EF) without advice from a neurologist or speech and language
therapist (12).

13. Rates for ASD/PDDs appear to be increasing rapidly. Eg. Brick
Township 2001 (15) 67/10,000 or 1 in 149 and Kadesio et al 1999 (6)
121/10,000 or 1in 83 in 7 year olds may have further increased in the last
4/5 years (16); Raymond Gallup’s statistics show rapid increases in rates
of ASD for the US. Those trends may now see Brick Township with ASD rate
of 1 in 83, and Kadesio et al’s ‘7 year old’ area with 1 in 46 of children
suffering ASDs. An acquaintance of mine has a young daughter with autism
who attends a West Yorkshire junior school where her child’s class has an
‘identified autism’ rate of 1 in 14 children.

Not only are Smeeth et al unlikely to have come close to assessing
the real probability of MMR causing autism, they may have contributed more
bad science to an already questionable library of sustained ignorance that
condemns increasing numbers of our children to avoidable serious
disability.

Regards

John H.

References

1. “GPRD case control with or without control control” Heptonstall,
and “Re: Re: GPRD case control with or without control control”
(Heptonstall) eBMJ Responses above.

2. “A case-control study of autism and MMR vaccination using the
GPRD: design and methodology”, Smeeth et al, BMC Public Health, 2001;
1(1): 2

3. “A population based study of MMR vaccination and autism”, Madsen
KM, Hvlid et al, NEJM 2002; 347: 1477-1482

4. “An investigation of the association between MMR vaccination and
autism in Denmark”, Goldman GS, Yazbak FE, JAmPhysSurg 2004; 9(3): 70-75

5. “Association between thimerosal containing vaccines and autism”,
Hvlid et al, JAMA 2003; 290: 1763-1766

6. “Autism and Asperger syndrome in 7 year old children: a total
population study”, Kadesio B et al, J Autism Dev Disord 1999; 29: 327-331

7. “Autism and other PDDs: exploring the dimensional view”, Myhr G,
can J Pshych 1998 Aug; 43(6): 589-95

8. “Does DSM-IV Aspergers disorder exist?”, Mayes et al, J Abnorm
Child Psychol 2001 June;29(3):263-71

9. “PDD in children presenting as possible hearing loss”, HO PT et
al, Laryngoscope 1999 Jan; 109(1): 129-35

10. “Exploring the borderlands of autistic disorder and specific
language impairment: a study using standardised diagnostic instruments”,
Bishop DV, Norbury CF, J Child Pshychol Psychitr 2002 Oct; 43(7): 917-29

11. “The epidemiology of Asperger syndrome: A total population
study” Ehlers S, Gillberg C, J Child Psychol Psychitr 1993 Nov; 34(8):
1327-50

12. “Autism, Asperger’s syndrome and semantic pragmatic disorder:
Where are the boundaries?”, DVM Bishop, http://www.mugsy.org/bishop.htm

13. “PDD among children and adolescents attending a psychiatric day
treatment”, Sverd J et al, Pshyciatr Serv 2003 Nov; 54(11): 1519-25

14. “Juvenile and young mentally disordered offenders: the role of
child neuropsychaitric disorders”, Siponmaa L et al, J Am Acad Psychiatr
Law 2001; 29(4): 420-6

15. “Prevalence of autism in the US population: the Brick Township,
New Jersey, investigation”, Bertrand J et al, Pediatrics 2001 Nov; 108(5):
1155-61

16. http://www.ideadata.org/tables27th/ar_aa3.htm

Competing interests:
None declared

1) I note that Adam Jacobs has not responded to the questions I posed
in my last response, above ['Re: John Stones's Response' 4 October].

2) It may (or possibly may not) be a relatively marginal point at a
technical level but it does reflect on the integrity of the study, since
there is a general claim by Smeeth et al that the GPRD is accurate and
representative (see my response above: 'Adam Jacobs' Response', 20
September). If they had reported what John Heptonstall and I believe about
the quality of the record, rather than the general claim of high
reliability which they make (but do not substantiate), the undertaking
would have looked absurd.

3) We are not dealing with only one problem. Smeeth et al in fact
state "We were not able to separately identify a subgroup with regressive
symptoms to investigate the hypothesis that only some children are
vulnerable to MMR induced disease and this is always regressive". Since
there was never a claim that autism was exclusively caused by MMR, this
would obviously substantially weaken any discernible effect.

4) If as Jacobs noted in his response ("Re: Danish MMR study") 22
September above, you cannot infer anything from the correlation between
MMR and autism incidence in Goldman and Yazbak, how can you infer anything
from the absence of a correlation here. As I have repeatedly said the
thing is to look at the children not the data, but Jacobs is not arguing
consistently.

5) The trouble with most of the data brought to bear on this issue,
apart from anything else, is that it is out of reach and scientifically
unverifiable.

6) The resort to data, rather than investigating the medical
condition of children is an act of bad faith in itself.

Competing interests:
As above

It is not clear to me whether Stone and Heptonstall have genuinely
not understood the point I have been trying to make, or whether they are
just going off on tangents to distract attention from the flaw I pointed
out in their original arguments. If the latter, I don't suppose anything I
say will make much difference, but if the former, then I'll try once more
to explain.

The thrust of their argument seems to be that reporting of autism in
the GPRD is incomplete. I have no idea whether that is true or not. But,
and here is the really important thing, incomplete reporting of autism
would make no difference to the interpretation of Smeeth at al's study.

If we assume that recording of autism is incomplete, the chances are
that there will be some children in the control group who are autistic.
The thing is, and again, this is important, there will be very few of
them, because autism is rare. Even if under-reporting of autism is so
atrocious that most cases of autism are unrecorded in the GPRD, there is
still no reason whatever to think that the control group of Smeeth et al's
study will have a higher prevalence of autism than the general population.
Even Heptonstall now seems to accept that fewer than 1% of the general
population are autistic. If 1% of the control group in Smeeth et al's
study were misclassified, it is hard to see how that would make much
difference.

As far as I can tell, Heptonstall's argument now seems to be based on
the idea that the control group of Smeeth et al's study could somehow have
a much higher prevalence than this. Why? What mechanism could possibly
account for over-representation of children with autism in the control
group?

Competing interests:
As stated previously