Q. What are preservatives and why are they added to vaccines?

A. Preservatives are compounds that kill or prevent the growth of microorganisms, such as bacteria or fungi. They are used in vaccines to prevent bacterial or fungal growth in the event that the vaccine is accidentally contaminated, as might occur with repeated puncture of multi-dose vials. Vaccines, both in the United States and throughout other parts of the world, are commonly packaged in multi-dose vials. In some cases, preservatives are added during manufacture to prevent microbial growth; with changes in manufacturing technology, however, the need to add preservatives during the manufacturing process has decreased markedly.

Preservatives have been used in vaccines for over 70 years. The requirement for a preservative in multi-dose, multi-entry vials was placed into the Code of Federal Regulations (21 CFR 610.15) in January 1968.

The general need for preservatives in multi-dose vials has been attested to by a number of examples of multi-dose vials being formulated without preservatives becoming contaminated during use, and causing the death of vaccine recipients.

[Back to Top]

Q. What is thimerosal?

A. Thimerosal is a preservative that has been used in some vaccines since the 1930's, when it was first introduced by the Eli Lilly Company. Thimerosal has been the most widely used preservative in vaccines, present in over 30 licensed vaccines. It is 50% mercury by weight and is metabolized or degraded into ethylmercury and thiosalicylate. At concentrations found in vaccines, it meets the requirements for a preservative as set forth by the United States Pharmacopeia; that is, it kills the specified challenge organisms and is able to prevent the growth of the challenge fungi. Prior to its introduction in the 1930's, data were available in several animal species and humans providing evidence for its safety and effectiveness as a preservative. Since then, thimerosal has a long record of safe and effective use preventing bacterial and fungal contamination of vaccines, with no ill effects established other than minor local reactions like redness and swelling at the site of injection.

In July 1999, the Public Health Service agencies, the American Academy of Pediatrics, and vaccine companies agreed that thimerosal should be reduced or eliminated in vaccines as a precautionary measure to minimize the exposure to mercury from all sources. Today, all routinely recommended licensed pediatric vaccines that are currently being manufactured for the U.S. market contain no thimerosal or only trace amounts.

A. All routinely administered pediatric vaccines are now being manufactured either in thimerosal-free or thimerosal-reduced (>95% reduction) presentations. The Food and Drug Administration (FDA) expedited reviews of manufacturers' supplements to their product license applications to eliminate or reduce the mercury content of vaccines to help assure that the Public Health Service goal of replacement of thimerosal-containing vaccines takes place as quickly as possible. In August 1999, the FDA approved a thimerosal-free hepatitis B vaccine and in March 2000, a hepatitis B vaccine containing only trace thimerosal (less than 0.5 micrograms mercury per dose) was approved. Thus, as of March 2000, all U.S. children had access to hepatitis B vaccines that are free of thimerosal as a preservative. All Haemophilus influenzae type b (Hib) vaccines currently distributed in the U.S. do not contain thimerosal. With the March 2001 approval of a second diphtheria and tetanus toxoid and pertussis vaccine (DTaP) that does not contain thimerosal as a preservative, all DTaP vaccines currently being produced in the U.S. are either thimerosal free or contain greatly reduced amounts (less than 0.5 micrograms of mercury per vaccine dose).

Based on this progress, the most likely maximum amount of ethylmercury that an infant may be exposed to from currently manufactured vaccines has been reduced from approximately 187.5 mcg to < 3 mcg. The measles, mumps, rubella, varicella, inactivated polio, and pneumococcal conjugate vaccines that are now in use have never contained thimerosal.

A. Thimerosal may be added at the end of the manufacturing process to act as a preservative to prevent bacterial or fungal growth in the event that the vaccine is accidentally contaminated, as might occur with repeated puncture of multi-dose vials. When thimerosal is used as a preservative in vaccines, it is present in concentrations up to 0.01 % (50 micrograms thimerosal per 0.5 mL dose or 25 micrograms mercury per 0.5 mL dose). In some cases, preservatives are added during manufacture to prevent microbial growth. Use of thimerosal during the manufacturing process contributes considerably less to the final content of vaccines (less than 0.5 micrograms mercury per 0.5 mL dose).

Q. What is the difference between "thimerosal-free," "thimerosal-reduced," and "preservative-free"?

A. Vaccines may be termed "thimerosal-free" if no thimerosal can be measured, i.e., thimerosal content is below the limit of detection. The term "thimerosal-reduced" usually indicates that thimerosal is not added as a vaccine preservative, but trace amounts (less than 0.5 micrograms mercury per 0.5 mL dose) may remain from use in the manufacturing process. Such trace amounts are not felt to be clinically significant, nor would they result in exposure exceeding any federal guidance for mercury exposure. The term "preservative-free" indicates that no preservative (thimerosal or otherwise) is used in the vaccine; however, traces used during the manufacturing process may be present in the final formulation. For example, some vaccines may be preservative-free but may contain traces of thimerosal (less than 0.5 micrograms mercury per 0.5 mL dose); in such settings, this information is noted in the package insert.

A. Two options are available to manufacturers seeking to remove thimerosal as a vaccine preservative: reformulating the vaccine in single dose containers that do not contain a preservative, or replacing thimerosal with an alternative preservative.

Since 1999, license supplements have been approved for one DTaP and two pediatric formulations of hepatitis B vaccine. In each of these cases, removal of thimerosal as a preservative has been accomplished by changing presentations from multi-dose to single dose vials that do not require a preservative. Preservatives are used in multi-dose vials to prevent bacterial or fungal growth in the event that the vaccine is accidentally contaminated, as might occur with repeated puncture. When thimerosal is removed as a preservative, the manufacturer submits relevant information to the FDA to demonstrate that the safety and effectiveness of this product has not been affected by the change in formulation.

A manufacturer may seek to replace thimerosal with an alternate preservative; however, additional data establishing its safety and effectiveness may be required.

A. Mercury is an element that is dispersed widely around the earth. Most of the mercury in the water, soil, plants, and animals is found as inorganic mercury salts. Mercury accumulates in the aquatic food chain, primarily in the form of methylmercury, an organomercurial. Methylmercury is more easily absorbed and is less readily eliminated from the body than inorganic mercury. Exposure to methylmercury has been shown to pose a variety of health risks to humans. Extremely high levels, such as that observed in poisoning episodes in Japan and Iraq, has caused neurological damage and death. The fetus is considered more sensitive to health effects of methylmercury than adults. In recent years, some studies have found adverse health effects of methylmercury at levels previously thought to be safe. Other studies, however, have shown conflicting results. It is important to note that the preservative thimerosal contains ethylmercury, a related though distinct organomercurial from methylmercury. Information on the toxicity of ethylmercury, especially at low doses, is limited.

A. The FDA is continuing its efforts to reduce exposure to infants, children, and pregnant women to mercury from all sources. Discussions with the manufacturers of influenza vaccines (which are routinely recommended for pregnant women) regarding thimerosal-reduced and thimerosal-free presentations are ongoing. Discussions are also underway with regard to other vaccines, in particular, the diphtheria and tetanus vaccines and one manufacturer's adolescent/adult formulation of the hepatitis B vaccine (a second manufacturer's hepatitis B vaccine is formulated without thimerosal as a preservative for both the pediatric and adult presentations.)

As part of the FDAMA review, the FDA evaluated the amount of mercury an infant might receive in the form of ethylmercury from vaccines under the U.S. recommended childhood immunization schedule and compared these levels with existing guidelines for exposure to methylmercury, as there are no existing guidelines for ethylmercury, the metabolite of thimerosal. At the time of this review in 1999, the maximum cumulative exposure to mercury from vaccines in the recommended childhood immunization schedule was within acceptable limits for methylmercury exposure guidelines set by FDA, the Agency for Toxic Substances and Disease Registry (ATSDR), and the World Health Organization (WHO). However, depending on the vaccine formulations used and the weight of the infant, some infants could have been exposed to cumulative levels of mercury during the first six months of life that exceeded EPA recommended guidelines for safe intake of methylmercury. As a precautionary measure, the Public Health Service (including the FDA, National Institutes of Health [NIH], Centers for Disease Control and Prevention [CDC] and Health Resources and Services Administration [HRSA]) and the American Academy of Pediatrics issued a Joint Statement, urging vaccine manufacturers to reduce or eliminate thimerosal in vaccines as soon as possible. The U.S. Public Health Service agencies have collaborated with various investigators to initiate further studies to better understand any possible health effects from exposure to thimerosal in vaccines.

Available data has been reviewed in several public forums including the Workshop on Thimerosal, held in Bethesda in August 1999 and sponsored by the National Vaccine Advisory Committee, two meetings of the Advisory Committee on Immunization Practices of the CDC, held in October 1999 and June 2000, and by the Institute of Medicine's Immunization Safety Review Committee in July 2001. Data reviewed did not demonstrate convincing evidence of toxicity from doses of thimerosal used in vaccines. In case reports of accidental high-dose exposures in humans to thimerosal or ethyl mercury, toxicity was demonstrated only at exposures that were 100 to1000 times that found in vaccines.

The FDA is encouraging the reduction or removal of thimerosal from all existing vaccines. Much progress has been made to date. The FDA has been actively working with manufacturers, particularly those that manufacture childhood vaccines, to reach the goal of eliminating or reducing thimerosal from vaccines, and has been collaborating with other PHS agencies to further evaluate the potential health effects of thimerosal. In this regard, all of the routinely recommended pediatric vaccines are now manufactured as either thimerosal free or thimerosal reduced (less than 0.5 microgram of mercury per dose) presentations.

Q. Why did the FDA wait until mandated by Congress under FDAMA 1997 to examine the use of preservatives containing mercury?

A. The FDA had previously reviewed thimerosal use in biological products, including vaccines, in 1976. This review evaluated exposure to thimerosal from biological products using the 1974 American Academy of Pediatrics "Red Book" immunization schedule and concluded that, with the exception of long term immune globulin replacement therapy, "no dangerous quantity of mercury is likely to be received from biologic products in a lifetime."⁷ Of note, immune globulin products manufactured in the U.S. no longer use thimerosal as a preservative.

Several factors led to examination of mercury-containing preservatives in childhood vaccines. Over the past decade there has been increased attention focused on the health effects of human exposure to mercury, particularly methylmercury. In 1994, the EPA revised its Reference Dose (RfD) for methylmercury exposure, lowering its guideline for safe exposures from 0.34 to 0.1 microgram per kilogram body weight per day. Prospective studies (primarily in the Seychelles and Faroe Islands) of the effects of low dose exposure to methylmercury in the diet have been published during the past few years. Studies carried out in the Faroe Islands have raised concern that neurodevelopmental outcomes in children may be subtly affected when their mothers were exposed to methylmercury from dietary sources at levels that were previously thought to be safe. However, studies carried out in the Seychelles Islands did not show adverse neurodevelopmental outcomes at comparable exposure levels to mercury. Also, in the past decade, the CDC's Advisory Committee on Immunization Practices (ACIP) and other recommending bodies have added new vaccines containing thimerosal as a preservative such as hepatitis B and Hib vaccines to the routine childhood immunization schedule. Additionally, beginning in 1996, the replacement of whole cell DTP-Hib combination vaccines with separately administered DTaP and Hib vaccines increased the amount of thimerosal that some infants might receive (depending on vaccine formulation(s) received). In light of efforts by various Federal agencies to decrease human exposure to mercury from all sources, and the potential increase in infant exposure to thimerosal from vaccines, FDA was beginning to review of this issue.

Thus, while enactment of FDAMA 1997 provided an official mechanism for review of mercury in FDA regulated products, the use of thimerosal as a preservative in vaccines had already begun to be considered by the FDA. During the past ten years, the FDA has provided advice to manufacturers recommending that new vaccines under development be formulated without thimerosal as a preservative.

[Back to Top]

A. The IOM's Immunization Safety Review Committee concluded that the evidence is inadequate to accept or reject a causal relationship between exposure to thimerosal from vaccines and the neurodevelopmental disorders of autism, attention deficit hyperactivity disorder (ADHD), and speech or language delay. The Committee also concluded that although the hypothesis that exposure to thimerosal-containing vaccines could be associated with neurodevelopmental disorders is not established and rests on indirect and incomplete information, primarily from analogies with methylmercury and levels of maximum mercury exposure from vaccines given in children, the hypothesis is biologically plausible.

The Committee believed that the effort to remove thimerosal from vaccines was "a prudent measure in support of the public health goal to reduce mercury exposure of infants and children as much as possible." Furthermore, in this regard, the Committee urged that "full consideration be given to removing thimerosal from any biological product to which infants, children, and pregnant women are exposed."

A. Issues involving the safety of vaccines, particularly childhood vaccines, continue to concern members of the public, health care professionals, the public health community, the media, Congress, vaccine companies, and Federal agencies.

In response to the concerns, the CDC and the National Institutes of Health (NIH) asked the National Academy of Sciences' Institute of Medicine to establish an independent expert committee to review hypotheses about existing and emerging immunization safety concerns. The first of these reviews was an examination of the possible link between the use of the Measles, Mumps, and Rubella (MMR) vaccine and autism. The Committee concluded that the evidence favors rejection of a causal relationship at the population level between MMR vaccines and autistic spectrum disorders (ASD). The Committee noted that its conclusion does not exclude the possibility that MMR vaccine could contribute to ASD in a small number of children, because the epidemiological evidence lacks the precision to assess rare occurrences of a response to MMR vaccine leading to ASD and the proposed biological models linking MMR vaccine to ASD, although far from established, are nevertheless not disproved.

The report on thimerosal-containing vaccines and neurodevelopmental disorders is the second review completed by the IOM's Immunization Safety Review Committee.

The Immunization Safety Review Committee is composed of 15 expert members from pediatrics, neurology, immunology, internal medicine, infectious diseases, genetics, epidemiology, biostatistics, risk perception and communications, decision analysis, public health, nursing, and ethics. The committee members were selected on the basis of a strict criteria to eliminate any potential or perceived conflict of interest.

A. The IOM's Immunization Safety Review Committee made several recommendations.

Policy Review and Analysis - The committee recommended the following:

• the use of thimerosal-free DTaP, hepatitis B, and Hib vaccines in the United States, despite the fact that there might be remaining supplies of thimerosal-containing vaccine available.
  
   
• that full consideration be given by appropriate professional societies and government agencies to removing thimerosal from vaccines administered to infants, children, or pregnant women in the United States.
  
   
• appropriate professional societies and governmental agencies review their policies about the non-vaccine biological and pharmaceutical products that contain thimerosal and are used by infants, children, and pregnant women in the United States.
  
   
• that policy analysis be conducted that will inform these discussions in the future.
  
   
• a review and assessment of how public health policy decisions are made under uncertainty.
  
   
• a review of the strategies used to communicate rapid changes in vaccine policy, and it recommends research on how to improve those strategies.

Public Health and Biomedical Research - The committee recommended a diverse public health and biomedical research portfolio. This will be most effective if it involves several different agencies (thus maximizing resources), provides some findings fairly quickly, and utilizes a variety of approaches.

Epidemiological Research recommendations included:

• case-control studies examining the potential link between neurodevelopmental disorders and thimerosal-containing vaccines.
  
   
• further analysis of neurodevelopmental outcomes in these populations.
  
   
• conducting epidemiological studies that compare the incidence and prevalence of neurodevelopmental disorders before and after the removal of thimerosal from vaccines.
  
   
• an increased effort to identify the primary sources and levels of prenatal and postnatal exposure to thimerosal (e.g., Rho (D) Immune Globulin, which is given to Rh-negative mothers during pregnancy) and other forms of mercury (e.g., maternal consumption of fish) in infants, children, and pregnant women.

Clinical Research recommendations included:

• research on how children, including those diagnosed with neurodevelopmental disorders, metabolize and excrete metals–particularly mercury.
  
   
• continued research on theoretical modeling of ethylmercury exposures, including the incremental burden of thimerosal on background mercury exposure from other sources.
  
   
• careful, rigorous and scientific investigations of chelation when used in children with neurodevelopmental disorders, especially autism.

Basic Science Research recommendations were:

• research to identify a safe, effective, and inexpensive alternative to thimerosal for countries that decide they need to switch.
  
   
• research in appropriate animal models on neurodevelopmental effects of ethylmercury.

A. Neurodevelopmental disorders are disorders affecting the nervous system, including the brain. For this report, the IOM's Immunization Safety Review Committee examined the hypothesis of whether or not vaccines containing thimerosal could have caused specific neurodevelopmental disorders, including autism, attention deficit/hyperactivity disorder (ADHD), and speech or language delay.

A. For each hypothesis to be examined, the committee assesses both the scientific plausibility of the issue and its significance in a broader societal context. The scientific plausibility is based on two parts: the biologic plausibility (if it is biologically possible) and causality (an examination of the evidence regarding a possible relation between the vaccine and the adverse event). The significance assessment considers the nature of the health risks associated with the vaccine-preventable disease and with the adverse event in question and other societal concerns. The findings of the plausibility and significance assessments provide the basis for the committee's recommendations.

A. A recall of thimerosal-containing vaccines is not warranted because currently, available data show that these products are safe and effective. Federal law is specific about the criteria that must be met before FDA can enforce a mandatory recall of a regulated product. Under section 351(d) of the Public Health Service Act, a licensed vaccine (or other biological product) shall be recalled if FDA determines that it "presents an imminent or substantial hazard to the public health..." FDA does not believe that thimerosal-containing vaccines present an imminent or substantial hazard to the public health because available scientific data do not establish that exposure to thimerosal in vaccines can cause neurodevelopmental disorders. Additional studies on the potential for adverse effects of mercury in vaccines are continuing. Results of these studies will be closely monitored.

FDA regulations also provide for a voluntary recall of products regulated by the FDA (21 CFR, Part 7). A firm may withdraw a product from the market, of its own volition, at any time. In addition, FDA may request a firm to recall a product that is in violation of FDA laws and regulations and that presents a risk of injury or gross deception, or is otherwise defective; an agency request for recall is reserved for urgent situations such as those that are necessary to protect the public health. FDA has concluded that voluntary recall is not warranted because vaccines that contain thimerosal as a preservative are not violative products and there are no conclusive data that they present a risk of injury.

However, the Department concurs with the IOM that it is prudent to avoid mercury exposure from vaccines, indeed, from all sources. Accordingly, the Department's Inter-Agency Vaccine Group has worked with manufacturers to remove or reduce thimerosal from vaccines. The FDA expedited reviews of manufacturers' supplements to their product license applications to eliminate or reduce the mercury content in vaccines to help assure that the Public Health Service goal of replacement of thimerosal-containing vaccines takes place as expeditiously as possible.

Thus, since March 2001, all routinely administered pediatric vaccines are now being manufactured either in thimerosal-free or thimerosal-reduced (> 95% reduction) presentations, and infant exposure to mercury from vaccines is unlikely to exceed any federal guidelines.

A. As of October 2001, the vast majority of the supplies of DTaP, Hib, and hep B vaccines are without thimerosal or with only trace amounts.

A. The ACIP met in June 2001 to review the progress in achieving the goal of removing thimerosal containing vaccines from the routinely recommended childhood immunization schedule. At that time, they chose not to make any changes to their previous recommendation which stated that thimerosal-containing or thimerosal-free vaccines were equally acceptable for use. The ACIP determined that the large risks of not vaccinating children for outweigh the unknown and probably much smaller risk, if any, of cumulative exposure to thimerosal-containing vaccines over the first six months of life. The ACIP will reconsider this issue at its next meeting on October 17-18 in Atlanta, GA.

A. The use of any DTaP, hepatitis B and Hib vaccine should continue according to the currently recommended schedule. The risk of not vaccinating children on time to protect them from these diseases is believed to far outweigh the slight risk, if any, of exposure to thimerosal-containing vaccines which are still available.

A. All influenza vaccines contain thimerosal; however, ACIP recommends no changes in the influenza vaccination guidelines, including those for children and pregnant women. Evidence suggests that children with certain medical conditions (e.g., cardiopulmonary disease, including asthma and immunodeficiency conditions) are at substantially increased risk for complications of influenza. During the influenza season, rates of hospitalizations for otherwise healthy women in their second or third trimester of pregnancy are similar to those for cardiopulmonary problems from influenza disease among persons aged greater than or equal to 65 years who do not have a chronic medical illness and for whom influenza vaccination also is recommended. Pregnant women with chronic medical conditions are at higher risk and have a hospitalization rate more than two times greater than among pregnant women without other high_risk medical conditions. A substantial safety margin has been incorporated into the health guidance values for organic mercury exposure developed by the Agency for Toxic Substances and Disease Registry and other agencies. ACIP concluded that the benefits of influenza vaccine outweigh the potential risks for thimerosal.

A. Efforts to remove thimerosal from the U.S. vaccine supply have been accompanied by research investigations to better assess the potential health effects of exposure to thimerosal-containing vaccines. The major studies and findings include: The NIH, in collaboration with researchers from the University of Rochester and the Bethesda Naval Hospital, undertook a study to determine how much mercury, if any, could be detected in the blood of infants following exposure to thimerosal-containing vaccines. There was no observed dose-dependent relationship between the level of thimerosal received through vaccination and the level of mercury in the infants.

The CDC has used large automated databases that link vaccination and International Classification of Disease codes (ICD_9) stored in the medical records in three managed care organizations (i.e., the Vaccine Safety Datalink project, or VSD) to screen for any possible associations between exposure to thimerosal-containing vaccines and a variety of neurologic, developmental, and renal outcomes. In phase I of this investigation, using the data from two of the managed care organizations, CDC and VSD researchers found statistically significant associations between thimerosal and some neurodevelopmental disorders, including language delays, speech delays, attention deficit hyperactivity disorder (ADHD), unspecified developmental delays, stammering, sleep disorders, emotional disorders, and tics. The consultants noted that two thirds of the ADHD cases in the data set were not later confirmed by specialists. Overall, the correlations were weak. However, the associations were not consistent between the two VSD sites examined.

In phase II of the investigation, CDC investigators obtained and examined data from a third managed care organization. Analyses of these data using the same methods and having similar limitations as in the above study, did not confirm results for speech or language delay and attention deficit disorder. The number of events was too small to examine the association with tics and the category of unspecified developmental delays was not defined clearly enough to permit reanalysis. No association was found between thimerosal in vaccines and autism in the one site that had enough children to test for a relationship.

The FDA and NIH are collaborating on development of protocols to evaluate the pharmacokinetics of ethylmercury vs. methylmercury vs. thimerosal in two animal studies. Pharmacokinetics is the study of how an agent is absorbed, distributed, metabolized (broken down), and excreted.

A. The review of the concerns that has been carried out by the independent expert panel assembled by the IOM will contribute to maintaining public confidence in our national immunization program and assuring the continued protection of U.S. children against vaccine preventable diseases in an effective and safe manner. The recommendations made by this expert panel are under review by the Department's Inter-Agency Vaccine Group. The ACIP will consider the IOM's recommendations at its next meeting, in Atlanta, GA, on October 17-18, 2001. In the meantime, ACIP childhood immunization recommendations remain unchanged.

[Back to Top]

A. The IOM used the following criteria for assessing whether evidence indicates the presence of an association between an adverse event and vaccine exposure (IOM 1991). Similar criteria were used recently by the IOM's Immunization Safety Review Committee in their assessment of thimerosal-containing vaccines and neurodevelopmental disorders (IOM 2001).

• Strength of Association – The strength of an association refers to the magnitude of the measure of effect of an exposure, usually the relative risk or odds ratio, in a study comparing an exposed and an unexposed group. The larger the magnitude of the effect the less likely any observed effect is due to chance, bias or confounding. In general, in observational studies, relative risks of 2 or less are considered to be evidence of a weak association, because the likelihood the effect is due to chance, bias or confounding is greater than if the effect is larger).
  
   
• Dose-Response Relation – The existence of a dose-response relation strengthens an inference that an association is causal. A dose-response relation is defined as an increased strength of association with increased magnitude of exposure.
  
   
• Temporally Correct Association – Exposure must precede the event by at least the duration of disease induction. This consideration may be limited by the fact that knowledge of the pathogenesis and natural history of an adverse event may be insufficient.
  
   
• Consistency of Association – This consideration requires that an association be found regularly in a variety of studies, using different study populations and study methods.
  
   
• Specificity of an Association –Uniqueness of an association between an exposure and an outcome provides a stronger justification for a causal interpretation than when the association is nonspecific. However, perfect specificity between an exposure and an effect cannot be expected in all cases because of the multifactorial etiology of many disorders.
  
   
• Biological Plausibility – The existence of a possible mechanism of action that fits existing biologic or medical knowledge is thought to increase the likelihood that an association is causal.

The final judgment on whether there is a causal relationship between an exposure, such as thimerosal, and an outcome, such as autism, is made by balancing between the strength of above considerations supporting a causal interpretation against the strength of alternative explanations.

A. These considerations may be applied as follows:

• Strength of Association – When evaluating this consideration 3 types of studies are generally included, controlled trials, cohort studies, and case-control studies. These studies have in common the ability to calculate an estimate of the relative risk of an effect from an exposure.
  
  There have been no controlled trials of the relationship between thimerosal and autism, and those controlled trials that have been done using thimerosal containing vaccines have not been conducted in a way that would allow an evaluation of the relationship between thimerosal and autism. There have been two retrospective cohort studies conducted recently by the CDC using data from 2 different health maintenance organization (HMO) databases. The first study did not show a statistically significant association between thimerosal exposure and autism. There were not enough cases of autism in the second database to study. There have been no case control studies of autism and thimerosal.
  
  An association that has been noted by some concerned parents of autistic children is that the increase in the prevalence of autism over the last few decades "closely matches the introduction and spread of thimerosal-containing vaccines". This type of comparison is known as an ecological study. Ecological studies alone are generally not accepted as strong evidence of causality, because they do not link individual exposure to individual outcome, and can be subject to confounding by unknown or uncontrollable factors. In addition, it has been noted that some children with autism have high levels of mercury in hair, urine and blood. This observation cannot be interpreted without information on the levels of mercury in individuals without autism (i.e. case-control study). However, such observations do indicate that the hypothesis should be studied further.
  
   
• Dose-Response Relation – In the CDC cohort study that was conducted to examine the relationship between thimerosal and autism, no increase in relative risk of autism was observed with increasing doses of thimerosal, making a causal relationship less likely.
  
   
• Temporally Correct Association – A detailed understanding of the time course of the pathological changes that lead to autism is not available. Since autism is not usually diagnosed at least until 18 months of age, it is difficult to judge whether the exposure to the highest levels of thimerosal on a body weight basis in the first 6 months to a year of life is consistent with a causal relationship between thimerosal exposure and autism.
  
   
• Consistency of Association – The one CDC cohort study in which autism could be examined failed to show an association between thimerosal and autism, but because only one study has been conducted we cannot evaluate consistency across studies. Thus, a causal relationship between thimerosal and autism is neither supported nor refuted by this consideration.
  
   
• Specificity of Association – There is not a clearly identified unique relationship between thimerosal and autism, or for methylmercury exposure and autism.
  
   
• Biological Plausibility – In a general sense it is biologically plausible that thimerosal could cause autism since thimerosal has caused neurological abnormalities at high doses. However, it has not been shown that mercury at the much lower doses found in childhood vaccines can cause the specific neurological injuries that are found in autism.

In summary, the CDC cohort study did not show an association between thimerosal exposure and autism and no dose-response relationship was observed, thus the existing evidence does not support a causal relationship between thimerosal and autism. However, additional studies to fill in gaps in our knowledge, such as whether the regressive subtype of autism is causally related to thimerosal in vaccines, may be warranted.

A. Analogous clinical features have been described between autistic individuals and those suffering from mercury toxicity (Bernard et al. 2001). Hill (1965) included this "analogy" as an additional consideration in his original discussion of causal inference. This consideration has not been accepted as strong evidence of causality, and was not used by the Institute of Medicine in its evaluation, because it is quite easy to draw analogies among exposure and disease relationships, even when causal relationships do not exist. The analogies between the neurological illness of mercury poisoning on the one hand and autism and thimerosal exposure on the other are not strong evidence of a causal relationship, but suggest more definitive studies should be conducted.

A. Intrauterine and postnatal development of the nervous system can be affected by many different toxins. Mercury, typically in the form of ingested methylmercury compounds, has been shown to induce abnormalities in the brain of humans and experimental animals. Physical damage to the areas of brain undergoing development at the time of exposure has been found in autopsy studies and in experimental animal models. Neurochemical transmitter abnormalities in the brain associated with toxic compounds, such as mercury, has been identified in experimental animal systems, and, indirectly, in human samples of blood or cerebrospinal fluid.

Available evidence indicates that autism is a developmental disorder of the nervous system. Many etiologies have been suggested or proven to lead to autism or autism spectrum disorders, from genetic (e.g., Rett's Syndrome, Fragile X Syndrome) to exposure to drugs in utero (e.g., thalidomide). Only a small number of brains from autistic individuals have been available for pathological studies. Those studies reveal some consistency of damage in specific areas (e.g., the cerebellum and hippocampus). Further, brain imaging studies have been performed in autism, and the results suggest similar brain areas are affected, but these studies are less consistent. How the reported neuroanatomical damage relates to the expression of autistic disease is unknown. Indirect assessments of neurochemical changes from the cerebrospinal fluid, blood or by special brain scans (PET) of autistic subjects also suggest a variety of neurochemical abnormalities. However, in part because these indirect measurements (e.g., blood levels of norepinephrine) may not reflect accurately changes in the brain neurochemistry, the relationship of the proposed neurochemical abnormalities to disease expression is unknown.

Determining cause and effect relationships of early toxin exposures to the developing nervous system is very difficult. Subtle changes in exposure doses and timing of exposure of a single toxin (e.g., methylmercury) can cause a variety of different outcomes in the nervous system. Moreover, exposure to many different toxic treatments can result in similar types of damage to the brain (e.g., ethanol, X-irradiation). Currently, little information is available on the outcome of exposure of the developing human nervous system to toxic levels of ethylmercury.

In summary, the data are not sufficient to support the causal relationship between ethylmercury exposure and autism. More information is needed about autism and mercury toxicity in order to understand the relationship, if any, between thimerosal and autism.

A. The only well-recognized adverse effect associated with use of thimerosal in drugs and vaccines is a transient skin allergy or local hypersensitivity reaction. In most cases when such reactions are reported, it cannot be definitely established that thimerosal is the cause of the allergy. Serious hypersensitivity reactions following vaccinations such as anaphylaxis are rare. A predisposition to allergic reactions probably is linked to genetic factors, although such factors are not well understood.