Previous reports from the Institute of Medicine (IOM) have reviewed the evidence regarding individual immunizations and adverse health outcomes. The most recent comprehensive report was Adverse Effects of Vaccines: Evidence and Causality (IOM, 2012). Most IOM reviews of vaccine safety have examined the association between adverse events and individual vaccines. One prior IOM review examined the evidence for an association between three adverse outcomes and the overall recommended childhood immunization schedule: increased susceptibility to heterologous infection; autoimmunity, as reflected in type 1 diabetes; and allergy, as reflected in asthma (IOM, 2002). The statement of task for the present IOM committee requests a review of the available data on the relationship between the overall immunization schedule and health effects that might be of concern to stakeholders, including parents, health care providers, and the public health community.
To complete its task, the committee reviewed research on the health outcomes and safety of the immunization schedule. It sought to identify study designs for analysis of health outcomes following immunization and ways to define the health outcomes used in recent studies reviewing aspects of the immunization schedule. Finally, it sought to provide guidance on ways to define exposures and health outcomes in the study designs that the committee may propose.
The committee did not have the time or the resources to conduct formal reviews meeting all criteria for systematic reviews for each question of interest, nor did it find substantial evidence to conduct a quantitative synthesis (IOM, 2011). Therefore, the committee searched for, assembled,
and summarized information on the association between aspects of the immunization schedule and specific health conditions already available in the peer-reviewed literature. The health outcomes that the committee chose to review were selected on the basis of its examination of the peer-reviewed literature, previous IOM vaccine safety studies, and public presentations at open meetings of the committee. The number of studies of aspects of the schedule varied; for some outcomes, several studies examining the cumulative effects of vaccines and adjuvants or preservatives were found; for other outcomes, very few studies were found. The committee’s methods and reviews are briefly summarized below.
The committee members and IOM staff conducted searches of the English-language literature published in the past 10 years (2002 to 2012) for children ages 0 to 18 years using the medical subject headings (MeSH) “immunization” or “vaccines,” combined with the following terms for health outcomes of interest:
- “autoimmune diseases” (which captures “diabetes mellitus, type 1”),
- “seizures” or “epilepsy” or “febrile seizures,”
- “child developmental disorders, pervasive” (which captures “autistic disorders”),
- “learning disorders” or “communications disorders” or “intellectual disability” or “developmental disorders,”
- “attention deficit and disruptive behavior disorders,” and
- “tics” or “Tourette’s syndrome.”
The literature published in the past 10 years was chosen to fill the gap since the 2002 IOM review and because several changes to the immunization schedule have been made since 2000 (e.g., addition of the pneumococcal and rotavirus vaccines). Studies more than 10 years old would be of outcomes that occurred after use of an immunization schedule with less resemblance to the current one.
All searches were run against the Ovid MEDLINE database (1950 to present). The search excluded reviews, commentaries, editorials, and similar publications. The conventional electronic searches were supplemented with articles identified by committee members and staff and articles that were noted during committee discussions and public presentations at open meetings. Commentaries and reviews were reviewed but not analyzed in the
same detail as were original research papers. The searches initially yielded 748 references. This number was further reduced to 143 by exclusion of articles that reviewed vaccines not included in the current or recent childhood immunization schedule or included vaccines for adolescents, such as the human papillomavirus vaccine, and by elimination of references duplicated in more than one category. The number of articles reviewed was further reduced by limitation of the search to articles describing studies that examined at least one health outcome and at least one of the following elements of the schedule, including
- number of vaccines,
- frequency of administration,
- spacing between doses,
- cumulative doses,
- age of the recipient, and
- order of vaccine administration.
Though the committee did not undertake a formal systematic review, the quality of individual articles was judged by the validity of the study design, the method by which the research was conducted, and the transparency of methods. In the end, 37 articles were chosen, and these, organized by category of health outcome, are briefly summarized below.
A second search was performed by use of the MeSH “immunization schedule” without predefined headings to investigate specific diseases or conditions. This search was conducted to ensure that the committee’s review adequately addressed any demonstrated associations between components of the immunization schedule and adverse health outcomes. Again, the search was limited to articles published in the past 10 years and excluded reviews, commentaries, editorials, and similar publications. After application of the exclusionary criteria, 1,235 abstracts were reviewed, and this number was narrowed to 56 that were considered potentially relevant to the committee’s charge. The committee concluded that only four of these research papers covered aspects of the childhood immunization schedule and safety. Two were considered not helpful to an evaluation of safety. (One was an economic evaluation of the childhood immunization schedule and did not examine safety; the second had serious limitations and was not considered for this chapter.) Two of the papers provided useful information, so summaries are included under the appropriate outcome section below (one is included under allergy/atopy; the second is included under neurological outcomes).
A third search was done to examine studies of immunization in infants born prematurely. Although prematurity is not a “health outcome,” the committee’s efforts included collection of data on premature infants because
of concerns about this vulnerable population. The search included the English-language literature published in the past 10 years (2002 to 2012) and used the previously mentioned MeSH terms “vaccines” and “immunization,” combined with “infant, premature,” and “premature birth.” The search was further reduced to include only research on children 0 to 18 years of age and infants from birth to 23 months of age. The initial results yielded 143 abstracts. The committee reviewed the only seven articles that contained relevant data and that met the quality criteria.
Allergy and Asthma
The Ovid MEDLINE literature search identified 40 references to articles on the relationship between immunizations or vaccines and asthma or hypersensitivity. (Although “atopy” and “allergy” were not search terms, many papers identified by use of the search term “asthma” or “hypersensitivity” included “atopy” or “allergy” as outcomes.) After an initial review, a team of two reviewers determined that 13 papers focused on some aspect of the immunization schedule. The committee’s second search provided a 14th paper for review, described below. A number of studies reported in the past 10 years have addressed the association between various aspects of the immunization schedule and asthma, atopy, or allergy. As one author noted (McKeever et al., 2004), it is necessary to have a detailed understanding of the relationship between allergic disease and vaccination, because the effectiveness of the immunization program may be adversely impacted by a perception that vaccination is harmful.
The following summary categorizes papers into groups: (1) studies examining an entire immunization schedule, (2) studies examining pertussis-containing vaccines, and (3) ecological studies (defined in Appendix B) and other studies that do not fit into one of the other two categories. Several papers reported on cohort follow-up studies with asthma, allergy, or atopy as the outcome and cumulative immunizations (the entire schedule for the country and time of the study) as the independent variable.
A longitudinal cohort in Australia was examined for the association between early childhood infection and immunization with the development of allergic diseases, including asthma (Thomson et al., 2010). The cohort included 620 allergy-prone children enrolled in 1989 and monitored from birth to 6 years of age. All data, including immunizations (diphtheria and tetanus toxoids and pertussis vaccine or diphtheria and tetanus toxoids absorbed [DT], oral poliovirus [OPV] vaccine, and measles, mumps, rubella [MMR] vaccine), were collected by telephone interviews. There was no relationship between cumulative immunizations and asthma. Administration
of DT in the first year of life but not the second year of life was associated with asthma and eczema. The study was limited by the self-report nature of the data and the small sample size.
Matheson and colleagues (2010) reported on atopy in the most recent follow-up study of 5,729 adults in the Tasmanian Longitudinal Health Study cohort of 1968 in Australia. This most recent follow-up of 44-yearolds was done by use of a mailed survey and explored the effects of immunization on atopic conditions. Only DTP, polio, and smallpox immunizations were in use in the cohort in 1968. The study is limited by the self-reported nature of the information on atopy. Nevertheless, the long-term follow-up demonstrated no association between immunization and asthma or atopic conditions into middle age.
A small study in France examined the association between vaccines received before age 6 months and asthma, allergic rhinitis, and eczema (Martignon et al., 2005). This was a retrospective cross-sectional study of 718 adolescents. Data on the three vaccines that were received before age 6 months were obtained from the pediatric record: bacillus Calmette-Guérin (BCG), diphtheria-tetanus-poliomyelitis, and pertussis vaccines. Live and inactivated vaccines were administered separately. Vaccinated adolescents were significantly less likely to have asthma, allergic rhinitis, and eczema than those who were not vaccinated. Although no association was found between an increase in cases of asthma, allergy, or eczema and immunization with the vaccines, the sample may have been too small to account for confounders, such as exposure to environmental tobacco smoke.
Benke et al. (2004) studied 4,500 young adults enrolled in a study in Australia in 1992 to determine whether childhood vaccines were associated with atopy and asthma in the cohort. Data on symptoms and vaccinations (including MMR, DTP, OPV, the hepatitis B [HepB] vaccine, and BCG) were collected by a mailed questionnaire. Atopy was measured directly by a skin test. Recall bias due to the collection of data via a mailed questionnaire was a limitation of this study. Overall, this study found no significant association between cumulative vaccinations and asthma.
McKeever et al. (2004) reported on a study of the relationship between vaccination and allergic disease, including asthma and wheezing, in the United Kingdom in individuals born from 1988 to 1999. The study had a retrospective observational cohort design and used the United Kingdom’s General Practice Research Database (GPRD). The cohort included 29,238 children ages 0 to 11 years with at least a single visit to a general practitioner in the first 6 months of life. Outcomes examined were asthma, wheeze, and eczema. The analysis controlled for the frequency of physician visits (“consulting frequency”). They examined groups of vaccines and also the total number of vaccines in the recommended immunization schedule. Children diagnosed with allergy before full vaccination was completed
were excluded from part of the analysis. The authors found no relationship between age at the time of the first immunization with DTP or MMR and asthma or eczema and no relationship between the total number of immunizations and allergic diseases. A relationship was explained by ascertainment bias rather than a biological effect for the children with from zero to six office visits, who appeared to have a higher risk of a diagnosis of asthma. The study was limited by the small numbers of unvaccinated children and possible ascertainment bias (number of office visits). No association between vaccinations and allergic disease, including asthma, was found.
Gruber and colleagues (2003) conducted a prospective investigation of atopy among 7,609 infants born in Germany in 1990 and monitored to age 5 years. The objective was to determine prospectively if the number (percentile) of childhood immunizations was associated with atopy in 5-yearolds who had been identified to be a high-risk cohort (at least two family members had atopy and a detectable immunoglobulin E concentration of >0.9 kU/L at birth). Atopy was confirmed by clinical diagnosis. Vaccination history was by parental report. The study analyzed exposure to individual vaccines and the cumulative use of vaccines containing aluminum. Overall, the study reported a negative correlation between atopy and the cumulative number of vaccine doses received, including pertussis vaccine. The principal limitation was the self-reporting of vaccination history. However, the committee believes that this was a well-constructed and well-reported study and may serve as one example of a means by which the U.S. immunization schedule could be studied.
Four Studies of Pertussis Vaccine-Containing Vaccines
Spycher et al. (2009) studied the development of wheezing and asthma among 6,811 children born in the United Kingdom from 1993 to 1997 and monitored to 2003 in a population-based cohort study. Immunization data were obtained from the National Health Service database. Data on the outcomes of wheezing and asthma were collected from repeated questionnaire surveys. The analysis compared children with complete, partial, or no vaccination against pertussis with children who were immunized with the whole-cell pertussis vaccine included in DTP at the time. Limitations included the self-reported nature of the outcomes data by questionnaires and the fact that 96.9 percent of the children were fully immunized: very few children were not vaccinated or incompletely vaccinated. Overall, the authors found no association between vaccination against pertussis and asthma by age 7 years.
A retrospective, longitudinal study in Manitoba, Canada, reported in 2008 (McDonald et al., 2008) examined an association between the timing of immunization with DTP and the development of childhood asthma by
age 7 years. The study used data on asthma risk from health administration records and income data from Canada Census by neighborhood. Manitoba switched from the use of DTP to the use of diphtheria and tetanus toxoids acellular pertussis vaccine adsorbed (DTaP) in 1997; most of the approximately 14,000 children in that study had received DTP and not DTaP. The study reported a decrease in the incidence of asthma for each month of delay in the time of vaccination with the first dose of DTP. A similar but weaker association between the incidence of asthma and each month of delay was also found for the second dose of DTP. The study was limited by potential ascertainment bias: variations in the number of doctor visits; nonrandom reasons for a delay in DTP administration (e.g., because of fever, an infection might promote a T-helper type 1 response [antiviral] over a T-helper type 2 response [proallergy/asthma]); and variations in socioeconomic status. A prospective study of DTaP would be needed to confirm whether these findings can be repeated with DTaP.
A second longitudinal study in the United Kingdom (Maitra et al., 2004) examined the association between pertussis immunization and asthma or atopy by age 7.5 years in a large birth cohort of 13,971 children as part of the Avon Longitudinal Study of Parents and Children. The study used three approaches (symptoms, a doctor’s diagnosis, and questionnaires) to identify children with asthma (symptoms reported by the parent or a doctor) via questionnaires. The aspect of the schedule covered in this study was immunization with DTP; the study differentiated between full, partial (diphtheria and tetanus toxoids [DT] but not pertussis vaccine), and no immunization. No association between asthma and pertussis immunization was found in children with a high cumulative prevalence of asthma.
Nilsson et al. (2003) reported on allergic disease in Sweden among 538 children at the age of 7 years after pertussis vaccination during infancy. This analysis was based on a follow-up study of a randomized controlled trial of three vaccines. The objective was to prospectively assess sensitization rates and the development of allergic diseases in a follow-up of children included in a randomized controlled trial of the pertussis vaccine. The group analyzed data from three randomized controlled trials evaluating differences in outcomes by age 7 years after immunization with DT or DT plus pertussis vaccine in a study with four arms: a two-component experimental pertussis vaccine, a five-component pertussis vaccine, a whole-cell pertussis vaccine, or no pertussis vaccine arm. All vaccines had aluminum phosphate as an adjuvant. Rigorous definitions of allergic disease were used, and skin tests of the children were used to demonstrate atopy. Compared with the DT vaccine, none of the three pertussis vaccines was a risk factor for the development of allergy in the first 7 years of life. The two-component pertussis experimental vaccine was associated with increased allergic symptoms after booster vaccination. This vaccine was not subsequently used. No relationship
between pertussis vaccines and atopic diseases was detected in children with a history of allergies.
Four Studies That Used Other Methods
One ecological study was done to examine trends in asthma prevalence and the recommended number of childhood immunizations (Enriquez et al., 2007). The group used National Health Interview Survey (NHIS) data on asthma, the timing of immunization, and the number of recommended immunizations by age 2 to determine whether increases in asthma prevalence paralleled trends in the number of immunizations recommended; however, the increase in the incidence of asthma reported in NHIS preceded the increase in the recommended number of vaccines. This information did not support a relationship between the recommended number of childhood immunizations and the increase in the prevalence of asthma and, in fact, provided evidence of no association.
Mullooly et al. (2007) used a case-control study of 6- to 16-year-olds in an allergy clinic with proven new allergic conditions to determine whether the receipt of immunizations or oral antibiotics in the first 2 years of life affected the odds that they would have atopy (measured by skin test). Compared with the control subjects, atopy cases received fewer antigen doses and fewer different antigens, had less exposure to Haemophilus influenzae type b conjugate vaccine (Hib), and received fewer doses of the Hib and mumps and rubella vaccines during the first 2 years of life. The study was limited by the fact that data on immunizations and other variables (e.g., family history of atopy, smoking in the home) were collected by retrospective medical record review. Their power to detect associations was also limited by the fact that only 21 percent of eligible allergy patients could be classified as non-atopic, leaving 79 percent as atopic study subjects. Finally, there was limited variation in vaccine exposure, further reducing the power to detect differences. Nevertheless, despite limited statistical power, this study found no association between atopy and vaccine exposure.
Maher et al. (2004) conducted a follow-up of a cohort previously enrolled in a study performed by a U.S.-managed care organization (MCO) as part of the Vaccine Safety Datalink (VSD) project. The analysis examined the association between immunizations and asthma among 1,778 children enrolled from 1991 to 1994. The original study used a matched-pair case-control method. Five vaccines were included: HepB, whole-cell pertussis vaccine, Hib, OPV, and MMR. The analysis was limited by the high rate of vaccine coverage and the small sample size. Childhood immunizations were not associated with asthma by age 5 years, but asthma was related to wheezing episodes in infancy. This study provides useful evidence of no association between vaccinations and asthma.
Bremner et al. (2005) examined the association between allergic rhinitis (“hay fever”) and MMR, DTP, and BCG immunization. The study used a case-control design and data from GPRD and the Doctors Independent Network primary care database in the United Kingdom. Children who had been immunized with MMR and DTP did not have greater odds of being diagnosed with hay fever than those who were unvaccinated. Slightly decreased odds of a diagnosis of hay fever in association with delayed DTP administration were detected, however. The researchers suggested that it is possible that an immunization delay in some children is associated with febrile illness. Infectious illness in early childhood could potentially protect against the development of atopy, and the association with delayed immunization with DTP needs further investigation. The small number of children who received BCG had slightly increased odds of having hay fever. The study was limited by the source of the outcomes data, which were based on medical records in which the International Classification of Diseases, revision 9, code for allergic rhinitis was used and medicines commonly prescribed for hay fever were listed. The study has limited value for interpretation of the safety of the U.S. immunization schedule, as the researchers were examining the association between allergic rhinitis and separate vaccines, and neither DTP nor BCG is currently recommended for U.S. children.
In summary, research examining the association between the cumulative number of vaccines received and the timing of vaccination and asthma, atopy, and allergy has been limited; the findings from the research that has been conducted are reassuring, however. No data have demonstrated harm (an increased risk of atopy) from immunizations. Indeed, the opposite may be the case. No evidence is available from studies that have directly examined the current immunization schedule (most studies enrolled children in the 1990s, and most were not conducted in the United States), but no studies suggest harm (e.g., an accelerated or increased likelihood of the development of asthma or atopic diseases). The single study finding an association between age at the time of immunization with the first whole-cell pertussis-containing vaccine and a later diagnosis of asthma (McDonald et al., 2008) has not been extended to examine acellular pertussis vaccine. One publication (Thomson et al., 2010) noted the importance of confounding infectious episodes, especially gastroenteritis, suggesting that childhood infections (a target for future effective vaccines) and not childhood immunizations are associated with asthma.
Fifty papers describing studies of a relationship between immunization or vaccines and autoimmune diseases were identified in the initial search.
This list was reduced to six papers after the exclusion criteria described above were used. After further review, four of the papers were believed to focus on some aspect of the immunization schedule and were selected for a more in-depth review.
A study of five U.S. MCOs involving 1.8 million children evaluated the risk of development of immune thrombocytopenic purpura (ITP) after immunization with childhood vaccines other than MMR (O’Leary et al., 2012). The study involved a self-controlled case series and was able to confirm an association between ITP and MMR. It found no increased risk of ITP after immunization with vaccines other than MMR in young children but did find an association between ITP and immunization with HepA; tetanus toxoid, reduced diphtheria toxoid, and acellular pertussis vaccine adsorbed; and varicella vaccine in older children. However, because of the small number of reports of ITP and potential confounders, the researchers concluded that further investigation is needed. A limitation of this study was that ITP is a rare adverse event, and it is difficult to examine the risk of ITP in association with immunization with other vaccines independently when these vaccines are routinely given at the same time as MMR, which has been determined to be one possible cause of rare cases of ITP (IOM, 1994).
Yong et al. (2010) used data from the United Kingdom GPRD to assess the incidence of ITP in the pediatric population in the United Kingdom and to compare the incidence of ITP in children with that in adults in a large population-based study. The researchers examined the evidence of infection and a history of immunization among pediatric patients with ITP, focusing on infections recorded within 8 weeks and immunizations recorded within 6 weeks before the first recorded diagnosis of ITP. A limitation of this study was that the investigators identified cases of infection through computerized records instead of the questionnaires used in other studies, which may have failed to capture a number of mild infections that did not lead to prompt contact with a physician.
Hviid and colleagues (2004) evaluated whether a link exists between childhood vaccinations and the development of type 1 diabetes using data from a cohort of children born between 1990 and 2000. The researchers used Danish data and estimated rates of type 1 diabetes according to vaccination status, including the type and number of doses, among all children and a subgroup of children who had a sibling with type 1 diabetes. Rate ratios were also estimated for the period from 2 to 4 years after vaccination. During the time period of the study, the schedule varied with the introduction of Hib from 1993 to 1995, when it was administered at 5, 6, and 16 months of age, but administration of Hib was then changed to 5, 6, and 15 months of age starting in 1996 and 3, 5, and 12 months of age starting in 1997. The combined diphtheria, tetanus, and inactivated poliovirus vaccine
was given at ages 5, 6, and 15 months until 1996, and a whole-cell pertussis vaccine was given separately at 5 weeks (half-dose), 9 weeks, and 10 months of age. In 1997, the pertussis vaccine was modified to the acellular pertussis vaccine, which was incorporated into the diphtheria, tetanus, and inactivated poliovirus vaccine. The schedule of the combined vaccine was modified to be given at 3, 5, and 12 months of age. Boosters of oral polio vaccine were given at 2, 3, and 4 years of age. The study evaluated 739,694 children for 4,720,517 person years of follow-up. Overall, 681 cases of type 1 diabetes were identified from the Danish National Hospital Register, 26 of whom (4,208 person years) had a sibling with type 1 diabetes. This study found no association between childhood vaccination and the development of type 1 diabetes, even among children who had a sibling with diabetes. A limitation noted by the authors was the use of the Danish National Hospital Register rather than the National Diabetes Registry, which goes back only to 1996, to make sure that they had large enough numbers of children for analysis. A strength of the study is that it was a nationwide cohort with longitudinal, individual-level information on vaccinations and type 1 diabetes incidence, minimizing selection and recall bias.
Verstraeten and colleagues (2008) performed an integrated analysis of studies performed internationally to assess the safety of vaccines containing the AS04 adjuvant according to the incidence of adverse events of potential autoimmune etiology, particularly in adolescents and young adults. The study compared recipients who received vaccines with the AS04 adjuvant and a control group who received nonadjuvanted vaccine (i.e., control), vaccines with aluminum adjuvant, or aluminum hydroxide alone. Overall, the rate of reporting of autoimmune disorders was low, with an event rate of approximately 0.5 percent which did not differ between the groups receiving vaccines with the AS04 adjuvant and the control groups
The distribution of the reports by category did not suggest unusual patterns of autoimmune disorders. The authors concluded that these analyses do not suggest any statistically significant association between the development of autoimmune disorders and immunization with AS04-adjuvanted vaccines. This conclusion reinforces other reports in the literature concluding that no evidence exists for an association between autoimmune disorders and most vaccines. Limitations of the analysis mainly included a lack of validation of each diagnosis, which relied on investigator reports, and variability in the collection of adverse event data between studies (Verstraeten et al., 2008).
In summary, the literature that the committee found to examine the relationship between the overall immunization schedule and autoimmunity was limited. The evidence from a single large Danish study for diabetes is reassuring because it did not detect a relationship between the immunization schedule and autoimmunity. Evidence for ITP confirms prior evidence
of an association with immunization with MMR and is not clear about immunization with other vaccines.
The initial literature search identified 32 papers on the relationship between immunizations or vaccines and pervasive developmental disorder (PDD), which includes the diagnoses autistic spectrum disorder, autism, and Asperger’s syndrome. After an initial review, a team of two IOM committee members determined that 12 papers focused on some aspect of the immunization schedule. Three of the papers either addressed only one vaccine or had methodological limitations. The other nine studies examined the association between thimerosal and autism and other neurodevelopmental problems (Andrews et al., 2004; Fombonne et al., 2006; Geier and Geier, 2003, 2004a,b, 2006; Hviid et al., 2003; Madsen et al., 2003; Young et al., 2008). Five of the studies had serious methodological limitations and were not helpful with examination of the association between thimerosal and vaccines. Each of the other four papers might help with a study of the schedule.
Fombonne et al. (2006) examined the prevalence of PDD in relation to two aspects of the immunization schedule in Canada: cumulative thimerosal dose and a change in the MMR schedule from one to two doses in birth cohorts from 1987 to 1998. Thimerosal was eliminated in 1996, and a second MMR (administered at age 18 months) was added to the schedule in 1996. Data on autism were from school records. Vaccine data were in part from a registry and in part from provider records. The dose of thimerosal was calculated from the recommended immunization schedule by year (not the dose received by individual children). A continuous increase in the incidence of PDD occurred over time, despite the elimination of thimerosal, and a decrease in MMR coverage was also detected. The increased rate of PDD was the same before and after the addition of a second required dose of MMR. The study was limited by reliance on administrative codes for the diagnosis of PDD. The study was also conducted in one school board (district), and some PDD cases may have moved into that board, which would have inflated the numbers. This was an ecological study, but the data were interpreted carefully and the differences in appropriate trends were noted.
Andrews et al. (2004) used the United Kingdom GPRD to evaluate the risk of a variety of neurodevelopmental disorders, including autism, tics, speech and language delay, attention deficit disorder, and other developmental delays, in association with the calculated cumulative exposure to thimerosal to up to 4 months of age in more than 100,000 children born between 1988 and 1997. The retrospective cohort study found no evidence for an increased risk of neurodevelopmental disorders, with the possible
exception of tics, in association with thimerosal exposure. For general developmental disorders, unspecified developmental delay, and attention deficit disorder, increasing thimerosal exposure had an apparent protective effect. Although the study was limited by an inability to adjust for several confounding factors, such as socioeconomic status and other medical conditions, in general, it had a sound methodology. GPRD is a good source of linked data that may be used to look at other aspects of the vaccination schedule in the United Kingdom. The aspect of the schedule covered by this study included the cumulative doses of thimerosal received by children immunized with DTP and DT and whether these were received, for example, on time or late.
Two studies examined aspects of the Danish immunization schedule. Hviid et al. (2003) studied the relationship between cumulative thimerosal exposure via the whole-cell pertussis vaccine and autistic spectrum disorder. The study included a cohort of children with a diagnosis of autistic spectrum disorder born between 1990 and 1996. The diagnoses were taken from the Danish Psychiatric Central Research Registry and linked with the immunization history of each child. The study covered a period (1990 to 1992) when only one thimerosal-containing vaccine was in use. The study found no association between a diagnosis of autism and the presence of thimerosal but noted that the incidence of autism may have been underascertained, especially in earlier birth cohorts. This study did not demonstrate a relationship between thimerosal administration via the pertussis vaccine and the development of autism in a small country (Denmark) with high immunization rates and a good system of record keeping. The only aspect of the schedule covered was thimerosal exposure specifically via the pertussis vaccine.
The other Danish study evaluating an association between immunization and PDD (Madsen et al., 2003) also used data from the Danish Psychiatric Central Research Registry. The authors sought to evaluate the vaccine history for all Danish children identified with autism between 1971 and 2000 to assess the incidence of autism among children between 2 and 10 years of age before and after the removal of thimerosal from vaccines in 1992. The annual incidence of autism increased rapidly starting in 1990 and continued to do so through 1999, even though thimerosal was eliminated from DTP in 1992. The study was limited, as was the study by Hviid et al. (2003), by the fact that before 1995, diagnoses of autism were made only for hospitalized patients, whereas after 1995, outpatient diagnoses of autism were included. This study failed to demonstrate a correlation between the discontinuation of thimerosal in DTP and the incidence of autism in Danish children. This was an ecological study and so it cannot confirm an association. The paper provided no real information about the immunization schedule.
In summary, the evidence of an association between autism and the overall immunization schedule is limited both in quantity and in quality and does not suggest a causal association. The committee found the literature to be most useful in suggesting study designs that might be adapted and extended for the committee’s core task of suggesting further research.
Other Neurodevelopmental Disorders
Forty-one papers concerning a relationship among immunizations, immunization schedule, or vaccines and learning disorders, communication disorders, developmental disorders, intellectual disability, attention deficit disorder, disruptive behavior disorders, tics, and Tourette’s syndrome were identified via an Ovid MEDLINE database search. This list was reduced to eight papers after use of the exclusion criteria described above, including exclusion of papers on vaccines not currently recommended for administration to children under age 6 years. After an initial review, five of the papers were believed to focus on some aspect of the immunization schedule and were selected for more in-depth review. Each of these five studies focused on possible adverse effects of thimerosal (given via different schedules). Importantly, with the exception of the influenza vaccine, since 2001 thimerosal has been either removed from or substantially reduced in amount in vaccines given to U.S. children under 6 years of age. Although thimerosal is no longer a component of U.S. childhood vaccines, these studies may suggest methods to study variations due to use of alternative schedules, or to changes to the recommended immunization schedule made over time. The committee identified a sixth study through its second search effort.
A study conducted by Tozzi et al. (2009) in Italy also evaluated the effects of different doses of thimerosal during infancy on neurodevelopmental outcomes. These investigators conducted a late follow-up evaluation at 10 to 12 years of age of subjects who were initially enrolled in a study of the efficacies of two formulations of pertussis vaccine that contained different amounts of thimerosal. Twenty-four neurodevelopmental outcomes were measured via 11 standardized tests. Only two statistically significant differences, which were believed not to have been clinically significant, were noted in the female subjects. Specifically, girls with higher thimerosal exposure had lower mean scores in the Boston Naming Test and on finger tapping with the dominant hand. Given the large number of comparisons, these significant differences could be attributable to chance. In this study, the cumulative dose of thimerosal was low compared with the doses that had been used in the United States.
In a cohort study of 1,047 subjects enrolled in three MCOs as part of the VSD, Thompson et al. (2007) evaluated the effects of cumulative exposure to thimerosal on 42 neurodevelopmental outcome measures (excluding
autism). The subjects were between 7 and 10 years of age. Immunization status was retrospectively assessed, and the assessment included exposure to thimerosal both prenatally (via maternal immunization or immunoglobulin administration) and then during the first 7 months of life. Few significant associations between cumulative thimerosal exposure and a particular neurodevelopmental outcome were noted. These associations were few in number and were equally divided between positive and negative effects. Most were gender specific. For example, in boys, higher exposure to thimerosal prenatally was associated with a higher score on the Stanford-Binet copying test and a lower score on the Wechsler Intelligence Scale for Children III (WISC-III) digit-span test of backward recall. In girls, higher thimerosal exposure at between birth and 7 months of age was associated with a better performance on the Grooved Pegboard Test in the nondominant hand as well as on the WISC-III digit-span test of backward recall. Although this study was limited by only a 30 percent participation rate, which may have resulted in selection bias, it failed to demonstrate a causal association between early exposure to mercury via thimerosal-containing vaccines or immunoglobulins and neurodevelopment.
Smith and Woods (2010) used secondary data from the VSD cohort study of Thompson et al. (2007) to determine if on-time immunization by 1 year of age was associated with neuropsychological outcomes. The researchers performed two analyses using immunization and outcomes data from the VSD. The first analysis compared children who had received all vaccinations on time with those who had not. Complete immunization was defined as having received within 30 days of the recommended age at least two doses of HepB, three doses of DTaP, three doses of Hib, and two doses of polio vaccine (referred to as the 2:3:3:2 series) during the first year of life. The second analysis stratified children into five groups by age at the time of completion of the 2:3:3:2 series. Children with on-time immunizations consisted of those who received at least 10 vaccinations in the first 7 months of life, whereas the least vaccinated group comprised those who had received less than seven vaccine doses of any type during the same time period. Using the outcomes data obtained from the research of Thompson et al. (2007), Smith and Woods (2010) found that children who had received their immunizations on time and also those who had received at least 10 doses did not have better neuropsychological outcomes in this study than those who had received fewer doses, and no significant differences were found between those who received the least vaccines and those with the greatest vaccine exposure during the first 7 months of life.
In a cohort study conducted in Brazil, Marques et al. (2007) evaluated the effects of thimerosal exposure during the neonatal period on neurodevelopment measured by use of the Gesell battery of tests at 6 months of age. In their study, 84 infants were immunized with a thimerosal-containing
HepB either on the day of birth or later in the neonatal period (between days 2 and 4 of life). Before the neurodevelopmental assessments at 6 months of age, these infants also received additional doses of vaccines containing thimerosal (two doses of HepB and three doses of DTP). The researchers did not report any difference in neurodevelopmental measures between the two groups. In addition to the small sample size, this research focused on a minimal alteration in the immunization schedule that may have been so minor that an effect on neurodevelopment would not be expected.
In a longitudinal study of 14,000 infants in the United Kingdom, Heron et al. (2004) evaluated the relationship between cumulative exposure to thimerosal and several neurodevelopmental outcomes, including behavioral difficulties, tics, deficits in speech and fine motor development, and other “special needs.” At the time of this study, thimerosal-containing vaccines were administered in the United Kingdom at 2, 3, and 4 months of age, which represents an accelerated schedule of exposure compared with the schedule used in the United States. This study evaluated 69 specific behavioral and developmental outcomes via questionnaires that were sent to the parents of children born over a 15-month interval during 1991 and 1992. Only one outcome (poor prosocial behavior) was found to be associated with cumulative thimerosal exposure at 3 months of age. Interestingly, this study demonstrated that adverse neurodevelopmental outcomes were less likely in children who had higher thimerosal exposures.
In another VSD study, Verstraeten et al. (2003) also evaluated the association between the cumulative exposure to thimerosal at 1, 3, and 7 months of age and neurodevelopmental disorders such as autism, other speech and language disorders, disorders of attention, and tics. This was a large retrospective cohort study of subjects from three MCOs that participated in the VSD. In Phase 1 of the study, data from two MCOs were analyzed. A positive association between cumulative thimerosal exposure and the development of tics was found for subjects from one MCO, whereas a positive association with language delay was found for subjects from the other MCO. In Phase 2 of the study, the most common associations seen in Phase 1 were evaluated in a third MCO, and no significant associations were demonstrated. Therefore, no consistent significant association between cumulative thimerosal exposure and neurodevelopmental outcomes was found. Importantly, in no instance was a significant risk of cumulative thimerosal exposure and either autism or disorders of attention detected. This study was limited, as the investigators evaluated thimerosal only as opposed to the type of vaccine. Neurodevelopmental outcomes for the subjects were determined only by medical record designations (codes) and not by a review of the results of formal neuropsychological assessments.
In summary, the evidence regarding an association between the overall immunization schedule and other neurodevelopmental disorders is limited
in quantity and of limited usefulness because of its focus on a preservative no longer used in the United States.
Seizures, Febrile Seizures, and Epilepsy
Fifty-eight papers of studies of the association among immunizations, immunization schedule, or vaccines and seizures, epilepsy, or febrile seizure were identified via an Ovid MEDLINE search. This list was then reduced to 14 papers. After an initial review, four of the papers were believed to focus on some aspect of the immunization schedule and were selected for a more in-depth review.
A study from Denmark by Sun and colleagues (2012) determined the risk of cumulative doses of combined DTaP-inactivated poliovirus vaccine (IPV)-Hib on the development of both febrile seizures and the later development of epilepsy as well as the risk of these adverse events after pneumococcal vaccine was added to the combined DTaP-IPV-Hib. This was a self-controlled case series study based on children with febrile seizures during follow-up of the cohort. In Denmark, DTaP-IPV was introduced in 1997, Hib was added in September 2002, and pneumococcal vaccine was added in October 2007. Data were collected from January 1, 2003, to December 31, 2008, and the immunization schedule that was evaluated included vaccine administration at 3, 5, and 12 months of age. The analysis did not include the 5-year booster immunization. Compared with a reference cohort of children who were not within 0 to 7 days of receiving an immunization, the increased risk of febrile seizure on the day of immunization only (but not between days 0 and 7 after immunization) was minimal after the first or second dose of combined DTaP-IPV-Hib vaccine but not after the third dose. The overall incidence of febrile seizures in these cohorts was small. The vaccinated group had a lower risk of developing epilepsy in the first 15 months of life than the reference cohort of children did, whereas the risk of epilepsy later in life was unchanged. The estimates did not change when pneumococcal vaccine was added to the vaccination program. It is not clear why the immunized children had a decreased risk of epilepsy. This may have been due to unmeasured confounding factors, as the investigators did not address whether children with a high risk of developing febrile seizures or epilepsy (such as children with preexisting neurological disorders) were less likely to have been vaccinated.
A VSD surveillance study by Klein et al. (2010) evaluated the risk of development of febrile seizures after children received the combined measles, mumps, rubella, and varicella (MMRV) vaccine, MMR plus the varicella vaccine, MMR alone, or the varicella vaccine alone. The investigators compared the incidence of evaluations for seizures in the emergency department or hospital and for fever in the clinic that occurred in patients at between 12
and 23 months of age within 42 days of receiving any “measles-containing vaccine” as well as the varicella vaccine (either as a component of the measles vaccine, at the same time as the measles vaccine, or at a different time). The investigators determined that both MMRV and MMR, but not the varicella vaccine alone, are associated with increased outpatient visits for fever and seizures 7 to 10 days after vaccination, with MMRV increasing the risk of fever and seizures twice as much as MMR plus the varicella vaccine. A limitation of this study was that the cases of febrile seizure were determined by the presence of International Classification of Diseases, Ninth Revision, codes for febrile seizure within the medical record. This may have somewhat overestimated the risk of this adverse event.
Another VSD study (Tse et al., 2012) investigated the risk of febrile seizures that followed the receipt of trivalent inactivated influenza vaccine (TIV) which was administered during the 2010-2011 influenza season. The investigators conducted surveillance of adverse events in children between the ages of 6 and 59 months of age who had received a first dose of TIV. Cases of febrile seizures were identified through the analysis of ICD-9 codes and chart review, specifically for patients presenting to emergency departments or those who were hospitalized. In mid-November 2010, a signal was detected that indicated an increased risk of febrile seizures occurring between 0 and 1 days following the first dose of TIV. However, further analysis demonstrated that the risk of febrile seizure was higher after the concomitant administration of both TIV and 13-valent pneumococcal conjugate vaccine (PCV13) compared with the additive risk of febrile seizure after receiving either TIV or PCV13 alone. This risk was highest in children vaccinated at 16 months of age, which is not surprising as studies of the natural history of febrile seizures indicate that the background risk is greatest around this age and progressively falls off in older children. Limitations of this study were that the investigators did not evaluate the possible effects of the concomitant administration of other vaccines (such as DTaP), and due to limited information about attributable causes, the investigators were not able to exclude cases who had intercurrent infections as the cause of the febrile seizure. Importantly, given the results of this study, the vaccine information statement for TIV was updated for the 2011-2012 influenza season to include a statement about the possible increased risk of febrile seizure in young children who concomitantly receive both TIV and PCV13 (CDC, 2012).
A study conducted in The Netherlands (David et al., 2008) evaluated the frequency of adverse events that occurred after infants received pertussis vaccine. In The Netherlands, infants receive this vaccine at 2, 3, 4, and 11 months of age. The study compared the adverse events that occurred after patients received whole-cell pertussis vaccine, acellular pertussis vaccine, or acellular pertussis vaccine along with pneumococcal vaccine. The data were
acquired from 28,796 of approximately 53,000 questionnaires distributed to parents. The risks of prolonged crying, pallor, high fever, and “fits and jerks” were significantly reduced when the whole-cell pertussis vaccine was replaced by the acellular vaccine. The authors point out that although “fits and jerks” was meant to be an indicator for “seizures,” upon review of their data, it was apparent that this category mainly included chills, shivering, jitteriness, and myoclonus. Possible febrile seizures were noted only after the fourth dose of vaccine, with only two cases occurring in the group receiving the whole-cell pertussis vaccine and one case occurring in the group receiving the acellular pertussis vaccine. This was not a statistically significant finding. The addition of pneumococcal vaccine to the schedule did not change the risk of any adverse events. This study was limited by the 54 percent questionnaire return rate, with a probable bias of an increased rate of return from parents of children who had had reactions. In addition, some at-risk children (children of mothers with hepatitis B) received HepB at the same time as pertussis vaccine, but this clinical feature was not factored into the analysis.
In summary, the literature associating the overall immunization schedule with seizures, febrile seizures, and epilepsy is limited and inconclusive. With the exception of the study suggesting the increased risk of febrile seizure after concomitant TIV and PCV13 immunization (Tse et al., 2012), there is no suggestion of a causal relationship between the administration of multiple vaccines and a single seizure or the later development of epilepsy.
Immunization of Premature Infants
The committee reviewed six papers on the immunization of premature infants published since 2002. Five papers examined postvaccination cardiorespiratory events, and two papers examined C-reactive protein levels following the immunizations at 2 months of age. All papers included at least some very premature infants (.32 weeks of gestation), all examined aspects of the vaccines scheduled to be delivered at 2 months of age, and two reviewed longer-term effects. Because small numbers of infants were monitored for short periods of time, it is challenging to draw conclusions from this review. An increased risk of cardiorespiratory events after vaccination may exist, especially in infants with prior septicemia and the need for continuous positive airway pressure for a longer period of time earlier in their lives. The authors of several papers proposed that some infants be monitored in a hospital after the first and perhaps the second round of immunizations, but the authors had no consensus on how to identify which infants born prematurely are the most likely to benefit from monitoring. They did note, however, that risk factors include lower birth weight, ongoing complications, and underlying medical conditions.
The committee conducted a review directed by conventional electronic searches of the peer-reviewed literature, findings from searches conducted by committee members, committee member expertise, committee discussions, and information from public presentations at open committee meetings.
The committee’s review confirmed that research on immunization safety has mostly developed around studies examining potential associations between individual vaccines and single outcomes. Few studies have attempted more global assessments of entire sequence of immunizations or variations in the overall immunization schedule and categories of health outcomes, and none has squarely examined the issue of health outcomes and stakeholder concerns in quite the way that the committee was asked to do in its statement of task. None has compared entirely unimmunized populations with those fully immunized for the health outcomes of concern to stakeholders.
Queries of experts who addressed the committee in open session did not point toward a body of evidence that had been overlooked but, rather, pointed toward the fact that the research conducted to date has generally not been conceived with the overall immunization schedule in mind.
The available evidence is reassuring, but it is also fragmentary and inconclusive on many issues. Nevertheless, the committee found in its literature review useful perspectives on how to define exposures and outcomes and how conventional study designs might be expanded and adapted to more clearly address the question of health outcomes after immunization with the overall immunization schedule.
A challenge to the committee in its review of the scientific literature was uncertainty as to whether studies published in the scientific literature have addressed all health outcomes and safety concerns. The field needs valid and accepted metrics of the entire schedule (the “exposure”) and clearer definitions of the health outcomes linked to stakeholder concerns (the “outcomes”) in research that is sufficiently funded to ensure the collection of a large quantity of high-quality data.
Recommendation 5-1: To improve the utility of studies of the entire childhood immunization schedule, the committee recommends that the National Vaccine Program Office develop a framework that clarifies and standardizes definitions of
- key elements of the schedule,
- relevant health outcomes, and
- populations that are potentially susceptible to adverse events.
Andrews, N., E. Miller, A. Grant, J. Stowe, V. Osborne, and B. Taylor. 2004. Thimerosal exposure in infants and developmental disorders: A retrospective cohort study in the United Kingdom does not support a causal association. Pediatrics 114(3):584-591.
Benke, G., M. Abramson, J. Raven, F.C.K. Thien, and E.H. Walters. 2004. Asthma and vaccination history in a young adult cohort. Australian and New Zealand Journal of Public Health 28(4):336-338.
Bremner, S.A., I.M. Carey, S. DeWilde, N. Richards, W.C. Maier, S.R. Hilton, D.P. Strachan, and D.G. Cook. 2005. Timing of routine immunisations and subsequent hay fever risk. Archives of Diseases in Childhood 90(6):567-573.
CDC (Centers for Disease Control and Prevention). 2012. Influenza vaccine 2012—inactivated. Atlanta, GA: Centers for Disease Control and Prevention. http://www.cdc.gov/vaccines/pubs/vis/downloads/vis-flu.pdf (accessed December 6, 2012).
David, S., P.E. Vermeer-de Bondt, and N.A.T. van der Maas. 2008. Reactogenicity of infant whole cell pertussis combination vaccine compared with acellular pertussis vaccines with or without simultaneous pneumococcal vaccine in The Netherlands. Vaccine 26(46):5883-5887.
Enriquez, R., T. Hartert, and V. Persky. 2007. Trends in asthma prevalence and recommended number of childhood immunizations are not parallel. Pediatrics 119(1):222-223.
Fombonne, E., R. Zakarian, A. Bennett, L. Meng, and D. McLean-Heywood. 2006. Pervasive developmental disorders in Montreal, Quebec, Canada: Prevalence and links with immunizations. Pediatrics 118(1):e139-e150.
Geier, D.A., and M.R. Geier. 2003. An assessment of the impact of thimerosal on childhood neurodevelopmental disorders. Pediatric Rehabilitation 6(2):97-102.
Geier, D.A., and M.R. Geier. 2004a. Neurodevelopmental disorders following thimerosal-containing childhood immunizations: A follow-up analysis. International Journal of Toxicology 23(6):369-376.
Geier, D.A., and M.R. Geier. 2004b. A comparative evaluation of the effects of MMR immunization and mercury doses from thimerosal-containing childhood vaccines on the population prevalence of autism. Medical Science Monitor 10(3):PI33-PI39.
Geier, D.A., and M.R. Geier. 2006. An assessment of downward trends in neurodevelopmental disorders in the United States following removal of thimerosal from childhood vaccines. Medical Science Monitor 12(6):CR231-CR239.
Gruber, C., S. Illi, S. Lau, R. Nickel, J. Forster, W. Kamin, C.P. Bauer, V. Wahn, U. Wahn, and the MAS-90 Study Group. 2003. Transient suppression of atopy in early childhood is associated with high vaccination coverage. Pediatrics 111(3):e282-e288.
Heron, J., J. Golding, and the ALSPAC Study Team. 2004. Thimerosal exposure in infants and developmental disorders: A prospective cohort study in the United Kingdom does not support a causal association. Pediatrics 114(3):577-583.
Hviid, A., M. Stellfeld, J. Wohlfahrt, and M. Melbye. 2003. Association between thimerosal-containing vaccine and autism. Journal of the American Medical Association 290(13): 1763-1766.
Hviid, A., M. Stellfeld, J. Wohlfahrt, and M. Melbye. 2004. Childhood vaccination and type 1 diabetes. New England Journal of Medicine 350(14):1398-1404.
IOM (Institute of Medicine). 1994. Adverse events associated with childhood vaccines: Evidence bearing on causality. Washington, DC: National Academy Press.
IOM. 2002. Immunization safety review: Multiple immunizations and immune dysfunction. Washington, DC: National Academy Press.
IOM. 2011. Finding what works in health care: Standards of systematic reviews. Washington, DC: The National Academies Press.
IOM. 2012. Adverse effects of vaccines: Evidence and causality. Washington, DC: The National Academies Press.
Klein, N.P., B. Fireman, W.K. Yih, E. Lewis, M. Kulldorff, P. Ray, R. Baxter, S. Hambidge, J. Nordin, A. Naleway, E.A. Belongia, T. Lieu, J. Baggs, E. Weintraub, and Vaccine Safety Datalink. 2010. Measles-mumps-rubella-varicella combination vaccine and the risk of febrile seizures. Pediatrics 126(1):e1-e8.
Madsen, K.M., M.B. Lauritsen, C.B. Pedersen, P. Thorsen, A.M. Plesner, P.H. Andersen, and P.B. Mortensen. 2003. Thimerosal and the occurrence of autism: Negative ecological evidence from Danish population-based data. Pediatrics 112(3 Pt 1):604-606.
Maher, J.E., J.P. Mullooly, L. Drew, and F. DeStefano. 2004. Infant vaccinations and childhood asthma among full-term infants. Pharmacoepidemiology and Drug Safety 13(1):1-9.
Maitra, A., A. Sherriff, M. Griffiths, J. Henderson, and the Avon Longitudinal Study of Parents and Children Study Team. 2004. Pertussis vaccination in infancy and asthma or allergy in later childhood: Birth cohort study. BMJ 328(7445):925-926.
Marques, R.C., J.G. Dorea, A.G. Manzatto, W.R. Bastos, J.V.E. Bernardi, and O. Malm. 2007. Time of perinatal immunization, thimerosal exposure and neurodevelopment at 6 months in breastfed infants. Acta Paediatrica 96(6):864-868.
Martignon, G., M.-P. Oryszczyn, and I. Annesi-Maesano. 2005. Does childhood immunization against infectious diseases protect from the development of atopic disease? Pediatric Allergy and Immunology 16(3):193-200.
Matheson, M.C., E. Haydn Walters, J.A. Burgess, M.A. Jenkins, G.G. Giles, J.L. Hopper, M.J. Abramson, and S.C. Dharmage. 2010. Childhood immunization and atopic disease into middle-age—A prospective cohort study. Pediatric Allergy and Immunology 21(2 Pt 1):301-306.
McDonald, K.L., S.I. Huq, L.M. Lix, A.B. Becker, and A.L. Kozyrskyj. 2008. Delay in diphtheria, pertussis, tetanus vaccination is associated with a reduced risk of childhood asthma. Journal of Allergy and Clinical Immunology 121(3):626-631.
McKeever, T.M., S.A. Lewis, C. Smith, and R. Hubbard. 2004. Vaccination and allergic disease: A birth cohort study. American Journal of Public Health 94(6):985-989.
Mullooly, J.P., R. Schuler, M. Barrett, and J.E. Maher. 2007. Vaccines, antibiotics, and atopy. Pharmacoepidemiology and Drug Safety 16(3):275-288.
Nilsson, L., N.I.M. Kjellman, and B. Bjorksten. 2003. Allergic disease at the age of 7 years after pertussis vaccination in infancy: Results from the follow-up of a randomized controlled trial of 3 vaccines. Archives of Pediatrics and Adolescent Medicine 157(12):1184-1189.
O’Leary, S.T., J.M. Glanz, D.L. McClure, A. Akhtar, M.F. Daley, C. Nakasato, R. Baxter, R.L. Davis, H.S. Izurieta, T.A. Lieu, and R. Ball. 2012. The risk of immune thrombocytopenic purpura after vaccination in children and adolescents. Pediatrics 129(2):248-255.
Smith, M.J., and C.R. Woods. 2010. The risk of immune thrombocytopenic purpura after vaccination in children and adolescents. Pediatrics 129(2):248-255.
Spycher, B.D., M. Silverman, M. Egger, M. Zwahlen, and C.E. Kuehni. 2009. Routine vaccination against pertussis and the risk of childhood asthma: A population-based cohort study. Pediatrics 123(3):944-950. [Erratum, Pediatrics 123(5):1437, 2009.]
Sun, Y., J. Christensen, A. Hviid, J. Li, P. Vedsted, J. Olsen, and M. Vestergaard. 2012. Risk of febrile seizures and epilepsy after vaccination with diphtheria, tetanus, acellular pertussis, inactivated poliovirus, and Haemophilus influenzae type B. Journal of the American Medical Association 307(8):823-831.
Thompson, W.W., C. Price, B. Goodson, D.K. Shay, P. Benson, V.L. Hinrichsen, E. Lewis, E. Eriksen, P. Ray, S.M. Marcy, J. Dunn, L.A. Jackson, T.A. Lieu, S. Black, G. Stewart, E.S. Weintraub, R.L. Davis, F. DeStefano, and the Vaccine Safety Datalink Team. 2007. Early thimerosal exposure and neuropsychological outcomes at 7 to 10 years. New England Journal of Medicine 357(13):1281-1292.
Thomson, J.A., C. Widjaja, A.A.P. Darmaputra, A. Lowe, M.C. Matheson, C.M. Bennett, K. Allen, M.J. Abramson, C. Hosking, D. Hill, and S.C. Dharmage. 2010. Early childhood infections and immunisation and the development of allergic disease in particular asthma in a high-risk cohort: A prospective study of allergy-prone children from birth to six years. Pediatric Allergy and Immunology 21(7):1076-1085.
Tozzi, A.E., P. Bisiacchi, V. Tarantino, B. De Mei, L. D’Elia, F. Chiarotti, and S. Salmaso. 2009. Neuropsychological performance 10 years after immunization in infancy with thimerosal-containing vaccines. Pediatrics 123(2):475-482.
Tse, A., H.F. Tseng, S.K. Greene, C. Vellozzi, G.M. Lee, and the VSD Rapid Cycle Analysis Influenza Working Group. 2012. Signal identification and evaluation for risk of febrile seizures in children following trivalent inactivated influenza vaccine in the Vaccine Safety Datalink Project, 2010-2011. Vaccine 30 (11):2024-2031.
Verstraeten, T., R.L. Davis, F. DeStefano, T.A. Lieu, P.H. Rhodes, S.B. Black, H. Shinefield, R.T. Chen, and the Vaccine Safety Datalink Team. 2003. Safety of thimerosal-containing vaccines: A two-phased study of computerized health maintenance organization databases. Pediatrics 112(5):1039-1048. [Erratum, Pediatrics, 113(1):184, 2004.]
Verstraeten, T., D. Descamps, M.-P. David, T. Zahaf, K. Hardt, P. Izurieta, G. Dubin, and T. Breuer. 2008. Analysis of adverse events of potential autoimmune aetiology in a large integrated safety database of AS04 adjuvanted vaccines. Vaccine 26(51):6630-6638.
Yong, M., W.M. Schoonen, L. Li, G. Kanas, J. Coalson, F. Mowat, J. Fryzek, and J.A. Kaye. 2010. Epidemiology of paediatric immune thrombocytopenia in the general practice research database. British Journal of Haematology 149(6):855-864.
Young, H.A., D.A. Geier, and M.R. Geier. 2008. Thimerosal exposure in infants and neurodevelopmental disorders: An assessment of computerized medical records in the Vaccine Safety Datalink. Journal of the Neurological Sciences 271(1-2):110-118.