Headline:

Sometimes called the "Swedish Study." A summary paper that used the same methodologically flawed approach used in the original Danish study published in Pediatrics, which we discuss as study #2. Shockingly, Sweden's autism rate is noted to be 1 in 10,000, which is 60 times lower than the U.S. rate, but the authors don't address this discrepancy.

Actual Question This Study Asked & Answered:

Q: Did the discontinuation of thimerosal use in vaccines in Denmark, Sweden, and the U.S. lead to a decrease in autism?

A: No.

Did the study look at unvaccinated children?

No.

Conflict of Interest (from the study itself):

"Financial support for the compilation of the data used in this investigation and the preparation of this report was provided by the National Immunization Program, Centers for Disease Control and Prevention. We are grateful to Victoria Romanus of the Swedish Institute for Infectious Disease Control, Ingrid Trolin of the Swedish Medical Products Agency, Anne-Marie Plesner and Peter Andersen of the Danish Statens Serum Institut Institut [Denmark’s largest vaccine company], and Roger Bernier and Susan Chu of the Centers for Disease Control and Prevention for their contributions in the design and conduct of this investigation, and in the preparation and review of this manuscript."

Ability to Generalize:

None. As SafeMinds noted:

"Stehr-Green et al. 1 have misrepresented my work and confused the debate over autism and mercury exposure in the United States with ecological data from Sweden and Denmark. Their uninformative article in AJPM should not delay efforts by independent researchers to carry out the investigations into mercury and neuro-developmental disorders recommended by the Institute of Medicine (IOM).2 Their report has many flaws, but four stand out.."

Post-Publication Criticism:

Very high.

Choice Excerpt from the Study:

"Prior to 1992, the data in the national register did not include cases diagnosed in one large clinic in Copenhagen (which accounts for approximately 20% of cases occurring nationwide). Prior to 1995, the autism cases reported to the national register reflected only cases diagnosed in inpatient settings."

Meaning:

Highlights the "fatal flaw" of the Danish data set, that many autism cases simply were not counted until after 1995, when Thimerosal was removed from vaccines.

This is the equivalent of doing a study on "Divorce Rates in North America" and counting Mexico and Canada only for the first few years, then adding in the United States, and noting that divorce rates went up. To compound the problem, Denmark also changed the diagnostic code they used, to the more universal ICD10 code, beginning in 1993, which would have further raised the rates.

Guest Critic #1: SafeMinds, Mark Blaxill

FULL RESPONSE TO STEHR-GREEN PAPER IN AMERICAN JOURNAL OF PREVENTIVE MEDICINE

by Mark Blaxill, Safe Minds

Concerns continue over mercury and autism

To the editors,

Stehr-Green et al. 1 have misrepresented my work and confused the debate over autism and mercury exposure in the United States with ecological data from Sweden and Denmark. Their uninformative article in AJPM should not delay efforts by independent researchers to carry out the investigations into mercury and neuro-developmental disorders recommended by the Institute of Medicine (IOM).2 Their report has many flaws, but four stand out.

1. Their erroneous description of the California data promotes complacency regarding autism rates.

   For an IOM review, I presented an ecological analysis of autism rates and thimerosal exposure in vaccines that demonstrated an association between rising autism rates in California and thimerosal exposure in childhood vaccines. I neverattempted to publish these findings, since ecological analysis, as the IOM summary pointed out, is generally uninformative. I simply made the case for concern and more (and independent) research. In their re-use of my charts, the authors err in describing the rates of autism reported therein as rates for the larger class of "autism-like disorders." This is not true. California prevalence rates are reported based only on DSM IV autism cases.3,4 By suggesting these rates belong to a broader category, the authors’ description minimizes the severity of the situation in California. These high and rising autism rates point to an urgent public health crisis, one that requires accurate measurement and precise classification.

2. Their new autism trend analyses account for only a fraction of the autism population.

   The large majority of autism cases are found in outpatient populations. Yet, their autism analyses in Sweden (exclusively) and Denmark (for two-thirds of the study period) rely on inpatient populations for disease frequency calculations. One recent study5 in Denmark reported that over 93% of autistic children were outpatients. Clearly, this small minority of inpatient autism cases has little value for purposes of trend assessment. Investigating restricted populations for purposes of causal inference is a fool’s errand.

3. Their rate and exposure assessments contain multiple errors, inconsistencies and misrepresentations.

   For example, in addition to the inpatient population problem: they measure autism case counts, not rates in Denmark; they mischaracterize their autism cohorts in Sweden as birth year cohorts when they are actually moving average cohorts of 2-10 year olds; they report vaccine compliance levels in Denmark (over 90%) that are inconsistent with the mercury exposures in their display; they represent the Danish pertussis doses as standard when they are highly unusual (and therefore suspect); they acknowledge non-standard and changing diagnostic criteria and incomplete institutional coverage in their Danish case count, without attempting correction; and they report time at diagnosis, not year of birth, in their Danish autism population, a common mistake. Despite these flaws, they claim, inappropriately, that their choice of Swedish and Danish sources was based on "high quality records."

4. Their interpretation of the autism-mercury hypothesis is incorrect.

   Based on these flawed trend assumptions, the authors use the shift in Sweden and Denmark from comparatively low thimerosal exposures to thimerosal-free vaccines in an attempt to falsify the autism-mercury hypothesis. Absent a clear increase in autism rates in Denmark and Sweden, this attempt at falsification fails. The autism-mercury hypothesis I tested was that increases in mercury exposure are associated with increases in autism rates. Reductions in comparatively low thimerosal exposures need not produce decreasing autism rates in stable, low-prevalence populations for the autism-mercury hypothesis to hold. Having performed the ecological analysis with which the authors started, I fully recognize its failings. Most notably: I could not match the exposure population with the affected autism population (a revision I have since made with little impact on the results); I could not measure administration of thimerosal-containing vaccines, only their sales; and I did not correct for lagged ascertainment of autism cases. Compared to the problems in the Danish and Swedish analyses, these are minor issues.

   I do not, however, wish to stand in defense of ecological analysis. The authors’ awkward attempts at trend analysis stand as eloquent testimony to the potential misuse of ecological analysis, especially when based on shifting data sources and incomplete time series. Instead, resources should flow to more informative, original research. Credible evidence points to rapidly rising U.S. autism rates.4 Mercury exposure is temporally,1 epidemiologically,6 and clinically7 associable with U.S. autism cases and may help explain these increases. The IOM has called for an active research program that did not include additional ecological speculations. Independent scientists should heed their recommendations, remain concerned over the connection between autism and mercury and investigate the connection further with proper methods.

Mark F. Blaxill
Director
Safe Minds

References:

1. Stehr-Green P, Tull P, Stellfeld M, Mortenson PB, Simpson D. Autism and thimerosalcontaining vaccines: lack of consistent evidence for an association. Am J Prev Med. 2003;25(2):101-106.

2. Institute of Medicine. Immunization Safety Review: Thimerosal Containing Vaccines and Neurodevelopmental Disorders. Stratton K, Gable A, and McCormick M, eds. Washington, D.C.: National Academy Press; 2001

3. California Department of Developmental Services. Changes in the population of persons with autism and pervasive developmental disorders in California's Developmental Services System: 1987 through 1998. A Report to the Legislature, Department of Developmental Services, California Health and Human Services Agency, State of California, Sacramento. 1999.

4. California Department of Developmental Services. Autistic spectrum disorders: changes in the California caseload, an update: 1999 through 2002. Department of Developmental Services, California Health and Human Services Agency, State of California, Sacramento. 2003.

5. Madsen KM, Hviid A, Vestergaard M, Schendel D, Wohlfahrt J, Thorsen P, Olsen J, Melbye M. A population-based study of measles, mumps, and rubella vaccination and autism. N Engl J Med. 2002;347(19):1477-82.

6. Geier MR, Geier DA. Neurodevelopmental disorders after thimerosal-containing vaccines: a brief communication. Exp Biol Med. 2003;228(6):660-664

7. Holmes AS, Blaxill MF, Haley BE. Reduced levels of mercury in first baby haircuts of autistic children. Int J Toxic. 2003;111(4):277-285