Introduction
There has been a lot of controversy about the possibility of a positive relationship between vaccination, especially the measles-mumps-rubella (MMR) vaccine. The two points which are oft-quoted to support this possible association are that
- at the same time that infant immunization and vaccination coverage has increased, so have incidences of autism; and
- in some incidences of autism, there appears to be a temporal association in which traits which point towards autism start showing at any time between a few weeks to a few months post-vaccination.
Research however has shown that neither of these arguments is particularly compelling. While it is true that incidences of autism and similar disorders have increased in recent times, this could be an actual increase but could also be an increase as a result of increased awareness, recognition and changes in certain diagnostic criteria. Also, the appearance of autistic characteristics close to vaccination could also be a coincidence: the MMR vaccine is given when a child is two years old which is also when autism is generally first suspected or diagnosed (DeStefano & Chen, 2001).
In this paper, we attempt to critically discuss the literature available on the possible link between vaccines and autism, and come to a decisive conclusion regarding the causal relationship between the two. The MMR vaccine will be discussed primarily because it is the focal point in this controversy.
Vaccines (MMR) and Autism
As Taylor (2006) put it, the “main celebrity” in the MMR/Autism controversy is Andrew Wakefield. This is so because he wrote the paper titled “Ileal-lymphoid nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children” which was published in February 1998 in The Lancet. In this paper, he reported the cases of 12 children, who were autistic and had bowel symptoms, and in eight of these twelve children, behavioral symptoms had appeared after a mean interval of 6.3 days following MMR vaccination. Wakefield and his colleagues hypothesized that the bowel dysfunction (enterocolitis) and the consequent neurodevelopmental disorders had been caused by the MMR vaccine.
This paper grabbed media attention as well as sparked controversy and created such chaos that the UK saw a drastic decrease in MMR vaccination uptake: from 90% in all of UK in 1995 down to 70% in some parts in 2001. As a result, mumps, measles and congenital rubella returned and so did the possibility of these conditions turning into epidemics. The interest and attention invoked by this paper were largely due to the fact that it appeared to provide a biological means through which MMR vaccine resulted in autism.
There is negligible empirical data to provide evidence of a similar decline in Canada as witnessed in the UK to provide evidence of lower rates of MMR vaccination uptake as a result of this study. As Mary Appleton, the senior manager for the Canadian Coalition for Immunization Awareness and Promotion stated, “Right now, nobody is able to document or trace the true impact of the MMR study. Our registration system is simply insufficient.”According to the 2006 National Immunization Survey reported by the Public Health Agency of Canada (PHAC), approximately 91% of Canadian children had received the MMR vaccine, which was only a little lower compared with the 94% coverage that had been projected in 1997.
For Mahnaz Farhang Mehr, chief of immunization programs for PHAC, this means that overall, the coverage rates for the vaccine and immunization rates have not been impacted by this study in Canada and doctors have not been stopped from administering MMR vaccine by parents. What it has resulted in at maximum is a decrease in parent’s confidence in the vaccine (Vaccine–autism link discounted, but effect of “study” is unknown).
The case series published in The Lancet has since then been widely censured. Experts have criticized the biased sample (referrals to a tertiary pediatric gastroenterology unit), and the lack of controls and case validation. This paper as well as other papers published by this research group have been described as “poor science” by the Chief Medical Officer of the UK. Studies carried out in the U.S., U.K, Canada, and Europe could not confirm these reported associations and investigation has revealed further evidence that this study can not be trusted, such as the fact that most of the children part of this study were litigating against the manufacturers of MMR vaccine (Taylor, 2006; Meadows, 2004).
Research which took place to look into the suggestions of this paper, including that carried out by a special panel of the British Medical Research Council was not convinced about its accuracy and authenticity. The small number of cases referred to a gastroenterology facility were hardly representative of the general population. There was no confirmed laboratory evidence to prove the correlation between the bowel dysfunction and autism and there remained a definite possibility of there being a coincidental temporal association (which was not causal) with MMR vaccine, which was not addressed in this paper (DeStefano & Chen, 2001).
Studies conducted later by Wakefield and his colleagues did not support this hypothetical link between MMR vaccine and autism. A specific finding from later research was that for patients who have inflammatory bowel disease, which was claimed to be the biological mechanism for autism post MMR vaccination, specific laboratory analyses were negative for the incidence of measles virus.
An epidemiological follow-up study was also carried out by Wakefield’s colleagues of a 1970 British birth cohort and the results showed no link between measles disease or vaccine and the consequential occurrence of bowel dysfunction. Another study conducted as part of the Vaccine Safety Datalink project of the US Centers for Disease Control and Prevention (CDC) found no correlation between MMR vaccine and inflammatory bowel disease (DeStefano & Chen, 2001).
Since the publication of this paper in 1998, numerous epidemiological studies have been carried out but have not found any link between MMR vaccination and autism. The most thoroughly conducted study was one by Taylor et al (1999) where they identified children with autism (included classical autism, atypical autism and Asperger’s syndrome) born since 1979 from special needs/disability registers and special schools in eight North Thames health districts of London, UK.
First, the authors proved that the since 1979, the known cases of Autistic Spectrum Disorder (ASD) had been on the rise, and after MMR vaccination was introduced in 1988, there was no perceptible increase.
They also showed that vaccination did not lead to an earlier appearance of autistic traits, because the patients who had been vaccinated before they reached the age of 18 months old had similar ages at diagnosis compared to patients who had been vaccinated after they were 18 months old or not vaccinated at all. Another important finding of this study was that at the age of 2 years, the patients with ASD as well as children in the same birth cohorts of that district had a similar MMR vaccination coverage, hence there was an overall lack of any association between autism and MMR vaccine (DeStefano & Chen, 2001).
This study was highly regarded in scientific circles because of its thoroughness and scope. The authors had also used a ‘case series’ methodology by which they compared the occurrence of autism within various pre-specified time periods after vaccination. This assessment was done using three distinct measures of the onset of autism: date of diagnosis, recorded age at first parental concern and date of onset of regression, and two specific categories of vaccines: MMR and any general vaccine used to contain measles. This study did not find any significant associations between MMR vaccination and autism (Taylor et al., 1999).
There is no dearth of studies which were carried out to determine whether there was or wasn’t a link between MMR vaccine and autism. When a large database of British general medical practices was analyzed, it was found that between 1988 and 1999, the incidences of autism increased by 7 times but the coverage of MMR vaccination was constant at over 95% within this time span. An almost identical study was carried out in California, U.S., where there was an increase in the number of people diagnosed with autism and receiving help regarding it, but in that time period, the prevalence of MMR vaccination was fairly consistent.
Another published study was conducted in Sweden, where autism was compared in children born before with children born after the introduction of MMR vaccination. There was found to be no increasing trend in incidences of autism in both groups (DeStefano & Chen, 2001).
The epidemiological findings mentioned above are in line with what the current understanding of autism is, a “syndrome defined by certain behavioral and developmental characteristics that may have a variety of causes” (DeStefano & Chen, 2001).
There are however few cases in which a specific cause has been pointed out. Autism is said to have a “strong genetic component” and the neurological aspects which characterize this disorder probably appear in the very early stages of embryonic development. Hence, in a majority of cases, autism is something children are born with, even though it makes its appearance at a later stage when communication and behavioral problems become pronounced. Following this reasoning, it does not appear to be likely that techniques used for immunization, such as vaccines, which are given after birth could lead to autism.
There are a small number of cases where a child would seem to be developing like any other child, but would then regress and acquire or develop autistic behavioral traits. It was only in such cases, which were a minority, where a biologically sound link could be made between vaccination and autism, as had been suggested by Wakefield and his colleagues. Hence, the research conducted by Taylor et al (1999) was particularly significant because it refuted the possibility of any link between the onset of regression and vaccination by providing particularly compelling evidence against this hypothesis that MMR vaccination could lead to autism or intensify its symptomatology.
A study carried out by a Working Party on MMR Vaccines the British Committee on Safety of Medicines, which carried out a systematic review of parental and physician information (Medicines Commission Agency/Committee on Safety of Medicines). In this data, it was seen that some parents had reported that their children displayed autistic characteristics after they had an unfavorable reaction to vaccines. In Wakefield’s study, 6 of the 12 cases had reported such adverse reactions but according to DeStefano & Chen (2001), there are very rare cases in which reactions to vaccines are particularly severe or harmful. In the process of unearthing a possible link between vaccines and autism, the most important reaction would be one that has an impact on the brain, such as encephalopathy.
To this effect, the most critical study regarding encephalopathy has been conducted in Britain where all children were identified in the country who had been treated for encephalopathy or severe convulsions between 1976 and 1979. This was a comprehensive effort, a study of 940 cases, and its results did not show an increased risk of serious encephalopathy or any sustained neurological effects caused by measles vaccine. Hence, since measles vaccine is not linked with acute encephalopathy or lasting neurological problems, an adverse reaction to measles vaccination could not cause autism (DeStefano & Chen, 2001).
Aside from MMR vaccine, which has been the most prominent in the debate surrounding immunization and autism, there has been some talk regarding other vaccinations as well. Pertussis vaccination, thimerosal in vaccines, the administration of multiple vaccine antigens, or peripartum rubella vaccination of a mother have been said to increase the possibility of autism, but no scientific evidence has been found to this effect (Baker, 2008).
In 2003, A Vancouver law firm filed the first 2 lawsuits in British Columbia postulating an association between the vaccine preservative thimerosal and autism. It joined hands with lawyers in the U.S. who were out to claim the same. These class-action suits were filed based on allegations that “thimerosal which contains 50% ethylmercury, causes neurological damage in a subset of children” and were filed on behalf of children born either on or after 1980 who were administered vaccines containing thimerosal before they were 2 years old.
According to Dr. John Blatherwick, chief medical officer for the Vancouver Coastal Health Authority, thimerosal “has been a very good preservative over the years” with very small and safe mercury levels and was present in Canadian childhood vaccines from the 1970s until about the late 1990s, and today is used primarily just in influenza vaccines. Blatherwick did not think these lawsuits had any chance of being successful simply because a causal relationship with autism had not been proved (BC lawsuits try to link thimerosal to autism).
Research conducted in Canada offered even more evidence that vaccines can be ruled out as a cause of autism, as 28,000 children in Quebec were exposed to various dosages of MMR and then studied. There was found to be no link between MMR vaccine exposure, thimerosal exposure and incidence of autism. On the contrary, a higher rate of autism was found in Canadian children vaccinated from vaccines not containing thimerosal than among children who were administered immunizations containing thimerosal.
Eric Fombonne, MD, who directs the department of pediatric psychiatry at The Montreal Children’s Hospital stated, “Because both doses were given prior to age 2, this was a unique opportunity to study this vaccine at the time in which these disorders are usually first seen. We found that the prevalence of autism and other pervasive developmental disorders was higher among kids who had zero exposure to mercury than among kids with what would be considered medium and even high exposures” (Boyles, 2006).
Table I is a concise representation of the causality assessment of the postulated link between autism and MMR vaccination. The case reports which sparked the controversy, published by Wakefield and his colleagues, lacked population rates and comparison groups. Taylor et al. (1999) conducted the most comprehensive, population-based epidemiological study which did not find any relation between MMR vaccine and the development of autism.
Selection or referral bias becomes a possibility because the sole evidence which points towards a possible association was found in cases that were referred to a university gastroenterology clinic. As DeStefano & Chen (2001) stated, “Biological plausibility is probably the strongest argument against a causal association. In most cases of autism, neuroanatomical abnormalities probably develop in utero. It is thus unlikely that an exposure that occurs after birth, such as vaccination, could cause autism.” And while a link with developmental regression may have biological plausibility, there is new evidence that is suggestive of the presence of biological abnormalities at birth, even in examples of autistic regression.
Conclusion
The paper which started this notion, that MMR vaccine, or any other vaccine or vaccine constituent used for immunization purposes would lead to autism, was written by 13 authors, ten of whom have to date retracted the implication of this causal relationship. There is hardly any supporting evidence to this effect, and no presently available epidemiological and related evidence proves this hypothesis. Researchers have since then hoped that their studies will reassure parents and others who have been worried about this implication, and that public confidence in immunizations will be restored.
References
Baker, J. (2008). Mercury, Vaccines, and Autism: One Controversy, Three Histories. American Journal of Public Health, 98, 2, 244-253.
BC lawsuits try to link thimerosal to autism. (2003). Canadian Medical AssociationJournal, 168 (9).
Boyles, S. (2006). Study: Vaccines Don’t Cause Autism. WebMD Health News. Web.
DeStefano, F., & Chen, R. T. (2001). Autism and Measles-Mumps-Rubella Vaccination: Controversy Laid to Rest? CNS Drugs, 15 (11), 831-837.
Meadows, M. (2004). IOM Report: No Link between Vaccines and Autism. FDA Consumer, 18-19.
Medicines Commission Agency/Committee on Safety of Medicines. (1999). The safety of MMR vaccine. Current Problems in Pharmacovigilance, 25, 9-10.
Taylor, B. (2006). Vaccines and the changing epidemiology of autism. Child: care, health and development, 32 (5), 511–519.
Taylor B., Miller E., Farrington CP., et al. (1999). Autism and measles, mumps, and rubella vaccine: no epidemiological evidence for a causal association. Lancet, 353, 2026-9.
Vaccine–autism link discounted, but effect of “study” is unknown. (2007). Canadian Medical AssociationJournal, 177 (8).