Thimerosal Reduction and Autism Trends — Schechter & Grether (2008)
- Apr 2
- 5 min read
Updated: Apr 17
Study summary and critical review
Robert Schechter & Judith Grether (2008)
Continuing Increases in Autism Reported to California’s Developmental Services System: Mercury in Retrograde
Research Classification
Publication type: Study
Research focus: Thimerosal exposure
Study design: Ecological time-trend analysis
Evidence discussed: Population-level autism trends
Data sources: California Department of Developmental Services records
Key question: Did autism rates change following the reduction of thimerosal in vaccines?
Journal: Archives of General Psychiatry (2008)
What was this study about?
Schechter and Grether examined whether autism rates changed in relation to the reduction and removal of thimerosal from vaccines in the United States.
Thimerosal, a mercury-containing preservative, had been reduced or removed from many childhood vaccines starting around 1999–2001.
The study looked at whether autism trends changed following this shift.
Why was this study important?
Earlier concerns suggested that mercury exposure from vaccines might be linked to autism.
If this were the case, a reduction in thimerosal exposure might be expected to be followed by a change in autism rates.
This study examined population data to see whether such a pattern could be observed.
How was the study conducted?
The researchers analysed data from the California Department of Developmental Services (DDS).
They examined autism case counts over time and compared these trends with the period when thimerosal was reduced or removed from vaccines.
The analysis focused on identifying whether autism trends changed following this reduction in exposure.
What did the study find?
The study reported that autism cases continued to increase during the period after thimerosal was reduced and removed.
The researchers did not observe a decrease or change in trend that aligned with the reduction in thimerosal exposure.
In simple terms
This study looked at autism trends in California and asked whether rates changed after thimerosal was reduced in vaccines.
The researchers found that autism cases continued to rise during this period and did not observe a change in trend that matched the reduction in thimerosal exposure.
Critic 1
Ecological Study Design & Interpretation
What is this critique about?
This critique focuses on the ecological study design used in Schechter & Grether (2008).
The study examined population-level trends in autism diagnoses alongside changes in thimerosal exposure over time.
Why does this matter?
Different study designs answer different types of questions.
Ecological studies are designed to examine:
• broad population trends
• whether patterns change over time
However, they do not analyse:
• individual-level exposure
• individual-level outcomes
What are the key considerations?
Two main factors are important when interpreting this type of study:
• multiple changes over time — during the study period, factors such as diagnostic criteria, awareness, and access to services were also changing and may influence autism rates
• lack of individual-level linkage — the study does not connect vaccination exposure to outcomes in specific individuals
What does this mean for interpretation?
Because multiple factors are changing at the same time, it can be difficult to:
• isolate the effect of a single variable
• determine cause-and-effect relationships
The study can identify patterns at a population level, but it cannot establish how exposure relates to outcomes in individuals.
Context: Interpreting “primary cause”
The study discusses whether thimerosal could be a primary cause of autism.
In this context, “primary cause” refers to a large population-level effect.
Ecological studies are suited to detecting such large-scale patterns, but are less able to assess:
• smaller effects
• variation between individuals or subgroups
In simple terms
This study looks at overall trends in autism diagnoses and compares them with changes in thimerosal exposure.
Because it uses population-level data and multiple factors are changing at the same time, it cannot determine how exposure relates to autism in individual children.
Critic 2
Data Source & Case Definition
What is this critique about?
This critique focuses on the data source and case definition used in Schechter & Grether (2008).
The study analysed autism cases recorded in the California Department of Developmental Services (DDS), a system that tracks individuals receiving services for developmental conditions.
Why does this matter?
The way data is collected can influence how trends are observed and interpreted.
DDS data reflects:
• individuals who have been diagnosed
• and who are receiving state-supported services
It does not represent all autism cases in the general population.
How were autism cases identified?
In this study, autism cases were identified through:
• administrative records within the DDS system
• service eligibility and classification criteria
These records are based on individuals entering and being recorded within the service system.
What are the key considerations?
Two main factors can influence how autism cases appear in this type of data:
• service-based data — the number of recorded cases depends on access to services, referral pathways, and eligibility criteria
• changes in diagnosis — over time, shifts in diagnostic practices, awareness, and identification of milder cases can increase recorded case numbers
What does this mean for analysis?
Increases in recorded autism cases may reflect:
• greater access to services
• broader diagnostic inclusion
• earlier or more frequent identification
rather than changes in underlying incidence alone.
Context: Case definition
Autism cases in DDS data are based on administrative classification within a service system.
These may differ from:
• research-based diagnostic assessments
• standardised clinical evaluations
This can affect how cases are grouped and compared over time.
In simple terms
This study used data from a service system that records people receiving support for autism.
Because access to services and diagnostic practices can change over time, increases in recorded cases may reflect how the system works, not only changes in how common autism is.
Critic 3
Broader Academic & Methodological Questions
What is this critique about?
Not all critiques appear as direct responses to a single study.
For ecological time-trend studies like Schechter & Grether (2008), many methodological discussions come from the wider epidemiological literature rather than formal rebuttals.
This section summarises those broader academic considerations.
Interpreting trends over time
Academic discussions note that population trends can be influenced by:
• changes in diagnostic criteria
• increased awareness of autism
• expansion of services and support systems
These factors can affect how autism is identified and recorded over time.
Incidence vs recorded cases
The literature highlights an important distinction between:
• true incidence — how common autism is in the population
• recorded cases — how many cases appear in administrative systems
Service-based data may reflect changes in how cases are recorded rather than changes in underlying incidence.
Timing of diagnosis
Autism is often diagnosed some time after symptoms first appear.
Academic discussions note that:
• diagnosis may occur years after early developmental changes
• recorded dates may not reflect when symptoms began
• delays can affect how trends are interpreted
Multiple influencing factors
Population-level studies involve many variables changing at once.
The literature notes that:
• diagnostic, social, and environmental factors can overlap
• changes in one area may coincide with changes in others
• it can be difficult to isolate the effect of a single factor
Interpreting consistency across studies
Consistency across multiple studies is often used to support conclusions.
At the same time, the literature notes that:
• similar study designs
• shared data sources
• common methodological limitations
can produce similar patterns across studies.
Takeaway
Broader academic discussions show that studies like Schechter & Grether (2008) provide insight into population trends, while being shaped by:
• how autism cases are recorded
• how timing is interpreted
• the presence of multiple changing factors
Understanding these considerations helps clarify what this type of study can reasonably address.
