You can see it for yourself here: http://www.sciencedirect.com/science/article/pii/S0264410X14006367
The problem I have with this is that people keep saying it is a “study”.
A meta-analysis is not a study, it is number crunching previous studies and papers.
In section 3.1 they explain the selections process as thus
“The search of Medline, PubMed, and Embase returned 519, 718, and 1133 results, respectively. After adjusting for duplicates, 1112 papers in total remained, 953 were excluded immediately on inspection of the abstracts as they clearly did not meet inclusion criteria, leaving 159 papers whose methods sections were analysed in more detail to determine suitability. No unpublished relevant studies were obtained. Five additional papers were found on examination of relevant reference lists. A further 113 were identified as having no possible case-control or cohort data and were excluded, leaving 46 papers to which the inclusion criteria were applied (Fig. 1). A total of five case-control studies and five cohort studies were identified for inclusion in the review.”
So out of a total of 1112 papers they immediately rejected 953 based on the extracts. Why? Because they “didn’t meet inclusion criteria”. That is a very technical way of saying it didn’t meet the outcome we wanted, also known as selection bias.
Furthermore they go on in section 3.3.1 that some of the studies were high risk for bias.
“Using the NOS, two studies were rated as having low risk of bias  and , two as moderate risk and , and one was rated as having a high risk of bias . Specific ratings for each study are included in Table 1. Bias encompassed in the assessment of the study by Uchiyama included selection bias due to recruitment of all participants from a private clinic, poor definition and inadequate description of assessment of regression, and a lack of controlling for comparability between the “MMR Generations” and “pre- and post-MMR Generations”. The study by Madsen also has the potential for bias as a result of investigating MMR vaccination status as opposed to a cumulative dosage of thimerosal or Hg. As the Hg or thimerosal dosage in vaccinations varies, there is a degree of fluctuation in the amount of exposure to the individuals within a population studied. In contrast, when using the binary system of vaccinated versus non-vaccinated in a population with such high immunisation coverage to investigate the risk of ASD, the unvaccinated group is at much higher risk of being non-representative of the larger population for many additional reasons thus creating bias. We have continued to include it in our meta-analysis despite risk of bias as it still provides valuable evidence for the question of the increased risk of autism or ASD in the vaccinated population compared to those unvaccinated, despite bias affecting the implications that can be drawn about the causal nature of the relationship. Follow-up periods for each of the cohort studies varied with time periods of 5 years (at least 3 years of data per individual) , 8 years (at least 2 years of data per individual) , 8 years , 11 years (at least 2 years of data per individual) , and individuals followed from 1 to 11 years . The mean length of follow-up of the five cohort studies is 8.6 years, with the range being 5 years to 11 years.”
As explained above they are also only looking primarily at one vaccine (mmr, and measles monovalent) and one ingredient (mercury). This is another way of circumventing the fact that one particular vaccine has it listed as a possible adverse event (Tripedia, pg. 11).
And most inserts and vaccine trials have brain damage listed as a possible adverse event from the vaccine. They hide it under the technical term encephalopathy.
In section 4 they go on to say;
“Publication bias was not found in the study, which may be due to the important public health nature of the question. While we thought it more important to include only studies that strictly adhered to a case-control or cohort study protocol and drew it's participants from reliable sources, we recognise that there does exist data from VAERS that reported positive results, however, due to the aforementioned reasons these papers were excluded. It could be considered that duplicate data may be influencing the results as two of the five cohort studies were performed at the population level in Denmark with a crossover of birth cohorts. While the two studies looked at different interventions (one MMR and the other thimerosal-containing vaccines) the outcome data was the same, so while being an interesting comparison to one another, may not provide completely individual results to contribute to this meta-analysis. However, a sensitivity analysis of these studies from Denmark did not change the overall result. An important strength of this meta-analysis is the length of follow-up of the cohort studies, with an average of 8.6 years.”
Lets reread that
“...we recognise that there does exist data from VAERS that reported positive results, however, due to the aforementioned reasons these papers were excluded. It could be considered that duplicate data may be influencing the results as two of the five cohort studies were performed at the population level in Denmark with a crossover of birth cohorts.”
That would be a direct admission of selection bias. Prior to that statement they said
“The only review to suggest that a link could not be excluded was that by Ratajczak looking into the aetiology of autism and concluded that it is multifactorial involving genetics and/or inflammation of the brain caused by a wide variety of environmental toxins, one of which may be mercury.”
There are indeed many factors of potential causation for ASD (autism spectrum disorder).
This is a fantastic investigative article written in the NY Times, Nov 10, 2002:
The Not-So-Crackpot Autism Theory
However there is a much broader picture than just autism. The effects of an adverse event following vaccination can range from mild to deadly, short term or long term.
Mercury in vaccines is only one of the points of concern. Remember that the Heb B and the flu vaccines still contain mercury. You can check the inserts for yourself here.
Below I will link to studies and documents for your own discretion.
A Positive Association found between Autism Prevalence and Childhood Vaccination uptake across the U.S. Population
Selected vaccine authorities from CDC, FDA, and manufacturers discuss, in a closed meeting, the possibility of neurodevelopment disorders resulting from vaccine components.
Neurotoxic effects of thimerosal at vaccines doses on the encephalon and development in 7 day-old hamsters.
Comparison of blood and brain mercury levels in infant monkeys exposed to methylmercury or vaccines containing thimerosal.
Influence of pediatric vaccines on amygdala growth and opioid ligand binding in rhesus macaque infants: A pilot study
Aluminum Toxicity in Infants and Children
Aluminum Vaccine Adjuvants: Are they Safe?
Aluminum hydroxide injections lead to motor deficits and motor neuron degeneration.
Aluminum in the central nervous system (CNS): toxicity in humans and animals, vaccine adjuvants, and autoimmunity.
Hepatitis B vaccine and the risk of CNS inflammatory demyelination in childhood.
Acquired autoimmunity after viral vaccination is caused by molecular mimicry and antigen complimentarity in the presence of an immunologic adjuvant and specific HLA patterns.
Review of Vaccine Induced Immune Overload and the Resulting Epidemics of Type 1 Diabetes and Metabolic Syndrome, Emphasis on Explaining the Recent Accelerations in the Risk of Prediabetes and other Immune Mediated Diseases