|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 002007384
007 cr mnu|||uuuuu
008 090619s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002301
The association between the measles, mumps, and rubella vaccine and the development of autism :
b a meta-analysis
h [electronic resource] /
by Rashad Carlton.
[Tampa, Fla] :
University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 58 pages.
Thesis (M.S.P.H.)--University of South Florida, 2008.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
ABSTRACT: Autism is a childhood developmental disorder characterized by impaired social interaction, difficulty with verbal and nonverbal communication and limited activities. The root cause of autism is unknown. However it has been postulated that administration of the measles, mumps, and rubella (MMR) vaccine may be causally related to the development of autism. MMR vaccination is a requirement for entry into schools, so any increase in adverse events associated with the vaccine carries widespread public health importance. The primary objective of this study was to conduct a meta-analysis of the association between the MMR vaccination and the development of autism. The meta-analysis was limited to studies with an experimental design, unvaccinated control group, and odds ratio or relative risk as the effect measure. Both the fixed effects and random effects models were applied. A total of 29 studies were identified for possible inclusion in the meta-analysis. After applying the inclusion/exclusion criteria seven studies were included in the meta-analysis. The pooled treatment effect was weighted based on the width of the 95% confidence interval for each of the individual studies. The pooled effect measure for the seven studies was 1.052 (95% CI: 0.973, 1.138) (P=0.202). An association between the MMR vaccine and the development of autism was not found in this analysis. Public health initiatives should continue to support mandatory vaccination with MMR for entry into school and steps should be taken to eliminate any barriers to vaccination. Further epidemiological studies are needed to assess the root cause of autism
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Barabara Orban, Ph.D.
x Health Policy and Management
t USF Electronic Theses and Dissertations.
The Association between th e Measles, Mumps, and Rubella Vaccine and the Development of Autism: A Meta-Analysis by Rashad Carlton A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Public Health Department of Health Policy and Management College of Public Health University of South Florida Major Professor: Barbara Orban, Ph.D. Laurence Branch, Ph.D. Etienne Pracht, Ph.D. Date of Approval: March 19, 2008 Keywords: MMR, autism, vaccination, childhood immunization Copyright 2008, Rashad Carlton
i Table of Contents List of Tables ii List of Figures iii Abstract iv Introduction 1 Autism 1 Measles, Mumps, Rubella 6 Measles 6 Mumps 7 Rubella 8 Vaccine Preparations 9 MMR Adverse Events 12 Thimerosal 15 Healthy People 2010 16 Methods 20 Results 23 Descriptive Statistics 23 Meta-Analysis Results 26 Sensitivity Analysis 27 Discussion 28 Conclusion 32 References 33 Bibliography 35 Appendices 37 Appendix A Recommended Immunization Schedule 38 Appendix B Research Review 39
ii List of Tables Table 1. National Vaccine Injury Comp ensation ProgramPost 1998 Statistics Report of Petitions Filed 14 Table 2. National Vaccine Injury Comp ensation ProgramPost 1998 Statistics Report of Adjudication of Claims 15 Table 3. Synopsis of Included Studies 23 Table 4. Recommended Immuniza tion Schedule for Ages 0-6 38 Table 5. Research Review 39
iii List of Figures Figure 1. Meta-analysis re sults: Autism and MMR 26 Figure 2. Meta-analysis resu lts: Funnel Plot 27
iv The Association between the Measles, Mumps, and Rubella Vaccine and the Development of Autism: A Meta-Analysis Rashad Carlton ABSTRACT Autism is a childhood developmental disord er characterized by impaired social interaction, difficulty with ve rbal and nonverbal communicati on and limited activities. The root cause of autism is unknown. However it has been postulated that administration of the measles, mumps, and rubella (MMR ) vaccine may be causally related to the development of autism. MMR vaccination is a requirement for entry into schools, so any increase in adverse events associated with the vaccine carries wide spread public health importance. The primary objective of this study wa s to conduct a meta-analysis of the association between the MMR vaccination and the development of autism. The metaanalysis was limited to studies with an e xperimental design, unvaccinated control group, and odds ratio or relative risk as the effect measure. Both the fixed effects and random effects models were applied. A total of 29 studies were identified fo r possible inclusion in the meta-analysis. After applying the inclusion/exclusion criteria seven studies were included in the metaanalysis. The pooled treatment effect wa s weighted based on the width of the 95% confidence interval for each of the individual studies. The pooled effect measure for the seven studies was 1.052 (95% CI: 0.973, 1.138) ( P =0.202).
v An association between the MMR vaccine and the development of autism was not found in this analysis. Pub lic health initiatives should continue to support mandatory vaccination with MMR for entry into school a nd steps should be taken to eliminate any barriers to vaccination. Further epidemiol ogical studies are needed to assess the root cause of autism.
1 Introduction Autism Autism is a developmental disorder charac terized by impaired social interaction, difficulty with verbal and nonverbal communi cation, and limited activities and interests (NINDS 2006). Autism was first characte rized in 1943 by Dr. Leo Kanner of Johns Hopkins Hospital, who termed the disease earl y infantile autism after studying a group of 11 children (Strock 2007). Th e Diagnostic and Statistical Manual of Mental Disorders (DSM IV) classifies autism as one disease in a class of developmental disorders referred to as autism spectrum disorders (ASDs) or pervasive developmental disorders (PDDs) (Strock 2007, NINDS 2006). Asperger syndrome, Rett syndrome, childhood disintegrative disorder, and pe rvasive developmental disorder not otherwise specified along with autism make up the ASDs. Autism is the most common of the ASDs with an estimated three to six children out of 1,000 developing the disorder (NINDS 2006). The most common sign of autism is impair ed social interaction. Autistic children usually do not know how to play interactiv ely with other children (NINDS 2006). Autistic children commonly have difficulty learning and understanding the normal giveand-take of social interaction. Autistic child ren are typically slow to learn to interpret what others are thinking or fee ling. At times they may appear to be indifferent to others. This lack of social interaction results in the perception that they pref er to be alone (Strock 2007).
2 Other common signs include problems with verbal and nonverbal communication, repetitive behavior, and obs essive interests (NINDS 2006) In some cases autistic children remain mute their entire life. Some autistic children develop verbal skills at a much later age than their peers. Others will develop skills at a normal age but have difficulty communicating with ot hers (Strock 2007). Additi onally, autistic children often engage in repetitive behavior such as rock ing or twirling (NINDS 2006). In most cases autistic children need and seek consistency in their environment. The smallest changes in daily routines can be very di sturbing for autistic children. In addition to the developmental challenges associated with autism, autistic children are at higher risk fo r several comorbid conditions including fragile X syndrome, tuberous sclerosis, epileptic seizures, Tourette syndrome, learning disabilities, and attention deficit disorder. Fragile X syndrome is an inherited form of mental retardation named for the part of the X chromosome that has a defective piece that appears pinched and fragile when viewed under a microscope. Tuberous sclerosis is a genetic disorder that causes benign tumors to grow in the brain and other vital organs (Strock 2007). Approximately 20 to 30% of autistic children will develop epileptic seizures by adulthood (NINDS 2006). In most cases the seizures can be controlled with pharmacologic therapy. Most autistic ch ildren will have some degree of learning disability, with the degree of disability va rying by child. Often times, some areas of ability are normal, while one or more areas will show developmental delay. The diagnosis of autism is challenging as autism varies widely in symptoms. Symptoms can range from mild to severe. Parents are often th e first to notice any symptoms of autism in their child. There are se veral screening tools th at can assist in the
3 screening for autism such as the Checklist of Autism in Toddlers, the Screening Tool for Autism in Two-Year-Olds and the Social Co mmunication Questionnaire (Strock 2007). In order to establish an a ccurate diagnosis of autism, a comprehensive evaluation is required (NINDS 2006). The evaluation s hould be performed by a multidisciplinary team consisting of a psychologist, neurologist, psychiatrist and speech therapist (NINDS 2006). Evaluation tools such as the Autism Diagnosis Interview Revised and the Autism Diagnostic Observation Schedule help in ev aluating a child for autism. The Autism Diagnosis Interview Revised is a structured interv iew of 100 questions consisting of four major factors: communication, social interaction, repetitive behaviors, and age of onset. The Autism Diagnostic Observation Schedule is an observational measure used to detect behaviors that are absent, delayed, or abnormal in children (Strock 2007). The National Institute of Neurological Di sorders and Stroke list the following core behaviors that doctors rely on for a diagnosis of autism: Impaired ability to make friends with peers Impaired ability to initiate or sustain a conversation with others Absence or impairment of imaginative and social play Stereotyped, repetitive, or unusual use of language Restricted patterns of interest that are abnormal in intensity or focus Preoccupation with certain objects or subjects Inflexible adherence to specific routine or rituals The diagnosis of autism is often complicat ed by other autism spectrum disorders. Children with insufficient symptoms to meet th e criteria for autism are often classified as having pervasive developmental disorder not otherwise specified. Children with
4 Asperger syndrome present with autistic symptoms but have well developed language skills (NINDS 2006). Tools such as the Au tism Spectrum Screening Questionnaire, the Australian Scale for Aspergers Syndrome, and the Childhood Asperger Syndrome Test can help to distinguish Asperger syndrome from autism (Strock 2007). Children with childhood disintegrative disord er develop normally and then deteriorate between the ages of 3 to 10, showing signi ficant autistic symptoms (NIN DS 2006). After a diagnosis of autism is made, the next step is to identif y appropriate treatment fo r the autistic child. Currently there is no cure for autism and there is no single best treatment regimen for all children. Current therapy is designed to alleviate specific symptoms and improve functional ability. Treatment falls into tw o broad categories: educational/behavioral interventions and pharmacologic therapy. Educational interven tions include skilloriented sessions to help children develop so cial and language skills along with family counseling for parents and siblings of autisti c children. Pharmacologic therapy includes antidepressants to treat depression, anxiety, or obsessive compulsive disorder or antipsychotics to treat severe behavi oral symptoms (NINDS 2006). The selective serotonin reuptake inhibito rs such as fluoxetin e, sertaline, and fluvoxamine are commonly used off-label to treat depression, a nxiety, and obsessive compulsive disorder in autistic children. In 2004, the Food and Drug Administration issued a black box warning al erting the public of the increa sed risk of suicidal ideation or suicide attempts in children and adolescen ts taking antidepressants. In October 2006, the Food and Drug Administration approved ri speridone for the symptomatic treatment of irritability in autistic children, repres enting the first drug with an FDA approved indication to treat symptoms in autism (Strock 2007). In cases where au tistic children
5 experience seizures, anticonvulsants are indicated. Commonly prescribed anticonvulsants include: carbamazepine, lamotr igine, topiramate, and valproic acid. Autistic children with attention deficit diso rder or hyperactivity can be treated with stimulant drugs, such as methylphenidate (Strock 2007). There are also several controversial therapies available, but few if any have been scientifically proven to be effective (NINDS 2006). The root cause of autism is unknown to date and as of yet a biological marker for autism has not been found. Several causat ive factors have been offered including genetics, environmental factors, abnormal levels of serotonin or other neurotransmitters in the brain, and vaccine administration. It is strongly suspected that genetics plays some role because families with one autistic child have an approximately 5 percent risk of autism in a second child, which is much hi gher than the risk in the general population (NINDS 2006). MRI studies of the brain ha ve shown that many areas of the brain are abnormal in autism including the cerebellum, cerebral cortex, limbic system, corpus callosum, basal ganglia, and brain stem. Re search is ongoing to study what role these areas of the brain have in the developmen t of autism along with the possible role of neurotransmitters such as serotonin, dopa mine, and epinephrine (Strock 2007). In the past few years there has been widespread public concern regarding a proposed theory that autism was linked to vaccine administration. The proposed link between autism and vaccines was first reporte d by Wakefield and colleagues in a case series study of 12 children with chronic en terocolitis and regre ssive developmental disorder (Wakefield 1998). In 8 of the 12 ch ildren, the onset of beha vioral problems was retrospectively linked to the measles, mumps, rubella (MMR) vaccination by a parent or
6 physician. The Wakefield st udy gave rise to the proposed association of the MMR vaccine and the development of autism and cr eated a major debate on the safety of the MMR vaccination. Since the publication of the Wakefield study, se veral other studies have been conducted assessing the associ ation between MMR vaccination and the development of autism. Due to a lack of randomized controlled trials and conflicting results, there has not been enough evidence to support or definitively refute the hypothesized association betw een MMR and autism. Measles, Mumps, Rubella Measles Since the inception of vaccinations for m easles, mumps, and rubella the number of cases of these diseases has decreased by 99%. The success of the vaccinations has led to an attempt to eliminate these diseas es. A goal of the 1993 Childhood Immunization Initiative was to eliminate indigenous transmi ssion of measles and rubella in the United States by 1996 (CDC 1998). While these goals have not yet been achieved it appears that vaccination makes these goals feasible in the near future. The first of these three diseasesmeasles also referred to as rubeolais a disease characterized by a total body skin rash. Measles is a highly contagious disease with an incubation period of 10-12 days from exposur e to prodrome and 14 days from exposure to rash. In the United States 1-2 out of every 1,000 cases results in death. Infants and young children are at higher risk of death from measles and its complications. The most severe complications of measles infection are pneumonia and acute encephalitis. The rate of death from measles in less developed countries can be as high as 25%. In 1963
7 the first measles vaccine was licensed. Prio r to this time an average of 400,000 measles cases were reported each year in the United States. Since the inception of the measles vaccine, the number of cases of measles has decreased by 99% (CDC 1998). Measles experienced a resurgence in the United Stat es from 1989 to 1991. There were more than 55,000 cases of measles and over 12 0 measles related deaths. The resurgence was led to a large degree by a number of unvaccinated pr eschool aged children residing in urban areas. In 1989 the Advisory Committee on Imm unization Practices (ACIP) and the American Academy of Pediatrics (AAP) r ecommended that all children receive two doses of a measles-containing vaccine. The implementation of the vaccination policy led to a decrease in the number of measles cases from 2,237 cases in 1992 to only 312 cases in 1993. The challenges to eliminating and sust aining the elimination of measles are: 1) continuing to vaccinate all children aged 12-15 months with a first dose of MMR, 2) ensuring that all children have received a second dose of MMR before entering school, and 3) working with other countries to achie ve measles elimination goals (CDC 1998). Mumps Mumps is a disease characterized by bilate ral or unilateral parotitis. The average onset of mumps is 16-18 days after exposure with nonspecific symptoms such as fever, headache, malaise, myalgia, and anorexia preceding parotitis. Approximately 15%-20% of all mumps cases are asymptomatic and up to 50% of cases are associated with nonspecific respiratory symptoms The most serious complications of mumps such as male sterility and aseptic meningitis are more likely to occur in adults than children (CDC 1998).
8 The mumps vaccine was first licensed in the United States in 1967. Since the inception of the mumps vaccine, reported ca ses of mumps have decreased by 99% from 185,691 cases in 1968 to only 906 cases in 1995. State laws requiring that children be vaccinated before school entry have significantly contributed to the decrease in reported cases of mumps. The current two dose sc hedule of MMR likely further decreases the mumps incidence by immunizing children with a second dose as not all children generate an immune response following the first dose (CDC 1998). Rubella Rubella is a disease characterized by nonspecific signs and symptoms including a pruritic rash, arthralgia, and low-grade fever. The average incubation period is 12 to 23 days and between 25-50% of all cases are subclinical. Congenital rubella syndrome carries the risk of severe cons equences such as miscarriages, stillbirth, fetal anomalies, and therapeutic abortions. Children born w ith congenital rubella syndrome often have several abnormalities including sensory deficits, ophthalmic deficits, mental retardation, microcephaly, pulmonary artery stenosis, and atrial or ventricular septal defects (CDC 1998). During the last major outbreak of rube lla in the United States from 1964-1965, there were an estimated 20,000 cases of congenital rubella syndrome from 1964-1965. This led to the first rubella vaccine being licensed in 1969 with a target of children in kindergarten and the early grades of elemen tary school. Following the initial vaccination campaign, cases of congenital rubella decr eased by 69% from 1970 to 1976. However, rubella outbreaks continued to occur in older adolescents and young adults. ACIP modified the recommendations for immuni zation to include the vaccination of post
9 pubertal girls and women in 1977. Cases of congenital rubella syndrome and rubella have occurred at a relatively constant ende mic level with an average of less than 200 cases with an occasional outbreak in persons over the age of 20 years. An accurate incidence rate for rubella is difficult to estimate because rubella surveillance in the United States relies on a pa ssive system (CDC 1998). Vaccine Preparations Measles, mumps and rubella (MMR) v accines are available as a combination vaccine, as well as monovalent vaccine for each specific agent. Vaccines are also available as a combination of measles-rube lla and a combination of rubella-mumps. Excipients included in each dose of the co mbined or monovalent vaccines are 0.3 mg of human albumin, 25 g of neomycin, 14.5 mg of sorbitol and 14.5 mg of hydrolyzed gelatin. The measles and mumps compone nts are live vaccines produced in chick embryo cell culture and the live rubella vaccin e is grown in human diploid cell culture (CDC 1998). The strain of measles used in the va ccine has changed several times since the vaccine was first licensed in 1963. Originally, both a live and an attenuated strain of vaccine were available. Currently, only th e Enders-Edmonston strain of measles licensed in 1968 is available in the United States. Upon vaccination the measles vaccine produces a mild, noncommunicable infection with an tibodies developing in 95% of children vaccinated at 12 months and 98% of children va ccinated at age 15 months. Serologic and epidemiologic evidence indicates that the v accine produces lifelong immunity in most persons (CDC 1998)
10 The strain of rubella used in vaccines is a live strain of RA 27/3 first licensed in 1979. Serologic immunity is achieved in 95% of people aged 12 months or older who receive at least one dose of rubella vaccine. Greater than 90% of vaccinated individuals have protection against both clinical rubella a nd viremia for at least 15 years. There have been several reports of viremic reinf ection following exposure among vaccinated individuals with low levels of detectable antibodies. Addition ally, rare cases of congenital rubella syndrome have occurred in infants born to mothers with documented serologic evidence of rubella immunity be fore they became pregnant (CDC 1998). The mumps vaccine contains the live Je ryl-Lynn strain of mumps. Vaccination with mumps vaccine produces a s ubclinical infection with few si de effects. In controlled clinical trials, over 97% of those vaccinat ed develop measurable immunity following vaccination. Studies in the field report that the efficacy of the vaccine ranges from 75% to 95%. Serologic and epidemiologic evid ence suggest continuing protection against infection although the duration of immunity is unknown (CDC 1998). Vaccine Administration The Advisory Committee on Immunization Practices (ACIP) recommends the combined MMR vaccine as the vaccine of choi ce to protect agains t all three diseases unless contraindicated. Two doses should be administered separated by at least one month in all children on or af ter their first birthday and in high risk adolescents and adults. The second dose of MMR should elicit a response in the few individuals who do not elicit an immunological re sponse to at least one component following their first vaccination. Administering the vaccines in combination produces a similar response to receiving single-antigen vaccinations with measles, mumps, and rubella vaccines at
11 different sites or different times. ACIP s upports the administration of the MMR vaccine at the same time as other vaccines such as diphtheria, tetanus toxoid and acellular pertussis (DTaP), Haemophilus influenzae type b (Hib), oral polio vaccine or inactivated polio vaccine according to the schedule to receive vaccines (CDC 1998). (See Appendix A) ACIP recommends that all children recei ve their first dose of MMR vaccine at age 12-15 months. In certain high risk areas it is recommended that children receive their first dose by 12 months. High risk areas are defined as: A county with a large in ner city population, A county where a recent measles outbr eak has occurred among unvaccinated preschool-aged children, or A county in which more than five case of measles has occurred among preschoolaged children each of the last 5 years. ACIP recommends that the second dose of MMR vaccine be administered when children are aged 4-6 years. ACIP, the American Academy of Pediatrics (AAP), and the American Academy of Family Physicia ns (AAFP) have jointly adopted the recommended timing of the second dose of MMR. ACIP encourages all states to adopt the two dose MMR requirement for preschool-a ged children so that all 50 states will have a universal policy requiring: For preschool-aged children: documentation of at least one dose of MMR vaccine administered on or after the first birthday, and
12 For children in kindergarte n through grade 12: document ation of two doses of MMR vaccine separated by at least 28 days with the first dose administered no earlier than the first birthday (CDC 1998). ACIP recommends that all people bor n in 1957 or later who do not have a contraindication to the MMR vaccine should receive at le ast one dose. People born before 1957 are considered to be immune to measles, mumps, and rubella. People considered being at high risk such as international traveler s, persons attending colleges, and persons who work at healthcare facilities should receive special consideration for vaccination. Additionally, a ll women of childbearing age who do not have acceptable evidence of rubella immunity should be offe red vaccinations whenever they contact the healthcare system (CDC 1998). MMR Adverse Events The range of adverse effects following the administration of the MMR vaccine ranges from common adverse events such as local pain and edema to the rare case of anaphylaxis. An expert committee at the Inst itute of Medicine determined that evidence supports a causal relationship between MMR vaccination and anaphylaxis, thrombocytopenia, febrile seizures, and acute arthritis (CDC 1998). Other adverse events reported in the MMR package insert include: va sculitis, otitis media, conjunctivitis, optic neuritis, ocular palsies, Guillan-Barre s yndrome and ataxia (M-M-R II 2007). In response to the adverse events caused by all vaccines including the MMR vaccine, The National Vaccine Injury Compen sation Program (VICP) was created by the National Childhood Vaccine Injury Act of 1986. VICP works to resolve vaccine injury claims by providing compensation for people found to be injured by certain vaccines.
13 The US Department of Health and Human Serv ices, the US Department of Justice, and the US Court of Federal Claims all have a role in VICP. Compensation for injured parties as a result of vaccin e administration is determined by the US Court of Federal Claims (HRSA). VICP cove rs the following vaccines: Diptheria, tetanus, pertussis (DTaP) Haemophilus influenzae type b Hepatitis A and Hepatitis B Human Papillmovirus Influenza Measles, mumps, rubella (MMR) Meningococcal Polio Pneumococcal conjugate Rotavirus Varicella Any person who has received a vaccine cove red by VICP that believes that they were injured as a result of that vaccine may file a claim. To file a claim the effects of the injury must have: 1) lasted for more than 6 months after the vaccine was given; or 2) resulted in a hospital stay and surgery; or 3) resulted in death. There is no limit on the amount of compensation an injured party may receive. The amount compensated is determined as a reasonable amount for past and future medical and custodial care costs and related expenses, lost earnings, reasona ble lawyer fees and up to $250,000 for actual and projected pain and sufferi ng. A claim must be filled w ithin 3 years after the first
14 symptom of the vaccine related injury or within 2 years of death or 4 years after the first symptom of the vaccine rela ted injury (HRSA 2008). Since the inception of VICP there ha ve been 774 claims against the MMR vaccine. A total of 271 cases resulted in compensation being paid to the claimant, 323 claims were dismissed, and 180 claims are still pending. Since 1989, there have been 2,865 claims filed for non-autism related injuries and 5,263 claims for autism related injuries across all vaccina tions. Since 1990, only one autism case has been deemed compensable to date and 350 cases have been dismissed (HRSA 2008). The remaining autism-related cases are still aw aiting a ruling. The debate of whether autism is causally linked to vaccine administration and therefore should be compensated under VICP is still ongoing and has a potentially significan t impact on public health. Table 1. National Vaccine Injury Compensa tion Program-Post 1988 Statistics Report of Petitions Filed Fiscal Year Non-Autism Autism Total FY 1989 101 FY 1990 29029 FY 1991 1180118 FY 1992 1860186 FY 1993 1370137 FY 1994 1060106 FY 1995 1790179 FY 1996 84084 FY 1997 1030103 FY 1998 1160116 FY 1999 4051406 FY 2000 1610161 FY 2001 19618214 FY 2002 189768957 FY 2003 1522,4372,589 FY 2004 1261,0881,214 FY 2005 146587733 FY 2006 154169323 FY 2007 238168406 FY 2008 392766 Total 2,865 5,263 8,128
15 Table 2. National Vaccine Injury Compensati on Program Post 1988 St atistics Report for Adjudications of Claims Fiscal Year Non-Autism Autism Total Compensable Dismissed SubTotal CompensableDismissed SubTotal FY 1990 2 0200 02 FY 1991 10 223200 032 FY 1992 30 437300 073 FY 1993 22 577900 079 FY 1994 41 438400 084 FY 1995 48 519900 099 FY 1996 50 7812800 0128 FY 1997 60 5111100 0111 FY 1998 53 7212500 0125 FY 1999 38 7311100 0111 FY 2000 67 7514200 0143 FY 2001 66 7914500 0145 FY 2002 98 9519304 4197 FY 2003 52 74126021 21148 FY 2004 60 1201800113 113293 FY 2005 60 69130051 51181 FY 2006 67 761450107 107249 FY 2007 80 71151032 32183 FY 2008 21 930122 2353 Totals 924 1,158 2,082 1 350 351 2,433 Thimerosal A second hypothesis dealing with vaccine administration and the development of autism contends that thimerosal, a mercury cont aining preservative that has been used in some vaccines and other pharmaceutical product s since the 1930s, is the agent causing autism. Thimerosal was contained in over 30 childhood vaccines including Haemophilus influenzae type b, hepatitis B, and DTaP up until 1999. Thimerosal was never used as a preservative for the MMR vaccine because thim erosal would inactivate the live vaccine. The thimerosal hypothesis gained credibility after FDA researchers determined that the
16 childhood vaccination schedule might expose children to cumulative doses of ethylmercury that exceed some federal safety guidelines (IRSC 2004). In 2001, the Institute of Medicine concluded that the evidence was inadequate to accept or reject a causal relationship between exposure to thimerosal and the development of autism (IRSC 2004). As of March 2001, all formulations of v accines for childhood immunizations are free of thimerosal (IRSC 2004). Healthy People 2010 Healthy People 2010 is a d eclaration of the national he alth objectives designed to identify the most significant preventable threat s to health and to establish national goals to reduce these threats. The two primary goals of Healthy People 2010 are to increase the quality and duration of healthy life and to eliminate health di sparities. There are 28 focus area chapters in Healthy People 2010 and each ch apter contains a concis e goal statement. The goal of Healthy People 2010 pertaining to im munizations and infec tious disease is to prevent disease, disability, and death from infectious diseases, including vaccine preventable diseases (Healthy People 2010). Healthy People 2010 includes specific goa ls and objectives under each chapter outlining goals and methods for improving health. The overall goal of Chapter 14: Immunization and Infectious Diseases, is to reduce or eliminate indigenous cases of vaccine-preventable diseases. Healthy Peopl e 2010 attempts to build upon the progress that was made toward meeting the Healthy People 2000 objectives. As an example, the number of cases of measles decreased from 3,396 indigenous cases in 1988 to only 74
17 cases in 1998 (Healthy People 2010). A pr imary objective under this goal is the elimination of all cases of m easles, mumps, and rubella. The achievement of a high level of c overage for two doses of MMR vaccine makes the goal of eliminating measles, mumps, and rubella feasible. In 1998 the baseline immunization rate for at least one dose of the MMR vaccine was 92%. The Healthy People 2010 goal is to maintain at least 90% coverage for one dose of the MMR vaccine. It is believed that coverage of at least 90% is sufficient to prevent the circulation of viruses and bacteria-causing di sease. However, even with the achievement of 90% coverage levels it is imperative to mon itor subgroups of the population that are under vaccinated. These groups of under vaccinated individuals increase th e possibility of a major outbreak of a vaccine preventable disease. Eliminating potential health disparities is a necessary measure to achieving the goal of eliminating all cases of measles, mumps, and rubella (Healthy People 2010). Another objective of the chapter on immuni zation and infectious diseases from Healthy People 2010 that builds upon Healthy People 2000 is to maintain vaccination coverage levels for children in licensed da y care facilities and children in kindergarten through the first grade. The goal for the MMR vaccine is 95% for children in day cares and children in kindergarten to first grade. As measured in Healthy People 2000, the individual coverage rates for three or more doses of polio, three or more doses of diphtheria/tetanus/acellular pert ussis (DTaP), one or more doses of MMR and three or more doses of Haemophilus influenzae type b vaccines were each at or above 91 percent (Health People 2010). Increasing the coverage rate to 95% for all children entering a
18 licensed day care facility or en tering school represents a signifi cant step to eliminating all cases of vaccine preventable diseases. Healthy People 2010 also outlines strate gies to achieve the outlined goals and objectives. Major strategies to protect people from vaccine preventable diseases include: Improving the quality and quantity of vaccination delivery services, Minimizing financial burdens for needy persons, Increasing community participa tion, education and partnership, Improving monitoring of disease and vaccination coverage rates, and Developing new or improved vaccine s and improving vaccine use. Programs such as Vaccines for Children and SCHIP have made vaccinations available for children. These programs along with the requirements of vaccinations for entry into day care and schools have increased the vaccination rate (Healthy People 2010). Improvements in the vaccination rate not only extend benefits to those that are vaccinated, benefits of vaccinati on are gained by society as a whole. In cases where a few people cannot be vaccinated, if the vaccination level in the community is high these people are protected due to group or herd immunity, whereby the unvaccinated group has a decreased risk of cont racting a disease b ecause all those around them have been vaccinated. Achievement of the Healthy People 2010 MMR vaccination goals will go a long way toward eradicating meas les, mumps, and rubella. It is hypothesized that administration of the MMR vaccine is causally related to the development of autism. The hypothe sis was first proposed by Wakefield and associates (Wakefield1998) following a case stu dy of 12 children with gastrointestinal disorders and developmental regression. In 8 of the 12 children, the parents or the
19 physician retrospectively li nked the onset of the beha vioral problems with the administration of the MMR vaccine. While the authors concluded that there was insufficient evidence in their study to support a causal relationship, th e study generated a hypothesis warranting further resear ch. The purpose of this research is to evaluate if a causal relationship exists between childhood vaccination with MMR and the development of autism by applying the sta tistical methods of a meta-analysis of published primary literature.
20 Methods The primary objectives of this analysis are to: 1. Assess the causal relationship between th e MMR vaccine and the development of childhood autism, and 2. Discuss the impact of the association of the MMR vaccine and the development of autism on the ACIP vaccination guid elines and the Healthy People 2010 goal for immunizations against preventable diseases. A quantitative meta-analysis was conduc ted to assess the causal relationship between the MMR vaccine and the development of autism. A comprehensive literature search of PubMed and Medline assessing th e relationship between the MMR vaccine and the development of autism during the years 1998 to 2007 was conducted. Key search terms for the systematic literature review included: autism, autistic, MMR, measles, mumps, and rubella alone or in combination. The search was limited to articles published in the English language. A total of 29 studies were identified. The following study inclusion criteria were employed on the 29 studies. 1. Appropriate study design (e.g. prospective or retrospective case-control or cohort study 2. Study must contain a control group 3. Study endpoints must report or allow for th e calculation of the effect measure and a standard error or confid ence interval for the association of MMR vaccination and the development of autism (odds ratio or relative risk)
21 4. Experimental group must have received MMR vaccination 5. Adequate sample size (n 100) The following study exclusion cr iteria were employed: 1. Study does not have an unvaccinated comparison group (e.g. case report/case series) 2. Study endpoints do not allow for the calculati on of an odds ratio or relative risk 3. Study exposure does not incl ude the MMR vaccination 4. Study is not independent of another study in cluded in the analysis (e.g. re-analysis of the same study population) After applying the inclusion/exclusion criteria seven studies remained for inclusion in the meta-analysis (see Appendix B). Statistical analyses combining and interpreting the resu lts of independent studies were c onducted to integrate the results of several studies and arrive at a conclusion on the causal association of the MMR vaccine and autism. The patient population included children (<18 years) who have received the MMR vaccine under a normal vaccination schedule according to ACIP guidelines or appropriate guidelines for their country. Study Endpoints will include: Effect measure o Odds ratio/relative risk for each individual study o Pooled effect for the associati on between MMR vaccination and the development of autism Meta-regression
22 o Meta-regression to assess the impact of continuous predictors (e.g. age at vaccination, number of vaccinations) The odds ratio of the association between the MMR vaccine and the development of autism was measured for each individual study as the effect measure. The pooled effect measure was then calculated based on th e weighted average of the effect measures seen in each study. The weight for each study was determined based on the width of the 95% confidence interval for th e effect measure of each st udy. The meta-analysis was conducted using both a fixed and a random effe cts model. A sensitivity analysis was conducted to assess the impact of individua l studies on the pooled effect measure. Publication bias was assessed using a funnel plot and the trim and fill technique. Data analysis will be conducted using Comprehe nsive Meta Analysis (CMA) Version 2.0, a computer program for meta analysis develo ped by a team of experts from the US and U.K. and funded by the US Nati onal Institutes of Health.
23 Results Descriptive Statistics Seven out of 29 studies met all inclusion/ exclusion criteria. Fi fteen studies were excluded due to the lack of an unvaccinate d control group (Studies 1, 2, 3, 5, 6, 9, 12, 13, 15, 16, 17, 21, 22, 27, and 28). Three studies were excluded for the lack of MMR exposure in the study (Studies 7, 11, and 25). Three studies were excluded because they did not allow for the calculation of an eff ect size (Studies 18, 19, and 29). One study was excluded because it was a seconda ry analysis of a previous st udy that was included in the meta-analysis (Study 4). The following seven studies were incl uded in the meta-analysis: Madsen 2002, Honda 2005, Smeeth 2004, Taylor 1999, Uchi yama 2007, Goldman 2004 and Taylor 2002. Table 3. Synopsis of Included Studies Study Design Country Years Dependent Variable Madsen 2002 Retrospective cohort study Denmark 1991-1998 Autism or ASD Honda 2005 Cohort study Japan 1988-1996 ASD Smeeth 2004 Case-control study United Kingdom 1987-2001 Autism or ASD Taylor 1999 Epidemiological study United Kingdom 1979-1992 Autism Uchiyama 2007 Epidemiological study Japan 1976-1999 ASD Goldman 2004 Retrospective cohort study Denmark 1980-1992 Autism Taylor 2002 Population study United Kingdom 1979-1992 Autism
24 Madsen 2002 was a retrospective cohort study of all children born in Denmark including a total of 537,303 children. Vaccination status was obtained from the National Board of Health and information on autis m status was obtained from the Danish Psychiatric Central Register. A total of 316 children were diagnosed with autism and 422 with an autism spectrum disorder. The re lative risk for autistic disorder in the unvaccinated group compared to the vaccinated group was 0.919. Honda 2005 was a cohort study of childre n up to age seven born between 1988 and 1996 in Kohoku Ward of Japan. All children born in 1993 or after did not receive a single vaccination of MMR. The cumulativ e incidence of autism spectrum disorders increased significantly in the birth cohorts from 1988 through 1996 with the most notable increase beginning with the 1993 birth cohort. The odds ratio for the association between MMR and autism was 1.921. Smeeth 2004 was a matched case-control st udy using the U.K. General Practice Research Database. Cases were people born in 1973 or later who were diagnosed with a pervasive developmental disorder, between 1987 and 2001. Controls were matched on age, sex, and general practice. A total of 1294 cases and 4469 controls were included in the study. The odds ratio for the association of MMR and autism was 0.880. Taylor 1999 was an epidemiological study of children from eight North Thames health districts in the U.K. Information from clinical records of the children was linked to immunization data held on the child health computing system. Any clustering of onset of autism in post vaccination periods were investigated us ing the case series method. A
25 total of 498 cases of autism were identified. The odds ratio for the association of MMR and the development of autism was 0.940. Uchiyama 2007 was an epidemiological study of the association between MMR vaccination and the development of autism in Japan. Since the MMR vaccine was only used in Japan between 1989 and 1993, this time period affords a natural experiment for MMR and autism. The study included an an alysis on 904 patients with autism spectrum disorder. The analysis included cohorts fr om three periods: before, during, and after MMR usage. There were no significant differe nces in the incidence of autism in the cohort that received MMR and those who had not with an odds ratio of 0.744. Goldman 2004 investigated the prevalen ce of autism by age category from 1980 to 2002 using a nationwide computerized regist ration system with the Danish Psychiatric Central Register. The MMR vaccine was added to the immunization schedule of Denmark in 1987. Linear regression analysis was used to investigate compare the periods preceding and following the introduc tion of the MMR vaccine. Longitudinal trends in the prevalence data suggest a tem poral association between the introduction of MMR vaccine and the rise in autis m with an odds ratio of 4.700. Taylor 2002 was a population study with case note review of five health districts in north east London. The study population incl uded 278 children with core autism and 195 with atypical autism identified from co mputerized disability registers born between 1979 and 1998. The odds ratio for the associ ation of MMR and th e development of autism was 0.980.
26 Meta-Analysis Results Two of the seven studies included in the meta-analysis suggested a causal association between MMR and autism and five studies did not suggest a causal association. The combined effect measur e using a fixed effects model was 1.052 (95% CI: 0.973, 1.138; p value= 0.202). The weights a ssigned for the studies based on the size of the confidence intervals under a fixed model were: 6.99% for Madsen 2002, 5.22% for Honda 2005, 8.35% for Smeeth 2004, 3.04% for Taylor 1999, 1.13% for Uchiyama 2007, 3.43% for Goldman 2004, and 71.84% for Ta ylor 2002. When using a random effects model, the combined effect measure was 1.267 (95% CI: 0.879, 1.827; p value=0.204). In the random effects model the studies weig hts were: 15.31% for Madsen 2002, 14.81% for Honda 2005, 15.56% for Smeeth 2004, 13.55% for Taylor 1999, 10.08% for Uchiyama 2007, 13.87% for Gold man 2004, and 16.82% for Taylor 2002. Figure 1. Meta-analysis re sults: Autism and MMR Study name Statistics for each study Odds ratio and 95% CI Odds Lower Upper ratiolimitlimitZ-Valuep-Value Madsen 20020.9190.6841.234-0.5630.574 Honda 20051.9211.3652.7043.7430.000 Smeeth 20040.8800.6721.153-0.9280.354 Taylor 19990.9400.6011.471-0.2710.787 Uchiyama 20070.7440.3571.551-0.7890.430 Goldman 20044.7003.0847.1637.1990.000 Taylor 20020.9800.8941.075-0.4300.667 1.0520.9731.1381.2760.202 0.01 0.1 1 10 100 No AssociationAssociationMeta Analysis
27 Figure 2. Meta-analysis re sults: Funnel Plot -2.0-1.5-1.0-0.50.00.51.01.52.0 0.0 0.1 0.2 0.3 0.4 Standard ErrorLog odds ratioFunnel Plot of Standard Error by Log odds ratio Sensitivity Analysis The meta-analysis was conducted rem oving the outlier study, Goldman 2004, in the sensitivity analysis. The combined effect measure for the 6 studies excluding Goldman 2004 was 0.998 (95% CI: 0.921, 1.080; p value=0.953) when using a fixed effects model. When using a random effect s model the combined effect measure was 1.035 (95% CI: 0.835, 1.281; p value=0.755). The f unnel plot assessing publication bias revealed no publication bias as the studies were symmetrical. Most studies assessing the association between MMR and the developm ent of autism fall in the range of no association. The Goldman 2004 and Honda 2005 studies fall outside of the funnel, indicating that they were outliers.
28 Discussion The results of this meta-analysis did not find a causal asso ciation between MMR vaccination and the development of autism. The combined fixed effect measure of 1.052 includes a 95% CI from 0.973 to 1.138 that cr osses one, thus making the results nonsignificant. The same scenario occurs when using a random effects mo del with an effect measure of 1.267 and a 95% CI from 0.879 to 1.827. In both a fixed and random effects model the meta-analysis did not find a causa l association between MMR vaccination and the development of autism. Due to the sm all number of studies included the study may not have had adequate power to de tect a difference if one existed. The two studies included in the meta-a nalysis that supporte d an association between MMR vaccination and the development of autism are potentially biased by time. Goldman 2004 examines the prevalence of autism by age category from 1980 to 2002 using a computerized registration system in Denmark. The study provided a comparison of prevalence rates pre and post licensure of the MMR vaccine. Pr evalence rates in the post licensure period may have been much highe r than in the pre licensure period due to greater diagnostic awareness and an overall increasing prevalence of autism. Studies such as Dales 2001 and Kaye 2001 have found an increasing prevalence of autism over time. Dales 2001 was a time tr end analysis of the association of autism and MMR vaccination coverage in California. Dales 2001 showed a sustained increase in autism cases from 44 cases per 100,000 bi rths in 1980 to 208 cases per 100,000 live births in 1994 with minimal change in MMR vaccination rates. The changes in the
29 prevalence rates seen in this study provide evidence that MMR vaccination was not the cause of the increased prevalence of autis m. Similarly, Kaye 2001 was a time trend analysis of autism prevalence in the UK fr om 1988 to 1999. The incidence of autism increased sevenfold from 1988 to 1999 with th e vaccination rate st aying above 95% over the entire timeframe. This suggests that so me other factor was the root cause of the increasing prevalence of autism and not MMR vaccination. The association found in Honda 2005 may ha ve been influenced by the same time factor bias as Goldman 2004. In Honda 2005, the cumulative incidence of autism spectrum disorders were examined from 1988 to 1996 in the Kohoku ward of Japan. The population was divided into vaccinated and unvaccinated populations based on MMR vaccination being discontinued in 1993. The association that was found may have been influenced by time with the two cohorts being determined by the time when MMR vaccination was available. There are several possible explanations for the apparent increase in autism prevalence. Changes in the prevalence of autism may be influenced by: substantial migration of affected children into or out of a community, change in age of onset or recognition, large changes in denominator population, increased ascertainment of children with diagnoses of autism, a change in diagnostic criteria to include individuals with milder symptoms, and a true increase in the incidence of the disease, which can be attributable to new environmental exposures (Halsey 2001). Studies to determine the prevalence of autism are very susceptible to bias and therefore pr evalence rates in the literature vary considerably. The increased attention and better understanding of autism
30 has likely led to a greater di agnostic awareness accounting fo r a portion of the increased prevalence. Vaccination rates and vaccine safety are ma jor issues of concern for public health. An adequate vaccination history is routinel y a requirement for school entry. Reasonable concerns about the safety of childhood vacci nes would require an examination of the vaccine schedule and the legality of requiring children to be vaccinated before entering school. The evidence in this case did not find a causal association between MMR vaccination and the development of autism. As such, public health initiatives should focus on achieving the Healthy People 2010 goals of eliminating vaccine preventable diseases. Measles, mumps, and rubella pose significant risks to the health of unvaccinated children. Media attention and pa rental concern about the safety of the MMR vaccine has led some parents to wit hhold the vaccination from their child. Disseminating evidence on the safety of the MMR vaccine should be a major public health initiative. Educating parents on the sa fety of the vaccine and the potential dangers of measles, mumps, and rubella is the first step to increasing the vaccination rate and eventually eliminating th ese three diseases. In order to adequately address the prob lem of autism more information is needed on the root cause of the disease. Areas that require further study include factors associated with autism including genetics and environmental exposure in utero and during the first few months after birth, or temporally associated with the onset of symptoms (Halsey 2001). One of the primary reasons that MMR vaccination was believed to be a cause of autism was the close temporal relationship between MMR vaccination and the onset of symptoms. A ccording to the Recommended Vaccination
31 Schedule (Table 1), the MMR vaccine is give n at 12-15 months, which coincides with the typical onset of autism. During the 12-15 months of age period, children are receiving several vaccines that may have some role in the development of autism. Additionally, presentation with autism at this early age w ould suggest that there may be genetic or in utero factors that are causally associated with autism development. Weiss and associates reported an association between microde letion and microduplica tion on chromosome 16p11.2 (Weiss 2008). This novel microdele tion and microduplication carries a substantial susceptibility to autism, but it only accounts for approximately 1% of all autism cases. The identification of this one genetic factor respons ible for 1% of all autism cases may be just th e beginning of uncovering the genetic link to autism. Another area for future research is th e incidence of regression in people with autism. Wakefield and associates firs t hypothesized that the MMR vaccine was associated with a regressive phenotype of autism (Wakefield 1998). Further study on the nature and incidence of people with reg ressive autism may show a certain group of individuals who are genetically predisposed or have some other factor that triggers regressive autism. It may be possibl e that the MMR vaccine or some other environmental exposure around the time of MMR vaccination may trigger the onset of regressive autism. There is the potential that a previously unrecognized in fectious agent may affect persons with autism. A close temporal re lationship between MMR vaccination and onset of autism symptoms led to the proposed causa l association. In th e 12-15 months of life before MMR vaccination, children are exposed to several potentially infectious agents. Like previous infectious diseases such as HIV, it may be years before a causative
32 pathogen is identified. It is vital that continued scientific efforts be directed to the identification of the cause or causes of autism an d ways to avoid this debilitating disease. The strengths of this study include population-based data, inclusion of an unvaccinated control group, and large total sample size. Limitations of the study include small number of studies included, retrospe ctive data collecti on for some studies, differences in the dependent variable, differen ces in vaccination policies across countries, recording of data may have been incomplete or inaccurate, and the in ability to control for potential confounders such as time, greater di agnostic awareness and age at vaccination. The quality of the meta-analy sis is limited by the limitations of each individual study. Conclusion This study does not support a causal association between the MMR vaccine and the development of autism. Public health ini tiatives are needed to promote the safety of the MMR vaccine and to increase the vaccination rate. Increa sing the vaccination rate for MMR will generate progress toward elimin ating measles, mumps, and rubella as childhood diseases. Further epidemiological re search on the root cau se of autism should focus on a genetic link or environmental factors in the early stages of life.
33 References CDC. Measles, Mumps, Rubella-Vaccine Use and Strategies for Elimination of Measles Mumps, Rubella, and Congenital Rubella Syndrome and Control of Mumps: Recommendations of the Advisory Comm ittee on Immunization Practices (ACIP). MMWR 1998;47(RR-8): 1-57. CDC. Recommended Immunization Schedule for Pe rsons Aged 0-6 YearsUnited States 2008. Available at: www.cdc.gov/vaccines/recs/schedules Dales L, Hammer SJ, Smith NJ. Time trends in autism and MMR immunization coverage in California. JAMA 2001;285(9): 1183-1185. Goldman GS, Yazbak FE. An investigation of the association between MMR vaccination and autism in Denmark. Journal of American Physicians and Surgeons 2004;9(3): 70-75. Halsey NA, Hyman SL for the Conference Writing Committee. Measles-mumps-rubella vaccine and the autistic spectrum disorder: report from the New Challenges in Childhood Immunizations Conference convened in Oak Brook, Illinois, June 12-13, 2000. Pediatrics 2001;107(5). Health Resources and Services Administra tion (HRSA). Vaccine Injury Compensation Program. January 2008. Available at: www.hrsa.gov/vaccinecompensation Healthy People 2010. Vol. 1, Ch 14: Immunizati ons and Infectious Diseases. Centers for Disease Control and Prevention. Available at: www.healthypeople.gov Honda H, Shimuzu Y, Rutter M. No effect of MMR withdrawal on the incidence of autism: a total population study. J Child Psychol Psychiatry 2005;46(6): 572-579. Immunization Safety Review Committee (ISRC) Immunization Safety Review: Vaccines and Autism. National Academies Press, 2004; Washington D.C. Kaye JA, Melero-Montes M, Jick H. Mump s, measles, and rubella vaccine and the incidence of autism recorded by genera l practioners: a time trend analysis. BMJ 2001;322: 460-463. Madsen KM, Hviid A, Vestergaard M, et al. A population based study of measles, mumps, and rubella vaccination and autism. N Eng J Med. 2002;347(19): 1477-1482.
34 M-M-R II [prescribing information]. Merck & Co., Inc. Whitehouse Station, NJ 08889. December 2007. National Institute of Neurological Disorders and Stroke (NINDS). Autism Fact Sheet. NIH Publication No. 06-1877, April 2006. Available at: www.ninds.gov/disorders/autism/detail_autism.htm Smeeth L, Cook C, Fombonne E, et al. MMR vaccination and perv asive developmental disorders: a case-control study. Lancet 2004;364: 963-969. Strock M for the National Institute of Mental Health. Autism spectrum disorders (Pervasive Developmental Di sorders). Department of Health and Human Services, National Institutes of Health. February 2007. Available at: www.nimh.nih.gov Taylor B, Miller E, Farrington CP, et al. Autism and measles, mumps, rubella vaccine: no epidemiological evidence for a causal association. Lancet 1999;353: 2026-2029. Taylor B, Miller E, Lingam R, et al. Meas les, mumps, and rubella vaccination and bowel problems or developmental regression in children with autism: population study. BMJ 2002;324: 393-396. Uchiyama T, Kurosawa M, Inaba Y. MMR vaccination and regression in autism spectrum disorders: negative re sults presented from Japan. J Autism Dev Disord 2000;37: 210-217. Wakefield AJ, Murch SH, Linnell J, Ca sson DM, et al. Illeal-lymphoid-nodular hyperplasia, non-specific colitis, and pervas ive developmental disorder in children. Lancet 1998;351: 637-641. Weiss LA, Shen Y, Korn JM, et al. Association between microdeletion and microduplication at 16p11.2 and autism. N Engl J Med. 2008;358.
35 Bibliography DeFastano F, Bhasin TK, Thompson WW, et al. Age at first measles-mumps-rubella vaccination in children with autism and school-matched control subjects: a populationbased study in metropolitan Atlanta. Pediatrics 2004;113: 259-266. DeWilde S, Carey I, Richards N, et al. Do children who become autistic consult more often after MMR vaccination? Br J Gen Pract 2001;51: 226-227. Farrington CP, Miller E, Taylor B. MMR and autism: further evidence against a causal association. Vaccine 2001;19: 3632-3635. Fombonne E, Zakarian R, Bennett A, et al Pervasive developmental disorders in Montreal, Quebec, Canada: prevalen ce and links with immunizations. Pediatrics 2006;118: e139-e150. Geier DA, Geier MR. Nuerodevelopmental di sorders following thimerosal-containing childhood immunizations: a follow-up analysis. Int J Toxicol. 2004;23(6): 369-376. Geier DA, Geier MR. A comparative evalua tion of the effects of MMR immunization and mercury doses from thimerosal-cont aining childhood vaccines on the population prevalence of autism. Med Sci Monit 2004;10(3): P133-P139. Hilton S, Hunt K, Petticrew M. MMR: marginalized, misr epresented and rejected? Autism: a focus group study. Arch Dis Child 2007;92: 322-327. Hviid A, Stellfeld M, Wohlfahrt J, Mel bye M. Association between thimerosalcontaining vaccine and autism. JAMA 2003;290(13): 1763-1766. Madsen KM, Vestergaard M. MMR vaccination and autism: What is the evidence for a causal association? Drug Safety 2004;27(12): 831-840. Makela A, Nuorti JP, Peltola H. Neurologi c disorders after measles-mumps-rubella vaccination. Pediatrics 2002;110: 957-963. Parker SK, Schwartz B, Todd J, Picker ing LK. Thimerosal-containing vaccines and autism spectrum disorder: a critical review of published original data. Pediatrics 2004;114: 793-804. Patja A, Davidkin I, Kurki T, et al. Serious adverse events after measles-mumps-rubella vaccination during a fourteen-year follow-up. Pediatr Infect Dis J 2000;19(12): 11271134.
36 Richler J, Luyster R, Risi S, et al. Is th ere a regressive phenotype of autism spectrum disorder associated with the measles-mumps-rubella vaccine? A CPEA study. J Autism and Dev Disord. 2006;36(3): 299-316. Smeeth L, Hall AJ, Fombonne E, et al. A case-control study of autism and measlesmumps-rubella vaccination using the genera l practice research database: design and methodology. BMC Public Health 2001; 1:2. Takahashi H, Suzumura S, Shikakizawa F, et al. An epidemiological study of Japanese autism concerning routine childhood immunization history. Jpn J Infect Dis 2003;56: 114-117. Taylor B. Vaccines and the chan ging epidemiology of autism. Child Care Health Dev 2006;32(5): 511-519. Thompson WW, Price C, Goodson B, et al. Early thimerosal exposure and neuropsychological outcome s at 7 to 10 years. N Engl J Med 2007;357(13): 1281-1292. Wilson K, Mills E, Ross C, et al. Associa tion of autistic spectrum disorder and the measles, mumps, and rubella vaccine. Arch Pediatr Adoles Med 2003;157: 628-634. Woo EJ, Ball R, Bostrom A, et al. Vaccine risk perception among reporters of autism after vaccination: vaccine advers e event reporting system 1990-2001. Am J Public Health 2004;94(6): 990-995.
38 Appendix A Recommended Immunization Schedule Table 4. Recommended Immuniza tion Schedule for Ages 0-6 Vaccine Birth 1 month 2 months 4 months 6 months 12 months 15 months 18 months 19-23 months 2-3 years 4-6 years Hepatitis B Hep B Hep B Hep B Rotavirus Rota Rota Rota Diphtheria, Tetanus, Pertussis DTaP DTaP DTaP DTaP Haemophilus influenzae type b Hib Hib Hib Hib Pnuemococcal PCV PCV PCV PCV PPV Inactivated Poliovirus IPV IPV IPV Influenza Yearly Measles, Mumps, Rubella MMR MMR Varicella Varicella Hepatitis A HepA (2 doses) Hep A Series Meningococcal MCV4 Key: = Range of recommend ages = Certain high-risk groups Adapted from: Recommended Immunization Schedule for Persons Aged 0-6 YearsUnited States 2008. Approved by the Advisory Commit tee on Immunization Practices, the American Academy of Pediatrics and the Am erican Academy of Family Physicians. Available at: www.cdc.gov/vaccines.recs/shedules
39 Appendix B Research Review Table 5. Research Review Citation Design Information Outcome Measures Results Key Assumptions/ Limitations Inclusion/ Exclusion 1. Dales L, Hammer SJ, Smith NJ. Time trends in autism and MMR immunization coverage in California. JAMA 2001;285(9): 11831185. Objective: To determine if a correlation exists in secular trends of MMR immunization coverage among young children and autism occurrence Methods: Retrospective analysis of MMR immunization coverage rates in children in California kindergartens and of autism caseloads among children who were enrolled in the California Department of Development Services MMR immunization coverage rates as of 17 months and 24 months and numbers of autism cases, grouped by year of birth No correlation was found; sustained increase in autism cases, while changes in immunization coverage where much smaller and of shorter duration Immunization at 17 months used because this is just before the mean age when parents first notice developmental disorders Immunization and autism records on the same individuals could not be linked Excluded Autism and immunization records not linked to be able to determine a control group
Appendix B (Continued) 40 2. DeFastano F, Bhasin TK, Thompson WW, et al. Age at first measles-mumpsrubella vaccination in children with autism and school-matched control subjects: a populationbased study in metropolitan Atlanta. Pediatrics 2004;113: 259266. Objective: To compare age at first MMR vaccination between children with autism and children who did not have autism Methods: Casecontrol study using vaccination data abstracted from immunization forms for school entry Odds Ratio determined by conditional logistic regression No significant associations for any age cutoff were found for specific case subgroups; More cases than control were vaccinated before 36 months (OR 1.49; OR 1.23 in the birth certificate cohort) MMR vaccine increases the risk of autism, which usually develops before 24 months of age, if so then children vaccinated at younger ages would have a higher risk of developing autism Compared cases and controls because they did not have an unvaccinated group and had incomplete information for determining the date of onset of autism Excluded -Study did not have a control group that was unvaccinated
Appendix B (Continued) 41 3. DeWilde S, Carey I, Richards N, et al. Do children who become autistic consult more often after MMR vaccination? Br J Gen Pract. 2001;51: 226227. Objective: To assess changes in parental concern as measured by a change in consultation behavior following MMR vaccination Methods: Casecontrol study using the Doctors Independent Network database to count the number of consultations Number of consultations before and after the MMR vaccination No significant differences in the numbers of consultations in the 6 months before and two months after MMR between cases and control Cases of autism could not be confirmed Cases and controls both received the MMR vaccination Excluded -Study does not have an unvaccinated control group 4. Farrington CP, Miller E, Taylor B. MMR and autism: further evidence against a causal association. Vaccine 2001;19: 3632-3635. Objective: To test the hypothesis that MMR vaccination causes autism without prespecifying any fixed time interval after vaccination in which the risk of autism might be increased Methods: Selfmatched case series study Relative incidence for autism diagnosis or regression following MMR vaccination In all instances the relative incidences did not differ significantly from 1, indicating no association between vaccination and autism in the subsequent risk periods Study includes data on all MMR vaccines, including those given later than the recommended schedules Excluded -Second analysis of same study population; follow-up analysis of Taylor et al. (Lancet 1998)
Appendix B (Continued) 42 5. Fombonne E, Zakarian R, Bennett A, et al. Pervasive developmental disorders in Montreal, Quebec, Canada: prevalence and links with immunizations. Pediatrics 2006;118: e139e150. Objective: To estimate the pervasive developmental disorders in Montreal from 1987 to 1998 and to evaluate the relationship with changes in cumulative exposure to ethylmercury (thimerosal) and trends in MMR vaccination use rates Methods: Cohort study of children born between 1987 and 1996 using surveys of vaccination rates and thimerosal exposure Odds Ratio for thimerosal exposure Odds Ratio for MMR vaccination use rates No significant effects of thimerosal exposure on the risk of developing a pervasive developmental disorder (OR 1.39 for exposure free vs. exposure); No association with MMR, as pervasive developmental disorders increased while MMR vaccination rates decreased (OR 1.10) Children with a PPD diagnosis were identified by school officials and this diagnosis was not verified by direct assessment Individual immunization data was not available Excluded -Study did not have an unvaccinated control group as individual immunization data was not available 6. Geier DA, Geier MR. A comparative evaluation of the effects of MMR immunization and mercury doses from thimerosalObjective: To evaluate the effects of MMR immunization and mercury from thimerosal-containing vaccines on the prevalence of autism Methods: Linear regression coefficients -Prevalence of autism in relation to average mercury doses -Number of doses of primary pediatric Close correlation between mercury doses from thimerosal-containing vaccines and prevalence of autism; Potential correlation between MMR vaccines and the Examined cohorts of children and did not look at individuals therefore can only detect large universal effects Evidence to support association with MMR vaccine Excluded -Study did not have an unvaccinated control group
Appendix B (Continued) 43 containing childhood vaccines on the population prevalence of autism. Med Sci Monit 2004;10(3): P133-P139. Retrospective cohort study of the Biological Surveillance Summaries of the CDC, the US Department of Education dataset and the CDCs yearly live birth estimates MMR vaccine to the prevalence of autism Odds Ratios from average increased mercury exposure received from thimerosalcontaining vaccines prevalence of autism is lacking (OR are not reported) 7. Geier DA, Geier MR. Nuerodevelopme ntal disorders following thimerosalcontaining childhood immunizations: a follow-up analysis. Int J Toxicol. 2004;23(6): 369376. Objective: To determine if the previously observed effect between thimerosal-containing childhood vaccines and neurodevelopmental disorders are still present after children have had time to further mature Methods: Cohort study of children receiving thimerosal-containing vaccines to those not receiving thimerosalcontaining vaccines from 1997 to 2000 Odds ratio for development of neurodevelopmental disorders Significantly increased OR for developmental disorders for autism (OR 1.8), mental retardation (OR 2.6), speech disorder (OR 2.1), personality disorder (OR 2.6) and thinking abnormality (OR 8.2) Limitations of the Vaccine Adverse Events Reporting Dataset include: underreporting, erroneous reporting, frequent multiple exposures, multiple outcomes, and lack of precise denominators Study of DTaP vaccine and not MMR Excluded -Study exposure does not include MMR vaccination
Appendix B (Continued) 44 8. Goldman GS, Yazbak FE. An investigation of the association between MMR vaccination and autism in Denmark. Journal of American Physicians and Surgeons 2004;9(3): 7075. Objective: To compare the prevalence of autism in the periods preceding an following the introduction of the MMR vaccine in Denmark Methods: Cohort study using data from nationwide computerized registration system in Denmark Linear regression on the prevalence of autism by age cohorts Relative Risk of autism (adjusted for greater diagnostic awareness) RR of autism following introduction of MMR vaccine is 8.5 and adjusted risk is 4.7; suggests a temporal association between the introduction of the MMR vaccine and the rise in autism The 20-24 year old cohort potentially could have received the monovalent vaccine as toddlers and the MMR booster dose when it became available Adjusted RR is highly sensitive to the increasing trend during 3 years, 1990 to 1992 Included 9. Hilton S, Hunt K, Petticrew M. MMR: marginalized, misrepresented and rejected? Autism: a focus group study [abstract]. Arch Dis Child 2007; 92: 322-327. Objective: To explore how the MMR vaccine controversy impacted the lives of parents caring for children with autism Methods: Qualitative focus group study conducted in the UK Parents perception of the MMR vaccine and their response to the controversy surrounding the vaccine and autism Many parents felt the vaccine could be too potent for children who are susceptible to developing autism; Many felt guilty that they may have contributed to their childs autism Nonexperimental design (focus group study) Excluded -Study does not have an unvaccinated control group (nonexperimental design)
Appendix B (Continued) 45 10. Honda H, Shimizu Y, Rutter M. No effect of MMR withdrawal on the incidence of autism: a total population study. J Child Psychol Psychiatry 2005;46(6): 572579 Objective: To compare autism frequency before and after the termination of the MMR vaccination program Methods: Analysis of annual trends in autism spectrum disorder incidence in Japan MMR vaccination rates and the corresponding incidence of autism cases Risk factor analysis using logistic regression The incidence of autism increased in birth cohorts both before and after MMR vaccinations were discontinued; statistically significant increase in autism incidence after the discontinuation of the MMR vaccination program In Japan children only receive one dose of MMR as opposed to the initial vaccination and a booster shot as is the custom in most countries Provides epidemiological data on the effect of the removal of the MMR vaccine (hypothesized offending agent) Included 11. Hviid A, Stellfeld M, Wohlfahrt J, Melbye M. Association between thimerosalcontaining vaccine and autism. JAMA 2003; 290(13): 17631766. Objective: To determine whether vaccination with a thimerosal-containing vaccine is associated with development of autism Methods: Cohort study of all children born in Denmark from 1990 to 1996 comparing children vaccinated RR for autism and other autisticspectrum disorders, including trend with dose of ethylmercury Risk of autism did not differ significantly between children vaccinated with thimerosalcontaining vaccines and those without (RR 0.85 for autism; RR 1.12 for autisticspectrum disorders) No evidence of a dose response (RR Examined children vaccinated with a thimerosalcontaining formulation of the pertussis vaccine compared with pertussis vaccine without thimerosal Excluded -Study exposure does not include MMR vaccination
Appendix B (Continued) 46 with a thimerosalcontaining vaccine and a thimerosal-free vaccine formulation 0.98) 12. Kaye JA, Melero-Montes M, Jick H. Mumps, measles, and rubella vaccine and the incidence of autism recorded by general practioners: a time trend analysis. BMJ 2001;322: 460463. Objective: To estimate the changes in the risk of autism and assess the relation of autism to the MMR vaccine Methods: Time trend analysis of data from the UK general practice research database (GRPD) Annual and age specific incidence for first recorded diagnosis of autism and coverage rates for the MMR vaccine Incidence of autism increased sevenfold in the birth cohorts, while the vaccine coverage rate remained above 95%; Suggests no association between MMR vaccine and autism incidence Initially intended to do a case-control study, but only 3% of cases and controls were not vaccinated Did not obtain and evaluate full clinical record information from general practioners to describe more fully the characteristics of children diagnosed with autism Excluded -Study does not contain an unvaccinated control group
Appendix B (Continued) 47 13. Madsen KM, Vestergaard M. MMR vaccination and autism: What is the evidence for a causal association? Drug Safety 2004;27(12): 831-840. Objective: To study the evidence on MMR vaccines and the occurrence of autism with particular focus on the epidemiological literature Methods: Literature review Case series Case-control studies Ecological studies Crosssectional studies Cohort studies Review of the literature does not support a causal relationship between MMR vaccines and autism; No primates models exist and the biologic plausibility remains questionable Literature review that identified 10 epidemiological studies with none reporting a causal association Excluded -Study does not contain an unvaccinated control group 14. Madsen KM, Hviid A, Vestergaard M, et al. A population-based study of measles, mumps, and rubella vaccination and autism. N Eng J Med. 2002;347(19): 1477-1482. Objective: To evaluate the hypothesized association of MMR vaccine and autism Methods: Retrospective cohort study of all children born in Denmark from January 1991 to December 1998 Relative risk of autistic disorder The RR of autistic disorder was 0.92 and the RR of another autismspectrum disorder was 0.83; strong evidence against a causal association Exposure data was collected prospectively, independent of parental recall and before the diagnosis of autism Assumed that the data on MMR vaccination is complete because practioners in Denmark are reimbursed only after reporting immunizations to the National Board of Health Included
Appendix B (Continued) 48 15. Makela A, Nuorti JP, Peltola H. Neurologic disorders after measles-mumpsrubella vaccination. Pediatrics 2002;110: 957963. Objective: To assess whether an association prevails between MMR vaccination and encephalitis, aseptic meningitis and autism Methods: Retrospective cohort study of individual MMR vaccination data from a hospital discharge register of children vaccinated between November 1982 and June 1986 Total number of hospitalizations for autism after MMR vaccination during the study period Number of events within 3 month risk intervals post-vaccination compared with the number expected for encephalitis and aseptic meningitis Did not detect any clustering of hospitalizations for autism after vaccination; Did not identify any association between MMR vaccination and encephalitis, aseptic meningitis, and autism Follow-up was extended to the end of the study for each child vaccinated because of the undefined latency of disease Authors did not have access to outpatient records and not all autism cases are treated as inpatients Exact incidence of autism could not be determined because autism develops subtly over long periods of time and some patients may be effected at birth but not symptomatic until later in life Excluded -Study did not have an unvaccinated control group 16. Parker SK, Schwartz B, Todd J, Pickering LK. Thimerosalcontaining vaccines and Objective: To assess the quality of evidence assessing a potential association between thimerosal-containing vaccines and autism 10 epidemiological studies and 2 pharmacokinetic studies of ethylmercury Studies do not demonstrate a link between thimerosalcontaining vaccines and autism, and the pharmacokinetics of ethylmercury make Results of epidemiological studies have several inherent limitations such as: differences in study population, multiple potential Excluded -Study does not have an unvaccinated control group
Appendix B (Continued) 49 autism spectrum disorder: a critical review of published original data. Pediatrics 2004;114: 793804. Methods: Literature review of original published articles from 1996 to 2004 such an association less likely sources of bias, and the potential effects of confounding 17. Patja A, Davidkin I, Kurki T, et al. Serious adverse events after measles-mumpsrubella vaccination during a fourteen-year prospective follow-up. Pediatr Infect Dis J 2000;19(12): 1127-1134. Objective: To distinguish events having a causal relation with MMR vaccination from those with only a temporal relation Methods: Prospective follow-up study using a passive surveillance system launched by the Finnish National Board of Health Serious adverse events following MMR vaccination (no time frame was imposed) 30 cases of anaphylaxis were reported and febrile seizures were the most commonly reported event; There were no cases of autism associated with MMR vaccination during the 14 year period Lack of a control group due to a near fully vaccinated population The 14 year follow-up allows for ample time to identify cases of autism in the 3 million vaccine doses Excluded -Study does not have an unvaccinated control group
Appendix B (Continued) 50 18. Richler J, Luyster R, Risi S, et al. Is there a regressive phenotype of autism spectrum disorder associated with the measlesmumps-rubella vaccine? A CPEA study. J Autism Dev Dis 2006; 36(3): 299-316. Objective: To assess whether a new phenotype of regressive autism is present and to assess the association with MMR vaccination Methods: A multi-site case-control study using data collected from 10 sites across the US as part of the Collaborative Program for Excellence in Autism (CPEA) Analyses from the initial ADIR and algorithm scores from the most recent ADI-R and ADOS as well as standard scores from the Vineland Adaptive Behavior Scales and verbal and non-verbal IQ scores Findings of present study provides no evidence that regression in autism is associated with MMR vaccination; Much like children with autism in general, children with autism and regression are a heterogeneous group with varying trajectories of development Children with autism and regression should have more social and communicative skills prior to loss but still show signs of atypical early development Children with regressive autism should show different outcomes in terms of social and communicative skills If regressive autism is associated with GI symptoms, then regressive autism patients should have more GI disorders/ symptoms than normal autism patients Age at onset of autistic symptoms should more closely follow age at MMR vaccination Limitation was Excluded -Study endpoints do not allow for the calculation of a RR or OR for the association of MMR and autism
Appendix B (Continued) 51 that information about childs early development was provided by retrospective parent report, which is open to bias 19. Smeeth L, Hall AJ, Fombonne E, et al. A casecontrol study of autism and mumps-measlesrubella vaccination using the general practice research database: design and methodology. BMC Public Health 2001; 1:2. Objective: To determine if autistic children are more likely to have received MMR vaccination prior to disease onset and to examine whether there is any association between clinical onset of disease and the timing of MMR vaccination Methods: A matched case-control study using data derived from the UK General Practice Research Database Conditional logistic regression to assess the association between MMR vaccination and autism (Odds ratio) Case series analyses to estimate the relative incidence of autism in defined time intervals after vaccination No results reported Copies of all hospital summaries will be requested to validate the diagnosis of autism Excluded -Study does not allow for the calculation of an OR or RR for the association of MMR and autism
Appendix B (Continued) 52 20. Smeeth L, Cook C, Fombonne E, et al. MMR vaccination and pervasive developmental disorders: a casecontrol study. Lancet 2004;364: 963-969. Methods and objectives described above Described above The OR for the association between MMR and pervasive developmental disorder was 0.86 (0.68-1.09) suggesting no association between MMR vaccination and increased risk of pervasive developmental disorders Case reports were obtained for 87% of patients to validate the diagnosis of a pervasive developmental disorder Limitation was that when children joined a participating general practice after the date of MMR vaccination, their previous vaccination history was recorded retrospectively Included 21. Takahashi H, Suzumura S, Shirakizawa F, et al. An epidemiological study of Japanese autism concerning routine childhood immunization history. Jpn J Infect Dis 2003;56: 114-117. Objective: To assess the causal association of autism with the MMR vaccine Methods: A casecontrol study of adjusted logistic regression analysis on subjects growing up in the Tokyo area between 1988 and Logistic regression analysis (Odds ratio) The OR for monovalent measles immunization (OR 5.33), non-mumps immunization (OR 8) and non-rubella immunization (OR 8.57) with the development of autistic spectrum disorders; Suggests a decreased risk of Only 21 cases were enrolled in the study Results may be biased by the informed consent process, which may have discouraged caregivers from enrolling children with a poor or incomplete immunization history Compared Excluded -Study did not have an unvaccinated control group
Appendix B (Continued) 53 1992 autism with MMR vaccines compared with monovalent antigens MMR vaccine to monovalent antigens 22. Taylor B. Vaccines and the changing epidemiology of autism. Child Care Health Dev 2006;32(5): 511519. Objective: To review the available literature on the association between autism and MMR vaccination Methods: Literature review and interpretation Interpretation of published studies on the association between autism and MMR vaccination The recorded prevalence of autism has increased considerably in recent years. The strong genetic component makes a post-natal cause unlikely Diagnostic procedures for determining special educational disabilities such as autism are not standardized nor have they been uniform over time Excluded -Study did not have an unvaccinated control group 23. Taylor B, Miller E, Farrington CP, et al. Autism and measles, mumps, and rubella vaccine: no epidemiological evidence for a causal association. Lancet 1999; 353: 2026-29. Objective: To investigate whether MMR vaccine may be causally associated with autism Methods: Epidemiological study of children born with autism since 1979 in the UK using the caseseries method to investigate the clustering of onsets in the post vaccination period Time trend analysis Relative risk of temporal association of autism and MMR vaccination Case series for temporal association Steady increase in cases by year of birth with no sudden step up or change in trend line; no temporal association between autism onset with one or two years of MMR vaccination (0.94, 1.09) Diagnosis of autism was confirmed based on medical records Limitation was not that the diagnosis could not be verified in all cases and the ascertainment may have been incomplete Clinical notes were of variable quality and many did not contain updated information Included
Appendix B (Continued) 54 24. Taylor B, Miller E, Lingam R, et al. Measles, mumps, and rubella vaccination and bowel problems or developmental regression in children with autism: population study. BMJ 2002;324: 393-396. Objective: To investigate whether MMR vaccination is associated with bowel problems and developmental regression in children with autism Methods: Population study with case note review to independently recorded vaccine data Recorded bowel problems lasting at least three months, age of reported regression and relation to MMR vaccination Single and multivariate logistic regression The proportion of children with developmental regression (OR 0.98) or bowel symptoms did not change significantly during the 20 years from 1979; findings provide no support for an MMR associated new variant form of autism with developmental regression and bowel problems The percentage of patients with regression (25%) is likely an overestimation Limitation was that data were not recorded systematically and there was variability in the level of detail Included 25. Thompson WW, Price C, Goodson B, et al. Early thimerosal exposure and neuropsychologic al outcomes at 7 to 10 years. N Engl J Med. 2007; 357(13): 12811292. Objective: To assess the association between current neuropsychological performance and exposure to mercury during the prenatal period, postnatal period, and the first 7 months of life Methods: Cohort Neuropsychol ogical outcomes associated with thimerosal (mercury) exposure Measures of association measured with ordinary leastsquares regression and logistic regression Only a few significant associations with exposure to mercury from thimerosal were found; the detected associations were small and almost equally divided between positive and negative effects Study did not assess autismspectrum disorders Mercury exposure was determined based on published data and the data from the FDA on the vaccines that the children received, from HMO records, immunization records, Excluded -Study exposure did not include MMR vaccination
Appendix B (Continued) 55 study of 1047 children between the ages of 7 and 10 years administered standardized tests assessing 42 neuropsychological outcomes and maternal interview Limitations include a potential selection bias due to a majority of selected families declining to participate and the inability to control for interventions such as speech therapy that may have biased results toward the null 26. Uchiyama T, Kurosawa M, Inaba Y. MMR vaccination and regression in autism spectrum disorders: negative results presented from Japan. J Autism Dev Disord 2007;37: 210-217. Objective: To investigate whether the MMR vaccination is associated with regressive autism Methods: Casecontrol comparison of children who received and did not receive the MMR vaccination and the change over time in the proportion of children who showed regressive symptoms across the pre-MMR, MMR and post-MMR Odds ratio for the association between regression and MMR exposure The OR for those in the MMR group who received the vaccine was 0.744 (0.3491.517); The odds ratio for children who did not receive the vaccine across the three generations was 0.626 (0.3231.200); the rate of regression did not vary across the preMMR, the MMR and the post-MMR cohorts In Japan the MMR vaccine was only given between 1985 and 1991 and thus allows for comparison of the before, during, and after effect of MMR vaccination Assumed that information in the Maternal and Child Handbook was accurate as most of the information is entered by health professionals Limitations: Included
Appendix B (Continued) 56 time frames in Japan study conducted in one private clinic, the sample size of subjects who received the MMR vaccine was small, and several issues concerning definition and measurement of regression (information was highly dependant on parental report using an instrument which asks limited questions on regression) 27. Wakefield AJ, Murch SH, Linnell J, Casson DM, et al. Illeallymphoid-nodular hyperplasia, nonspecific colitis, and pervasive developmental disorder in children. Lancet 1998;351: 637641. Objective: To investigate a series of children presenting with chronic enterocolitis and regressive developmental disorder following vaccination with the MMR vaccine Methods: Consecutive series Temporal association between symptom onset and vaccination In 8 of the 12 children the onset of behavioral problems was linked to MMR vaccination by the parent or childs physician; Once child received the monovalent measles vaccine after which his development slowed Potential selection bias in selfreferred group Small sample size (12 children) Lack of control group Temporal association largely depends on parent recall and when they first notice signs of developmental Excluded -Study did not have an unvaccinated control group
Appendix B (Continued) 57 analysis of 12 children with chronic enterocolitis and regressive developmental disorder regression 28. Wilson K, Mills E, Ross C, et al. Association of autistic spectrum disorder and the measles, mumps, and rubella vaccine. Arch Pediatr Adoles Med 2003;157: 628634. Objective: To systematically review the evidence for and against the existence of an association between autistic spectrum disorder and the MMR vaccine Methods: Systematic review of medical literature to identify all controlled epidemiological articles examining the association between autistic spectrum disorder and MMR vaccination 12 epidemiological studies assessing the association between autistic spectrum disorder and MMR vaccination The results of all of the studies showed no association between autistic spectrum disorders and the MMR vaccine Studies included: case-series, cross-sectional studies, case-control studies, and retrospective cohort studies Studies varied in how autistic spectrum disorder was diagnosed Excluded -Study did not have an unvaccinated control group
Appendix B (Continued) 58 29. Woo EJ, Ball R, Bostrom A, et al. Vaccine risk perception among reporters of autism after vaccination: vaccine adverse event reporting system 19902001. Am J Public Health 2004; 94(6): 990-995. Objective: To investigate vaccine risk perception among reporters of autism to the Vaccine Adverse Event Reporting System (VAERS) Methods: Structured interviews were conducted of 124 patients who reported autism to VAERS between 1990 and 2001 compared to published survey results of parents in the general population Perceptions of parents of children with autism who believe that vaccinations may have been a cause Only 15% deemed immunizations extremely important for childrens health; two-thirds had withheld vaccines from their children Passive surveillance systems such as VAERS are subject to many limitations including: underreporting, incomplete information, inadequate denominator data, and lack of an unbiased comparison group Excluded -Study did not allow for the calculation of an OR or RR for the association between autism and MMR