Vaccine Autoimmune Project for Research and Education (VAP)
  Archived on 2Apr09

How PEDIATRICS Validated Erroneous British Mercury Data
John Stone

‘Mercury and Autism in the United Kingdom’ a 3-part investigation by John Stone has just been published on VAP. Originally featured on ‘Red Flags’, it evidenced fraud and high-level collusion between the CDC, the WHO and British health officials.  The revealed information was presented to PEDIATRICS and turned down.

On February 8, 2004, the Immunization Safety Review Committee of the IOM met to re-examine the issue of vaccines and autism.  There were five thimerosal-related epidemiological studies that convinced the committee that thimerosal was not a trigger for regressive autism.  One of those studies was conducted in the United Kingdom.  It had not yet been published when the committee met but its data was presented by co-author Dr. Elizabeth Miller of the United Kingdom Health Protection Agency (formerly Public Health Laboratory Service).

The study was subsequently published in the September 2004 issue of PEDIATRICS.  Nick Andrews, Elizabeth Miller, Andrew Grant, Julia Stowe, Velda Osborne, and Brent Taylor were listed as authors, in that order, and the publication was titled 'Thimerosal Exposure in Infants and Developmental Disorders: A Retrospective Cohort Study in the United Kingdom Does Not Support a Causal Association' (1).  However, it was not widely discussed in the UK where the subject of thimerosal has never been as exposed to public attention as it has been in the United States, and never competed in media and public interest with the MMR controversy.

It was only months after publication that I began to look carefully at the historical record and realized that the study contained numerous anomalies in addition to those already cited in electronic responses by John Heptonstall and Mark and David Geier.

It is well known that electronic letters to a medical journal are usually published unless they are not relevant, contain offensive language or constitute a personal attack on someone.  In fact, PEDIATRICS did accept three P3Rs that I sent concerning the Andrews study.  However, as the months went by and I awaited a response from the authors ever more problems presented themselves, and I made three further attempts to address these to the journal with letters written in August 2005, February 2006 (after the publication of my three part investigation Mercury and Autism in the United Kingdom, originally published in Red Flags (2)) and November 2006.  None of these were published.  And as it has now emerged through publications on VAP this was not an isolated matter (3, 4).

Below are the three submissions which were refused.  They contain information about the mercury in vaccines issue in the United Kingdom that may not be widely known in the United States.


19 August 2005:

ID: 114/3/584

Why a "retrospective cohort study" rather than a time-trend analysis?

Having noted the continuing silence of Nick Andrews, Elizabeth
Miller, Andrew Grant, Julia Stowe, Velda Osborne and Brent Taylor [1] (not
to mention Jon Heron and Jean Golding [2]) regarding the mysterious issue
of the double dose of thimerosal [3], there is another event in the
history of the UK dosage of thimerosal which is recorded by Andrews et al
but has yet to be discussed: the change in the UK DPT schedule in 1990
from 3, 5 and 10 months to 2, 3 and 4 months [1].  In addition to the
greater immaturity of the infant metabolism in the new schedule - and
presuming the thimerosal content to have remained the same - this
represents an approximately 20% rise in dosage when measured against body-
weight [4].

Without going once again into the hugely problematic usage of the UK
General Practitioner's Research Database [5, 6] I would like to suggest
that there is another important source of data available in relation to
this event which ought to have been considered and which is published in a
contemporaneous study of MMR and autism in Archives of Disease in
Childhood which shared 4 of the same authors with the present study [7].
The study covers 5 districts of North-East London between 1979 and 1998
(perhaps my own son is one of the cases), and very speculatively the
authors seem to attribute the entire rising trend to better diagnosis [8].
However, the cohorts, of 1990 and 1991 show a particularly alarming growth
in the autism rate.  According to my reading of their graph cases rose from
21 for the birth cohort of 1989, to 37 in 1990 and 47 in 1991.
(In 1992 incidentally, the trend continued to rise and that was the year
HiB was introduced in the UK.)

This is highly unsatisfactory.  It is apparent that the authors of the
present paper chose the ground on which to explore the possible
relationship between thimerosal and autism, and it was not autism
prevalence in 5 north east London districts 1979-98.  If they had done they
would have had a little more explaining to do.  As it is they did not
mention the hike in thimerosal exposure in the ADC paper, or allude to the
ADC paper here.  Once again we are owed an explanation.

[1] Nick Andrews, Elizabeth Miller, Andrew Grant, Julia Stowe, Velda
Osborne, and Brent Taylor, 'Thimerosal Exposure in Infants and
Developmental Disorders: A Retrospective Cohort Study in the United
Kingdom Does Not Support a Causal Association', PEDIATRICS Vol. 114 No. 3
September 2004, pp. 584-591 (doi:10.1542/peds.2003-1177-L)

[2] Jon Heron, and Jean Golding, and the ALSPAC Study
Team,'Thimerosal Exposure in Infants and Developmental Disorders: A
Prospective Cohort Study in the United Kingdom Does Not Support a Causal
Association', PEDIATRICS Vol. 114 No. 3 September 2004, pp. 577-583

[3] John Stone, 'Please can we have the correct information about
dosage?', 12 June 2005,


[5] John Heptonstall: 'Does weight confound?' 30 October 2004,

[6] John Heptonstall: 'Re: Re: Does weight confound?' 24 November

[7] R Lingam, A Simmons, N Andrews, E Miller, J Stowe and B Taylor:
'Prevalence of autism and parentally reported triggers in a north east
London population', Archive of Disease in Childhood 2003; 88: 666-670,

[8] John Daniel Stone: Prevalence of autism: no evidence for
conclusion', 8 August 2003,

Conflict of Interest: Autistic son

7 February 2006:

ID: 114/3/584

I have just completed a 6 month long three part investigation into
the present study [1] published in the on-line medical journal Red
Flags[2, 3, 4, 5].  In addition to the many unresolved points in the above
correspondences (and I include not only my previous e-letters but those
from Mark and David Geier and John Heptonstall) I note the following

i) The mercury load of the WHO program in 2001 exposed infants to
187.5 µg of mercury between birth and 14 weeks, and was not "the same" as
the official U.K. load, as stated here [1, 3, 4].

ii) The study was under the control of the CDC, which was not
transparent [4].

iii) The project was reviewed by the CDC after it had been announced
in the British press, and was given the green light despite the fact that
it was recognised that exposure in the official U.K. programme was a
fraction of the WHO exposure, and despite recognised deficiencies in the
U.K. database [4].

iv) Evidence exists from other Department of Health documents and a
parliamentary answer that mercury exposure in the period examined varied
greatly in the U.K. and may frequently have exceeded 75µg 2-4 month dose
which is given as standard in the study [3, 5].

v) Four of the authors of the present study (Andrews, Miller, Taylor
and Stowe) worked contemporaneously on a separate MMR focussed study
published in Archives of Diseases in Childhood which disclosed a
diagnosed 109% increase in lower continuum ASD between 1989 and 1991 in
northeast London synchronous with the introduction of the accelerated DPT
schedule in 1990 [6].  This dramatic trend is neither mentioned in the
present study, nor is the introduction of the accelerated DPT schedule
mentioned as a possible confounder in the MMR study [5].

vi) Recording of developmental disorders and particularly autism is
haphazard in the General Practitioners' Research Database.  Results vary
alarmingly between studies and seem to be below actual population levels
by several orders of magnitude [5].

vii) It was British Government policy to deny there was any concern
regarding vaccine mercury from the time the issue arose in the US in 1999
and before any studies at all were carried out.  This position was
maintained while studies were carried out, and even after mercury had been
removed from the programme in October 2004 [5].

Supporting documents supplied.

[1] Andrews N, Miller E, Grant A, Stowe J, Osborne V, Taylor B.
Thimerosal Exposure in Infants and Developmental Disorders: a
retrospective cohort study in the United Kingdom does not support a causal
association. PEDIATRICS 2004;114:584-591.

[2] Stone J, Mercury ad Autism in the U.K. I-III, February 1-6, 2006,

[3] Stone J, Mercury and Autism in the U.K. I: The British Government
and the WHO,

[4] Stone J, Mercury and Autism in the U.K. II: The Long Arm of the

[5] Stone J, Mercury and Autism in the U.K. III: The PHLS Study

[6] Lingam R, Simmons A, Andrews N, Miller E, Stowe J, Taylor B.
Prevalence of autism and parentally reported triggers in a north east
London population. ADC 88;2003:666-70. 

Conflict of Interest: Autistic son

26 November 2006:

ID: 114/3/584


Nick Andrews, Elizabeth Miller, Andrew Grant, Julia Stowe, Velda
Osborne and Brent Taylor state concerning the mercury exposure in the UK:

"Because the United Kingdom changed to an accelerated 2/3/4 month DTP
immunization schedule in 1990 (replacing the former 3/5/10 month schedule)
and because vaccinations are generally given on time in the United
Kingdom, a substantial proportion of children in the GPRD cohort will have
had a cumulative Hg exposure of 150 µg of thimerosal (75 µg of Hg) by 4
months of age.  This level of Hg exposure, although lower than the maximum
of 187.5 µg received in the United States by 6 months of age, is similar
to the level received by 3 to 4 months of age in the United States.  It is
also the same as the amount of thimerosal used by developing countries
that follow the expanded immunization schedule." [1]

It is odd that they do not give WHO schedule as a straight point of
comparison with US and UK practice.  The WHO schedule is not widely
publicised, however a UK Committee on the Safety of Medicines (CSM)
document from 2001 gives the WHO mercury exposure as 187.5 µg between
birth and 14 weeks, being similar in content but not time span to the US
schedule.  It is hard to see how by intent the UK and WHO schedules are in
any way "the same".  This is exceptionally disturbing given that this study
is presented as evidence in support of WHO practice, and was indeed
commissioned by the WHO for this purpose.

I note additionally a contemporaneous study also co-authored [by] Nick
Andrews, Elizabeth Miller, Julia Stowe and Brent Taylor [2].  This study,
while focussing on the MMR and autism, documents a rise in incidence in
lower continuum autistic disorder in a north east London population from
22 cases in the cohort of 1989 to 46 cases in the cohort of 1991 coincident
– though this is not mentioned - with the introduction of the
accelerated DPT schedule in 1990.  It must be a matter for grave concern
that the authors failed to consider anywhere this evidence of a possible
population effect for DPT and thimerosal, either here or as a potential
confounder in the MMR study.  Indeed, it [is] not obvious why a time-trend
analysis in relation to the thimerosal problem was not undertaken.  If
there were strong arguments to discount the trend it would surely have
looked better if they had been rehearsed at the time, as the authors were
patently aware of it.  Unfortunately, it is far more telling than the
impenetrably crunched data of the probably inadequate GPRD
(I commend John Heptonstall's analysis, above).

(CSM document submitted).

John Stone

[1] Andrews N, Miller E, Grant A, Stowe J, Osborne V, and Taylor B,
'Thimerosal Exposure in Infants and Developmental Disorders: A
Retrospective Cohort Study in the United Kingdom Does Not Support a Causal
Association', PEDIATRICS Vol. 114 No. 3 September 2004, pp. 584-591

[2] Lingam R, Simmons A, Andrews N, Miller E, Stowe J and Taylor B,
'Prevalence of autism and parentally reported triggers in a north east
London population', Archives of Disease in Childhood 2003;88:666-670,

Conflict of Interest:
Autistic son


Contemporaneously with last submission I sent one to Archive of Diseases in Childhood (27 November 2006) which was in fact published on-line, although not answered (5):

Dear Editor,

Figure 1 in this study shows a rise in autism incidence (recorded) from 22 cases in the birth cohort of 1989 to 46 in 1991.  R Lingam, A Simmons, N Andrews, E Miller, J Stowe and B Taylor fail to take into consideration the introduction of the accelerated Diptheria, Pertussis, Tetanus (DPT) as a potential confounder, advancing the doses from 3, 5 and 10 months to 2, 3 and 4 months [1]: each dose also containing 50 µg Thiomersal (Thimerosal) including 25 µg mercury [1].  This is a surprising omission as N Andrews, E Miller, J Stowe and B Taylor were also co-authors of [1], which though not published at the time of this study had already been delivered privately to the World Health Organization in June 2002 [1].  Equally in [1] they failed to note the rise aforementioned in this present study as possible evidence of a population effect of the accelerated schedule.

John Stone


[1] Andrews N, Miller E, Grant A, Stowe J, Osborne V, and Taylor B, 'Thimerosal Exposure in Infants and Developmental Disorders: A Retrospective Cohort Study in the United Kingdom Does Not Support a Causal Association', PEDIATRICS Vol. 114 No. 3 September 2004, pp. 584-591 (doi:10.1542/peds.2003-1177-L),

However, in November 2006 a further potential confounder to the Andrews study also came to light. A story appeared in the Sun newspaper that due to poor recording many British infants may have received the vaccine schedule more than once thus significantly increasing their mercury exposure. On 9 November 2006, John Heptonstall dispatched the following P3R submission to PEDIATRICS. Once again, it was not published.

ID: 114/3/584 -->


The UK media today (8th November 2006) contains information
suggesting that the UK government and medical establishment is presiding
over a major national scandal with respect to its vaccine programmes

"Child immunisation records are in such chaos that health chiefs
across Britain have no idea what jabs have been given to hundreds of
thousands of kiddies" says The Sun..."The scandal is exposed in top secret
documents leaked to The Sun, which reveal records held by the NHS are a
farce....up to 60% more shots have been given to children than should have
been...some youngsters appear to have received THIRTY jabs - when they
should have had just 13....amazingly some children are shown as having
nine jabs in a single day.....the shambolic records - dating back an
astonishingly TWENTY years - are mainly blamed on incompetent paperwork
in surgeries......six records taken at random were ALL found to contain
errors.....That means records for one million children in London alone are insider said there is no reason to think that the rest of
the country is any different, it's just that London is the first place any
of this work has been done".

Dr. Elizabeth Miller, one of the authors and a noted employee and
spokesperson for the UK Department of Health during the past two decades,
could comment on this extremely serious developing situation and how it
might affect the integrity and validity of this study and all others
reliant on the accuracy of UK vaccination records and respective databases
developed this past twenty years, not least the UK GPRD, a counterpart of
which has long been maintained by the Boston Collaboration using constant
updates from UK computerised medical records?

Assuming The Sun story holds merit, would it not now be wise for
every parent to seriously consider withholding vaccination from their
children until validated evidence-based studies are available to prove
safety and efficacy of this medical intervention?  Perhaps Dr. Miller can
convey to the readers her expert opinion as to whether vaccines, which may
now be devoid of proper evidence-based support from UK based research,
ought to be suspended until they can be scientifically declared safe?


John H.

Conflict of Interest: None declared

I also wrote querying this matter with health minister, Caroline Flint.  I received the following reply on her behalf from Dr Philip Bryan of the Medicines and Healthcare products Regulatory Agency on 18 January 2007:


Dear Mr Stone,

Thank you for your e-mail of 10 November 2006 to Caroline Flint, Minister of State for Public Health, in relation to thiomersal in vaccines.  As vaccine safety falls within the remit of Medicines and Healthcare products Regulatory Agency (MHRA), I have been asked to reply on her behalf.

As I have stated in our previous correspondence, the Commission on Human Medicines (CHM) and its Vaccine Expert Group has reviewed the available evidence in relation to the safety of thiomersal-containing vaccines and considers all relevant new evidence as it be comes available.  The advice of the CHM, and the MHRA, remains that there is no evidence of adverse effects caused by levels of thiomersal in vaccines.

We have noted the information provided in your latest correspondence, in particular that in relation to the validity of the study by Andrews et al 2004 and the amount of thiomersal contained in some UK vaccines.  As you are aware, through direct correspondence between yourself and MHRA during 2005 and 2006, we have tried our best with the information available to us to answer in full your many request for information and clarification around issues relating to the safety of thiomersal in vaccines.  Several of the points in our past correspondence also concerned the study by Andrews et al 2004.

We wish to re-iterate that the study by Andrews et al was published in a peer-reviewed journal and was given full consideration by the CHM (then the Committee on Safety of Medicines) Vaccine Safety Working Group.  The Group considered that the study was robust, that the methodology and results were valid and the conclusions justified.  This also applied to the GPRD database in the study which has been used in many peer-reviewed vaccine studies previously.  The MHRA has nothing further to add on the methodology of the study.

We are aware that the authors of this study have previously tried to address your, and others’, concerns via correspondence on the Pediatrics journal on-line message board.  If you have any further concerns over the methodology, you may wish to raise these with the authors directly.

Finally, we have clarified the issues around thiomersal content of DTP vaccines used in the UK in previous correspondence (I refer you to my letters of 3 August and 5 September 2005).  I also understand that Professor Miller has answered similar questions via the Pediatrics journal on-line message board.  We have nothing further to add on this matter.

Yours sincerely,

Dr Philip Bryan
Senior Scientific Assessor
Vigilance and Risk Management of Medicines (VRMM)

The implications of this letter are manifold:

1)   A supposedly independent agency answers on behalf of a government minister.
2)   Further evidence of potential flaws in study data are airily waved aside without reasons being offered.
3)   The robustness of the methodology and data source of the study is once again asserted without contrary evidence ever having been addressed, or at best only peripherally.  Scientific dialogue has once again been avoided.
4)   Peer review status is held to validate the study irrespective of further evidence coming to light.


Conclusion:  The Andrews study was fraught with severe flaws from its inception.  The authors had undisclosed and to this day unacknowledged competing interests which should at least have been made known, and which should have effectively disqualified them from undertaking the research.  They stated in reply to Mark and David Geier that: “The requirements by Pediatrics for the conflict of interest declaration were complied with…” (6), so it would seem that the commercial conflicts were made known to the journal, but these were not initially passed on for the information of the readers.  The question of the hidden commercial conflicts, though serious, was less bad than the even more fundamental problem of the historical involvement of the authors with the policy they were reviewing.

PEDIATRICS initially allowed serious criticism of the paper in electronic peer to peer review, which the authors responded to in part, though far from satisfactorily.  However, when further and even more serious flaws were exposed, the editor simply refused to publish the documentary evidence.  This has left the journal profoundly compromised.

The study lies at the heart of the defence of a policy which was established and longstanding before the study was even undertaken.  It was commissioned in a triangular arrangement between British, United States and World Health Organization health officials, the very people who had heaviest responsibility for the policy in the first place.  We now live with the official fiction that the science was correct.  In 2004, when the spotlight fell on the famous Wakefield Lancet study of 1998 (7), there was so little wrong with it that the ten authors who signed a letter of retraction could only dissociate themselves from “an interpretation” and not something which was stated in the paper (8).  On the other hand the Andrews study is so flawed it is hard to know where to begin or where to end, but it remains a cornerstone of international health policy.

Scarcely a day passes without someone repeating that based on reliable epidemiological evidence the Immunization Review Safety Committee has decreed that thimerosal in pediatric vaccines is not and never was related to autistic regressions.  When the British data was presented at the committee meeting in February 2004, it had not been subjected to peer-review, which in itself is problematic.  After the study was published, the authors only responded in a selective and incomplete way to published criticisms, while the journal blanked out the most serious criticisms of all.  The editor and any US reviewer could hardly have known all the details that I revealed, and the study certainly should not have been published in the first place if they did.  Indeed, it is even possible that British reviewers were not aware of some of the details.  The suppression of material criticism is inexcusable, and places infants across the world in jeopardy.

(3) F E Yazbak, A Tale of Two Cities: Flawed Epidemiology, March 7, 2007,
(4) J A Trelka, How Mercury was Absolved: Creativity, Collusion and Censorship, March 19, 2007, 
(7) A J Wakefield, S H Murch, A Anthony, J Linnell, D M Casson, M Malik, M Berelowitz, A P Dhillon, M A Thomson, P Harvey, A Valentine, S E Davies, J A Walker-Smith, ‘Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children’, Lancet, Volume 351, Number 9103, 28 February 1998.
(8) ‘Retraction of an Interpretation’, Volume 363, Number 9411, 6 March 2004.

John Stone
London, UK


VAP Home  |   When 1/150 is 1/67  |   cdc-spinach-autism  |   How-mercury-was-absolved
Justice-Davis-Complaint  |   let-justice-be-done  |   Letter-to-Fineberg
pediatrics-validated-erroneous-mercury-data  |   tale-of-two-cities  |   uk-mmr-vaccination-rates-before-1998
mmr-briefing  |   vaccination-of-very-premature-infants  |   publicletters  |   Erics-story  |   Jonny-Sierra.htm