Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
"We can now move on..." Oh no you don't Prof Smeeth!
- John Stone (17 September 2004)
-
Same old, same old..
- Jamie S Robertson (17 September 2004)
-
IS THERE ANYBODY THERE?
- John Stone (19 September 2004)
-
Environmental reasons for increases in autism
- Ellen c G Grant (19 September 2004)
-
Re: IS THERE ANYBODY THERE?
- Adam Jacobs (20 September 2004)
-
Adam Jacobs' response
- John Stone (20 September 2004)
-
The Bigger They Are…
- F. Edward Yazbak (20 September 2004)
-
Danish MMR study
- Hilary L Davidson (20 September 2004)
-
Re: Danish MMR study
- Adam Jacobs (22 September 2004)
-
Re: Danish MMR study
- John Stone (22 September 2004)
-
Re: Re: Danish MMR study
- L. Travis Haws (22 September 2004)
-
Re: Re: Re: Danish MMR study
- Adam Jacobs (23 September 2004)
-
Re: Re: Re: Re: Danish MMR study
- L. Travis Haws (24 September 2004)
-
Re: Re: Re: Re: Re: Danish MMR study
- Adam Jacobs (24 September 2004)
-
GPRD case control with or without control control?
- John P Heptonstall (25 September 2004)
-
Re: Danish MMR study
- Graeme Johnston (27 September 2004)
-
Re: GPRD case control with or without control control?
- Adam Jacobs (27 September 2004)
-
Re: Re: GPRD case control with or without control control?
- John P. Heptonstall (28 September 2004)
-
Re: Re: GPRD case control with or without control control?
- Carol Johnston (28 September 2004)
-
Re: Re: Re: GPRD case control with or without control control?
- Adam Jacobs (28 September 2004)
-
Re: Re: Re: Re: GPRD case control with or without control control?
- John P Heptonstall (29 September 2004)
-
Re: Re: Re: Re: Re: GPRD case control with or without control control?
- Adam Jacobs (30 September 2004)
-
Re: Danish MMR study
- Raymond Gallup (30 September 2004)
-
Re: Re: Re: Re: Re: Re: GPRD case control with or without control control?
- John Stone (30 September 2004)
-
Re: Re: Re: Re: Re: Re: Re: GPRD case control with or without control control?
- Adam Jacobs (1 October 2004)
-
Re: Re: Re: Re: Re: Re: Re: Re:
- John Stone (3 October 2004)
-
Re: Re: Re: Re: Re: Re: GPRD case control with or without control control?
- John P Heptonstall (4 October 2004)
-
John Stone's response
- Adam Jacobs (4 October 2004)
-
Re: Re: Re: Re: Re: Re: Re: GPRD case control with or without control control?
- Adam Jacobs (4 October 2004)
-
Re: Re: Re: Re: Re: Re: Re: Re: GPRD case control with or without control control?
- John P Heptonstall (4 October 2004)
-
Re: John Stone's response
- John Stone (4 October 2004)
-
Response to the two Johns
- Adam Jacobs (5 October 2004)
-
Re: Response to the two Johns
- John P Heptonstall (5 October 2004)
-
Re: Response to the two Johns
- John Stone (5 October 2004)
-
Re: Re: Response to the two Johns
- Adam Jacobs (5 October 2004)
-
Criticisms confounding Smeeth et al
- John P Heptonstall (7 October 2004)
-
Out of time
- John Stone (7 October 2004)
-
Re: Criticisms confounding Smeeth et al
- John P Heptonstall (9 November 2004)
-
Re: Out of time (Adam Jacobs's interests)
- John Stone (12 November 2004)
-
A letter the Lancet failed to publish
- John Stone (22 December 2004)
|
|
||||||||||||||||||||||||
|
John Stone, none London N22
Send response to journal:
|
There has never been anything to stop the Department of Health investigating the causes of autism, but why are we once again - and after so many years - looking at databases rather than children? The present paper notes: "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" [1], so the study admits that there might be such a subgroup within the entire population but the database does not allow them to differentiate it. Not finding a link between MMR and autism or PDDs is an unsurprising result of the exercise when the data was too generalised in the first place. Additionally, I note, that I have already pointed to anomalies regarding autism incidence in the General Practitioners Research Database [GPRD] as reported in the study of Hershel Jick and James Kaye [2], and no one has ever attempted to answer my observation. I wrote in Rapid Responses on 15 May 2004: "Using the UK General Practitioners Research Database the (Jick and Kaye)document the increased incidence in autism (some confusion between year of birth and year of diagnosis) from 1.6/10,000 in 1993 to 9.5/10,000 in 1999. During the same period (birth cohorts of 1993-6) Lingam et al record a steady incidence of 2.6/1,000 excluding Asperger Syndrome in a North East London population [3]. The probability is that all that Jick and Kaye have documented is improved but very inadequate recording over the period." [4] If this is the case then the GPRD not only has inadequate information about cases to draw any conclusions about MMR and autism incidence, but also apparently massive under recording of total incidence. I have also previously noted The Lancet's double standards in not identifying Prof Fombonne's competing interest as a defence witness in the MMR litigation for Aventis Pasteur, in relation to its extreme treatment of Andrew Wakefield [5]. [1]Op.cit p.967 [2] Jick and Kaye, 'Epidemiology and possible causes of autism', Pharmocotherapy vol 23 no 12 2003, p.1526-30. [3]Lingam et al, 'Prevalence of autism and parentally reported triggers in a North East London population', Archive of Diseases in Chilhood, vol 88, 2003, p.666-70 [4] Posting 'Re:Re:Re: Response to Hiliary (sic) Butler' in Watts 'The new MMR?' http://bmj.bmjjournals.com/cgi/eletters/328/7442/773 [5]'Prof Fombonne, Dr Horton and the Lancet: double standards on competing interests' (12 September 2004) in Bedford and Elliman: 'Misconceptions about the new combination vaccine', http://bmj.bmjjournals.com/cgi/eletters/329/7463/411 Competing interests: Parent of an autistic child |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Jamie S Robertson, Intercalating medical Student (Immunology) University of Glasgow, G12
Send response to journal:
|
I share Mr. Stone's concerns. Particularly as this study was plastered all over the media as 'showing that there is no link between MMR and autism' (very different from showing no link, I might add!), when quite clearly it does nothing of the sort. All it shows is that, as a population, you are no more likely to get autism if you have had the MMR jag. But as Mr. Stone says, it doesn't take into account the powerful individual cases, and fails to explain what HAS clearly happened in some children - as have all studies performed in this way. The relevant studies are surely those which look at affected children for an aetiology. Competing interests: None declared |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
John Stone, none London N22
Send response to journal:
|
Does anyone dispute that there is most likely massive under-recording of autism on the United Kingdom General Practitioners Research Database? Competing interests: As above |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Ellen c G Grant, physician and medical gynaecologist 20 Coome Ridings, Kngston-upon-Thames, KT2 7JU ,UK
Send response to journal:
|
We already know many of the environmental reasons for impaired brain function in children. These include deficiencies of essential nutrients, especially of zinc, during development and the effects of common social poisons which include alcohol, tobacco smoking and toxic metal exposures. Since lead-free petrol became available, blood lead levels in my patients have fallen. However, alcohol use, smoking and steroid hormone use in younger women in the form of contraceptives or hormonal medications, before their first pregnancies, and during pregnancy or at birth to “protect” the lungs of a premature child, have been increasing. Inexplicably, dental mercury amalgams continue to be inserted into the mouths of future parents. John McLaren-Howard’s biochemical findings in 61 autistic children give an important perpective into why childhood illnesses have increased so dramatically. Most of the autistic children had zinc and copper deficiencies and superoxide dismutase (SODase) dysfunction. 16 had DNA- adducts in leucocytes to malondialdehyde, 12 to cadmium, 9 to nickel, 3 to mercury and one to lead. Even if only 5% of autistic children show evidence of signs of mercury exposures, this still means large numbers of children are adversely affected by mercury but the main source is likely to be parental dental amalgams. It could be that any extra mercury in a vaccination is “the straw that breaks the camel’s back” for some vulnerable children. The relation between environmental factors and brain development and function has been investigated for decades. In 1982 Tanaka and colleagues discovered a beneficial effect of supplementary zinc on foetal growth in rats which could possibly help preventing the CNS dysfunctions of Foetal Alcohol Syndrome. They also noted that the effect of zinc supplementation was not as good compared to the results in the control animals not given alcohol. In 1994 Nishida and colleagues determined copper, zinc-superoxide dismutase (Cu,Zn-SOD) activities in 22 human brains from fetuses to adults. Cu/Zn-SOD activity of the cerebral cortex and white matter increased from 15% in fetuses to 50% of adult levels in neonates. The activity of the white matter was higher than that in the cortex in the foetal period, but was essentially the same as those of the cortex in the postnatal period. Cu/Zn-SOD activity in the central nervous system was highest in the spinal cord and higher in the order pons, medulla oblongata > cerebellum, midbrain, thalamus > putamen, pallidum and cerebrum. These low activities may be related to the vulnerability of cerebral cortex and white matter in premature infants. If even a little of the huge amount of money spent on “politically correct epidemiological studies” was used instead for basic biochemical investigations, so many of today’s “mysteries” would be unveiled. 1 Grant ECG, McLaren-Howard J. Re: The effects of toxic metals in autistic children. http://bmj.com/cgi/eletters/329/7466/588-b#74117, 13 Sep 2004 2 Tanaka H, Nakazawa K, Suzuki N, Arima M. Prevention possibility for brain dysfunction in rat with the fetal alcohol syndrome-low-zinc-status and hypoglycemia. Brain Dev.1982; 4: 429-38. 3 Nishida A, Misaki Y, Kuruta H, Takashima S. Developmental expression of copper, zinc-superoxide dismutase in human brain by chemiluminescence. Brain Dev. 1994; 16: 40-3. Competing interests: None declared |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Adam Jacobs, Director Dianthus Medical Limited, London SW19 3TZ
Send response to journal:
|
If autism is under-reported in the GPRD (and I don't feel qualified to comment on whether it is or not), that would make no difference to the interpretation of the study by Smeeth et al, because it was a case-control study. The design of case-control studies, in which cases are chosen not at random from the population, but because they are cases, makes the absolute prevalence of the condition under study irrelevant to the interpretation of the results. If one assumes that MMR vaccination really does cause autism, and that the failure of this study to find an association is because of incomplete reporting, then one would need to hypothesise that the probability of a case of autism being recorded in the GPRD is related to vaccination status. In other words, children who are vaccinated and autistic would not be recorded as autistic in the GPRD, whereas children who are autistic but not vaccinated would be recorded in the GPRD as autistic. Although this is theoretically possible, it does seem rather far fetched. Competing interests: My company provides consultancy services to pharmaceutical companies |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
John Stone, none London N22
Send response to journal:
|
We are assured under the section 'Methods' of Smeeth et al(p.963): "The practices are all broadly representative of practices in England and Wales, in terms of geographical distribution, practice size, and the age and sex of registered patients. The GPRD aims to include complete prescribing and diagnostic information for every registered patient. Data are anonymised. The quality of the information in the database has to be high in independent validation studies etc." But what if this is not so in this instance? The authors seem to admit (but do not highlight) that essential diagnostic data was unavailable (p.967) as quoted in my previous posting. It seems quite likely that if case registration was taking place at some fraction of actual incidence that detailed diagnostic information might also have been lacking in those cases that were reported. If the GPRD was not able to provide adequate data on autism it should surely never have been used as the basis of this and other studies. Contrary to Adam Jacobs I should say it calls the entire enterprise into question from its inception. Competing interests: As above |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
F. Edward Yazbak, Pediatrician, Director TL Autism Research, Falmouth, Massachusetts 02540, USA
Send response to journal:
|
1998: I. Andrew Wakefield publishes in the Lancet. The Public Health Officials: Only 12 cases II. Peltola and friends also publish in the Lancet; the study is
funded by Merck.
1999: Taylor and friends publish in the Lancet; the study is supported by
the MCA and PHLS.
2001: Kaye and friends publish in the BMJ. The study is not funded by
anyone but the authors are employed by the Boston Collaborative Drug
Surveillance Program which is supported by a dozen or more drug companies.
2002: Madsen and friends publish in the NEJM. The study is co-funded by the
CDC and criticizes the studies by Taylor, Kaye and others.
2004: Smeeth and friends publish in the Lancet … Meanwhile,
References: 1. Kielinen M, Linna SL, Moilanen I. : Autism in Northern Finland Eur Child Adolesc Psychiatry 2000 Sep; 9(3):162-7. 2. Goldman GS, Yazbak FE: An Investigation of the Association between MMR Vaccination and Autism in Denmark. JAmPhysSurg 2004; 9(3):70-75 http://www.jpands.org/vol9no3/goldman.pdf 3. Stott C, Blaxill M, Wakefield AJ: MMR and Autism in Perspective: The Denmark Story. JAmPhysSurg 2004; 9(3):89-91 http://www.jpands.org/vol9no3/stott.pdf 4. Bradstreet JJ, El-Dahr J, Anthony A, Kartzimel JJ, Wakefield AJ: Detection of Measles Virus Genomic RNA in Cerebrospinal Fluid of Children with Regressive Autism: A report of 3 cases. JAmPhysSurg 2004; 9(2):38-45 http://www.jpands.org/vol9no2/bradstreet.pdf 5. Yazbak FE:Autism in the United States: A Perspective. JAmPhysSurg 2003; 8(4):103-107 http://www.jpands.org/vol8no4/yazbak.pdf 6. Yazbak FE: Autism seems to be increasing worldwide, if not in London. BMJ 2004 Jan 24; 328(7433):226-7. http://bmj.bmjjournals.com/cgi/content/full/328/7433/226-c 7. Autism A.L.A.R.M., January 2004 http://www.medicalhomeinfo.org/screening/Autism%20downloads/AlarmFinal1.jpg Competing interests: Grandfather of a child with autism and evidence of measles virus genomic RNA in the ileal wall |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Hilary L Davidson, Staff Nurse Ipswich Hospital IP4 5SW
Send response to journal:
|
Further to the MMR debate,another new report on MMR and autism has recently been published (1) though ignored by the British Press. This study used 537,304 children in Denmark as opposed to the 5,763 children used in The Lancet study.The full report is available free on the Internet from reference 1 below. This study supports an autism-MMR association. Other questions which remain unresolved, are why the following occurred. A rise in autism a year after 1992 mass vaccination programme. A steep rise in autism throughout 1995-96, after a mass vaccination campaign in late 1994. A second round of MMR vaccinations was carried out in the autumn of 1996. Levels of autism rose dramatically in 1997. (2). It seems many of the questions of the relationship of MMR vaccination and autism remain unanswered. For those interested there is a conference, with speakers Dr Carol Shott of Cambridge University, and Paul Shattock of Sunderland University, at the University of London in Bloomsbury, London, on Saturday afternoon of October 30th. To find out more E-mail www.wddty.co.uk/shop/details.asp?products=405. (1) Journal of American Physicians and Surgeons Volume 9 Number 3 Fall 2004. (2) “WDDTY e-News <e-news©wddty.co.uk> (WDDTY=What Doctors Don't tell You) Competing interests: None declared |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Adam Jacobs, Director Dianthus Medical Limited, London SW19 3TZ
Send response to journal:
|
Readers of Hilary Davidson's response might like to note that the study she refers to did not, in fact, investigate an association between MMR vaccination and autism. It's an easy mistake to make, as the title of the paper is, rather misleadingly, 'An investigation of the association between MMR vaccination and autism in Denmark'. The study did not actually compare vaccinated and unvaccinated children. Instead, it looked at time trends in the population prevalence of autism. What the study found was that the prevalence of autism has increased in recent years: a finding which I don't think was ever in doubt, so I don't see how this study contributes anything new (which no doubt explains why it has been ignored by the British press). The paper then made a giant leap of logic by assuming that, since MMR vaccination has also been introduced in recent years, MMR must be the cause of this increase in autism. This is really not the same thing as investigating an association between the two. Competing interests: As stated previously |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
John Stone, none London N22
Send response to journal:
|
The Goldman and Yazbak study is primarily a critique of the much headlined study of Madsen et al, which found a negative relation between the introduction of MMR and autism, and the authors expose huge problems with Madsen's data assessment. Goldman and Yazbak did not claim to have definitively proved a causal connection between MMR and autism but they do claim to have overturned the Madsen study for which absurd claims were made by governments and health officials in the opposite direction. And I am absolutely certain that they would agree with me that correct way to resolve this issue is to investigate the children not the data. Competing interests: As above |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
L. Travis Haws, Dentist Lakewood CO 80228
Send response to journal:
|
Editor: I find the double standards some in the medical community enjoy quite astonishing. We are immediately told of the causal relationship when mass vaccination programs preceed so-called decreases in "preventative" disease epidemics. How then does this reasoning not hold true of mass vaccination programs immediately preceeding spikes in autism rates. This is then confirmed via isolation of vaccine strain compatible measles virus genomic RNA the gut wall and CSF. What more does one need to confirm such an obvious correlationship? Or how does such reasoning not apply to clustering of vaccine administration and SIDS, SBS, etc. etc. The case of Sally Clark's son so amply demonstrates the lack of medical competence as vaccines (that were administered four hours prior--what about delayed cell mediated reactions that so many "experts" ignore or have forgotten) were not even considered a factor! Few things give me such a visceral nauseating psychosomatic reaction as such. Some want to have their cake and eat it too. To their dismay, the cake is likely to end up on their face rather than in their G.I. tract. Competing interests: None declared |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Adam Jacobs, Director Dianthus Medical Limited, London SW19 3TZ
Send response to journal:
|
Travis Haws seems to be under the impression that the only evidence that vaccines work is of an ecological nature. Now, the ecological evidence in itself is pretty convincing (does anybody seriously believe that smallpox died out purely from natural causes?), but there is in fact much more than just ecological evidence for the efficacy of vaccines in preventing infectious diseases. I spent 5 minutes on Medline to identify some references showing that measles vaccines prevent measles, based on comparisons of vaccinated and unvaccinated children [1-4]. No doubt Haws could find similar evidence for other diseases if he wishes to look in the same place. In answer to Haws' question 'What more does one need to confirm such an obvious correlationship?' I would say that what is most needed is any scientific evidence whatever that autism is more common in vaccinated than in unvaccinated children. I have yet to see such evidence. References: 1. Aaby et al. A comparison of vaccine efficacy and mortality during routine use of high-titre Edmonston-Zagreb and Schwarz standard measles vaccines in rural Senegal. Trans R Soc Trop Med Hyg. 1996 May- Jun;90(3):326-30 2. Ramsay et al. Measles vaccine: a 27-year follow-up. Epidemiol Infect. 1994 Apr;112(2):409-12. 3. Samb et al. Protective efficacy of high-titre measles vaccines administered from the age of five months: a community study in rural Senegal. Trans R Soc Trop Med Hyg. 1993 Nov-Dec;87(6):697-701 4. King et al. Clinical efficacy of measles vaccine during the 1990 measles epidemic. Pediatr Infect Dis J. 1991 Dec;10(12):883-8 Competing interests: As stated previously |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
L. Travis Haws, Dentist Lakewood CO 80228
Send response to journal:
|
Editor: If Adam Jacobs had read any of my previous responses, he'd realize that I have quite extensively studied all the research and have pointed out numerous examples where epidemics occur in very highly "immunized" populations. And that measles deaths had decreased by 98% prior to vaccines. I do thank him for his 5 minutes and four references, however. Now there is a report of resurging Haemophilus Influenzae type b in many fully immunized children. (1) As stated in the reference “Since 1998 the number of Hib cases in children has almost doubled each year, most of them in children who have been fully immunised according to the United Kingdom's primary vaccination schedule for infants aged 2, 3, and 4 months.” Of course the authors find some spin to state we need better surveillance of immunization programs etc. I have never said vaccines “work”. I was reiterating what vaccine proponents constantly tell us re. vaccine efficacy and temporal relationships (which really doesn’t hold water) and then ignore such reasoning when the temporal relationships consistently point to serious adverse events. And then continue to ignore the relationship when vaccine strain measles virus are isolated in these children’s gut and CSF. I must admit that I’m just a little lost on his ecological reasoning, though. Adam Jacobs appeals to us regarding Smallpox. That had to be one of the worst examples to
use, as Smallpox vaccines clearly wrought much more havoc than aid if there
even was any aid (see responses by John Heptonstall). Other than what John Heptonstall
has pointedly shown in the Phillipines re. Smallpox vaccine use and subsequent maiming and death of
thousands, death from small pox in Official Statistic from
How does Adam Jacobs explain the above relationship? Surely Smallpox would have resurged during such times of low vaccine uptake? How did bubonic plague go away without vaccines? Oops I’m sorry, it hasn’t fully gone away, nor apparently has Smallpox, as there have been case reports here and there (very, very rarely) of veterinarians catching the disease from some alley cat that’s been playing with those dastardly rats, for example. Oh, I almost forgot the weird temporal association with a
more recent launch of Smallpox vaccine programs to eradicate the disease and
AIDS epidemics, as reported in the London Times in 1987 and 88. For example, greatest concentrations of AIDS cases
coincided with regions such as, Adam Jacobs states “I would say that what is most needed is any scientific evidence whatever that autism is more common in vaccinated than in unvaccinated children. I have yet to see such evidence.” How could you not see such striking evidence as a very high percentage of infants are immunized thanks to mandating? That is undisputable and autism is thus, far more common in vaccinated individuals. Never mind that, I’ll play along. I would challenge Adam Jacobs, or anyone, to produce some cases of autism in an unvaccinated child. Let’s leave the estimate extremely conservative at, let’s say, 1,000 cases of autism in an unvaccinated child. MMR isn’t the only culprit, as Mark Struthers has pointed out. It just seems to be the trigger man. There are many other vaccines that could very well be involved in loading the gun. People talk of environmental changes/factors etc. I know of no other more profound and widespread environmental change/factor in an infants early, and largely isolated, life than the introduction of mass immunization programs and the constant addition of various vaccines to the vaccine schedule. Hep B (for those IV drug abusing and sexually active babies), Pnuemococcal (“protects” against 7 of some 90 strains), Varicella (would you prefer chicken pox of shingles?) being a few of the most recent. Now they are recommending flu (which strain do you recommend they conjecture to target). I also support a call for a public enquiry that has been asked of for quite some time, by many. That also goes for things such as SBS, MSbP, “think dirty” of SIDS etc. These kids need looked at and are (as evidence of vaccine strain measles virus in the gut and CNS), but such results are glossed over by those claiming to care. 1) BMJ 2004;329:655-658 (18 September),
doi:10.1136/bmj.329.7467.655 2) Allen, Hannah. Don’t
Get Stuck: The Case Against Vaccinations. Oldsmar 3) Wright P. Smallpox vaccine ‘triggered AIDS virus’ London
Times. 4) Franks G. AIDS and Vaccinations: The Competing interests: None declared |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Adam Jacobs, Director Dianthus Medical Limited, London SW19 3TZ
Send response to journal:
|
Travis Haws seems to have missed the point. He still seems to think that vaccine proponents base their claims of vaccine efficacy solely on temporal relationships. As I said in my previous response, this just isn't true. There is also ample evidence from direct comparisons between vaccinated and unvaccinated individuals that vaccines prevent infectious diseases. He also claims that autism is more common in vaccinated individuals than in unvaccinated individuals, but does not provide evidence to back this up. As far as I am aware, there has not been a single study showing more autism in vaccinated than in unvaccinated individuals. If Haws knows of one, perhaps he could supply the reference. Competing interests: As stated previously |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
John P Heptonstall, Director of The Morley Acupuncture Clinic and Complementary Therapy Centre LS27 8EG
Send response to journal:
|
Sir Susan Mayor recounts Smeeth L. et al (1), following hard on the heels of Smeeth L. et al (2) which incorporates into (1). John Stone queries the validity of the GPRD as an epidemiological tool. Tyrer et al state (3) "In reality the aims of GPs and epidemiologists differ and the use of GP computer systems is not always compatible with the information needs of epidemiologists" - the former tend to focus on information storage which assists prescribing whilst tending to ignore personal and demographic data. The GPRD has a chequered history since its inception in 1987 when VAMP had GPs volunteer information from their practice systems that were found to be sporting some 119 different types of generally incompatible software (GP Computing Survey 1991 DoH) for inclusion into the VAMP database -later known as GPRD. Problems emerged of coding causing lost, inaccurate and missing data (Tyrer 1996), improper or misrepresented diagnoses (Tyrer 1996), mapping (Hawkings 1996), disappearance and duplication of records through re-registration procedures for patients (Tyrer 1996), data from commonly used ViSion software was excluded from analysis, EMIS had allowed GPs to create their own codes and so on, plus it was thought that the range of general practices represented on GPRD may not be representative of the population as a whole as there was an emphasis on larger practices (fundholders with 4 and more partners) possibly unrepresentative of smaller practices and the general patient population - a shambolic situation that is being gradually rectified. Although the potential inaccuracies exist with the GPRD, extracting individual cases and cross-checking medical records reduces the impact of errors in database recording - that is for the case subjects in particular for whom great attention to detail is evident from the research team. But what of the larger control group, what measures were taken to confirm the diagnoses of 'no-PDDs' amongst this group of almost 4,500 persons? It is not clear from the publication. Knowing the high percentage risk the GPRD poses to accuracy of diagnosis in its records, I would ask the authors how they can guarantee that none of the control group suffer/ed PDDs or ASDs during the period covered by their study - did they sample the controls to any extent (as with case subjects) or merely represent them as they were represented on the GPRD? The team expects us to accept the re-validation of autism and PDD diagnoses applied to the case subjects by EF and LA, an experienced psychiatrist and psychologist, who disagreed with about 10% of the original diagnoses recorded in both medical notes and GPRD. Yet it is possible their professional opinions for revalidation diagnoses might not always concur with those of other professionals say neurologists or speech and language therapists (SALT), or even others from their own professions (4) the outcome from diagnostic assessments can be chaotic (Rutter 1978a)(Kanner 1965)(Wing 1976)- there are areas of overlap in diagnosis which could easily have altered the final figures for amount of autism, aspergers and PDD/PDDNOS had a more varied group of professionals re- validated diagnoses. Accurate diagnosis probably requires more than courageous efforts from Fombonne and Heavey (5)-(10) and can the presence of PDDs be ruled out amongst controls? Opinion generates bias and revalidations were the opinions of a psychiatrist and a psychologist; are they more likely to arrive at similar diagnoses - ASDs - than might say a neurologist or SALT? There must also be difficulty diagnosing ASDs where clinical pictures change with age - the range within this study being from 0 to about 30 years old; and further complication when records of immigrants (GPRD has no record of ethnicity or place of birth) such that original diagnoses may remain aloof of new registration data, further complicated if re-registration occurs due to practitioner retirement leading to whole swathes of registrants being misrepresented on the GPRD. Tyrer et al found that reprocessing for re-registrations increased patient numbers by 9.5% in 1994, 6.8% in 1995, 5.3% in 1996 and 3.4% in 1997 and 1998 - and it was not until Autumn 1999 that the GPRD structure was altered to capture 'date of move to practice' replacing the single field available for almost 12 years for date of registration. The use of the GPRD and validation of diagnoses present a great challenge, not least when one wishes to present a case-control study with a very large control group. Any group of 4,500 children/adolescents might have about 2-400 with ADHD (as many as 50% ADHD sufferers are never diagnosed), 15-20% could have an undiagnosed specific learning disability, 450 might suffer conduct disorder, 20 epilepsy any of whom might therefore hide disorders that if diagnosed could alter the outcome for Smeeth et al studies where case subjects and controls are considered. Perhaps the GPRD does not provide the kind of detail or accuracy required for a study of the kind described, over the long period (1973- 2001 by age, 1987 to 2001 by record) into complex conditions like ASDs, Aspergers and PDDs especialy if the picture is further complicated with attempts to make sense of the effects of complex prescribing of vaccinations against particular diagnoses. Could the team, or Susan Mayor, reassure us that problems highlighted above have been either confounded out or do not exist? A relatively small swing in the numbers of controls found to carry undiagnosed PDDs or ASDs could completely reverse the published outcome. Then one would have to consider the impact these studies had on public finance and opinion, and what if any damage might have been done? Regards John H. References 1. "MMR vaccination and PDDs: a control study" Smeeth L. et al, Lancet 2004 Sep 11; 364 (9438): 963-9 2. "Validation of the diagnosis of autism in general practitioner records", Fombonne, Smeeth L. et al, BMC Public Health, 2004; 4 (1): 5 3. "Building a research database from computerised general practice records" Tyrer F et al, Journal of Informatics in Primary Care 1996 Sep; 8-13 4. "Autism, Asperger's syndrome and semantic-pragmatic disorder: Where are the boundaries?" D.V.M Bishop, Brit. Jnl of Disorders of Communication 24, 107-121 (1989), http://www.mugsy.org/bishop.htm 5. Exploring the borderlands of autistic disorder and specific language impairment: a study using standardised diagnostic instruments" BishopDV, Norbury CF, J Child Psychol Psychitary, 2002 Oct; 43(7): 917-29 6. "Autistic symptoms in children with ADHD" Clark T et al, Eur Child Adol Psychiatry. 1999 mar; 8(1):50-5 7. "Does DSM-IV Asperger's disorder exists?", Mayes SD et al, J Abnor Child Psychol 2001 Jun; 29(3): 263-71 8. "Social communication deficits in conduct disorder: a clinical community survey" Gilmour et al, J Child psychiatry 2004 Jul; 45(5): 967- 78 9. "ASDs in children with normal intellectual levels: associated impairments and subgroups", Sturm H et al, Dev Med Child Neurol 2004 Jul; 46(7): 444-7 10. "The comorbidity of PDD and ADHD: results of a retrospective chart review" Goldstein S, Schwebach AJ, J Autism Dev Disord 2004 Jun; 34(3): 329-39 Competing interests: None declared |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Graeme Johnston, Student MK7 6AA
Send response to journal:
|
In his "rapid response" of 25th September, Travis Haws wrote: Competing interests: None declared |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Adam Jacobs, Director Dianthus Medical Limited, London SW19 3TZ
Send response to journal:
|
John Heptonstall's criticism of Smeeth et al's study seems to be based on the supposition that the control group would have contained some children with PDDs or ASDs, as a result of the diagnosis having been missed in the GPRD. If that is true, it is unlikely to affect more than a tiny fraction of the control group, as such disorders are rare in the population, and a randomly chosen sample such as the control group would therefore also have a low prevalence of the disorders, even if many of them were unrecorded in the GPRD. This seems hard to reconcile with Heptonstall's assertion that 'A relatively small swing in the numbers of controls found to carry undiagnosed PDDs or ASDs could completely reverse the published outcome.' Perhaps Heptonstall could illustrate his claim with some numbers? How many undiagnosed PDDs or ASDs in the control group would there need to be to change the conclusions of the study? Would this be compatible with the population prevalence of the disorders? Does it require the assumption that the probability of recording of PDDs and ASDs in the GPRD depends on vaccination status? If so, what is the basis for that assumption? Just as an aside, anyone worried about the validity of GPRD data might wish to look at some studies that have specifically examined the accuracy and completeness of the database [1-4]. References: 1. Lewis JD, Brensinger C, Bilker WB, Strom BL. Validity and completeness of the General Practice Research Database for studies of inflammatory bowel disease. Pharmacoepidemiol Drug Saf 2002;11:211–218 2. Kaye JA, Derby LE, del Mar Melero-Montes M, Quinn M, Jick H. The incidence of breast cancer in the General Practice Research Database compared with national cancer registration data. Br J Cancer 2000;83:1556–1558 3. Garcia Rodriguez LA, Perez Gutthann S. Use of the UK General Practice Research Database for pharmacoepidemiology. Br J Clin Pharmacol 1998;45:419–425 4. Hollowell J. The General Practice Research Database: quality of morbidity data. Popul Trends 1997;87:36–40 Competing interests: As stated previously |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
John P. Heptonstall, Director of the Morley Acupuncture Clinic and Complementary Therapy Centre Leeds LS27 8EG
Send response to journal:
|
Sir Adam Jacobs misrepresents my response: - I have not criticised Smeeth et al as yet, I have questioned, in relation to evidence given, if the control group was afforded stringent analysis for the presence of ASDs and PDDs as the case subjects appear to have been. Nor were my questions based on “the supposition that the control group would have contained some children with PDDs or ASDs, as a result of the diagnosis having been missed in the GPRD”, they were the basis for ascertaining if the control group had been treated differently from the case subject group – whether or not that group contained children with unidentified PDDs or ASDs. Adam Jacob’s confusion is surprising in light of his companies role as a provider of ‘high quality medical writing and statistical services to the pharmaceutical industry and others in the biomedical field’ (1); as is his apparent lack of concern that a control group might have been subjected to a different regime than a case study group. I would ask him what he means by ‘rare’ and ‘low prevalence’ which he says make it hard to “reconcile my assertion that a relatively small swing in the numbers of controls found to carry undiagnosed PDDs and ASDs could completely reverse the published outcome”. The difference of 78% case subjects to 82% controls is not huge and may move at least towards parity after stringent analysis of GPRD records, raters’ opinions, control subjects health status, etc., the outcome might require an altered announcement from ‘MMR is unrelated to autism’ to ‘More research is needed’ at least. I could argue that applying any figures, as Adam requests of me, is pointless. There are major problems inherent in diagnosing ASD/PDD and other disorders which can ‘mimic’ the former – not least SPD, CDD, ADHD, Conduct Disorder, severe anxiety disorders, and various forms of communication and social disorders that might present with an apparent lack of social and communication skills and thence result in an ASD or PDD label and vice versa. I already gave some examples of statistical incidences amongst a general population of children from which inferences may be drawn about a control group of 4,469 children (bearing in mind the ‘children’ taken from the GPRD were born from 1973) – especially if that group was not subjected to formal analysis. For example, '1 in 10 may have anxiety disorders' (447), 'mental health problems affect 1 in 5 young people at any time' (893), 'an estimated 66% of young people with mental health problems are not getting the help they need' (589), 'ADHD is the most common psychiatric condition affecting children', 'prevalence being 5 -10%', 223-446; 'as many as 50% with ADHD are never diagnosed', 111-223; 'as many as 1 in 10 children and adolescents may have conduct disorder', 447; 'anywhere from 15-20% of children with ADHD have a “specific learning disability”', 33-45, or 66-90; and 'perhaps 50% with learning disability have ADHD'; 'schizophrenia occurs in 3 per 1000 adolescents' -13. (2) I appreciate these are figures from across the Atlantic but the UK often follows their trends. Just how rare is Adam's rare? I refer to these disorders as they carry characteristics that arguably could lead to misrepresentation as PDD. For example PDD and ADHD can be interchangeably misdiagnosed, as can SPD and ASD/PDD. How many controls labelled ADHD were passed over by a PDD check, how many PDDs were actually ADHD? How many young ADHDs are now, in adolescence or adulthood, re-labelled PDD or ASD since re-registration or second/third opinion? One can only wonder at how many of these statistical probabilities adorn the control group and, without a sample, I am sure Adam Jacobs is well aware one can only guestimate – not what stringent scientific scrutiny requires, hence my questions for Smeeth et al. Finally, none of Adam Jacobs’ four references are likely to inspire confidence in GPRD ‘accuracy and completeness’ for ASD/PDD recording; they are for “diseases with clear diagnostic criteria so are generally better recorded, as were data on specific procedures” (3). I would further refer him to Thiru et al (3) whose 'criticisms' of electronic patient data in primary care includes “data validity was expressed under a range of terms (completeness, correctness, accuracy, consistency, and appropriateness) which were rarely defined”. Regards John H. References 1. http://www.dianthus.co.uk 2. http://www.nmha.org/children/prevent/stats.cfm 3. “Systematic review of scope and quality of electronic patient record data in primary care”, Thiru et al, BMJ 2003 May 17; 326: 1070 Competing interests: None declared |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Carol Johnston, Carer Carshalton, Surrey
Send response to journal:
|
Sir Reading through this thread and spectulations as to the data used by Smeeth et al, it seems that unless access is given to raw data we do not know just how or why subjects were chosen for inclusion in the study. Were they completely random? There seems to be confusion over GPRD accuracy especially amongst the control group. I agree with John Heptonstall that indeed I believe the net should be cast wider than autism and PDDs. Incidents of ADHD, ADD, OCD, learning difficulties Speech and Language, dyslexia, childhood diabetes, asthma, IBDs, behavourial problems, would have probably been significant in the control group. Onset of these conditions have been linked by parents to vaccination. There are many children who have no formal diagnosis but show signs of communication, learning and behavioural symptoms. This can be for a number of reasons, ie parental reluctance to "label" a child - professional reluctance to statement a child because of costs incurred to LEAs. Indeed from my own experience its rare these days to come across a perfectly healthy "normal" child. With 1 in 5 children having a form of allergy/food intollerance it seems that a "normal" child is the exception rather than the rule these days. The authors of this study actually did fantastically well to find a group of 4,500 individuals without any disorders whatsoever! From my own experience my daughter was originally diagnosed with behavioural problems - Oppositional Definance Disorder and Elective Mutism, this diagnosis ignored her learning delay and her bowel problems. She would not have been represented on medical record as having High Functioning Autism or PDD. Many indviduals fight for years to get diagnosis, many dont get diagnosis until adulthood so theoretically for the control group it would be interesting to see if any were awaiting diagnosis. Adam Jacobs presumes that if a number were missed it would be a small number of the control group. In my opinion it would in all probability be a much larger number. I know of many children who show signs of PDDs and Autism yet because they are high functioning they are dismissed as "just a little different" but not given diagnosis, for reasons stated above or parental influence or emotional disturbances are blamed. If the study had looked at auto-immunity and IBDs the picture would have been much different. These studies are self-limiting. By casting a net to look at specific diagnosis, this will not give us a true reflection of the picture. The study should have tested and examined all subjects and submitted these subjects to tests to determine possible undiagnosed ASDs and PDDs amongst control group. These tests do exist. A study which looks at subjects, tests subjects by recognised diagnosis techniques would I feel have given us a very different picture. These "arms length" studies leave unanswered too many questions. A study which examines and looks at subjects directly would give us a clearer picture. Also amongst the group of diagnosed ASDs and PDDs, were IBDs, late onset type of autism, classic autism noted? This study tells me that my children would have regressed without or without MMR. There I would disagree from my own observations in my own children and others this study leaves unanswered the question "Just why do some children regress post MMR vaccination?". This study in fact does not deny that children do regress but says MMR is not a factor. Also the whole vaccination history of an individual needs to be taken into account. Children are given MMR as part of vaccination programme. Sensitisation starts from the earliest courses of vaccines. Also is it recorded when symptoms became apparent or just when or if diagnosis has been given. When will the study be carried out that looks at children who have regressed and some answers given. Just why do the MRC ignore our children and concentrate on the wider picture of the majority seem fine with MMR. The idea of an undiagnosed amount of individuals amongst the control group is not beyond the realms of possibilities and until we have access to raw data, selection criteria then all we can do is summarise. It is a fact that autism and PDDs are under diagnosed. I am afraid this study like many others before it has led to more questions rather than given reassurance because unless all cards (raw data) are laid on the table for independent scrutiny and the subjects all tested with recognised diagnostic tests, to see any have missed diagnosis then we can just guess as to what really is going on. If another team were to examine the data in this study would they indeed come up with different results, not unike the recent Danish study. If every child or 50 per cent. of children regressed immediately after MMR then the MMR would have been withdrawn. If the medical profession recognises that the individual's genetic makeup has a bearing on the outcome of any drug then surely vaccination would affect individuals in the same way. Reactions to vaccination happens, it is a fact,and in some cases it is fatal. What is so far unproven or is being denied is that PDDs, developmental and behavioural problems associated with autism can result. If epilepsy, nerve deafness, encaphalitis can result then so can neurological and behavourial and learning disorders. Parents like myself need to be listened to. We saw our children regress developmentally following vaccination. Ihave video tape, photographs, child development records which show a regression. Why is this important "anecdotal" evidence continuously ignored in studies such as this? Also to Adam Jacobs I would say that autism is more common amongst vaccinated individuals, simply because the majority of individuals have received vaccinations. I agree with Adam Jacobs that what is needed is a study of unvaccinated populations against vaccinated populations. I for one would welcome such a study. Competing interests: 2 ASD children post MMR vaccination |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Adam Jacobs, Director Dianthus Medical Limited, London SW19 3TZ
Send response to journal:
|
I am pleased to hear that John Heptonstall has not criticised the Smeeth et al study, and offer my apologies for the misrepresentation. As for Heptonstall's question about what I mean by 'rare' and 'low prevalence', I hesitate to supply exact figures, but I dare say I am on safe ground if I say 'less than 1%'. I am sure that it would make precious little difference to the results if 1% of the control group actually had ASDs. If Heptonstall can demonstrate otherwise, I would still be interested to see if he can suggest some numbers, as I asked in my previous post. Competing interests: As stated previously |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
John P Heptonstall, Director of The Morley Acupuncture Clinic and Complementary Therapy Centre LS27 8EG
Send response to journal:
|
Sir I appreciate Adam Jacob's request for figures from me but note that a considerable proportion of his responses to various responders have minimal fact and maximal requests for figures - yet coming from a stats background his use of definitive figures to contest responders views, not least mine, are conspicious by their absence. I would like to know how he finds himself on 'safe ground' with '1%'; and how he finds it possible to limit the likely rate of disorders amongst the control group to within 1%? Does he have access to study data, unpublished data, or is he merely 'supposing' out loud? I welcome his demonstration of the 1% hypothesis at this stage. Regards John H. Competing interests: None declared |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Adam Jacobs, Director Dianthus Medical Limited, London SW19 3TZ
Send response to journal:
|
Sorry, it was lazy of me not to give references for my assertion that the prevalence of ASD is less than 1%. I am happy to do so now [1-5]. Bear in mind also, that unless the completeness of recording of ASDs in the GPRD is truly shocking, most children with ASDs would have a diagnosis recorded in the GPRD, and thus be ineligible for Smeeth et al's control group. The prevalence of unrecorded ASDs must therefore be even lower than the prevalence of all ASDs. Now, how about seeing some numbers to back up John Heptonstall's argument about how undiagnosed ASDs in the control group could invalidate Smeeth et al's study? References: 1. Fombonne E. Epidemiological Surveys of Autism and Other Pervasive Developmental Disorders: An Update. Journal of Autism and Developmental Disorders 2003;33(4):365-382 2. Yeargin-Allsopp M, Rice C, Karapurkar T, Doernberg N, Boyle C, Murphy C. Prevalence of autism in a US metropolitan area. JAMA 2003;289(1):49-55 3. Merrick J, Kandel I, Morad M. Trends in autism. Int J Adolesc Med Health. 2004;16(1):75-8 4. Keen D, Ward S. Autistic spectrum disorder: a child population profile. Autism. 2004;8(1):39-48 5. Dalton R, Forman MA, Boris NW. Pervasive Developmental Disorders and Childhood Psychosis. In: Behrman RE, Kliegman R, Jenson HB. Nelson textbook of pediatrics. 17th ed. Philadelphia: Elsevier Science; 2004. p 93-95 Competing interests: As stated previously |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Raymond Gallup, Founder of The Autism Autoimmunity Project 45 Iroquois Avenue, Lake Hiawatha, NJ 07034, USA
Send response to journal:
|
Autism figures from IDEA, Part B
50 states, DC and PR
1992/1993 1996/1997
Total Total % Increase
12,222 34,354 181%
1992/1993 1997/1998
Total Total % Increase
12,222 42,487 248%
1992/1993 1998/1999
Total % Increase
12,222 53,561 339%
1992/1993 1999/2000
Total Total % Increase
12,222 65,396 435%
1992/1993 2000/2001
Total Total % Increase
12,222 78,717 544%
1992/1993 2001/2002
Total Total % Increase
12,222 97,847 701%
1992/1993 2002/2003
Total Total % Increase
12,222 118,602 870%
and the latest figures
1992/1993 2003/2004
Total Total % Increase
12,222 140,920 1,054%
See................
http://www.ideadata.org/tables27th/ar_aa3.htm
The IOM has again denied a vaccine link between autism and the epidemic continues. Instead, the IOM, NIH, and CDC want to pursue a "genetic epidemic". If they really knew anything about science they would realize there never has been and there never will be a "genetic epidemic". This from the people at the IOM, CDC and NIH, who "know science." Meanwhile, on a daily basis more and more children are regressing into autism while the IOM, NIH and CDC want to continue to be whistling pass the graveyard. Very sad and shameful. If these government agencies were in charge of child care facilities, then child abuse would be rampant and epidemic. We will not go away but will continue to remind people year-in and year-out about this epidemic. Raymond Gallup, Founder The Autism Autoimmunity Project http://www.taap.info/ Competing interests: Founder of The Autism Autoimmunity Project and father to Eric Gallup, who was born normal and regressed into autism after receiving the MMR vaccine |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
John Stone, none London N22
Send response to journal:
|
What does Adam Jacobs mean: "unless the completeness of recording of ASDs in the GPRD is truly shocking"? It is "truly shocking" as I pointed out in the first response under this topic and again as short a time ago as yesterday under another topic. I posted first on the matter on 15 May. In all this time no one has come forward to set the record straight, or put me right. I even remember challenging Adam Jacobs back in June to explain and defend it, and met with silence. http://bmj.bmjjournals.com/cgi/eletters/329/7467/642#74567 http://bmj.bmjjournals.com/cgi/eletters/328/7442/773#59578 http://bmj.bmjjournals.com/cgi/eletters/329/7466/588-b#76273 http;//bmj.bmjjournals.com/cgi/eletters/328/7442/773#62031 I thiink he treats this all as a joke. I wish I could. Competing interests: As above |
||||||||||||||||||||||||
|
|
||||||||||||||||||||||||
|
Adam Jacobs, Director Dianthus Medical Limited, London SW19 3TZ
Send response to journal:
|
I am not an expert on diagnosing autism, so, as I pointed out in my response of 20 September, I really don't feel qualified to comment on the completeness or otherwise of diagnoses of autism in the GPRD. John Stone will have to look elsewhere for answers on that one. I do, however, know a little bit about statistics, and as I pointed out in the same response, under-reporting of autism in the GPRD would make precious little difference to the interpretation of the results of Smeeth et al's study, other than in the bizarrely unlikely event that whether a diagnosis was recorded in the GPRD was somehow related to whether the child had been vaccinated. I can assure John Stone that I don't think autism is a joke. I must confess, however, that I do feel that some of the arguments used by anti- vaccinationists, such as the notion that smallpox died out from natural causes, tend more towards the category of comedy than of science. Competing interests: As stated p | ||||||||||||||||||||||||