Results

Courtney Gidengil; Matthew Bidwell Goetz; Margaret Maglione; Sydne J. Newberry; Peggy Chen; Kelsey O’Hollaren; Nabeel Qureshi; Keller Scholl; Alicia Ruelaz Maher; Olamigoke Akinniranye; Tim M. Kim; Oluwafemi Jimoh; Lea Xenakis; Weilong (David) Kong; Zichun Xu; Owen Hall; Jody Larkin; Aneesa Motala; Susanne Hempel

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Gidengil C, Goetz MB, Maglione M, et al. Safety of Vaccines Used for Routine Immunization in the United States: An Update [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2021 May. (Comparative Effectiveness Review, No. 244.)

Cover of Safety of Vaccines Used for Routine Immunization in the United States: An Update

Safety of Vaccines Used for Routine Immunization in the United States: An Update [Internet].

Show details

Contents

< Prev Next >

Results

This chapter summarizes the results across studies included in the systematic review. For each Key Question (KQ), we list key points, summarize the update findings, and synthesize strength of evidence (SoE) across all available data.

Appendix B lists excluded studies and the reasons for exclusion. Appendix C details the literature search results and included studies. The evidence table of included studies is provided in Appendix D.

Description of Included Evidence

This update identified 189 studies reported in 338 publications.⁵¹^–³⁸⁸ Studies were identified through database and grey literature searches, public comments, and reference mining of relevant systematic reviews and package inserts.²^, ³^, ⁵^–⁷^, ¹³^–¹⁵^, ¹⁸^, ¹⁹^, ²¹^, ²³^, ²⁵^, ²⁶^, ²⁸^–³⁰^, ³²^–³⁷^, ³⁸⁹^–⁹⁰⁶ As search filters for adverse events are often insufficient, we used a broad search strategy and screened 56,608 citations for this report. Of these, 7,573 publications were reviewed in detail as full text (Figure 2).

The figure shows the PRISMA study flow starting with the initially identified citations down to the number of included studies. The figure also includes the breakdown of reasons for exclusion. The flow describes in parentheses the breakdown of the number of newly identified citations for the update.

Figure 2

Literature flow diagram.

In total, 338 studies reported in 518 publications across both reports met inclusion criteria.⁵¹^–³⁸⁸^, ⁹⁰⁷^–¹⁰⁸⁶ The most common reason for exclusion was the intervention type (n=1,724); some studies were excluded because they were included in the 2011 Institute of Medicine (IOM) report or were published prior to IOM report for vaccines included in that report (n=164). The most frequent study designs identified in the update were randomized controlled trials (RCTs) (n=104), followed by cohort studies (n=40), pre-post designs (n=15), case-control designs (n=12), and one non-randomized controlled clinical trial, along with 17 others that used self-controlled designs (either self-controlled risk interval or self-controlled case series analyses; two of these used self-controlled designs in conjunction with a cohort approach). Studies reported on a variety of datasets, ranging in size from fewer than 50 to millions of data points (Figure 3).

The figure displays the range of the sample size in a column graph. A description of the figure is provided right above the figure.

Figure 3

Range and frequency of sample sizes across studies.

The number of studies for each vaccine is listed in Figure 4.

The figure is a column graph displaying the frequency of the included studies by vaccines. A description of the figure is provided under the figure.

Figure 4

Number of studies evaluated for vaccines identified in this update. Notes: Data given for the report update at the time of the draft report. aIIV—Adjuvanted inactivated influenza vaccine; DTaP—Diphtheria, tetanus, and acellular pertussis (more...)

Studies most frequently evaluated influenza vaccines and rotavirus vaccines. Post-vaccination follow-up periods varied (Figure 5). A large number of studies followed patients for 6 months to record emerging adverse events, and some studies reported data of up to 15 years for some of the included study participants. Of all studies that reported a follow-up period, 23 percent reported only on events occurring 42 days or less from vaccine administration.

Figure 5

Length of followup in included studies.

The methodological quality and the reporting of the adverse events varied widely across studies over 15 assessed critical appraisal domains based on the McHarm tool (Figure 6). Details regarding these domains and their results are documented in more detail in Appendix C. Most studies reported the timing and frequency of the adverse events assessment, but only a few studies reported on the training and background of the outcome assessors.

The figure is a critical appraisal of the 189 included studies using the McHarms tool for assessing quality of harms. The figure shows the distribution of high risk of bias and low risk of bias studies within the sample.

Figure 6

Critical appraisal of included studies.

In the remainder of this chapter, we describe the results stratified by population (KQ1 addresses adult populations; KQ2 addresses children as well as evidence from studies that included both children and adults; and KQ3 focuses on pregnant women and their offspring, specifically). Within populations, we stratify by vaccine.

We focus on the synthesis of findings across studies; more information on individual studies is reported in Appendix C. We summarize results for key adverse events selected with the help of content experts and other findings. Results for all other outcomes are shown in the evidence tables stratified by population in Appendix D. All included studies report on the presence or absence of at least one adverse event, but the number of adverse events varies across studies, and a small number of studies assess several hundred adverse events.

Many of the findings we describe are from RCTs, of which some were placebo-controlled (or in which both intervention and control groups received the same base treatment, such as other routine vaccines) while others used an active comparator. We note the comparator for each study directly in the results section, as well as in the evidence tables in Appendix D.

This chapter provides responses to each of the KQs. The responses are divided into subquestions that are parallel across each KQ, as follows.

KQ1a, 2a, and 3a describe the spectrum of types of adverse events collected across all the included studies, and KQ1b, 2b, and 3b identify the specific adverse events reported for each vaccine.

KQ1c, 2c, 3c, and 3d are further divided into three subquestions. In KQ1c1, 2c1, 3c1, and 3d1, we examine the average severity and frequency of adverse events for which statistically significant associations were observed with a specific vaccine; KQ3c1 includes reports of associations that involve the pregnant woman, and KQ3d1 reports on associations that involve the fetus or infant. In KQ1c2, 2c2, 3c2, and 3d2, we assess the range of apparent associations between a vaccine and a specific type of event, including where the relative risk (RR) did not favor the intervention (meaning the RR was >1) but that did not reach statistical significance given that some of the adverse events are rare. In KQ1c3, 2c3, 3c3, and 3d3 we describe the risk factors for adverse events. At the end of each Key Question section, the summary of findings is presented.

Key Question 1. What is the evidence that vaccines included in the immunization schedule recommended for adults are safe in the short term or long term?

This section describes the evidence for the safety of vaccines routinely recommended for use in adults.

Key Points

Hepatitis B vaccines: No evidence of increased risk^b of diabetes or multiple sclerosis onset or exacerbation (moderate SoE) both remain unchanged from prior report. Anaphylaxis in patients allergic to yeast (moderate SoE) also remains unchanged from prior report. No evidence of increased risk of asthma, autoimmune disease, cardiovascular events, death, diabetes, herpes zoster, reproductive system events, or stroke associated with the new hepatitis B vaccine with a novel immunostimulatory adjuvant (low SoE).
Quadrivalent inactivated influenza vaccines: No evidence of increased risk of asthma, cardiovascular events, death, myocardial infarction, reproductive system events, seizures, or stroke (low SoE).
Adjuvanted inactivated influenza vaccine (trivalent or quadrivalent): No evidence of increased risk of asthma, autoimmune disease, cardiovascular events, death, encephalitis/encephalopathy, Guillain-Barré syndrome, idiopathic thrombocytopenic purpura, myocardial infarction, seizures, or stroke (all low SoE except for cardiovascular events and stroke, which were moderate SoE).
Quadrivalent recombinant influenza vaccine: No evidence of increased risk of cardiovascular events, death, encephalitis/encephalopathy, myocardial infarction, reproductive system events, or stroke (low SoE).
Measles, mumps, rubella vaccine: No evidence of increased risk of type 1 diabetes mellitus (moderate SoE) remains unchanged from prior report.
Serogroups A, C, W, and Y meningococcal vaccines: No evidence of increased risk of cardiovascular events, death, myocardial infarction, or stroke (low SoE except for death, which was moderate SoE).
13-valent pneumococcal conjugate vaccine: No evidence of increased risk of acute disseminated encephalomyelitis, anaphylaxis or systemic allergic reactions, asthma, autoimmune disease, cardiovascular events, death, encephalitis/encephalopathy, herpes zoster, idiopathic thrombocytopenic purpura, meningitis, myocardial infarction, reproductive system events, seizures, or stroke (low or moderate SoE).
23-valent pneumococcal polysaccharide vaccine: No evidence of increased risk of death (moderate SoE). No evidence of increased risk of cardiovascular or cerebrovascular events in adults (high SoE).
Tetanus, diphtheria, and acellular pertussis and tetanus and diphtheria vaccines: Anaphylaxis to tetanus toxoid (high SoE) remains unchanged from prior report.
Recombinant zoster vaccine: No evidence of increased risk of amyotrophic lateral sclerosis, anaphylaxis or systemic allergic reaction, asthma, cardiovascular events, death, diabetes, encephalitis/encephalopathy, Guillain-Barré syndrome, herpes zoster, idiopathic thrombocytopenic purpura, meningitis, myocardial infarction, reproductive system events, seizures, or stroke (moderate or high SoE). No evidence of increased risk of acute disseminated encephalomyelitis, angioedema, ataxia, autoimmune disease, or autoimmune thyroiditis (Hashimoto’s disease) (low SoE).
Insufficient evidence to permit evidence statements for the following vaccines in adults: hepatitis A vaccine, combination hepatitis A and hepatitis B vaccine, 9-valent human papillomavirus vaccine, live attenuated influenza vaccine, serogroup B meningococcal vaccine, and varicella vaccine.

KQ1a. What adverse events are collected in clinical studies and in observational studies containing a control/comparison group?

The information on the data collection for vaccines in adults comes from trial records, published protocols, and all published papers available for an individual study. The information was collected from experimental and observational studies. The number and reporting detail of collected adverse events varied widely, and studies often assessed general, non-specific adverse events, such as hospitalizations, along with specific adverse events, such as Guillain-Barré syndrome. Some studies assessed dozens of adverse events, while others assessed several hundred. The table in Appendix C lists the adverse events collected in the included studies by vaccine to document the presence and the absence of adverse events. The table lists all assessed adverse events for a vaccine group (e.g., HepA ) in alphabetical order.

The methods used to collect and assess adverse events are summarized for each study in the evidence tables in Appendix D. The risk of bias table in Appendix C documents the methods used to ascertain and report adverse events.

KQ1b. What adverse events are reported in clinical studies and in observational studies containing a control/comparison group?

The evidence tables in Appendix D contain all serious and severe adverse events reported in studies of adults.

KQ1c. What adverse events are associated with these vaccines?

This section further characterizes the risk of adverse events identified in studies of adults who received routinely recommended vaccines.

KQ1c1. For each adverse event associated with a particular vaccine, what is the average severity and frequency?

The prior 2014 report found several associations, which are detailed in that report and summarized at the end of this KQ section in tables that integrate the findings from both reports. Several of these associations were in trivalent inactivated influenza vaccine (IIV) or monovalent H1N1 influenza vaccine that are no longer in use because they have been replaced by quadrivalent influenza vaccines. For those vaccines that are still in use, the associations between adverse events and specific vaccines established in the prior 2014 report and/or Institute of Medicine (IOM) report remain unchanged because there were no new studies or the outcome was not considered a key adverse event (in the case of transient arthralgia). These include anaphylaxis in patients allergic to yeast for hepatitis B vaccines (HepB; moderate SoE), transient arthralgia in women for measles, mumps, and rubella vaccine (MMR; moderate SoE), and anaphylaxis to tetanus toxoid for tetanus toxoid-containing vaccines (high SoE).

We did not find any new associations between any of the studied vaccines and the key adverse events, including for new vaccines not reviewed in the prior 2014 report. For all events, the severity for each event is documented in the evidence table in Appendix D.

KQ1c2. For adverse events without statistically significant associations with a particular vaccine, what is the range of possible effects?

We summarize the range of possible effects by vaccine below for all adverse events for which the primary relative risk estimate did not favor the intervention (i.e., RR>1), but the association was not statistically significant. We also note any instances where an individual study contributing to a pooled risk estimate indicated a statistically significant increased risk of an adverse event. Throughout this section, where possible, we describe caveats to contextualize the finding, such as noting the confidence interval (e.g., many confidence intervals were extremely wide because only a single event was reported in the study or several studies with rare events were combined for meta-analysis) or assessing whether study authors attributed the event to something other the vaccine (e.g., death due to a motor vehicle accident or drug overdose).

Hepatitis Vaccines

Hepatitis A Vaccine

We did not identify new studies reporting on hepatitis A vaccine (HepA) in this update.

Hepatitis B Vaccines

We identified five studies⁹⁹^, ¹⁰⁰^, ¹³²^, ¹³⁴^, ¹⁵¹ evaluating HepB in this update. Three RCTs⁹⁹^, ¹⁰⁰^, ¹³² compared HepB with a novel immunostimulatory adjuvant (HEPLISAV-B^®) to the previously available HepB (Engerix-B^®), one cohort study¹³⁴ compared HepB (Engerix-B or Recombivax HB^®) to no vaccine, and one case-control epidemiological study¹⁵¹ compared HepB (Engerix-B or Recombivax HB) to no vaccine.

Across the three RCTs⁹⁹^, ¹⁰⁰^, ¹³² of HepB with novel adjuvant, we found no evidence of increased risk of herpes zoster (RR 1.95; confidence interval [CI] 0.45, 8.51), though one RCT⁹⁹ showed a significantly increased risk (RR 2.10; CI 1.02, 4.34; 38/5587 vs 9/2781). The relative risk of cardiovascular events across these same RCTs was 1.55 (CI 0.46, 5.18). Across two of the RCTs,⁹⁹^, ¹³² we found no evidence of increased risk of myocardial infarction (RR 2.00; CI 0.00, 41600564). However, the effect estimate was imprecise and with substantial heterogeneity (I² 64%), possibly due to a non-statistically significant imbalance of myocardial infarction (14/5887 [0.25%] vs 1/2781 [0.04%]) in one of the RCTs.⁹⁹ We found no evidence of increased risk of death (RR 0.97; 0.04, 23.18), although again one of the RCTs found a non-statistically significant imbalance in deaths (25/5587 [0.44%] vs 7/2781 [0.25%]). These events—and particularly an imbalance in the rate of death and myocardial infarction in one of the RCTs⁹⁹ though not the other two RCTs—were noted to be of concern to the Food and Drug Administration (FDA).³⁷⁹ For myocardial infarction in particular, the rate remained similar even once cases were adjudicated by experts (0.25% vs 0.04%), and when looking more broadly at cardiac disorders (0.9% vs 0.5%), despite baseline risk factors for cardiovascular disease having been balanced across the intervention and control groups. The manufacturer proposed two post-marketing observational surveillance studies of approximately 60,000 people—one assessing herpes zoster and other outcomes and the other acute myocardial infarction specifically—which started in August 2018.¹⁰⁸⁷ According to the manufacturer in a statement¹⁰⁸⁸ released in December 2019, interim results from the acute myocardial infarction study suggested no increased risk with the vaccine.

Across the three RCTs,⁹⁹^, ¹⁰⁰^, ¹³² we found no evidence of increased risk of asthma (RR 0.83; CI 0.03, 22.31). We found no evidence of increased risk of autoimmune disease (RR 0.70; CI 0.15, 3.21) or reproductive system events (RR 0.21; CI 0.00, 13.46). Across two of these RCTs,⁹⁹^, ¹⁰⁰ we found no evidence of increased risk of stroke (RR 1.27; CI 0.00, 1430); the effect estimate was imprecise and based on two studies reporting 11/5587 cases in the vaccine group vs 4/2781 cases in the comparator group in one study and 1/1821 vs 0/607 cases in the other. One of these RCTs⁹⁹ found RR>1 for encephalitis/encephalopathy (RR 1.99; CI 0.09, 44.14), meningitis (RR 1.99; 0.09, 44.14), and seizures (RR 1.99; CI 0.22, 17.81). For encephalitis/encephalopathy and meningitis, the risk estimates were imprecise as there were only two cases each in the vaccinated group (2/5587 vs 0/2781). There were also few cases of seizures, resulting in a wide confidence interval (4/5587 vs 1/2781).

Another RCT²⁸⁸ reported no significant difference in medically attended (meaning the subject sought medical attention for the adverse event) grade 3 or 4 adverse events (HEPLISAV-B 23.5%, Engerix-B 22.2%) in people aged 60 to 70 years with diabetes mellitus but the study reported not enough detail for further analyses. No serious adverse events related to vaccines were reported in either group. In one cohort study¹³⁴ comparing HepB (presumed to be Engerix-B or Recombivax HB as it took place prior to use of HEPLISAV-B) to no vaccine, there was no evidence of increased risk of diabetes (RR 0.61; CI 0.55, 0.67). When combined with an RCT⁹⁹ that compared HEPLISAV-B to a non-adjuvanted HepB, the risk estimate remained similar (RR 0.61; CI 0.33, 1.11). In the epidemiological study for HepB (Engerix-B or Recombivax HB),¹⁵¹ vaccination was not associated with increased risk for central nervous system demyelinating syndromes (odds ratio [OR] 0.8; CI 0.4, 2.0), including acute disseminated encephalomyelitis (OR 1.5; CI 0.2, 12.1) and multiple sclerosis (OR 0.4; CI 0.1, 1.9) for up to three years post-vaccination.

Combined Hepatitis A-Hepatitis B Vaccine

This update identified two studies⁶²^, ⁶³ of HepA-HepB (Twinrix^®), which included both children and adults; we presumed that only adults received HepA-HepB as the ages were not specified by vaccine. The case-centered analyses compared the vaccination status of each case to vaccination of all matched persons in the study population who received the same type of vaccine with respect to the exposure interval. One case-centered analysis⁶² showed no evidence of increased risk of optic neuritis in the 2- to 42-day risk interval after HepA-HepB administration (adjusted OR was not estimable). Another case-centered analysis⁶³ detected no association between sudden sensorineural hearing loss and HepA-HepB within one week of vaccine administration (OR 2.39; CI 0.37, 9.14).

9-Valent Human Papillomavirus Vaccine

We did not identify studies assessing 9-valent human papillomavirus vaccine (HPV9) in adults in this update. More evidence on adverse events reported for HPV9 in studies of children and adults together is documented in KQ2.

Influenza Vaccines

We identified 17 studies assessing influenza vaccines currently in use (all quadrivalent, except for trivalent adjuvanted inactivated influenza vaccine [aIIV] as it is still in use) in this update. Thirteen studies (reported in 24 publications)⁶⁰^, ⁶⁶^, ⁷⁷^, ¹⁰³^, ¹¹⁰^, ¹¹⁸^, ¹¹⁹^, ¹⁴⁵^, ¹⁸⁵^, ²⁰⁴^, ²¹⁶^, ²¹⁸^, ²³²^, ²⁵⁹^, ²⁶²^, ²⁶⁴^, ²⁹⁶^, ³²⁹^, ³³⁰^, ³⁶⁰^, ³⁶²^, ³⁶⁸^, ³⁷³^, ³⁷⁴ assessed quadrivalent inactivated influenza vaccines (IIV) and four studies (reported in five publications)⁸⁴^, ⁹⁷^, ⁹⁸^, ¹⁹²^, ³⁷⁸ assessed quadrivalent recombinant influenza vaccines (RIV). Studies of IIV included one study⁶⁰ of Flucelvax Quadrivalent^®, two studies⁶⁶^, ¹⁴⁵ of Fluarix Quadrivalent^®, two studies¹¹⁸^, ¹¹⁹ of Fluzone Quadrivalent^®, one study⁷⁷ of Fluzone High Dose Quadrivalent^® one study²¹⁶ of Flulval Quadrivalent^®, and one study of Afluria Quadrivalent^®.²¹⁸ We identified six studies¹⁰³^, ¹¹⁰^, ¹⁸⁰^, ¹⁸⁵^, ²⁰⁴^, ²³² of aIIV (four of Fluad^® and two of Fluad Quadrivalent^®). In addition, we identified one study⁷¹ of quadrivalent live attenuated influenza vaccine (LAIV) in adults.

Quadrivalent Inactivated Influenza Vaccines (Non-Adjuvanted)

Across six studies⁶⁰^, ⁶⁶^, ⁷⁷^, ¹¹⁸^, ¹⁴⁵^, ²¹⁸ that compared quadrivalent IIV to trivalent IIV, we found no evidence of increased risk of death (RR 1.03; CI 0.41, 2.59). Of the six studies, two studies reported no deaths in either group; two reported deaths that were explicitly not considered attributable to the vaccine by the study authors; one did not comment on whether deaths were related to the vaccine; and one found that one of the deaths was considered related to the intervention vaccine.

Across five studies⁶⁰^, ⁷⁷^, ¹⁴⁵^, ²¹⁶^, ²¹⁸ that compared quadrivalent IIV to trivalent IIV, we found no evidence of increased risk of cardiovascular events (RR 1.11; CI 0.34, 3.68) or of myocardial infarction specifically (RR 1.15; CI 0.27, 4.86). Although the RR was greater than 1, the estimate is not statistically significant. We also note that these studies compare quadrivalent to trivalent IIV; the prior report noted no increased risk of cardiovascular events (in adults aged 65 years and older; high SoE) when comparing trivalent IIV to a non-active comparator. Across four studies⁷⁷^, ¹⁴⁵^, ²¹⁶^, ²¹⁸ that compared quadrivalent IIV to trivalent IIV, we found no evidence of increased risk of stroke (RR 0.87; CI 0.13, 5.89). Across four studies,⁶⁰^, ⁷⁷^, ¹⁴⁵^, ²¹⁶ one that compared a quadrivalent cell-cultured IIV to trivalent IIV,⁶⁰ one⁷⁷ that compared quadrivalent IIV to two other trivalent IIVs, and two¹⁴⁵^, ²¹⁶ that compared quadrivalent IIV to one trivalent IIV, we found no evidence of increased risk of reproductive system events (RR 0.75; CI 0.06, 8.76). Across three studies,¹⁴⁵^, ²¹⁶^, ²¹⁸ each comparing quadrivalent IIV to trivalent IIV, we found no evidence of increased risk of asthma (RR 0.46; CI 0.01, 24.59). Across two studies,⁶⁰^, ⁷⁷ one that compared quadrivalent cell-cultured IIV to trivalent IIV⁶⁰ and one⁷⁷ that compared quadrivalent IIV to two other trivalent IIVs, we found no evidence of increased risk of seizure (RR 0.17; CI 0.00, 478774); the effect estimate was imprecise and based on with few events occurring across the two studies (0/1324 vs 2/673 and 0/1777 vs 1/893). Across another two studies,⁶⁶^, ²¹⁸ one⁶⁶ that compared quadrivalent IIV with trivalent IIV and one²¹⁸ that compared quadrivalent IIV with two trivalent IIVs containing alternate B strains, we found no evidence of increased risk of autoimmune disease (RR 1.49; CI 0.00, 24522910); the effect estimate was again imprecise due to only one event occurring across both studies (0/105 vs 0/105 and 1/1721 vs 0/1728). One study⁶⁰ found an RR greater than 1 when assessing amyotrophic lateral sclerosis (RR 2.02; CI 0.07, 60.19) and meningitis (RR 2.02; CI 0.07, 60.19); the risk estimates are imprecise as only one event each was reported in the vaccinated group (1/1324 vs 0/1338).

A sensitivity analysis of the new quadrivalent cell-based IIV (Flucelvax Quadrivalent, assessed in one study⁶⁰ in adults) compared to trivalent cell-based IIV showed no evidence of increased risk of amyotrophic lateral sclerosis (RR 2.02; CI 0.07, 60.19; 1/1324 vs 0/1338), anaphylaxis or systemic allergic reaction (RR 0.51; CI 0.02, 15.05), cardiovascular outcomes (RR 2.53; CI 0.49, 13.00), meningitis (RR 2.02; CI 0.07, 60.19; 1/1324 vs 0/1338), myocardial infarction (RR 4.04; CI 0.45, 36.12), reproductive system events (RR 0.51; CI 0.02, 15.05), seizures (RR 0.13; CI 0.01, 2.81), or death (RR 0.72; CI 0.23, 2.27). A sensitivity analysis of the high-dose quadrivalent IIV for adults aged 65 years and older (Fluzone Quadrivalent High Dose, assessed in one study⁷⁷) compared to high-dose trivalent IIV showed no evidence of increased risk of cardiovascular events (RR 0.34; CI 0.06, 2.00), myocardial infarction (RR 0.50; CI 0.07, 3.56), reproductive system events (RR 1.01; CI 0.09, 11.07), seizures (RR 0.25; CI 0.01, 7.48), stroke (RR 0.25; CI 0.01, 7.48), or death (RR 0.75; CI 0.13, 4.50).

Adjuvanted Inactivated Influenza Vaccines

For all outcomes except for seizures and death, the risk estimates were imprecise because a small number of events were being reported across only two studies (see rates of each event following the risk estimates).

Across three studies¹⁰³^, ¹¹⁰^, ²³² (one¹⁰³ that compared quadrivalent aIIV to two strains of trivalent aIIV and two¹¹⁰^, ²³² that compared trivalent aIIV to trivalent IIV), there was no evidence of increased risk of seizures (RR 0.93; CI 0.36, 2.38). Across two studies¹⁰³^, ¹¹⁰ (one¹⁰³ that compared quadrivalent aIIV to two strains of trivalent aIIV, and one¹¹⁰ that compared trivalent aIIV with trivalent IIV) there was no evidence of increased risk of death (RR 1.15; CI 0.09, 14.54). Although the RR was greater than 1, in neither study were any of the deaths attributed to vaccines.

Across the same two studies,¹⁰³^, ¹¹⁰ there was no evidence of increased risk of cardiovascular events (RR 0.47; CI 0.00, 81.39; 1/888 vs 3/888 and 8/3545 vs 16/3537) or stroke (RR 1.18; CI 0.00, 33607; 0/888 vs 1/888 and 3/3545 vs 2/3537). A case-control study¹⁸⁵ of trivalent aIIV also showed no increased risk of acute coronary syndrome (adjusted odds ratio [aOR] 0.13; 0.03, 0.65) or stroke (aOR 0.07; CI 0.01, 0.48). There was also no evidence of increased risk of asthma (RR 0.76; CI 0.00, 836505; 0/888 vs 1/888 and 1/3545 vs 1/3537), myocardial infarction (RR 1.05; CI 0.00, 295; 0/888 vs 1/888 and 10/3545 vs 9/3537).

Across two studies,¹⁰³^, ²³² one¹⁰³ that compared quadrivalent aIIV to two strains of trivalent aIIV and one²³² that compared trivalent aIIV to trivalent IIV, there was no evidence of increased risk of autoimmune disease (RR 0.98; CI 0.00, 2288259; 1/888 vs 1/888, 0/88449 vs 0/82539) or encephalitis/encephalopathy (RR 0.48; CI 0.00, 2761252; 0/888 vs 1/888, 0/88449 vs 1/82539). Across another two studies¹¹⁰^, ²³² that compared trivalent aIIV to trivalent IIV, there was no evidence of increased risk of Guillain-Barré syndrome (RR 0.29; CI 0.00, 44494; 0/3545 vs 1/3537, 1/88449 vs 4/82539). Across the same two studies,¹¹⁰^, ²³² there was also no evidence of increased risk of idiopathic thrombocytopenic purpura (RR 2.52; CI 0.00, 505129; 1/3545 vs 0/3537, 3/88449 vs 1/82539). While both studies showed more events in the vaccinated group, in the first study¹¹⁰ the investigators did not consider the event to be related to vaccine; the second study²³² did not comment on causality. In one active surveillance study¹⁸⁰ of Medicare beneficiaries following influenza vaccination that used a self-controlled risk interval analysis, there was an increased risk of Guillain-Barré syndrome following trivalent aIIV (OR 3.75; CI 1.01, 13.96) with an attributable risk of 2.5 episodes per million vaccinations. However, when adjusted for multiplicity the result was no longer robust (q=0.15). One RCT²³² found an RR greater than 1 for anaphylaxis (RR 1.87; CI 0.06, 55.63), but the risk estimate was imprecise due to one event occurring in the aIIV group (1/88449 vs 0/82539). When analyzed separately, trivalent aIIV and quadrivalent aIIV still showed no evidence of increased risk of any of the above outcomes. One study²⁰⁴ of trivalent aIIV plus PPSV23 versus PPSV23 alone reported no vaccine-related serious adverse events.

Quadrivalent Recombinant Influenza Vaccine

Across three studies⁹⁷^, ⁹⁸^, ¹⁹² of quadrivalent RIV compared with quadrivalent IIV, we found no evidence of increased risk of cardiovascular events (RR 1.23; CI 0.34, 4.46). The RR was greater than 1, largely due to one study with four excess events (0/335 vs 0/508, 23/4328 vs 19/4344, 2/998 vs 0/332); no risk estimates from individual studies were statistically significant on their own. Across the same three studies, there was no evidence of increased risk of myocardial infarction specifically (RR 1.35; CI 0.08, 22.21); the risk estimate was imprecise due to few events occurring across three studies of varying sizes (0/335 vs 0/508; 4/4328 vs 3/4344; 2/998 vs 0/332). Across the same three studies, there was no evidence of increased risk of death (RR 0.66; CI 0.10, 4.26). Across two of the studies,⁹⁷^, ⁹⁸ we found no evidence of increased risk of encephalitis/encephalopathy (RR 0.58; CI 0.00, 3294190), or reproductive system events (RR 0.41; CI 0.00, 82495). The risk estimates for both outcomes were imprecise due to few events occurring across each of the two studies (0/4328 vs 1/4344 and 1/998 vs 0/332 for encephalitis/encephalopathy; 1/4328 vs 3/4344 and 1/998 vs 0/332 for reproductive system events).

Quadrivalent Live Attenuated Influenza Vaccine

No outcomes were assessed in more than one study of quadrivalent LAIV in adults. In one study⁷¹ comparing quadrivalent LAIV to two different trivalent LAIV, we found no evidence of increased risk of anaphylaxis (RR 0.25; CI 0.01, 7.43), asthma (RR 0.25; CI 0.01, 7.43), cardiovascular events (RR 0.25; CI 0.01, 7.43), myocardial infarction (RR 0.25; CI 0.01, 7.43), reproductive system events (RR 0.12; CI 0.01, 2.76), or death (RR 0.50; 0.01, 25.13).

Measles, Mumps, and Rubella Vaccine

We did not identify studies assessing MMR in adults in this update. Results in children and mixed samples (adults and children) are documented in KQ2.

Meningococcal Vaccines

Meningococcal A, C, W, and Y Vaccines

We identified four studies⁵³^, ⁵⁴^, ¹⁰⁴^, ¹⁴⁷ assessing serogroups A, C, W, and Y meningococcal vaccine (MenACWY) in adults in this update. Studies compared MenACWY-TT versus meningococcal polysaccharide vaccine (MPSV),¹⁰⁴^, ¹⁴⁷ MenACWY-CRM plus typhoid and yellow fever vaccines versus typhoid and yellow fever vaccines alone,⁵³ and MenACWY-CRM plus HepA-HepB versus HepA-HepB alone.⁵⁴ No findings were reported for MenACWY in adults in the prior 2014 report.

Both studies⁵³^, ⁵⁴ of MenACWY-CRM were performed in adults aged 18–60 or 18–64 years of age; this vaccine is approved through 55 years of age, but the majority of adults in these studies fell within this range. Both studies¹⁰⁴^, ¹⁴⁷ of MenACWY-TT were performed in adults aged 56 years and older; this vaccine has no upper limit of age. Across the two studies⁵³^, ⁵⁴ that compared MenACWY-CRM to base treatment received by both the intervention and control groups, we found no evidence of increased risk of death (RR 0.99; CI 0.00, 60563320). The risk estimate was imprecise due to no events occurring in either study (0/99 vs 0/100 and 0/85 vs 0/84). Combining the four MenACWY studies also showed no evidence of increased risk of death (RR 0.53; CI 0.03, 10.33).

In a sensitivity analysis of the newer vaccine, MenACWY-TT, two studies¹⁰⁴^, ¹⁴⁷ that used an existing meningococcal polysaccharide vaccine as the comparator showed no evidence of increased risk of death (RR 0.33; CI 0.00, 2256832). In a study of MenACWY-TT compared to MenACWY-CRM,¹⁰⁴ there were no systematic difference in cardiovascular events (RR 0.25; CI 0.01, 5.59), myocardial infarction (RR 0.25; CI 0.01, 5.59), reproductive system events (RR 2.02; CI 0.07, 60.13), or stroke (RR 0.51; CI 0.02, 15.03). The RR was greater than 1 for reproductive events but was imprecise due to only one event occurring in the study (1/448 vs 0/453).

Meningococcal B Vaccines

We did not identify any evaluations of serogroup B meningococcal vaccine (MenB) in adults in this update or in the original 2014 report.

Pneumococcal Vaccines

We identified 31 studies⁷²^, ⁸⁰^, ⁸¹^, ¹⁰¹^, ¹⁰⁷^, ¹⁰⁹^, ¹²⁰^, ¹³⁷^, ¹⁴⁰^, ¹⁴²^, ¹⁴³^, ¹⁵⁹^, ¹⁶²^, ¹⁷¹^, ¹⁷⁶^, ¹⁸⁴^, ¹⁹⁵^, ¹⁹⁷^, ¹⁹⁹^–²⁰⁵^, ²¹⁰^, ²¹⁴^, ²²²^, ²²⁷^, ²³¹^, ²³⁹ assessing pneumococcal vaccines (13-valent pneumococcal conjugate vaccine [PCV13] or 23-valent pneumococcal polysaccharide vaccine [PPSV23]). One study⁹³⁵ of PPSV23 from the prior report that examined cardiovascular events and death could be pooled statistically with new studies from the update.

13-Valent Pneumococcal Vaccine

PCV13 was not assessed in the prior 2014 report for adults. In the current report, one study assessed PCV13 compared to placebo,⁷² one assessed PCV13 plus trivalent aIIV versus trivalent aIIV alone,²⁰² three assessed PCV13 plus trivalent IIV (Fluzone^® in one study, Fluarix^® in two other studies) versus trivalent IIV plus placebo,¹⁰⁹^, ¹⁹⁵^, ²¹⁴ and one assessed PCV13 plus Td versus Td alone.²⁰³ One large epidemiological study compared individuals vaccinated with PCV13 to individuals who were not vaccinated.²³¹ Twelve studies compared PCV13 to PPSV23.¹⁰¹^, ¹²⁰^, ¹⁴⁰^, ¹⁴²^, ¹⁴³^, ¹⁶²^, ¹⁸⁴^, ¹⁹⁷^, ²⁰⁰^, ²¹⁰^, ²²²^, ²²⁷

Across four studies,⁷²^, ¹⁰⁹^, ¹⁹⁵^, ²¹⁴ we found no evidence of increased risk of cardiovascular events (RR 0.97; CI 0.58, 1.64) or myocardial infarction specifically (RR 1.76; CI 0.42, 7.39) and this finding remained consistent when we combined these studies with another two studies¹²⁰^, ¹⁸⁴ that used PPSV23 as an active comparator (RR 0.97 for cardiovascular events [CI 0.64, 146] and RR 1.51 for myocardial infarction [CI 0.52, 4.33]). Of the six studies⁷²^, ¹⁰⁹^, ¹²⁰^, ¹⁸⁴^, ¹⁹⁵^, ²¹⁴ that together found an RR greater than 1 for myocardial infarction, all but one of these studies had either no events or one events in the intervention and/or control group (1/576 vs 0/575, 0/551 vs 1/560, 12/42237 vs 6/42255, 1/478 vs 1/237, 0/439 vs 0/437, 1/417 vs 1/414).

Across four PCV13 studies,⁷²^, ¹⁹⁵^, ²¹⁴^, ²³¹ we found no evidence of increased risk of death (RR 1.85; CI 0.37, 9.31). We also found no increased risk of death when we combined these studies together with another four studies¹²⁰^, ¹⁴⁰^, ²⁰⁰^, ²²⁷ that used active pneumococcal vaccine comparators (RR 1.22; CI 0.40, 3.69), but there was substantial heterogeneity (I² 99%). We note that one study²³¹ in particular had a significantly increased RR of 3.62 (CI 3.30, 3.96) based on unadjusted rates from the cohort (420/5010 vs 46845/2020720), which contributed to the pooled estimate. When the same study adjusted for age, sex, history of pneumococcal disease or pneumonia, and comorbidities, the risk estimate was lower and no longer significant (multivariate hazard ratio [mHR] 1.07 [0.97–1.18]). Of the eight studies⁷²^, ¹²⁰^, ¹⁴⁰^, ¹⁹⁵^, ²⁰⁰^, ²¹⁴^, ²²⁷^, ²³¹ that examined the risk of death, four did not comment on whether the deaths could be attributed to the vaccine, one study noted that the one death was not attributable to the vaccine, and three studies reported no deaths. Although the relative risk is not statistically significant which precludes conclusions about the risk, increased mortality in conjunction with this vaccine would be surprising given that the PCV13 decreases pneumonia. It is also possible that the studies’ time frames were too short to capture the decreased mortality over the longer term as a result of vaccination.

Across three studies,⁷²^, ¹⁰⁹^, ²¹⁴ we found no evidence of increased risk of reproductive system events (RR 0.59; CI 0.01, 42.46); this was consistent when we included two other studies¹²⁰^, ¹⁸⁴ that used PPSV23 as an active comparator (RR 0.71; CI 0.08, 6.42). Across two studies,⁷²^, ¹⁹⁵ we found no evidence of increased risk of herpes zoster (RR 1.49; CI 0.00, 24855526); the risk estimate was imprecise as it was based on two studies with only one event across the intervention and control groups (0/576 vs 0/575 and 1/42337 vs 0/42255). Across another two studies,⁷²^, ¹⁰⁹ we found no evidence of increased risk of stroke (RR 1.12; CI 0.00, 451); the risk estimate was imprecise given that the two studies were of different sizes and one study had no events (0/551 vs 0/560 and 9/42237 vs 8/42255). In one study,⁷² we found no evidence of autoimmune disease (RR 1.00; CI 0.02, 50.42); of note this risk estimate was imprecise as it was based on one study with no events in either the intervention or control groups (0/42337 vs 0/42255). The finding remained similar when combined with another study¹⁸⁴ that used PPSV23 as an active comparator (RR 0.67; CI 0.00, 11165626).

One study⁷² noted an RR greater than 1 for acute disseminated encephalomyelitis and meningitis following PCV13 (RR 2.00; CI 0.07, 59.64 for both outcomes, due to one event each in the intervention and control groups). Another study¹⁰⁹ noted an RR greater than 1 for encephalitis (RR 1.02; CI 0.02, 51.13) but neither group had any events of encephalitis. A cohort study²²² of over 545,000 people comparing PCV13 to PPSV23 found no evidence of increased risk of pre-specified adverse events with PCV13 compared to PPSV23. However, only anaphylaxis as an adverse event was verified by chart review; out of 9 anaphylaxis events (5 in the PCV13 group and 4 in the PPSV23 group), only one was verified as having occurred after vaccination, in a patient who received four other vaccines concomitantly. The study did not report sufficient detail for further analyses for other adverse events.

Regarding other outcomes that were not pre-specified key adverse events for the current report, a study¹⁴³ of PCV13 versus PPSV23 among individuals with Crohn’s disease reported no serious adverse events in the PCV13 group compared to two serious adverse events in the PPSV23 group (infection with enterococcus faecalis and sinusitis). Adverse events were assessed as unrelated to vaccination. A study¹⁶² of PCV13 compared to PPSV23 among individuals with HIV reported no serious adverse events in either group. A small RCT²¹⁰ of PCV13 versus PPSV23 in adults with leukemia reported no severe adverse events in either group. A small RCT²²⁷ of PCV13 versus PPSV23 in hemodialysis patients reported no difference in the risk of developing pneumonia or undergoing a kidney transplant between the two groups. No specific adverse events data were reported. Finally, an RCT¹⁰¹ of PCV13 versus PPSV23 in solid organ transplant recipients showed more serious adverse events following PCV13 within the first week following vaccination (7/66 among PCV13 recipients compared to 1/66 among PPSV23 recipients), but no events were considered related to vaccination and no vaccine-related allograft rejection was noted.

One study²⁰³ reported that no serious adverse events occurred, while three others¹⁴²^, ¹⁹⁷^, ²⁰² reported that no vaccine-related serious adverse events occurred (though did not specify what other serious adverse events occurred).

23-Valent Pneumococcal Vaccine

Seven studies assessed PPSV23 versus placebo or no vaccine,⁸¹^, ¹⁵⁹^, ¹⁷¹^, ¹⁷⁶^, ²⁰¹^, ²³⁹^, ⁹³⁵ two⁸⁰^, ¹³⁷ assessed PPSV23 plus trivalent IIV (one used Vaxigrip^®, a formulation of IIV not available in US and the other did not specify brand) versus trivalent IIV alone, and one each assessed PPSV23 plus trivalent IIV (a Japanese formulation not available in the US) versus trivalent IIV plus placebo,¹⁹⁹ PPSV23 plus quadrivalent IIV (Influsplit, a formulation of IIV not available in US) versus quadrivalent IIV alone,²⁷³ and PPSV23 plus trivalent aIIV versus trivalent aIIV alone.²⁰⁴^, ²⁰⁵ A small cohort study¹⁰⁷ of individuals with a diagnosis of immune-mediated inflammatory disorder undergoing treatment with high-dose systemic corticosteroids and/or immunosuppressive drugs compared those vaccinated with PPSV23 with those not vaccinated against pnuemococcal disease.

Across four studies of PPSV23 (two cohort studies²³⁹^, ⁹³⁵ comparing to vaccinated to unvaccinated patients, one cohort study⁸⁰ comparing to trivalent IIV alone, and one RCT¹⁹⁹ comparing PPSV23 plus trivalent IIV to trivalent IIV plus placebo), we found no evidence of increased risk of cardiovascular events (RR 0.46; CI 0.27, 0.76). A cohort study¹⁷⁶ of people aged 60 years and older in Spain that compared PPSV23 to no vaccine reported no increased risk of no increased risk of myocardial infarction (mHR 0.95; CI 0.76, 1.18).

The same four studies⁸⁰^, ¹⁹⁹^, ²³⁹^, ⁹³⁵ found no evidence of increased risk of death (RR 0.62; CI 0.16, 2.44). A cohort study¹³⁷ comparing PPSV23 plus trivalent IIV to trivalent IIV alone reported that dual-vaccinees experienced fewer deaths (hazard ratio [HR] 0.65; CI 0.55, 0.77) and fewer cases of pneumonia (HR 0.57; CI 0.51, 0.64), ischemic stroke (HR 0.67; CI 0.54, 0.83), and acute myocardial infarction (HR 0.52; CI 0.38, 0.71). The study did not provide sufficient detail for further analyses. Another cohort study¹⁵⁹ specifically of patients with prostate cancer in Taiwan found that the seven-year overall survival rate was significantly higher in the PPSV23-vaccinated group than in the unvaccinated group (47.5% vs 42.3%, p=0.0003). Finally, a cohort study¹⁷⁶ of people aged 60 years and older in Spain that compared PPSV23 to no vaccine reported no increased risk of death (mHR 0.97; CI 0.89, 1.05). The same cohort study¹⁷⁶ found no increased risk of myocardial infarction (mHR 0.95; CI 0.76, 1.18) nor of stroke (mHR 1.04; CI 0.83,1.30).

One cohort study⁸¹ found no increased risk of stroke (mHR 0.57; CI 0.31, 1.03 for the cohort study). A case-control study²⁰¹ also found no increased risk of stroke (aOR 0.94; CI 0.89 to 1.00) but found an increased risk of transient ischemic attacks (OR 1.16; 1.09, 1.23).

Regarding other outcomes, the cohort study¹⁰⁷ of PPSV23 in patients with immune-mediated inflammatory disorder found vaccinated and unvaccinated groups had a similar decrease in disease activity, and the study reported no adverse events. The RCT¹⁷¹ of PPSV23 versus placebo reported an increased risk for extensive arm swelling among participants who received PPSV23 compared to saline (0.55% vs 0.00%, p<0.001). The remaining RCTs¹⁷⁷^, ²⁰⁵ reported that no vaccine-related serious adverse events occurred, but did not provide more specific details about specific adverse events.

Tetanus, Diphtheria, and Acellular Pertussis Vaccines

We identified one study¹²⁴ that assessed Tdap in this update. This was an RCT of Tdap (Adacel^®) compared to Td (Tenivac^®) in adults who had received one dose of Tdap 8 to 12 years earlier. There was no evidence of increased risk for autoimmune disease or spontaneous abortion in either group. There were no deaths in either group and no significant differences in serious adverse events overall.

Varicella Vaccine

We did not identify studies assessing varicella vaccines in studies in adults, in this update or the original report. Results from studies of children and adults together are documented in KQ2.

Zoster Vaccine

We identified six studies⁸²^, ⁸⁵^, ¹⁵⁰^, ¹⁶⁵^, ¹⁹⁴^, ²⁰⁹ that assessed recombinant zoster vaccine (RZV) (which was not in use at the time of the prior report), all of which were RCTs comparing the vaccine to either placebo or a base treatment also received by the intervention group. One RCT²⁰⁹ compared RZV plus Tdap to Tdap alone, but the control group later also received RZV; thus, we could only include events that occurred after the first visit, which did not include any key adverse events (not discussed further here).

Across five studies,⁸²^, ⁸⁵^, ¹⁵⁰^, ¹⁶⁵^, ¹⁹⁴ we found no evidence of increased risk for herpes zoster occurring as an adverse event after vaccination (RR 0.09; CI 0.02, 0.30), but there was heterogeneity across studies (I² 58%). Across four studies⁸²^, ⁸⁵^, ¹⁵⁰^, ¹⁹⁴ we found no evidence of increased risk of death (RR 0.93; CI 0.78, 1.11). Across three studies,⁸²^, ⁸⁵^, ¹⁵⁰ we found no evidence of increased risk for cardiovascular events (RR 0.89; CI 0.66, 1.21) or myocardial infarction specifically (RR 0.89; CI 0.38, 2.05).

Many of the estimates across only two studies⁸⁵^, ¹⁵⁰ were imprecise given the low numbers of events across the intervention and control groups; we provide the rates as well for these estimates. We found no evidence of increased risk for anaphylaxis (RR 1.32; CI 0.00, 1463200; 1/6950 vs 1/6950 and 1/7695 vs 0/7710), asthma (RR 0.90; CI 0.00, 493; 2/6950 vs 4/6950 and 6/7695 vs 5/7710), autoimmune disease (RR 0.88; CI 0.23, 3.31), diabetes (RR 1.00; CI 0.00, 606; 5/6950 vs 6/6950 and 3/7695 vs 2/7710), encephalitis/encephalopathy (RR 0.50; CI 0.00, 2867570; 0/6950 vs 1/6950 and 0/7695 vs 1/7710), Guillain-Barré syndrome (RR 0.67; CI 0.00, 865; 1/6950 vs 2/6950 vs 1/7695 vs 1/7710), meningitis (RR 0.50; CI 0.00, 2867570; 0/6950 vs 1/6950 and 0/7695 vs 1/7710), reproductive system events (RR 1.04; CI 0.03, 37.17), or seizures (RR 1.34; CI 0.00, 13492; 2/6950 vs 0/6950 and 3/7695 vs 3/7710). For some of the key adverse events, there was a higher rate in the vaccine arm compared to the placebo arm in each of the two studies. For amyotrophic lateral sclerosis (RR 2.60; CI 0.00, 5715367; 2/6950 vs 0/6950 and 2/7695 vs 1/7710), none of the events were considered intervention-related by the investigators. For idiopathic thrombocytopenic purpura (RR 2.65; CI 0.00, 530690; 1/6950 vs 0/6950 and 3/7695 vs 1/7710), investigators blinded to the study arm thought that one case in the vaccine arm and one case in the placebo arm were intervention-related. For stroke (RR 1.44; CI 0.03, 71.52; 7/6950 vs 6/6950 and 19/7695 vs 12/7710), no events were considered intervention-related in the vaccine arm, and 1 was considered intervention-related in the placebo arm.

One RCT⁸⁵ identified an RR greater than 1 for angioedema (RR 2.00; CI 0.07, 59.61) based on one case among those vaccinated with RZV (1/6950 vs 0/6950), resulting in an imprecise risk estimate.

KQ1c3. For each adverse event associated with a particular vaccine, what are the risk factors for the adverse event?

No adverse events were significantly associated with a vaccine in our meta-analyses, but we discuss any risk factors that were examined for adverse events below.

In one RCT⁸⁵ of RZV conducted in multiple countries, the investigators performed adverse event analyses by race/ethnicity, gender, and age. Unsolicited adverse events were least commonly reported by people who are Black, most commonly reported by people who are Asian, more commonly reported by women than by men, and very slightly more commonly reported by people aged 50 to 69 years than by those aged 70 years and older.

Four studies identified risk factors for adverse events for PPSV23. An RCT¹⁷¹ of PPSV23 versus placebo reported an increased risk for extensive arm swelling, which was significantly more likely in women (female 10/1138 vs male 3/1202, p=0.04). Another study¹⁶⁵ compared PPSV23 administered with RZV to PPSV23 alone. Solicited general adverse events were more frequently reported when the first dose of RZV and PPSV23 were co-administered than when PPSV23 was administered alone. A third study¹⁷⁶ that assessed myocardial infarction, stroke, deaths from myocardial infarction or stroke, and deaths from all causes found no increased risk of these outcomes when looking at those with and without history of prior coronary artery disease and those with and without a history of cerebrovascular disease. A fourth study²⁰¹ looked at risk of stroke following PPSV23, and found no difference in risk of stroke among those who were younger than 65 years compared to those who were 65 years and older. The same study looked at transient ischemic attacks and found an increased risk among those younger than 65 years (aOR 1.61; 1.40, 1.85) but not those 65 years and older (aOR 1.03; 0.96, 1.10).

Three studies¹⁶²^, ²²²^, ²³¹ examined risk factors or sub-groups for adverse events associated with PCV13. One large cohort study²³¹ of over two million people found PCV13 to be significantly associated with increased risk of all-cause pneumonia among adults aged 65 years and older (mHR 1.76; CI 1.5, 2.04), immunocompromised persons (mHR 1.51; CI 1.24, 1.83), and immunocompetent persons (mHR 1.86; CI 1.55, 2.25) compared to no vaccination for pneumococcal disease but did not report sufficient detail for further analyses. Among people aged 50 to 64 years, there was no significant difference in all-cause pneumonia if vaccinated (mHR 1.21; CI 0.85, 1.72). Among all sub-groups, there was no significant association between vaccination and pneumococcal pneumonia or death. These findings are at odds with those from the large, high-quality study⁷² of PCV13 in older adults, which evaluated 84,496 adults 65 years of age or older. This RCT found that PCV13 was not associated with increased risk, and in fact prevented, vaccine-type pneumococcal, bacteremic, and nonbacteremic community-acquired pneumonia and vaccine-type invasive pneumococcal disease, and found no increased risk of all-cause community-acquired pneumonia. Of note, some of the unvaccinated population in the cohort study was assumed to have potentially received PPSV23, which may have affected the relationship observed as older adults may have been more likely to have received PPSV23. In addition, the cohort study was an observational one that could have had some residual confounding by indication, despite multivariate analyses to adjust for this. Another cohort study²²² of over 545,000 people comparing PCV13 to PPSV23 found no evidence of increased risk of several pre-specified adverse events; however, anaphylaxis was the only adverse event verified by chart review, with only one event confirmed as having occurred after vaccination out of 9 events in total. Stratified analyses by age group (65–69 years, ≥70 years) were consistent with the findings from the main analysis. Finally, a study¹⁶² of pneumococcal vaccine-naïve adults with HIV found no serious adverse events associated with PCV13, though they were not compared to adults without HIV.

Three studies looked at risk factors for adverse events among recipients of different influenza vaccines. An RCT of quadrivalent IIV (Fluzone Quadrivalent, compared to its trivalent counterpart) found that solicited systemic reactions were reported more frequently by participants 18 to 60 years of age than by participants 61 years of age or over.¹¹⁸ Another RCT of IIV (Flucelvax Quadrivalent, again compared to its trivalent counterpart) stratified risk of adverse events by age, sex, and race/ethnicity.⁶⁰ These subgroup analyses did not show any notable differences in terms of safety. Unsolicited adverse events and medically attended adverse events were reported by somewhat higher percentages in participants 65 years of age and older than in those 18 to 64 years of age, although rates of possibly vaccine-related adverse events were similar between the two subgroups. Another study was itself a subgroup analysis, examining the administration of PPSV23 with IIV compared to IIV alone.¹⁷⁷ There were no safety concerns, except that pain (Grade 3 and up, meaning more severe) occurred more often with co-administration of the vaccines compared to IIV alone (3.5% vs 0.0%).

For quadrivalent aIIV, one RCT¹⁰³ conducted in the United States found that rates of any solicited adverse events were higher in the 65 to 74 year age subgroup than in the 75 to 84 year age subgroup across both intervention and comparator arms of the study. No notable differences were observed in the subgroups by gender and race.

In an RCT⁹⁹ of HepB with novel adjuvant (HEPLISAV-B) compared to an existing HepB (Engerix-B), a subgroup analysis²⁸⁸ of adults aged 60 to 70 years with diabetes mellitus found no increase in risk of adverse events when comparing the vaccines.

KQ1. Summary of Findings for Safety of Vaccines in Adults

The summary of findings tables document the results across studies grouped by vaccines. These tables show the number of RCTs, the number of other studies, the number of participants across pooled analyses, the studies contributing to the risk estimate, findings for the outcomes of interest, the criteria used to downgrade the SoE, and the SoE summary statement. The relative risk of an adverse event was derived by comparing the reported event rates in vaccinated participants compared to a control group across all studies that reported the data for that outcome. The absolute rates of adverse events (number of events, number of assessed participants) for the vaccine and the control group are also shown. In many instances, results were based on single occurrences of a specific adverse event. Where studies reported insufficient detail and did not contribute to the effect size estimates, the tables report the results as reported by the study authors.

Hepatitis Vaccines

Table 2 shows the evidence for key adverse events associated with HepA, HepB, and HepA-HepB vaccines.

Table 2

KQ1: Update summary of findings and SoE for safety of hepatitis vaccines in adults.

HepA. The prior 2014 report had concluded that there was insufficient evidence to make evidence statements regarding an association between HepA and acute disseminated encephalomyelitis, transverse myelitis, multiple sclerosis, Guillain-Barré syndrome, chronic inflammatory demyelinating polyneuropathy, Bell’s palsy, anaphylaxis, and autoimmune hepatitis across all identified studies (including samples in children) based on a review of the evidence by the IOM, and one additional post-licensure study with no association with adverse events. We identified no new studies evaluating key adverse events since the prior 2014 report. Thus, the evidence remains insufficient to draw conclusions about HepA.

HepB. The prior 2014 report found moderate SoE for no association with multiple sclerosis onset or exacerbation based on a 2002 IOM report¹⁰⁸⁹ on HepB and demyelinating disorders (HEPLISAV-B was not in use at that time). This finding remains unchanged for HepB (update identified no new studies except for HEPLISAV-B, for which the SoE was insufficient).

The prior 2014 report also found moderate SoE for no association of type 1 diabetes mellitus with HepB. Studies identified in the update found no evidence of increased risk of diabetes across all HepB (type of diabetes was not specified). Thus, across the prior 2014 report and this update, there was moderate SoE for no evidence of increased risk of diabetes (including type 1 diabetes mellitus).

The prior 2014 report found moderate SoE supporting a causal relationship between hepatitis B vaccine and anaphylaxis in patients allergic to yeast, which was based on mechanistic evidence from the IOM report; the estimate of the magnitude of increased risk was not determined. Of note, HEPLISAV-B was not in use at that time. This finding remains unchanged (update identified no new studies on anaphylaxis, including for HEPLISAV-B).

There was also low SoE for no evidence of increased risk for asthma, autoimmune disease, cardiovascular events, death, herpes zoster, reproductive system events, or stroke (all studies were of HEPLISAV-B).

The prior 2014 report found insufficient evidence for an association between the vaccine and optic neuritis, first demyelinating events in multiple sclerosis, Guillain-Barré syndrome, systemic lupus erythematosus, onset or exacerbation of vasculitis, polyarteritis nodosa, and onset or exacerbation of rheumatoid arthritis because there were no studies. The review of studies from the update found insufficient evidence for key adverse events either because the evidence was graded as insufficient or there were no studies.

HepA-HepB. In this update, we found no studies examining the key adverse events of interest for HepA-HepB, except for optic neuritis (which was judged as insufficient evidence). The prior 2014 report also did not identify any studies. Thus, there was insufficient evidence to draw conclusions about HepA-HepB.

Table 2a summarizes the findings across the prior 2014 report and the update.

Table 2a

KQ1: Safety of hepatitis vaccines in adults.

9-Valent Human Papillomavirus Vaccine

The SoE for adverse events associated with HPV9 across all identified studies in adults is documented in Table 3.

Table 3

KQ1: Update summary of findings and SoE for safety of HPV9 in adults.

There was insufficient evidence for all outcomes of interest for HPV9 because no study that evaluated the vaccine in adults only met inclusion criteria (Table 3a). Studies that combined children and adults are reported in KQ2. HPV9 was not available at the time of the prior 2014 report. Table 3a summarizes the findings across the prior 2014 report and the update.

Table 3a

KQ1: Safety of HPV9 in adults.

Influenza Vaccines

We identified a large number of studies evaluating influenza vaccines in adults. Table 4 documents results for the key adverse events.

Table 4

KQ1: Update summary of findings and SoE for safety of influenza vaccines in adults.

IIV (non-adjuvanted). Quadrivalent IIV was not reviewed in the prior 2014 report, only trivalent IIV and monovalent H1N1 influenza vaccine. The prior 2014 report found high SoE for anaphylaxis in allergic persons with prior influenza vaccines based on the IOM report for trivalent IIV only. The prior 2014 report also found high SoE for arthralgia, myalgia, malaise, fever, and pain at the injection site for trivalent IIV. There was also high SoE for no association with cardiovascular events in adults aged 65 years and older in the prior 2014 report for trivalent IIV (based on post-licensure studies). Finally, the prior 2014 report found high SoE that the monovalent H1N1 influenza vaccine was associated with increased risk of Guillain-Barré syndrome, but this was not examined in the update as the vaccine is no longer in use. There was insufficient evidence to conclude anything regarding influenza vaccines and multiple sclerosis onset and exacerbation in the prior 2014 report.

All evidence statements for quadrivalent IIV currently in use come only from the current report, which identified low SoE for no evidence of increased risk for asthma, cardiovascular events, death, myocardial infarction, reproductive system events, seizures, or stroke for quadrivalent IIV compared to existing vaccines. There was insufficient evidence or no studies for several other key adverse events.

aIIV. aIIV (Fluad and Fluad Quadrivalent) was not in use at time of prior 2014 report; all evidence statements are based on the current report. The current report found moderate SoE for no evidence of increased risk of cardiovascular events or stroke. The current report also found low SoE for no evidence of increased risk for asthma, autoimmune disease, death, encephalitis/encephalopathy, Guillain-Barré syndrome, idiopathic thrombocytopenic purpura, myocardial infarction, or seizures compared to existing influenza vaccines. There was insufficient evidence to judge the risk of other key adverse events.

RIV. Quadrivalent RIV was not in use at the time of prior 2014 report (and trivalent RIV results were combined with all IIV); all evidence statements are based on the current report. The current report found low SoE for no evidence of increased risk for cardiovascular events, death, encephalitis/encephalopathy, myocardial infarction, reproductive system events, or stroke for quadrivalent RIV compared to quadrivalent IIV. For some key adverse events, the evidence was judged to be insufficient or we did not find studies reporting on the outcome.

LAIV. We reviewed one study of quadrivalent LAIV in the current report, with insufficient evidence or no studies for key adverse events. Table 4a summarizes the findings across the prior 2014 report and the update.

Table 4a

KQ1: Safety of influenza vaccines in adults.

Measles, Mumps, and Rubella Vaccine

Table 5 summarizes the SoE for studies of MMR in adults.

Table 5

KQ1: Update summary of findings and SoE for safety of MMR in adults.

There was insufficient evidence for all outcomes of interest for MMR in the update because no study that evaluated the vaccine in adults met inclusion criteria. The prior 2014 report found moderate strength evidence supporting a causal relationship between the rubella component of MMR and transient arthralgia in women from the IOM report, which was based on mechanistic evidence; the estimate of the magnitude of increased risk was not determined. This finding remains unchanged, as transient arthralgia was not examined as a key adverse event in the current report. The prior 2014 report found moderate strength evidence of no association of onset of type 1 diabetes and insufficient evidence for Guillain-Barré Syndrome, onset of multiple sclerosis, chronic arthralgia in women, and chronic arthritis and arthropathy in men. These findings also remain unchanged given that there were no new studies in the update and/or these were not examined as key adverse events. Table 5a summarizes the findings across the prior 2014 report and the update.

Table 5a

KQ1: Safety of MMR in adults.

Meningococcal Vaccines

Table 6 documents the findings for meningococcal vaccines in adults.

Table 6

KQ1: Update summary of findings and SoE for safety of meningococcal vaccines in adults.

MenACWY. The prior 2014 report reported no findings on MenACWY in adults; all evidence statements are based on the current report. There was moderate SoE for no evidence of increased risk for death and low SoE for no evidence of increased risk for cardiovascular events, myocardial infarction, or stroke. The evidence was graded as insufficient or there were no studies for several other key adverse events.

MenB. There was insufficient evidence for evidence statements regarding MenB in adults as no studies of adults alone were identified in the update; studies of children and adults are under KQ2. MenB was not in use at time of prior 2014 report, so there is insufficient evidence to draw conclusions. Table 6a summarizes the findings across the prior 2014 report and the update.

Table 6a

KQ1: Safety of meningococcal vaccines in adults.

Pneumococcal Vaccines

Table 7 documents results for pneumococcal vaccines in adults.

Table 7

KQ1: Update summary of findings and SoE for safety of pneumococcal vaccines in adults.

PCV13. There were no findings in the prior 2014 report for the safety of PCV13 in adults. Based on the current report, there was moderate SoE for no evidence of increased risk for cardiovascular events, herpes zoster, myocardial infarction, reproductive system events or stroke. Based on the current report, there was low SoE for no evidence of increased risk for acute disseminated encephalomyelitis, anaphylaxis or systemic allergic reaction, asthma, autoimmune disease, death, encephalitis/encephalopathy, herpes zoster, idiopathic thrombocytopenic purpura, meningitis, or seizures. For several key adverse events, we found either evidence graded as insufficient or no studies reporting on these adverse events.

PPSV23. The prior 2014 report found high strength evidence for no association of PPSV23 with cardiovascular or cerebrovascular events in adults aged 65 years and older. This finding remains unchanged given that the update continues to find no evidence of increased risk of cardiovascular events (including myocardial infarction) or stroke across all studies that are now available. Based on the update, there was moderate SoE for no evidence of increased risk of death following PPSV23. Evidence was graded as insufficient or there were no studies for a number of other key adverse events.

Table 7a summarizes and synthesizes the findings across the prior 2014 report and the update.

Table 7a

KQ1: Safety of pneumococcal vaccines in adults.

Tetanus, Diphtheria, and Acellular Pertussis Vaccines

Results regarding the safety of Tdap in adults are documented in Table 8.

Table 8

KQ1: Update summary of findings and SoE for safety of Tdap in adults.

The prior 2014 report described high strength of evidence supporting a causal relationship between tetanus toxoid vaccine and anaphylaxis from the IOM report, which was based on mechanistic evidence; the estimate of the magnitude of increased risk was not determined. This finding remains unchanged as there were no studies examining anaphylaxis in the update. There were two additional RCTs in adults in the prior 2014 report, but they did not contribute to further evidence statements.

Although we identified one study evaluating Tdap in adults, we graded the evidence as insufficient or found no studies for a number of other key adverse events.

Table 8a summarizes the findings across the prior 2014 report and the update.

Table 8a

KQ1: Safety of Tdap in adults.

Varicella Vaccine

Evidence regarding the safety of varicella vaccines in adults is documented in Table 9.

Table 9

KQ1: Update summary of findings and SoE for safety of varicella vaccines in adults.

The SoE was insufficient for evidence statements regarding varicella vaccines in adults as no study was identified that reported on the outcome in this update and the prior 2014 report. There were no studies of varicella vaccine in either report, and thus insufficient evidence to draw conclusions about risk. Table 9a summarizes the findings across the prior 2014 report and the update.

Table 9a

KQ1: Safety of varicella vaccine in adults.

Zoster Vaccine

Evidence regarding the safety of zoster vaccine in adults is documented in Table 10.

Table 10

KQ1: Update summary of findings and SoE for safety of zoster vaccines in adults.

RZV was approved after the publication of the prior 2014 report, so all evidence statements come from the current report. There was high SoE for no evidence of increased risk for herpes zoster. There was moderate SoE for no evidence of increased risk for amyotrophic lateral sclerosis, anaphylaxis or systemic allergic reaction, asthma, cardiovascular events, death, diabetes, encephalitis/encephalopathy, Guillain-Barré syndrome, idiopathic thrombocytopenic purpura, meningitis, myocardial infarction, reproductive system events, seizures, or stroke. There was low SoE for no evidence of increased risk for acute disseminated encephalomyelitis, angioedema, ataxia, autoimmune disease, or autoimmune thyroiditis (Hashimoto’s disease). Table 10a summarizes the findings across the prior 2014 report and the update.

Table 10a

KQ1: Safety of zoster vaccines in adults.

Key Question 2. What is the evidence that vaccines included in the immunization schedules recommended for children and adolescents are safe in the short term or long term?

This section describes the evidence for the safety of vaccines routinely recommended for use in children, including evidence from studies of children and adults together.

Key Points

Diphtheria, tetanus, and pertussis vaccine: No evidence of increased risk^c of asthma or death (low SoE). No evidence of increased risk of type 1 diabetes mellitus (moderate SoE) remains unchanged from prior report.
Tetanus, diphtheria and acellular pertussis vaccine: No evidence of increased risk of cardiovascular events or death (low SoE).
Hepatitis A vaccine: Increased risk of idiopathic thrombocytopenic purpura in children aged 7 to 17 years (moderate SoE) remains unchanged from prior report.
Hepatitis B vaccine: No evidence of increased risk of multiple sclerosis (moderate SoE) remains unchanged from prior report.
9-valent human papillomavirus vaccine: No evidence of increased risk of autoimmune disease, birth defects, death, reproductive system events, seizures, or spontaneous abortion (low SoE).
Quadrivalent inactivated influenza vaccine: No evidence of increased risk of anaphylaxis or systemic allergic reaction, asthma, autoimmune disease, cardiovascular events, death, febrile seizures, or seizures (low SoE, except for death which was moderate SoE).
Quadrivalent live attenuated influenza vaccine: No evidence of increased risk of death or seizures (low SoE).
Measles, mumps, and rubella vaccine: No evidence of increased risk of autism (high SoE remains consistent from prior report). Increased risk of febrile seizures (high SoE) and idiopathic thrombocytopenia (moderate SoE) remains unchanged from prior report. Anaphylaxis (high SoE) remains unchanged from prior report. No evidence of increased risk of asthma (low SoE).
Serogroup A, C, W, and Y meningococcal vaccines: No evidence of increased risk of acute disseminated encephalomyelitis, asthma, autoimmune disease, cardiovascular events, death, diabetes, encephalitis/encephalopathy, febrile seizures, intussusception, idiopathic thrombocytopenic purpura, Kawasaki disease, meningitis, multiple sclerosis, reproductive system events, seizures, or transverse myelitis (low or moderate SoE). Anaphylaxis in children with allergies remains moderate SoE, but there was no evidence of increased risk among all children (low SoE) based on the update.
Serogroup B meningococcal vaccine: No evidence of increased risk of anaphylaxis or systemic allergic reaction, asthma, death, reproductive system events, or seizures (low or moderate SoE).
13-valent pneumococcal vaccine: No evidence of increased risk of asthma, cardiovascular events, death, intussusception, meningitis, reproductive system events, or seizures associated with pneumococcal conjugate vaccine (low SoE except for death, which was moderate SoE). Increased risk of febrile seizures (low SoE, which was downgraded from moderate from prior report).
Rotavirus vaccine: No evidence of increased risk of intussusception across studies (moderate SoE, which was downgraded from moderate SoE for increased risk from prior report to moderate SoE for no increased risk when combining all available studies), though some observational studies indicated increased risk. No evidence of increased risk of anaphylaxis or systemic allergic reaction, asthma, autoimmune disease, autoimmune thyroiditis (Hashimoto’s disease), death, diabetes, encephalitis/encephalopathy, febrile seizures, idiopathic thrombocytopenic purpura, Kawasaki disease, meningitis, reproductive system events, seizures, or stroke (varied SoE).
Varicella vaccine: Anaphylaxis, disseminated varicella zoster virus, vaccine strain viral reactivation without other organ involvement, and vaccine strain viral reactivation with subsequent infection resulting in meningitis or encephalitis (high SoE) all remain unchanged from prior report, as does idiopathic thrombocytopenic purpura in children aged 11 to 17 years (moderate SoE).
Combination vaccines: For DTaP-IPV-Hib-HepB: no evidence of increased risk of asthma, death, febrile seizures, or meningitis (low SoE). For DTaP-IPV/Hib: no evidence of increased risk for anaphylaxis or systemic allergic reaction (low SoE). For DTaP-IPV: no evidence of increased risk for encephalitis/encephalopathy (low SoE).
Insufficient evidence to permit evidence statements for the following vaccines in children: Haemophilus influenzae type b vaccine, inactivated poliovirus vaccine, 23-valent pneumococcal polysaccharide vaccine, and DTaP-HepB-IPV.

KQ2a. What adverse events are collected in clinical studies and in observational studies containing a control/comparison group?

Given the many types of events assessed across studies of vaccines in children, the complete list is shown in Appendix C. We differentiated adverse events collected in studies that included only children and those in studies that included both children and adults. The assessed adverse events include both prespecified and incidentally collected adverse events. The information comes from a variety of experimental and observational designs. Details of the studies, including the method used to assess adverse events, are documented in the evidence tables (Appendix D) and the risk of bias table (Appendix C).

Studies collected information about the presence and absence of a variety of different events encompassing the range of mild and transient events to serious adverse events with permanent consequences.

KQ2b. What adverse events are reported in clinical studies and in observational studies containing a control/comparison group?

The evidence tables in Appendix D contain all serious and severe adverse events reported in studies of children and studies of children and adults together.

KQ2c. What adverse events are associated with these vaccines?

This section further characterizes the risk of adverse events identified in studies of children who received routinely recommended vaccines.

KQ2c1. For each adverse event associated with a particular vaccine, what is the average severity and frequency?

The prior 2014 report found several associations, which are detailed in that report and summarized at the end of this Key Question section in tables that integrate the findings from both reports. Several of these associations were in vaccines that are no longer in use because they have been replaced by newer vaccines, including trivalent IIV and monovalent H1N1 influenza vaccine, and 2-valent and 4-valent human papillomavirus vaccines. For those vaccines that are still in use, the associations between adverse events and specific vaccines established in the prior 2014 report and/or IOM report remain largely unchanged because there were no new studies or the outcome was not considered a key adverse event. These include idiopathic thrombocytopenic purpura with HepA and MMR (moderate SoE); anaphylaxis in children with allergies with MMR, MenACWY, and varicella vaccine (moderate or high SoE); transient arthralgia with MMR (moderate SoE); and disseminated varicella zoster virus, vaccine strain viral reactivation without other organ involvement (i.e., herpes zoster), and vaccine strain viral reactivation with subsequent infection resulting in meningitis or encephalitis with varicella vaccine.

In the update, we found new studies of febrile seizures (or seizures that were presumed to be febrile seizures) that did not contradict the increased risk of febrile seizures with MMR found in the prior 2014 report (high SoE) and we downgraded the SoE for febrile seizures with PCV13 (from moderate to low SoE); see KQ2c2 and summary of findings tables for more details.

We did not find any new statistically significant associations between any of the studied vaccines (including for new vaccines not reviewed in the prior 2014 report) and the key adverse events. Adverse events that are no longer considered to be associated with a particular vaccine are discussed in KQ2c2. For all events, the severity for each event is documented in the evidence table in Appendix D.

KQ2c2. For adverse events without statistically significant associations with a particular vaccine, what is the range of possible effects?

Although many studies contributed to the analyses, only a small number of studies reported sufficient detail about key adverse events to allow effect size estimates. Below we report all pooled risk estimates for adverse events that were reported on by more than one study. We note any instances where individual studies contributing to a pooled risk estimate were statistically significant on their own, indicating increased risk of an adverse event. We also report on studies with risk estimates that could not be computed and pooled statistically. Finally, as for KQ1, we summarize the range of possible effects by vaccine in brief below for all adverse events for which the primary relative risk estimate did not favor the intervention (i.e., RR>1), but the association was not statistically significant. Where appropriate, we contextualize the finding to understand the range of possible effects.

Diphtheria, Tetanus, and Acellular Pertussis Vaccines (DTaP, Tdap)

DTaP

We identified one RCT,¹⁸³ one pre-post analysis,⁹⁰ two self-controlled risk interval analyses,⁹⁶^, ²³⁵ and two studies⁶²^, ⁶³ using case-centered methods assessing diphtheria, tetanus, and acellular pertussis vaccines (DTaP) in children in this update. One study¹⁸³ assessed the Infanrix^® brand but the others did not specify the vaccine (vaccination status was determined using administrative data).

One study¹⁸³ of children that compared DTaP plus routine vaccines to vaccine routines alone reported one case of asthma in 145 intervention participants and none in the 146 control participants, which resulted in an imprecise risk estimate (RR 2.01; CI 0.07, 59.56). The same study also reported that no deaths occurred in the intervention or the control group, again with an imprecise risk estimate (RR 1.01; CI 0.02, 50.40). A pre-post study⁹⁰ reported an increased incidence of sepsis evaluation (adjusted rate ratio 3.2; CI 2.2, 4.5), respiratory support (adjusted rate ratio 1.9; CI 1.4, 2.6), and intubation (adjusted rate ratio 2.5; CI 1.3, 4.8) in extremely low birth weight infants after receipt of DTaP compared to the period before vaccination. One self-controlled risk interval analysis²³⁵ using Vaccine Safety Datalink data from 1995 to 2015 identified a risk of seizures following DTaP among children aged 11 to 23 months, but subsequent analyses adjusting for age at vaccination and concomitant vaccines showed no evidence of increased risk (incidence risk ratio [IRR] 1.2; CI 0.9, 1.6). A related study using the same population for a subset of those same years (2004–2008) for children under 24 months of age showed no increased risk of seizures following DTaP for either children three months and younger (IRR 1.26; CI 0.65, 2.45) or 3–23 months (IRR 1.56; CI 0.19, 12.92). Another self-controlled risk interval analysis⁹⁶ looking at febrile seizures specifically found no increased risk during the first day following vaccination compared to a control period of 14 to 20 days (IRR 1.17; CI 0.52, 2.60). This same study found increased risk of febrile seizures when a DTaP-containing vaccine was given with trivalent IIV compared to trivalent IIV given on a separate day from DTaP-containing vaccines (ratio of IRRs 7.57; CI 2.40, 23.89), but the study only reported across all DTaP-containing vaccines (including combination vaccines such as Pediarix^® and Pentacel^®) for this analysis. One study⁶² using case-centered methods reported no increased risk of acute disseminated encephalomyelitis in the 5- to 28-day risk interval following DTaP, compared to the remaining nine months after vaccination (aOR 0; CI 0.0, 270.4). Another case-centered analysis⁶³ detected no association between sudden sensorineural hearing loss and DTaP within one week of vaccination (OR 0; CI 0, 36.61). No other adverse events associated with DTaP were reported.

Tdap

We also identified four studies⁶²^, ⁶³^, ¹⁷³^, ¹⁸⁷ assessing Tdap and Td in samples containing children and adults. One was a study of Tdap plus MenACWY (Nimenrix^®, which is a brand used in Europe) versus MenACWY alone;¹⁸⁷ the other three studies were case-centered analyses⁶²^, ⁶³^, ¹⁷³ that used a large dataset of participants who received Tdap or Td.

The study¹⁸⁷ of Tdap plus MenACWY versus MenACWY alone reported no elevated risk of cardiovascular events or death (RR 0.99; CI 0.02, 49.53 for both outcomes); the risk estimate was imprecise for both outcomes as no cases occurred among either intervention or control groups (0/231 vs 0/228).

One study⁶² reported an increased risk of acute disseminated encephalomyelitis in the 5 to 28-day risk interval following Tdap, compared to the remaining nine months after vaccination (aOR 15.8; CI 1.2, 471.6). No other study investigated this outcome after Tdap, and the effect has not been replicated. The study reported insufficient detail to compute effect sizes, which would have allowed comparisons to other studies, so we could not pool results to better understand the frequency of this adverse event. In addition, severity was not reported, although in general acute disseminated encephalomyelitis—inflammation in the brain and spinal cord that damages myelin, which is the protective covering of nerve fibers—can have a variable course that can include severe illness. Overall, the number of cases (two) was very small. The investigators noted that one of the cases was a healthy adult who had been vaccinated 11 days prior to symptom onset, and the second case was a child who received both Tdap and meningococcal polysaccharide vaccine (which was not recommended for the child’s age group at that time). One case would have been expected statistically, and thus it was the presence of a second case that made the result significant and as a result the confidence intervals were very wide (with the lower confidence interval being closer to 1). Finally, the study had a large number of statistical comparisons without adjusting for multiple testing, meaning that the result could also be due to chance alone. Assuming the result was real, the absolute estimated excess risk was extremely small at 0.385 (CI −.04, 1.16) cases per million doses. No risk factors were explored in this study given the small number of cases, but this could also be an area for future investigation.

The same study⁶² found no increased risk of transverse myelitis during the immediate post vaccination period or optic neuritis in the 2- to 42-day risk interval after vaccination with Tdap or Td versus the remainder of the nine months after vaccination. Another case-centered analysis⁶³ detected no association between sudden sensorineural hearing loss and Tdap (OR 0.84; CI 0.39, 1.62) or Td (OR 0; CI 0.00, 2.67) within one week of vaccination. Finally, one study¹⁷³ reported no increased risk of primary ovarian failure after Tdap (aHR 0.88; CI 0.37, 2.10).

Haemophilus influenzae Type b Vaccines

We identified two self-controlled risk interval analyses,⁹⁶^, ²³⁵ and one pre-post study⁹⁰ reporting on Haemophilus influenzae type b vaccine (Hib) in children in this update. The authors did not provide the specific brand names of Hib used. The pre-post study⁹⁰ found an increased incidence of sepsis evaluation (adjusted rate ratio 4.0; CI 3.3, 4.8), respiratory support (adjusted rate ratio 2.1; CI 1.8, 2.5), and intubations (adjusted rate ratio 1.6; CI 1.2, 2.7) in extremely low birth weight infants after vaccination with Hib compared to the period before vaccination; this was not replicated in other studies. One self-controlled risk interval analysis²³⁵ using Vaccine Safety Datalink data from 1995 to 2015 identified no risk of seizures following Hib among children aged 11 to 23 months (IRR 1.0; CI 0.8, 1.2), nor did a related study using the same population for a subset of those same years (2004–2008) for children under 24 months of age. Another study⁹⁶ looking at febrile seizures specifically found no increased risk during the first day following vaccination compared to a control period of 14 to 20 days (IRR 1.53; CI 0.87, 2.72).

We identified one case-centered analysis⁶² assessing Hib in a sample of children and adults. The study reported no increased risk of acute disseminated encephalomyelitis or transverse myelitis in the 5- to 28-day risk interval following vaccination with Hib (brand not specified) versus the remainder of the nine months after vaccination.

Hepatitis Vaccines

Hepatitis A Vaccine

We identified one case control study¹⁵⁸ and one self-controlled risk interval analysis⁹⁶ evaluating HepA (among other vaccines) in children in this update.⁹⁰ We also identified two case-centered analyses⁶²^, ⁶³ assessing HepA in mixed samples of children and adults.

The case-control study¹⁵⁸ assessed the relationship of HepA to adverse events in children with new neuropsychiatric disorders. The study reported a slight association between a broken bone (HR 1.08; CI 1.02, 1.13), obsessive compulsive disorder (HR 1.40; CI 1.07, 1.82), and attention deficit hyperactive disorder (HR 1.09; CI 1.02, 1.18) diagnosis but no association with open wound, anorexia, anxiety disorder, tic disorder, major depression, or bipolar disorder for HepA.¹⁵⁸ The self-controlled risk interval analysis⁹⁶ found no increased risk of febrile seizures during the first day following vaccination with HepA compared to a control period of 14 to 20 days (IRR 0.88; 0.56, 1.38).

Of studies that combined children and adults, one case-centered analysis⁶² assessed HepA over a 10-month follow up. The authors reported no increased risk of acute disseminated encephalomyelitis, optic neuritis, or transverse myelitis in the short term interval following vaccines versus the remainder of the nine months after vaccination. Another case-centered analysis⁶³ detected no association between sudden sensorineural hearing loss and HepA (OR 0; CI 0.00, 3.42) within one week of vaccination.

Hepatitis B Vaccines

We identified one case-control study,¹⁵⁸ one self-controlled risk interval analysis,⁹⁶ and one pre-post study evaluating HepB (among other vaccines) in children in this update.⁹⁰ We also identified two case-centered analyses⁶²^, ⁶³ assessing HepB in mixed samples of children and adults.

The case-control study¹⁵⁸ assessed the relationship of HepB to adverse events in children with new neuropsychiatric disorders. There were no associations of adverse events with HepB. The pre-post study⁹⁰ found an increased incidence of sepsis evaluation (adjusted rate ratio 3.1; CI 2.3, 4.1) and use of respiratory support (adjusted rate ratio 2.1; CI 1.6, 2.8), but not intubations (adjusted rate ratio 1.5; CI 0.9, 2.6) in extremely low birth weight infants after vaccination with HepB compared to the period before vaccination. The self-controlled risk interval analysis⁹⁶ found no increased risk of febrile seizures during the first day following vaccination with HepB compared to a control period of 14 to 20 days (IRR 1.17; CI 0.31, 4.40).

One case-centered analysis⁶² combined children and adults and assessed HepB over a 10-month follow up. The authors reported no increased risk of acute disseminated encephalomyelitis, optic neuritis, or transverse myelitis in the short term interval following vaccines versus the remainder of the nine months after vaccination. Another case-centered analysis⁶³ detected no association between sudden sensorineural hearing loss and HepB (0.67; CI 0.03, 3.51) within one week of vaccination.

9-Valent Human Papillomavirus Vaccine

We identified two RCTs¹¹⁴^, ²²⁸ that assessed the risk for adverse events among children who received HPV9. HPV9 was not available at the time of the prior 2014 report.

Both RCTs¹¹⁴^, ²²⁸ assessing HPV9 in children compared the vaccine with HPV2 and HPV4 respectively. One outcome, death, was assessed in both studies (RR 0.58; CI 0.00, 362688); the risk estimate was imprecise as neither study reported any deaths (0/27 vs 0/93 and 0/299 vs 0/300). One study²²⁸ reported one incidence of Henoch-Schönlein purpura and complex partial seizure each in the control group but not in the HPV9 group (RR 0.5; CI 0.02, 14.90 for both adverse events).

We identified four studies⁹⁵^, ¹¹¹^, ¹³⁶^, ²²⁶ of HPV9 that included children as well as adults. Two studies were RCTs¹³⁶^, ²²⁶ that compared HPV9 to HPV4, while another RCT¹¹¹ compared HPV9 to placebo. The fourth study⁹⁵ was a post-marketing real-time surveillance study.

In one study¹¹¹ that compared HPV9 to placebo, we found no evidence of increased risk of birth defects (RR 0.50; CI 0.01, 25.22). When combined with another study¹³⁶ that compared HPV9 to another HPV vaccine, there was still no evidence of increased risk (RR 0.33; CI 0.00, 2276574). In one study¹¹¹ that compared HPV9 to placebo, we found no evidence of increased risk of reproductive system events (RR 0.25; CI 0.01, 7.46). When combined with another study¹³⁶ that compared HPV9 to another HPV vaccine, there was still no evidence of increased risk (RR 1.01; CI 0.00, 383). Although the RR was greater than 1, this was due to one excess event in the intervention group in one¹³⁶ of the two studies (0/608 vs 1/305; 9/7071 vs 8/7078), resulting in extremely wide confidence intervals.

In one study¹¹¹ that compared HPV9 to placebo, we found no evidence of increased risk of death (RR 0.50; CI 0.01, 25.22). When combined with two other studies that compared HPV9 to another HPV vaccine, we still found no evidence of increased risk of death (RR 1.11; CI 0.10, 12.13) across the three studies.¹¹¹^, ¹³⁶^, ²²⁶ Two of the RCTs¹¹¹^, ²²⁶ had no deaths in either the intervention or control groups, and one RCT¹³⁶ had one excess death (6/7071 vs 5/7078; none were considered to be vaccine-related by the investigators).

One RCT¹³⁶ of HPV9 found an RR of 2.00 (CI 0.07, 59.66) for anaphylaxis or systemic allergic reaction, an RR of 2.00 (CI 0.07, 59.66) for asthma and an RR of 2.00 (CI 0.18, 22.07) for multiple sclerosis. All three of these estimates had extremely wide confidence intervals that suggest a wide range of either possible protective or higher risk. For anaphylaxis or allergic reaction and asthma, the RR estimate was due to one event each in the vaccine-treated group (out of 7071) and none among controls. For multiple sclerosis, there was only one excess event among the vaccine-treated group (2/7071 vs 1/7078). The same RCT¹³⁶ identified an RR of 4.00 (CI 0.18, 88.77) for cardiovascular events (in this case, orthostatic tachycardia syndrome), which occurred in two patients in the intervention group (out of 7071) and none in the control group (out of 7078), resulting in the extremely wide confidence intervals.

A post-hoc analysis¹⁰⁹¹ of spontaneous abortions following HPV9 did note an imbalance (28.4% [19/67] vs 12.7% [7/55]) when compared to HPV4 among women who became pregnant within 30 days of vaccination. No other meaningful differences in adverse events were reported in these studies.

One post-marketing real-time surveillance study⁹⁵ of HPV9 found no signals for any adverse events, including for pre-specified adverse events, including anaphylaxis, chronic inflammatory demyelinating polyneuropathy, Guillain-Barré syndrome, pancreatitis, seizure, stroke, or venous thromboembolism.

Inactivated Poliovirus Vaccine

We identified two self-controlled risk interval analyses,⁹⁶^, ²³⁵ and one pre-post study⁹⁰ evaluating inactivated poliovirus vaccine (IPV) in children in this update. The authors did not provide the brand name of the IPV used. One self-controlled risk interval analysis²³⁵ using Vaccine Safety Datalink data from 1995 to 2015 identified no risk of seizures following IPV among children aged 11 to 23 months (IRR 1.0; CI 0.7, 1.4), nor did a related study using the same population for a subset of those same years (2004–2008) for children under 24 months of age. Another study⁹⁶ found no increased risk of febrile seizures during the first day following vaccination compared to a control period of 14 to 20 days (IRR 1.41; 0.08, 24.53). A pre-post evaluation⁹⁰ reported an increased incidence of sepsis evaluation (adjusted rate ratio 3.0; CI 2.1, 4.2), use of respiratory support (adjusted rate ratio 2.1; CI 1.5, 2.9), and intubations (adjusted rate ratio 2.5; CI 1.3, 4.9) in extremely low birth weight infants before and after IPV. No other outcomes were assessed in studies in children.

We identified two case-centered analyses⁶²^, ⁶³ assessing IPV in a broader sample of children and adults. One of these studies⁶² reported no increased risk of acute disseminated encephalomyelitis in the 5- to 28-day risk interval following administration of IPV compared with the remainder of the nine months after vaccination (aOR 0; CI 0.00, 270.1). The other case-centered analysis⁶³ detected no association between sudden sensorineural hearing loss and IPV within one week of vaccination (OR 0; CI 0, 92.95).

Influenza Vaccines

Quadrivalent Inactivated Influenza Vaccines

We identified eight RCTs⁹⁴^, ¹¹⁷^, ¹²⁹^, ¹⁵²^, ¹⁵³^, ¹⁸⁸^, ²³⁴ and one historically-controlled cohort design study combined with a self-controlled risk interval analysis¹⁶⁰ assessing quadrivalent IIV in children in this update. Studies assessed Fluzone Quadrivalent,¹¹⁷^, ²³³ Flucelvax Quadrivalent,¹²⁹ Fluarix Quadrivalent,⁹⁴^, ¹⁵²^, ¹⁸⁸ Flulaval Quadrivalent,¹⁵³^, ²³⁴ and quadrivalent IIV not otherwise specified (study in the Vaccine Safety Datalink of available quadrivalent IIV in the United States).¹⁶⁰ Quadrivalent IIV was compared to trivalent IIV in all but one of the trials;⁹⁴^, ¹¹⁷^, ¹²⁹^, ¹⁵²^, ¹⁵³^, ¹⁸⁸^, ²³⁴ in that one trial, the control group received the same base treatment as the intervention group (PCV13 and DTaP).²³³

Across five studies,⁹⁴^, ¹¹⁷^, ¹²⁹^, ¹⁵²^, ²³⁴ there was no evidence of increased risk of seizures (RR 1.08; CI 0.11, 10.40). All five studies had one or no events occur—two studies⁹⁴^, ¹⁵² had no events in either group, two studies¹¹⁷^, ¹²⁹ had one event in the intervention group, and none in the control group, and one study²³⁴ had one event in the control group only. Across four studies,⁹⁴^, ¹¹⁷^, ¹⁵²^, ¹⁵³ there was no evidence of increased risk of febrile seizures (RR 1.28; CI 0.17, 9.83). In the study¹¹⁷ with the most events (8/2892 vs 2/734), none of the febrile seizures in the intervention group were attributed to the vaccine (one event was attributed to the control vaccine in the control group). Two studies⁹⁴^, ¹⁵² had no febrile seizures in either group, and a fourth study¹⁵³ had one event in the intervention group (a simple partial febrile seizure 6 hours after vaccination that resolved without sequelae).

Across three studies,⁹⁴^, ¹¹⁷^, ¹⁵² there was no evidence of increased risk of cardiovascular events (RR 2.01; CI 0.03, 148). The relative risk estimate was imprecise as only one event (bradycardia not related to the vaccine in one study,¹¹⁷ myocarditis not related to the vaccine in another study,⁹⁴ and hypertension in the third study¹⁵²) occurred in the intervention group in each of the studies (1/915 vs 0/1823; 1/2892 vs 0/734; 1/932 vs 0/1861). Across another three studies,¹¹⁷^, ¹⁵²^, ¹⁵³ we also found no evidence of increased risk of asthma (RR 1.77; CI 0.03, 117); the risk estimate was imprecise due to differing sizes of the intervention and control groups (8/2892 vs 0/734; 0/932 vs 0/1861; 0/299 vs 1/302). We detected no evidence of increased risk of autoimmune disease (RR 0.50; CI 0.00, 2849944) across two studies;¹²⁹^, ¹⁵³ the risk estimate was imprecise due to one event in each study (0/1149 vs 1/1149 and 0/299 vs 1/302). We found no evidence of increased risk of anaphylaxis or systemic allergic reaction (RR 1.41; CI 0.00, 8066971) across two studies;¹²⁹^, ¹⁵³ again the risk estimate was imprecise with only one case in each study (1/1149 vs 0/1149 and 0/932 vs 1/1861).

There was no evidence of increased risk of death in one RCT²³³ when comparing to base treatment received by both intervention and control groups (RR 1.08; CI 0.02, 53.95), with no deaths occurring in either group in the study (0/99 vs 0/107). When combined with five RCTs¹¹⁷^, ¹²⁹^, ¹⁵²^, ¹⁸⁸^, ²³⁴ comparing to trivalent IIV, there was still no evidence of increased risk of death and the RR was no longer greater than 1 (RR 0.73; CI 0.10, 5.63).

In one study,¹¹⁷ the RR for Kawasaki disease was greater than 1 but not statistically significant (RR 1.52; CI 0.08, 30.37) due to 3 events in the intervention group; none of these events were considered to be related to the vaccine by the study authors.

A self-controlled risk interval analysis¹⁶⁰ found no increased risk of febrile seizures following IIV4 when alone given without concomitant vaccines (OR 1.2; CI 0.12, 11.20) (the study also examined other outcomes using a larger cohort and found no increased risk, but risk estimates were not provided).

A sensitivity analysis of the new cell-based quadrivalent IIV (Flucelvax Quadrivalent) approved for children 4 years of age and older based on one study¹²⁹ that compared to trivalent cell-based IIV showed no evidence of increased risk of anaphylaxis or systemic allergic reaction (RR 2.00; CI 0.07, 59.56), autoimmune disease (RR 0.50; CI 0.02, 14.89), seizures (RR 2.00; CI 0.07, 59.56), or death (RR 1.00; CI 0.02, 50.35).

Quadrivalent Live Attenuated Influenza Vaccine

We identified three RCTs,⁶⁹^, ¹⁶³^, ¹⁶⁴ one pre-post study,⁷⁶ and one cohort study²⁰⁷ assessing quadrivalent LAIV in this update. All studies were of FluMist Quadrivalent. Quadrivalent LAIV was compared to trivalent LAIV in two RCTs,⁶⁹^, ¹⁶³ placebo in one RCT,¹⁶⁴ IIV (trivalent or quadrivalent) in one cohort study,²⁰⁷ and matched unvaccinated controls in one study.⁷⁶

We found no evidence of increased risk of seizure across two studies⁷⁶^, ¹⁶⁴ that compared quadrivalent LAIV to placebo or no vaccine (RR 1.80; CI 0.06, 53.57); the risk estimate was imprecise as it was across only two studies, and the intervention and control groups differed in size (22/6745 vs 36/20163 and 1/868 vs 0/433). We also found no evidence of increased risk of death across another two studies,⁶⁹^, ¹⁶³ both of which compared quadrivalent LAIV to trivalent LAIV (RR 0.82; CI 0.00, 50615229); the risk estimate was imprecise as no deaths occurred in either study (0/1382 vs 0/923 and 0/66 vs 0/67).

In one study⁶⁹ of quadrivalent LAIV, the RR for diabetes following vaccination was 1.34 (CI 0.04, 39.77), due to one event in the intervention group only which was not thought to be due to vaccination. In another study,²⁰⁷ the RR of febrile seizures was 1.39 (CI 0.03, 69.76) but no events occurred in either the intervention or control groups (0/226 vs 0/314).

We identified one study⁶¹ assessing quadrivalent LAIV that did not differentiate between children and adults. This study compared quadrivalent LAIV recipients to unvaccinated controls and controls who received IIV, and also performed a within-cohort analysis comparing risk interval and control windows. The study reported no cases of seizure or convulsions, Guillain-Barré syndrome, Bell’s palsy, or encephalitis among vaccinated or unvaccinated participants. Significant findings included a lower risk of respiratory infection (adjusted HR 0.61; CI 0.50, 0.75), wheezing (adjusted HR 0.66; CI 0.52, 0.84) or any hospitalization (RR 0.35; CI 0.24, 0.50) compared to matched unvaccinated controls. There was a higher risk of wheezing (but not asthma specifically) among quadrivalent LAIV recipients two to four years of age than among unvaccinated controls (adjusted HR 1.50; CI 1.03, 2.20); none were hospitalized. There was no increased risk of wheezing among children with a history of wheezing or asthma.

Measles, Mumps, and Rubella Vaccine

We identified one case-control study,²²⁴ three self-controlled risk interval analyses,⁹⁶^, ¹⁶⁷^, ²³⁵ and four cohort studies¹¹³^, ¹³⁸^, ¹⁴¹^, ²¹⁵ evaluating MMR in children in this update. None of the studies specified the brand name of the specific vaccine used (all studies used administrative data to determine vaccination status and were presumed to use MMR-II based on the setting).

One self-controlled case series²³⁵ of children who received MMR reported an increased risk of seizures following vaccination at 12 months (IRR 2.9; CI 2.1, 4.1), 15 months (IRR 4.3; CI 3.2, 5.9), and 18 months (IRR 6.4; CI 4.4, 9.3), but did not provide sufficient detail for further analyses. Of note, the majority of the seizure events were likely to be acute febrile seizures per the authors of a related study²⁸² considered to be a multiple publication (due to entirely overlapping population and time frame). A self-controlled risk interval analysis¹⁶⁷ comparing the risk of seizures during the seven to ten days following receipt of MMR to a control window of 15 to 42 days found significantly increased risk of seizures both in children who had been born full-term (IRR 2.7; CI 2.2, 3.2) and pre-term (IRR 3.2; CI 1.9, 5.3). Concomitant vaccines, including varicella, were not controlled for in this analysis. These authors also noted that the events represented primarily febrile seizures. However, this study is not included as independent evidence for the risk of seizures and febrile seizures due to the overlap entirely with the time frame and population in the prior study. A study¹¹³ comparing a cohort of children who had received MMR to a cohort of those who had not found an increased risk of seizures (HR 5.94; 2.81, 12.58) and seizure disorders (HR 17.4; 2.23, 136) in adjusted analyses, and the finding was consistent in a self-controlled case series and additional sensitivity analyses. The authors did not distinguish between febrile and afebrile seizures. Another study⁹⁶ assessing febrile seizures specifically found no increased risk during the first day following vaccination compared to a control period of 14 to 20 days (IRR 0.78; 0.44,1.40). However, fever and febrile seizures tend to be seen later with vaccines such as MMR, hence the risk interval may have been too soon after vaccination to detect events.

Two cohort studies¹³⁸^, ¹⁴¹ compared those who received MMR to those who had not and reported no evidence of increased risk of autism (RR 0.60; CI 0.09, 4.12). The studies did not report on other adverse events. A third study,²²⁴ which used a case-control approach, also found no association between MMR and autism at any age, including up to 36 months of age (OR 1.04; CI 0.65, 1.68). A cohort study²¹⁵ examined the risk of asthma after MMR and found no evidence of increased risk (RR 0.37; CI 0.20, 0.67).

We identified two case-centered analyses⁶²^, ⁶³ assessing MMR in children and adults over a ten-month follow up period. One study⁶² evaluated MMR (brand unspecified but presumed to be MMR-II) and reported adjusted risk ratios. The authors reported no increased risk of acute disseminated encephalomyelitis, optic neuritis, or transverse myelitis in the short term interval following vaccines versus the remainder of the nine months after vaccination. The other study⁶³ detected no association between sudden sensorineural hearing loss and MMR within one week of vaccination (OR 0; CI 0, 33.29).

Meningococcal Vaccines

Meningococcal A, C, W, and Y Vaccines

We identified eight RCTs⁵¹^, ⁵⁷^, ⁶⁸^, ⁷⁹^, ¹⁰⁵^, ¹¹²^, ²¹⁹^, ⁹⁷⁸ across both reports and one self-controlled case series¹²⁷ assessing MenACWY in children in this update. Studies evaluated MenACWY-D (Menactra),¹²⁷ MenACWY-CRM (Menveo),⁵¹^, ⁶⁸^, ¹⁰⁵^, ¹¹²^, ²¹⁹^, ⁹⁷⁸ and MenACWY-TT (MenQuadfi).⁵⁷^, ⁷⁹ Of studies that did not use a self-controlled design, four RCTs⁵¹^, ¹¹²^, ²¹⁹^, ⁹⁷⁸ compared MenACWY-CRM to base treatment received by both the intervention and control groups and two RCTs⁶⁸^, ¹⁰⁵ compared MenACWY-CRM to meningococcal polysaccharide vaccine (MPSV; Menomune^®, which is no longer in use). Of the two RCTs of MenACWY-TT, one RCT⁵⁷ compared to MPSV and one RCT⁷⁹ compared to MenACWY-CRM and to base treatment received by both the intervention and control groups. In terms of age groups studied, three studies⁵¹^, ²¹⁹^, ⁹⁷⁸ examined the use of MenACWY in young infants (55–89 days of age), four studies⁵⁷^, ⁶⁸^, ¹⁰⁵^, ¹²⁷ in children (2–9 or 2–10 years of age), and two studies⁷⁹^, ¹¹² in older children and adolescents (10–17 or 10–18 years of age.

One RCT²¹⁹ did not report on adverse events in a way that could be interpreted across intervention and comparator, except for severe local and systemic adverse events and is not discussed further here. One Phase II RCT⁶⁸ found no serious AEs comparing those vaccinated with MenACWY-CRM versus meningococcal polysaccharide vaccine.

Across four RCTs⁵¹^, ⁷⁹^, ¹¹²^, ⁹⁷⁸ that compared MenACWY to a non-meningococcal vaccine comparator, there was no evidence of increased risk of death (RR 1.37; CI 0.11, 16.65). When including all RCTs even if the comparator was another meningococcal vaccine,⁵¹^, ⁵⁷^, ⁷⁸^, ⁷⁹^, ¹⁰⁵^, ¹¹²^, ⁹⁷⁸ there was still no evidence of increased risk of death (RR 1.17; CI 0.24; 5.73). Of note, although the RR was greater than 1, only one⁵¹ of the four studies had any deaths (7/5772 in vaccinated participants versus 1/1968 in the control group). The authors of this study examined the causes of these deaths and concluded that none were vaccine-related. When combined with another three RCTs⁵⁷^, ⁷⁸^, ¹⁰⁵ that compared to another meningococcal vaccine, the relative risk was similar (RR 1.17; CI 0.24, 5.73); there were no deaths in any of these three additional studies.

There was no evidence of increased risk of asthma in one RCT⁷⁹ (RR 1.51; CI 0.05, 44.86) when comparing MenACWY-TT plus routine vaccines to routine vaccines alone. The relative risk remained similar when combined with another group within the same RCT¹⁰⁵ that compared MenACWY-TT to MenACWY-CRM, and another two RCTs that compared MenACWY-CRM to another meningococcal vaccine (one¹⁰⁵ to MPSV and the other⁵⁷ to MenACWY-CRM) (RR 1.14; CI 0.08, 16.49). The RR was greater than 1, but based on only two cases of asthma across the three studies (2343 vaccinated individuals in total). Given the prevalence of asthma among children, the observed incidence appears to be low and the severity of the events was not clear.

In one RCT⁵¹ there was no evidence of increased risk of cardiovascular events (RR 0.34; CI 0.02, 5.46) or idiopathic thrombocytopenic purpura (RR 0.17; CI 0.01, 5.09). These results remained non-significant when combined with another RCT⁵⁷ that compared MenACWY-TT to MenACWY-CRM for cardiovascular events (RR 0.69; CI 0.00, 762141) and idiopathic thrombocytopenic purpura (RR 0.36; 0, 6069454). One RCT⁵¹ with routine vaccines as a comparator reported no evidence of increased risk of febrile seizures (RR 0.51; CI 0.18, 1.44); the risk estimate did not change when combined with one RCT¹⁰⁵ with meningococcal polysaccharide vaccine as a comparator and another RCT⁵⁷ with MenACWY-CRM as a comparator (RR 0.57; CI 0.07, 4.65).

One RCT⁷⁹ that compared MenACWY plus routine vaccines to routine vaccines alone showed no evidence of increased risk of seizures (RR 1.51; CI 0.05, 44.86). When combined across another group from the same RCT⁷⁸ and two other RCTs⁵⁷^, ¹⁰⁵ where the comparators were other meningococcal vaccines, no evidence of increased risk was found (RR 0.97; CI 0.07, 12.98) and the RR no longer favored the control for seizures.

Across two RCTs,⁷⁹^, ¹¹² there was no evidence of increased risk of diabetes (RR 1.32; CI 0.00, 21861366); the risk estimate was imprecise because there was only one event across both studies (1/396 vs 0/397 and 0/392 vs 0/296). The relative risk of diabetes remained similar when including a second population from one of the RCTs where the comparison was another meningococcal vaccine (RR 1.53; CI 0.02, 137).

One RCT⁵¹ examined the risk of Kawasaki disease among recipients of MenACWY, finding an RR of 1.37 (CI 0.15, 12.22). Of four cases observed in the vaccinated group, the authors believed that two could be related to vaccination, but there was also one case seen in the control group. Another RCT⁵⁷ that used an active comparator found no cases of Kawasaki disease; when combined the RR was 1.27 (CI 0.00, 306194). One RCT⁷⁹ found an RR of 1.51 for both asthma and seizures (CI 0.05, 44.86), due to one excess case in the intervention group (1/392 vs 0/296) resulting in extremely wide confidence intervals.

One study¹²⁷ of children through 10 years of age looked at the rates of outcomes per 1,000 person-months during a risk window compared to a control window and found no increased risk, including for asthma among children aged 2–10 years (RR 0.98; CI 0.03, 38.35), asthma among those aged 9–23 months (RR 1.48; 0.04, 57.69), convulsions among those aged 2–10 years (RR 0.00; 0.00, 18.68), and febrile seizures among those aged 9–23 months (RR not evaluable).

Looking specifically at the newest vaccine, MenACWY-TT (MenQuadfi), one RCT⁷⁹ had four groups constituting two comparisons: MenACWY-TT plus routine vaccines (Tdap and HPV4) versus routine vaccines alone, and MenACWY-TT versus MenACWY-CRM. Another RCT⁵⁷ compared MenACWY-TT to MenACWY-CRM. When comparing MenACWY-TT plus routine vaccines to routine vaccines alone, there were no deaths in either group (RR 0.76; CI 0.02, 37.94), nor when including the comparisons between MenACWY-TT and MenACWY-CRM (RR 0.91; CI 0.01, 130). There was no evidence of increased risk of asthma (RR 1.51; CI 0.05, 44.86), nor when including the two comparisons to MenACWY-CRM (RR 1.13; CI 0.02, 70.54). When compared to routine vaccines alone, there was no evidence of increased risk of diabetes (RR 0.76; CI 0.02, 37.94), nor was there evidence of increased risk when combined with a pair of groups from the same trial that compared MenACWY-TT to MenACWY-CRM (RR 1.31; CI 0.00, 21815040). When compared to routine vaccines alone, there was no evidence of increased risk of seizures (RR 1.51; CI 0.05, 44.86), nor was there evidence of increased risk when including the two comparisons of MenACWY-TT to MenACWY-CRM (RR 0.92; CI 0.02, 49.19). Only one RCT⁵⁷ comparing MenACWY-TT to MenACWY-CRM looked at cardiovascular events, and found no evidence of increased risk (RR 1.98; 0.07, 59.00). The same RCT looked at febrile seizures, Guillain-Barré syndrome, idiopathic thrombocytopenic purpura, and Kawasaki disease and found no evidence of increased risk (RR 0.99; CI 0.02, 49.89 for all).

We identified eight studies⁵⁶^, ⁶²^, ⁶³^, ⁹³^, ¹²⁸^, ¹⁵⁷^, ¹⁷³^, ²²⁰ assessing meningococcal A, C, W and Y vaccines in samples of children and adults. Studies assessed MenACWY-D (Menactra), MenACWY-CRM (Menveo), and MenACWY-TT (MenQuadfi). One study¹⁷³ did not report the brand used, but took place in the United States. Of those studies that were not self-controlled, two⁵⁶^, ⁹³ compared MenACWY-TT to MenACWY-D, and one¹⁵⁷ compared MenACWY-CRM to placebo. In terms of age groups studied, three studies⁹³^, ¹²⁸^, ¹⁵⁷ examined children and adults 10 or 11 years of age through 55 years of age, one²²⁰ examined children and adults 11 to 21 years of age, one examined children and adults 15 to 59 years of age,⁵⁶ one¹⁷³ examined female children and adults aged 11 through 34 years, and two⁶²^, ⁶³ did not specify the ages studied (used a large administrative database and presumably included all those individuals for whom the vaccine was indicated).

Three RCTs⁵⁶^, ⁹³^, ¹⁵⁷ reported on death as the outcome and found no evidence of increased risk (RR 0.50; CI 0.00, 71.42); the risk estimate was imprecise as deaths did not occur in any of the studies (0/297 vs 0/153; 0/2676 vs 0/635; 0/402 vs 0/407). An RCT¹⁵⁷ of MenACWY-CRM (Menveo) reported no serious adverse events in either the vaccine or placebo group.

Looking specifically at the newest vaccine, MenACWY-TT, (MenQuadfi), one RCT⁹³ that enrolled children and adults and compared MenACWY-TT to MenACWY-D reported no evidence of increased risk of anaphylaxis (RR 0.24; 0.00, 11.95); asthma, cardiovascular events (coronary artery disease), diabetes, multiple sclerosis, seizures (status epilepticus), or spontaneous abortion (RR 0.47; CI 0.02, 14.13 for all); or of autoimmune disease (Henoch-Schönlein purpura) (RR 0.12; CI 0.00, 3.53). In another RCT,⁵⁶ there was one case of pulmonary embolism in the MenACWY-TT group and no cases in the MenACWY-D group. Across the two RCTs⁵⁶^, ⁹³ of MenACWY-TT (both compared to MenACWY-D) there was no evidence of increased risk of death, with no deaths in either trial.

A self-controlled case series²²⁰ of MenACWY-CRM (Menveo) in children and adults reported an increased risk for Bell’s palsy in participants receiving concomitant vaccines (risk incidence [RI] 5.0; CI 1.4, 17.8) and no increased risk for those without concomitant vaccine (RI 1.1; CI 0.2, 5.5). The same study reported no elevated risk for other outcomes including anaphylaxis (confirmed by chart review), aseptic meningitis, asthma, autoimmune hemolytic anemia, Hashimoto’s disease (confirmed by chart review), Henöch-Schonlein purpura, idiopathic thrombocytopenic purpura, juvenile diabetes, multiple sclerosis, rheumatoid arthritis, seizure (confirmed by chart review), or transverse myelitis. In a retrospective cohort study¹²⁸ of MenACWY-D (Menactra), the study investigator considered two serious adverse events possibly related to vaccination: new-onset diabetes mellitus in a participant with family history and new-onset juvenile rheumatoid arthritis in a patient with family history. After review of medical records, investigators concluded these conditions were not likely related to vaccination. This study also reported elevated short-term (0–30 days) risk (IRR 11.80; CI 2.04, 254.6) and long-term (6 months) risk (IRR 1.93; CI 1.18, 3.20) for febrile illness. Medical record review of outcomes with significantly elevated IRRs did not suggest any relationship with MenACWY-D vaccine.

One study⁶² of children and adults used a case-centered design to assess both MenACWY-D specifically as well as all MenACWY as a group over 10 months of follow up. For both vaccine types, the study reported no increased risk of acute disseminated encephalomyelitis or transverse myelitis, and for all MenACWY (Menactra alone not examined for this outcome) no increased risk of optic neuritis in the short term interval following vaccination versus the remainder of the nine months after vaccination. Another case-centered analysis⁶³ detected no association between sudden sensorineural hearing loss and MenACWY within one week of vaccination (OR 0; CI 0, 21.06). Finally, one study¹⁷³ reported no increased risk of primary ovarian failure after MenACWY (aHR 0.94; CI 0.27, 3.23).

Meningococcal B Vaccines

We identified five RCTs¹⁵⁶^, ¹⁸⁶^, ¹⁹³^, ¹⁹⁶^, ²³⁰ assessing MenB in children (not in use at the time of the prior report). Studies assessed either Bexsero or Trumenba. Four studies¹⁸⁶^, ¹⁹³^, ¹⁹⁶^, ²³⁰ compared to a non-active comparator (placebo or routine vaccines that the intervention group received); one study¹⁵⁶ of Bexsero compared the vaccine to placebo (for dose 1) and MenACWY-CRM (for dose 2). One study¹⁹³ reported that there were no vaccine-related serious adverse events.

The risk estimates were imprecise for all three outcomes reported across more than one study as there were only two or three studies and the events were infrequent (none or one per group); rates are reported after each risk estimate. Across two MenB studies¹⁸⁶^, ¹⁹⁶ in children, there was no evidence of increased risk of anaphylaxis or systemic allergic reaction (RR 0.56; CI 0.00, 34735108; 0/198 vs 0/121 and 0/992 vs 0/501) There was also no evidence of increased risk of death across another two studies¹⁹⁶^, ²³⁰ (RR 1.12; CI 0.00, 18531125; 0/992 vs 0/501 and 1/374 vs 0/378); the one death that occurred in the vaccinated group was due to a motor vehicle accident. Across three studies,¹⁵⁶^, ¹⁸⁶^, ²³⁰ there was no evidence of increased risk of reproductive system events (RR 0.89; CI 0.01, 65.20; 1/198 vs 0/121; 1/174 vs 0/99; 0/374 vs 1/378). One of the RCTs²³⁰ also identified one case of idiopathic thrombocytopenic purpura among those vaccinated (N=374) compared to none in the control group (N=378), for an RR of 2.02 (CI 0.07, 60.07); the confidence intervals were extremely wide as a result.

We identified one RCT¹⁷⁸ in children and adults that compared MenB (Trumenba) to placebo. The study did not assess any key adverse events, but showed more reactogenicity events with the vaccine.

Pneumococcal Vaccines

13-Valent Pneumococcal Conjugate Vaccine

We identified five RCTs,⁵⁵^, ⁸⁶^, ⁸⁷^, ¹⁴⁶^, ²¹⁷ one cohort study,²⁰⁶ two self-controlled risk interval analyses,⁵⁹^, ⁹⁶ one self-controlled case series analysis,²³⁸ and two pre-post studies⁵⁸^, ²²¹ assessing PCV13 in children in this update. Of the studies that were not self-controlled, four RCTs⁵⁵^, ⁸⁷^, ¹⁴⁶^, ²¹⁷ compared PCV13 to PCV7 plus a base treatment received by both the intervention and control groups, and two studies (one RCT and one cohort study) compared PCV13 to a base treatment received by both groups (trivalent IIV in one study²⁰⁶ and MenACWY-TT in the other⁸⁶).

Across three RCTs,⁸⁷^, ¹⁴⁶^, ²¹⁷ there was no evidence of increased risk of asthma (RR 1.49; CI 0.15, 14.39), although all three identified studies reported more events in the PCV13 group compared to PCV7 or routine vaccines received by all participants. The risk estimate across studies was greater than 1, but was imprecise due to few events occurring across three studies with variable sample sizes (1/84 vs 0/88; 7/882 vs 5/875; 1/183 vs 0/183). None of these events were attributed to PCV13 by the investigators. Across another three RCTs⁵⁵^, ⁸⁷^, ¹⁴⁶ there was no evidence of increased risk of cardiovascular events (RR 0.60; CI 0.01, 38.23) or seizures (RR 0.91; CI 0.05, 15.11).

For risk estimates across two studies, the risk estimate was imprecise due to the small number of studies and of events in each study. Across two RCTs,⁵⁵^, ⁸⁷ there was no evidence of increased risk of meningitis (RR 1.32; CI 0.00, 1451908; 1/354 vs 0/355 and 1/930 vs 1/931). Across two RCTs,⁸⁷^, ¹⁴⁶ there was no evidence of increased risk of intussusception (RR 1.33; CI 0.00, 1437199; 1/88 vs 0/89, 1/930 vs 1/933) or reproductive system events (RR 1.49; CI 0.00, 24350819; 0/88 vs 0/89, 1/882 vs 0/875).

Across two RCTs,¹⁴⁶^, ²¹⁷ there was no evidence of increased risk of Kawasaki disease (RR 1.33; CI 0.00, 1407345; 1/88 vs 0/89, 1/183 vs 1/183). In the study¹⁴⁶ with one excess case in the intervention group, the event was not attributed to PCV13. One self-controlled case series analysis²³⁸ found no increased risk following any of three doses of PCV13, but did find an increased risk of complete Kawasaki disease specifically following the first dose of PCV13 (RI 2.59; CI 1.16, 5.81) based on seven patients who served as their own controls. One pre-post study⁵⁸ reported an age-adjusted risk estimate for Kawasaki disease of 1.07 (CI 0.70, 1.63) in a self-controlled case series and 0.97 (CI 0.79, 1.19) compared to an unvaccinated cohort. Another pre-post study²²¹ reported no evidence of increased risk of Kawasaki disease when comparing those who received PCV13 with those who received PCV7 (RR 1.94; 0.79, 4.86).

There was no evidence of increased risk of death in an RCT⁸⁶ that compared PCV13 to a base treatment received by both intervention and control groups (RR 2.02; CI 0.07, 59.88; 1/193 vs 0/195); the death that occurred was considered unrelated to the vaccine by study authors as the cause was asphyxiation 121 days post-vaccination. When combined with four other RCTs⁵⁵^, ⁸⁷^, ¹⁴⁶^, ²¹⁷ that compared PCV13 to PCV7, the risk estimate remained non-significant (RR 1.18; CI 0.12, 11.32). Across all five studies,⁵⁵^, ⁸⁶^, ⁸⁷^, ¹⁴⁶^, ²¹⁷ there were two deaths in the vaccinated group and one in the control group (0/84 vs 0/88; 0/354 vs 1/355; 1/193 vs 0/195; 1/884 vs 0/877; 0/183 vs 0/183). At least one of the two deaths in the vaccinated group was thought to be unrelated to the vaccine according to the study authors as above; the other was a sudden death of unknown cause.

One RCT⁸⁷ examined the risk of autoimmune disease (Henoch-Schönlein purpura) and identified one case in the vaccinated group and none in the control group (1/929 vs 0/931), resulting in an RR of 2.00 (CI 0.07, 59.07) with low precision given the rare events.

One study⁵⁹ found a significant association between PCV13 and febrile seizures. This effect held through both unadjusted and adjusted models, with an IRR adjusted for age, calendar time and concomitant IIV of 1.80 (CI 1.29, 2.52). This increased risk was not reported in five other studies. In a cohort study²⁰⁶ comparing PCV13 plus IIV to IIV alone, there was no evidence of increased risk of febrile seizures (RR 0.98; CI 0.02, 49.22). When combined with the two RCTs⁵⁵^, ²¹⁷ that compared PCV13 to PCV7, the result remained non-significant (RR 1.24; CI 0.12, 12.81). A self-controlled risk interval analysis⁹⁶ reported no increased risk of febrile seizures during the first day following vaccination compared to a control period of 14 to 20 days (IRR 1.41; CI 0.08, 24.53). This same study found an increased risk of febrile seizures when a PCV vaccine was given with trivalent IIV compared to trivalent IIV given alone on a separate day from a PCV vaccine (ratio of IRRs IIV3 7.57; 1.91, 30.07, but the study mixed both PCV7 which is no longer in use and PCV13 in this analysis. A pre-post study²²¹ of PCV13 compared to PCV7 reported no increased risk of febrile seizures, but there was an overlap in population studied and time frame with the study⁹⁶ of febrile seizures detailed above.

One pre-post study²²¹ identified an increased risk for encephalopathy, but this was not confirmed following the medical record review. This same pre-post study²²¹ reported no increased risk of febrile seizures (as discussed above), urticaria or angioneurotic edema, asthma, thrombocytopenia, or anaphylaxis for PCV13 compared to PCV7.

There was one study of children and adults that used a case-centered design to assess PCV13 over ten months of follow up.⁶² The study reported no increased risk of acute disseminated encephalomyelitis, or transverse myelitis in the short term interval following vaccination compared to the remainder of the nine months after vaccination.

23-Valent Pneumococcal Polysaccharide Vaccine

Among two studies that included children and adults in this update, one study⁶² of children and adults used a case-centered design to assess PPSV23 over ten months of follow up. The study reported no increased risk of acute disseminated encephalomyelitis, optic neuritis, or transverse myelitis in the short term interval following vaccination compared to the remainder of the nine months after vaccination.

Another case-centered analysis⁶³ detected no association between sudden sensorineural hearing loss and PPSV23 within one week of vaccination (OR 0.81; CI 0.20, 2.26).

Rotavirus Vaccines

A large number of studies has addressed rotavirus vaccines. Across both reports, we identified 22 RCTs⁹²^, ¹³⁹^, ¹⁶¹^, ¹⁶⁹^, ²²⁹^, ⁹⁰⁹^, ⁹²⁴^, ⁹⁶⁷^, ⁹⁶⁹^, ⁹⁷²^, ⁹⁷³^, ⁹⁸⁸^, ¹⁰⁰⁵^, ¹⁰¹⁰^, ¹⁰¹¹^, ¹⁰¹⁶^, ¹⁰²⁵^, ¹⁰⁴¹^, ¹⁰⁴²^, ¹⁰⁴⁴^, ¹⁰⁴⁵^, ¹⁰⁵⁴ that assessed adverse events associated with the use of rotavirus vaccines in children. In this update we identified three case control studies,⁷⁵^, ¹⁰⁸^, ¹⁷⁵ nine pre-post studies,⁷³^, ⁸³^, ¹⁰⁶^, ¹³⁰^, ¹³¹^, ¹⁶⁸^, ¹⁹⁰^, ²¹³^, ²²³ and four cohort studies.¹¹⁵^, ¹⁵⁵^, ¹⁸⁹^, ²²⁵ We also identified eight self-controlled case series,¹⁰²^, ¹²²^, ¹³³^, ¹³⁵^, ²⁰⁸^, ²¹²^, ²³⁵^, ²³⁷ and two self-controlled risk interval analyses.⁹⁶^, ²³⁶ Studies assessed RotaTeq^® or Rotarix^® vaccines. In all studies that were not self-controlled, rotavirus vaccine was compared to either placebo or the same base treatment as was received by the intervention group (e.g., routine vaccines). One RCT⁹² only reported on Grade 3 diarrhea and the overall rate of serious adverse events, and is not discussed further here.

Rotavirus vaccines and intussusception. Across all unique RCTs¹³⁹^, ¹⁶¹^, ¹⁶⁹^, ²²⁹^, ⁹⁰⁹^, ⁹²⁴^, ⁹⁶⁷^, ⁹⁷²^, ⁹⁷³^, ⁹⁸⁸^, ¹⁰⁰⁵^, ¹⁰¹⁰^, ¹⁰¹¹^, ¹⁰¹⁶^, ¹⁰²⁵^, ¹⁰⁴¹^, ¹⁰⁴²^, ¹⁰⁴⁴^, ¹⁰⁴⁵ of rotavirus vaccine that examined intussusception as an outcome from both the prior report and the update, we found no evidence for increased risk of intussusception (RR 0.65; CI 0.41, 1.05) at the time of latest follow-up. Three additional studies¹⁰⁶^, ¹³¹^, ¹⁶⁸ demonstrated no increase at the population level in intussusception rates before and after introduction of the rotavirus vaccine, and another two studies¹²²^, ¹³³ showed no significant difference in intussusception rates in a group of patients before and after receiving rotavirus vaccine. Another study²¹² demonstrated no significant increase in intussusception risk following any dose of rotavirus vaccine. One study¹⁰⁸ showed no increased risk of intussusception with the first dose or any dose (though the vaccine coverage was low in the region, which may have precluded detecting an increased risk). A study¹⁵⁵ of infants who received rotavirus vaccine together with DTaP compared to infants who received DTaP alone found no increased risk of intussusception (nor of Kawasaki disease, febrile seizures, seizures, meningitis, or encephalitis). Finally, a study⁷⁴ of rotavirus vaccine found no increased risk of intussusception, whether looking at children who were fully vaccinated (aHR 0.79; CI 0.57, 1.09) or partially vaccinated (aHR 0.89; CI 0.66, 1.19).

However, the authors of other studies identified in this update reported an increased risk of intussusception (primarily after the first dose). One study⁷⁵ showed an increase in the relative incidence of intussusception in infants seven days following the first dose of RotaTeq (RI 9.9; CI 3.7, 26.4) and Rotarix (RI 6.8; CI 2.4, 19.0). One study¹⁰² reported an increased IRR in the days following the first RotaTeq vaccination (IRR 3.45; CI 1.85, 6.55). One study¹⁷⁵ showed a significantly increased adjusted odds of intussusception in individuals who received the first dose of rotavirus vaccine compared to those who did not (aOR 5.74; CI 1.51, 21.79). Another study²³⁶ found an attributable risk of 1.5 excess cases intussusception per 100,000 recipients (CI 0.2, 3.2) in the 21 days following the first dose of RotaTeq but not following Rotarix (although the study was underpowered for Rotarix). A study²³⁷ of infants in Singapore reporting an increased relative incidence of intussusception in the seven days following the first dose of Rotarix vaccination (RI 8.36; CI 2.42, 28.96). A self-controlled case series¹³⁵ in Taiwan showed an increase in intussusception risk in the first seven days (IRR 12.59; CI 8.07, 19.66) and eight to 21 days (IRR 1.78; CI 1.00–3.16) following the first dose of Rotarix, but not RotaTeq. Another self-controlled case series study²⁰⁸ in England showed significantly increased relative incidence of intussusception in the 21 days after both the first (RI 4.53; CI 2.34, 8.58) and second (RI 2.60; CI 1.43, 4.81) doses of Rotarix, with the peak risk occurring one to seven days after the first dose (RI 13.81; CI 6.44, 28.32).

Several studies reported no significant increase in intussusception risk when looking at the second dose only,⁷⁵ and following the second or third doses.¹⁰²^, ¹⁷⁵^, ²³⁶^, ²³⁷ Finally, two studies⁸³^, ²¹³ showed an increase in the number of cases of intussusception at a population level following the introduction of rotavirus vaccine. A study⁸³ of infants reported an upward trend in yearly intussusception-related surgery during 2006 to 2010 (+2 excess cases per 100,000 births per year; p=0.023). Another study²¹³ of infants in the United States found significantly elevated rates of intussusception in 2007 (40.7 per 100 000 children <12 months of age; RR 1.13; CI 1.07, 1.20) and in 2010 (40.3 per 100 000 children <12 months of age; RR 1.12; CI 1.06, 1.20) but also reported that in other postvaccine introduction years, intussusception rates were not significantly different from baseline before rotavirus vaccine introduction.

Rotavirus vaccines and death. Across fourteen RCTs,¹³⁹^, ¹⁶¹^, ¹⁶⁹^, ⁹⁰⁹^, ⁹⁶⁷^, ⁹⁶⁹^, ⁹⁷³^, ¹⁰⁰⁵^, ¹⁰¹¹^, ¹⁰¹⁶^, ¹⁰²⁵^, ¹⁰⁴¹^, ¹⁰⁴⁵^, ¹⁰⁵⁴ there was no evidence of increased risk of death (RR 1.05; CI 0.82, 1.35). All but six¹³⁹^, ¹⁰¹¹^, ¹⁰¹⁶^, ¹⁰²⁵^, ¹⁰⁴⁵^, ¹⁰⁵⁴ of these RCTs had fewer deaths in the vaccinated group than in the unvaccinated groups (with roughly equal sample sizes in each), and none of the six studies had statistically significant risk estimates on their own. In one of these RCTs,¹⁰¹⁶ there was an imbalance of deaths due to pneumonia that was further investigated and determined not to be of concern. In two RCTs,¹⁰¹¹^, ¹⁰⁴⁵ the causes of death were not specified but the study authors note that none were attributed to rotavirus vaccine. In another three RCTs,¹³⁹^, ¹⁰²⁵^, ¹⁰⁵⁴ no deaths in the vaccinated group were considered attributable to rotavirus vaccine (deaths were due to respiratory syncytial virus [RSV] bronchiolitis 36 days after dose 2, gastroenteritis of unknown etiology and bronchopneumonia 46 days after last dose, bronchopneumonia 124 days after last dose, and pneumonia 16 days after dose 1).

Rotavirus vaccines and cardiovascular events. Across seven studies,⁷³^, ¹⁶¹^, ¹⁶⁹^, ⁹⁰⁹^, ⁹⁶⁹^, ⁹⁸⁸^, ¹⁰⁴⁵ the relative risk of cardiovascular events was not significant (RR 1.79; CI 1.00, 3.20), but the lower CI was 1.00. We note that this relative risk reflected multiple types of events and was driven by one study⁷³ of rotavirus vaccine in premature infants who received RotaTeq and experienced bradycardia (RR 1.87; CI 1.13, 3.10; 38/201 vs 20/198). In general, bradycardia events in premature infants are mild and in this study were considered clinically insignificant by the authors. There was no evidence of increased incidence of sepsis evaluation or elevation in respiratory support, which further suggests that these bradycardic events were mild. Most importantly, the study’s results were inconsistent as this finding was apparent only when comparing vaccine recipients to historic controls. In conducting a pre-post analysis in the same study, there was no difference in the proportion of infants experiencing an increase in apnea and bradycardia events with and without stimulation in the week after rotavirus vaccination compared with the week prior. Moreover, in the same pre-post analysis, the total number of bradycardia (1,523 vs 1,236, p=0.003) and bradycardia with stimulation (882 vs 699, p=0.02) events was higher in the seven days before rotavirus vaccination than after vaccination (with none in the 1- and 3-day time periods). Finally, we note that many experts would consider the physiology of bradycardia in premature infants to be neurologic in nature rather than cardiovascular. Overall, this was a mild event that did not appear to be associated with more severe sequelae, and results within the same study using a different analytic approach varied. Adverse events in the other six studies occurred at a frequency of one or no excess events in each study, and included congestive heart failure, tachycardia, cardiomyopathy, myocarditis, neonatal hypertension.

Rotavirus vaccines and febrile seizures or seizures. Across seven RCTs,¹⁶¹^, ¹⁶⁹^, ⁹⁶⁷^, ⁹⁶⁹^, ⁹⁸⁸^, ¹⁰¹⁰^, ¹⁰⁴⁵ there was no evidence of increased risk of febrile seizures (RR 0.82; CI 0.33, 2.05). Across five RCTs,¹⁶⁹^, ⁹⁰⁹^, ⁹⁶⁷^, ⁹⁸⁸^, ¹⁰⁴⁵ we found no evidence of increased risk of seizures (RR 1.02; CI 0.25, 4.16). Although the RR was greater than 1, the event was rare and occurred more frequently in the vaccinated group in only two of the five studies (1/508 vs 2/257; 1/1647 vs 3/1641; 1/3744 vs 1/3745; 17/34904 vs 9/34862; 1/2015 vs 0/2019). An additional study that reported insufficient detail for our analyses evaluated seizures among infants in Spain before and after rotavirus vaccine introduction found decreasing risk of hospitalizations for seizures following the introduction of rotavirus vaccine.¹⁹⁰ A self-controlled case series also found no association between rotavirus vaccine and seizures.²³⁵ Another study⁹⁶ found no increased risk of febrile seizures during the first days following vaccination with rotavirus vaccine (RotaTeq) compared to a control period of 14 to 20 days (IRR 1.18; 0.47, 2.99).

Other adverse events. Across five RCTs,¹⁶¹^, ¹⁶⁹^, ⁹⁶⁷^, ⁹⁸⁸^, ¹⁰⁴⁵ there was no evidence of increased risk of asthma (RR 1.33; CI 0.65, 2.72). The RR was greater than 1, but no risk estimates were statistically significant on their own and no studies noted the severity of asthma or the nature of these events (5/508 vs 3/257; 3/1647 vs 2/1641; 10/34904 vs 11/34862; 3/1666 vs 1/1667; 17/2015 vs 9/2019). Across another five RCTs.¹³⁹^, ¹⁶¹^, ⁹⁰⁹^, ⁹⁸⁸^, ¹⁰⁴⁵ there was no evidence of increased risk of meningitis (RR 1.18; CI 0.36, 3.89), nor any statistically significant increased risk when the studies were considered alone. Two of the studies⁹⁰⁹^, ¹⁰⁴⁵ had fewer events in the vaccinated group (4/3744 vs 6/3745; 3/34904 vs 4/34862); two of the studies¹³⁹^, ¹⁶¹ had one event in each of the vaccinated and control groups (1/380 vs 0/381, 1/1666 vs 0/1667; one event was specified to be due to coxsackie virus); and one study⁹⁸⁸ had more cases in the vaccinated group with no further details provided (7/1647 vs 2/1641).

Across three studies,¹¹⁵^, ²²⁵^, ¹⁰⁴⁵ we found no evidence of increased risk of diabetes (RR 0.74; CI 0.45, 1.22). A cohort study of rotavirus vaccine and type 1 diabetes found a 33 percent reduction in the risk of type 1 diabetes among those who completed the rotavirus vaccine series compared to those not vaccinated.¹⁸⁹

Across three RCTs,⁹⁶⁷^, ¹⁰¹⁰^, ¹⁰⁴⁵ we found no evidence of increased risk of Kawasaki disease (RR 2.08; CI 0.04, 123); the risk estimate was imprecise as the studies were of disparate sizes and with few events (1/508 vs 1/257; 5/34904 vs 0/34862; 1/4359 vs 0/4328). One self-controlled case series analysis¹³⁵ found no increased risk of Kawasaki disease in a primary analysis looking at 1–28 days post-rotavirus vaccine, but found an increased risk following the first dose of Rotarix at 22–28 days (IRR 1.98; CI 1.16, 3.40) and following the second dose of RotaTeq at 15–21 days (IRR 2.33; CI 1.35, 4.00). One pre-post study²²³ that took place in Germany reported no increase in Kawasaki disease-related hospitalizations after introduction of rotavirus vaccine.

We found no evidence of increased risk of encephalitis/encephalopathy (RR 0.67; CI 0.00, 85995) across two RCTs,⁹⁸⁸^, ¹⁰⁴⁵ of idiopathic thrombocytopenic purpura (RR 0.64; CI 0.00, 1778394) across two RCTs,¹⁰¹⁰^, ¹⁰⁴⁵ or of stroke (RR 1.32; CI 0.00, 1459247) across two RCTs.¹⁶¹^, ¹⁰⁴⁵ In all three cases, the risk estimates were imprecise as they were across only two studies and events were infrequent (encephalitis/encephalopathy 1/1647 vs 2/1641 and 1/34904 vs 1/34682; idiopathic thrombocytopenic purpura 1/34904 vs 0/34682 and 0/4359 vs 2/4328; stroke 1/34904 vs 0/34962 and 1/1666 vs 1/667). A small study in Aichi prefecture in Japan found no increased risk of encephalitis/encephalopathy or sudden death following rotavirus vaccine by self-report before and after rotavirus vaccine introduction.¹³⁰

Across one RCT¹⁰⁴⁵ and one cohort study,²²⁵ we found no evidence of increased risk of autoimmune disease (RR 0.65; CI 0.16, 2.67).

One RCT¹⁰⁴⁵ noted an RR greater than 1 (RR 2.00; CI 0.07, 59.94) for reproductive system issues based on one case of testicular torsion in the intervention group (1/34904 vs 0/34862), resulting in an imprecise risk estimate.

We identified one study⁶² assessing rotavirus vaccine (RotaTeq) in children and adults in the Vaccine Safety Datalink population (presumably rotavirus vaccine was only administered to children, but the study did not specify). The study reported no increased risk of transverse myelitis (which was the only adverse event studied for rotavirus vaccine [RotaTeq]) in the 5- to 28-day risk interval following vaccination of versus the remainder of the nine months after vaccination.

Varicella Vaccine

We identified two self-controlled risk interval analyses,⁹⁶^, ²³⁵ and one case control study¹⁵⁸ assessing varicella vaccine in children in this update. Studies assessed Varivax¹⁵⁸ and varicella vaccine with brand not otherwise specified²³⁵ (study utilized administrative data to determine varicella vaccination status).

One self-controlled risk interval analysis²³⁵ from the Vaccine Safety Datalink (1995–2015) found no increased risk of seizures following administration of varicella vaccine (IRR 0.8; CI 0.6, 1.0), controlling for concomitant vaccines. A related and overlapping case series of children also from the Vaccine Safety Datalink population (2004–2008) reported an increased risk of seizures following varicella vaccination at 12 to 15 months (IRR 2.75; CI 2.05, 3.70), as well as at 16 to 23 months (IRR 3.64; CI 1.86, 7.12). However, varicella vaccine was likely to have been given concurrently with MMR in this study, which was not controlled for (unlike the prior, larger study of the same population²³⁵). Another study⁹⁶ looking at febrile seizures specifically found no increased risk during the first day following vaccination compared to a control period of 14 to 20 days (IRR 0.80; 0.45, 1.42); however, the risk interval may have been too soon after vaccination to detect a risk.

One study¹⁵⁸ reported no association with a diagnosis of broken bone, open wound, obsessive-compulsive disorder, anorexia, tic disorder, attention deficit hyperactive disorder, major depression, or bipolar disorder.

We identified two studies⁶²^, ⁶³ of children and adults reporting on varicella vaccine (presumed to be Varivax as it took place in the United States). In one study,⁶² the authors reported no increased risk of acute disseminated encephalomyelitis (aOR 4.3; CI 0.5, 25.4), optic neuritis (aOR 2.1; CI 0.1, 23.2), or transverse myelitis (aOR 0.0; CI 0 (0.0, 10.7) in the short term interval following vaccines versus the remainder of the nine months after vaccination. In a second study,⁶³ there was no association between sudden sensorineural hearing loss and varicella vaccine within one week of vaccination (OR 0; CI 0, 155.27).

Combination Vaccines

This section is subdivided according to the different combination vaccines.

DTaP-IPV-Hib-HepB

We identified three RCTs⁷⁰^, ¹⁶⁶^, ²¹¹ assessing the new hexavalent combination vaccine DTaP-IPV-Hib-HepB. All three RCTs compared DTaP-IPV-Hib-HepB to the older combination vaccine DTaP-IPV/Hib plus HepB as well as other routine vaccines received by both intervention and control groups.

Across three studies,⁷⁰^, ¹⁶⁶^, ²¹¹ there was no evidence of increased risk associated with DTaP-IPV-Hib-HepB, including for death (RR 0.90; CI 0.02, 45.23) and febrile seizures (RR 0.71 CI 0.03, 19.74). Across two studies,⁷⁰^, ¹⁶⁶ there was no evidence of increased risk of asthma (RR 0.68; CI 0.00, 222465); the risk estimate was imprecise due to studies of differing sizes, both with few events (1/2390 vs 0/397 and 2/983 vs 1/480). Across the same two studies, there was no evidence of increased risk of meningitis (RR 0.42; CI 0.00, 462552); again, the risk estimate was imprecise due to one or no cases in each study (1/2390 vs 0/397 and 1/980 vs 1/483).

One study⁷⁰ found a statistically insignificant reduction in the risk of diabetes, intussusception, and idiopathic thrombocytopenic purpura. A second study¹⁶⁶ reported a statistically insignificant reduction in the risk of Kawasaki disease.

DTaP-HepB-IPV

We identified one pre-post study⁹⁰ and one self-controlled risk interval analysis⁹⁶ that assessed DTaP-HepB-IPV in children in this update. In the pre-post study,⁹⁰ the authors reported that in a population of extremely low birth weight infants in the neonatal intensive care unit, the three-day period following DTaP-HepB-IPV vaccination was associated with an increase in incidence of sepsis evaluation (adjusted rate ratio 4.3; CI 3.6, 5.3), respiratory support (adjusted rate ratio 2.6; CI 2.1, 3.1), and intubation incidence (adjusted rate ratio 1.8; CI 1.3, 2.6) compared with the three day period prior to DTaP-HepB-IPV vaccination. The authors also noted that true bacteremia (sepsis) occurred rarely both before and after vaccination. A self-controlled risk interval analysis⁹⁶ looking at febrile seizures among children aged 6 to 23 months found no increased risk during the first day following vaccination compared to a control period of 14 to 20 days (IRR 1.64; 0.62, 4.35).

One case-centered analysis⁶² in children and adults assessed DTaP-HepB-IPV. The study reported no increased risk of transverse myelitis in the 5- to 28-day risk interval following vaccination versus the remainder of the nine months after vaccination. None of the identified studies reported sufficient data to allow for further analyses such as an independent assessment of the relative risk.

DTaP-IPV/Hib

We identified one observational retrospective study,¹²⁶ one self-controlled risk interval analysis,⁹⁶ one cohort study,¹⁷⁴ and one pre-post study⁹⁰ assessing DTaP-IPV/Hib in children in this update. All studies except for one assessed Pentacel; the one study⁹⁰ did not identify the specific brand name of the DTaP-IPV/Hib vaccine used in the study (vaccination status was determined using administrative data). Three of the studies used a self-controlled design⁹⁰^, ⁹⁶^, ¹²⁶ and the fourth¹⁷⁴ compared DTaP-IPV/Hib to DTaP-containing vaccines in use prior to its introduction.

The observational retrospective cohort study¹²⁶ compared rates of outcomes during the post-vaccination risk interval with rates from a comparison interval for all non-elective hospitalizations, and all outcomes in the emergency department setting, and found no safety concerns. The study examined pre-specified outcomes in the outpatient setting, including Guillain-Barré syndrome, encephalopathy, encephalitis, meningitis, hypersensitivity reactions (urticaria, angioedema, or anaphylaxis), new-onset autoimmune disease (immune thrombocytopenic purpura [ITP], hemolytic anemia), type 1 diabetes, and Kawasaki disease, but as none of these outcomes had an elevated risk following vaccination they were not further assessed with any sort of comparator as part of the risk-interval analysis (and thus are not included in analyses for this report). One study⁹⁶ looking at febrile seizures specifically found no increased risk during the first day following vaccination compared to a control period of 14 to 20 days (IRR 0.72; 0.13, 3.85). The cohort study¹⁷⁴ reported that no cases of Guillain-Barré syndrome, anaphylaxis, or invasive Hib disease occurred among DTaP-IPV/Hib recipients during the study period, and the study reported no increased risk in serious allergic reaction (OR 1; CI 0.34, 2.99), encephalitis (OR 0.78; CI 0.23, 2.69), seizure (OR 0.91; CI 0.52, 1.57), or hospitalizations (OR 0.87; CI 0.73, 1.04) but the study reported events per dose rather than per participant, and the concurrent control group received DTaP as an individual injection.

The authors of the pre-post study⁹⁰ reported that in a population of extremely low birth weight infants in the neonatal intensive care unit, DTaP-IPV/Hib vaccination was associated with an increase in incidence of sepsis evaluation (adjusted rate ratio 4.0; CI 2.3, 6.9) and respiratory support (adjusted rate ratio 2.3; CI 1.3, 4.0); as noted previously, very few cases of true bacteremia (sepsis) occurred either pre or post vaccination.

One case-centered analysis⁶² in children and adults also evaluated DTaP-IPV/Hib. The study reported no increased risk of acute disseminated encephalomyelitis in the 5- to 28-day risk interval interval following vaccination versus the remainder of the nine months after vaccination.

DTaP-IPV

We identified one pre-post study⁸⁸ assessing DTaP-IPV (Kinrix^®) among children in this update (and none of Quadracel^®). Among eight pre-specified events (meningitis/encephalitis/myelitis, seizures, stroke, Guillain-Barré syndrome, Stevens-Johnson syndrome, anaphylaxis, angioneurotic edema and other non-anaphylactic allergic reactions, and serious local reactions), there was no increased risk compared to historical incidence rates.

One case-centered analysis⁶² in children and adults reported no increased risk of acute disseminated encephalomyelitis associated with DTaP-IPV (Kinrix) in the 5- to 28-day risk interval interval following vaccination versus the remainder of the nine months after vaccination.

MMR-V

We identified one RCT,⁸⁹ two self-controlled risk interval analyses,⁹⁶^, ¹⁶⁷ and two cohort studies¹⁴⁸^, ¹⁴⁹ assessing measles, mumps, rubella, and varicella vaccine (MMR-V) in children in this update. All studies assessed the ProQuad^® vaccine.

An RCT⁸⁹ comparing MMR-V with a hexavalent vaccine to hexavalent vaccine alone identified two participants with febrile seizures in the intervention arm compared to none in the control arm. Both participants had concomitant infections and the difference was not statistically significant (RR 2.02; CI 0.09, 44.55). A cohort study¹⁴⁸ comparing children aged 12 to 23 months who received MMR-V to those who received MMR and varicella vaccine separately on the same day, reported an increased risk of febrile seizures (RR 1.98; CI 1.43, 2.73) during the seven to ten days following vaccination that was similar when looking only at chart-confirmed febrile seizures. The authors calculated that the excess risk for febrile seizures 7 to 10 days after MMR-V compared with separate MMR and varicella vaccination was 4.3 per 10,000 doses (CI 2.6, 5.6). Another study⁹⁶ looking at febrile seizures specifically found no increased risk during the first day following vaccination compared to a control period of 14 to 20 days (IRR 1.12; CI 0.49, 2.54). Of note, fever and the risk of febrile seizures tends to be later with vaccines such as MMR-V, so the risk interval was likely to have been too soon after vaccination to detect a risk. In addition, two of these studies overlapped in terms of the population studied (Vaccine Safety datalink) and time frame (2000–2008 for one study¹⁴⁸ and 2006–2011 for the other⁹⁶).

For seizures, which were not differentiated from febrile seizures, a self-controlled risk interval analysis¹⁶⁷ comparing the risk during the seven to ten days following MMR-V vaccination to a control window of 15 to 42 days found significantly increased risk of seizures both in children who had been born full-term (IRR 5.7; CI 4.1, 7.8) and pre-term (IRR 7.9; CI 3.0, 20). Another cohort study¹⁴⁹ compared risk of seizures following MMR-V in a number of ways, including through unadjusted risk differences, case-centered analysis comparing MMR-V to itself, and a case-centered analysis comparing MMR-V to MMR and varicella vaccine separately. All analyses showed an increased risk of seizures, with an adjusted RR of 1.99 (CI 1.08, 3.52) reported by the authors among children, as well as risk of fever. The authors of both of the studies of MMR-V and seizures note that based on prior analyses, the majority of seizures were presumed to be febrile seizures. Again, both of these studies overlapped in terms of the population studied (Vaccine Safety datalink) and time frame (2000–2012 for one study¹⁴⁹ and 2003 to 2015 for the other¹⁶⁷).

The same cohort study¹⁴⁹ also compared risk of acute demyelinating encephalomyelitis, anaphylaxis, arthritis/arthralgia, ataxia, idiopathic thrombocytopenia purpura, Kawasaki disease, meningitis/encephalitis. The risk of idiopathic thrombocytopenia purpura was elevated in the secondary analysis comparing MMR-V to itself (RR 11.28; CI 1.87, 68.2). Anaphylaxis risk was increased in the same secondary analysis based on two cases, but on chart review the two cases were confirmed not to be anaphylaxis.

Another study⁶² that included both children and adults also reported no increased risk of acute disseminated encephalomyelitis associated with MMR-V (ProQuad) in the 5- to 28-day risk interval interval following vaccination versus the remainder of the nine months after vaccination.

KQ2c3. For each adverse event associated with a particular vaccine, what are the risk factors for the adverse event?

No adverse events were significantly associated with a vaccine in our meta-analyses, but we discuss any risk factors that were examined for adverse events below.

Rotavirus Vaccines

In one study⁷³ of rotavirus vaccine, the relative risk of cardiovascular events was largely driven by bradycardia events in a study of premature infants who received RotaTeq, an important subpopulation. Another study¹³⁹ noted that premature infants and term infants had a comparable incidence of adverse events following administration of rotavirus vaccines.

Some other rotavirus studies included subgroup analyses as well, with almost all looking at age as a risk factor or at results stratified by age group. In one study⁸³ that compared intussusception before and after rotavirus vaccine introduction, results were reported separately for three age groups. There was a 2-fold higher rate of intussusception among six to 14 week-olds in 2006 to 2010 compared with the 2000 to 2005 time period [IRR 1.90; CI 1.33, 2.74). A statistically significant difference between the last 2 periods was also detected for the 33 to 52 weeks of age subgroup; however, when the narrower 33 to 40 weeks age subgroup was considered, no difference was seen in the risk of intussusception among time periods. This is in keeping with other studies⁷⁵^, ¹⁰²^, ¹³⁵^, ¹⁷⁵^, ²⁰⁸^, ²³⁶^, ²³⁷ that noted higher risk of intussusception after dose 1 (typically given to younger infants) compared to doses 2 and/or 3. In a study²¹³ of infants in the United States that found significantly elevated rates of intussusception in some years after rotavirus vaccine introduction, no consistent change in intussusception hospitalization rates was observed among all children under 12 months of age and among children 15 to 24 weeks and 25 to 34 weeks of age. The intussusception hospitalization rate for children aged 8 to 11 weeks was significantly elevated by 46 to 101 percent in all postvaccine years except 2011 and 2013 compared with the pre-vaccine baseline, further indicating that risk for intussusception after rotavirus vaccine may be age-related. However, a study¹³¹ of the Canadian population that found no increase in incidence of intussusception following implementation of universal rotavirus (Rotarix) immunization programs also found no difference when limiting the age of intussusception incidence to infants two to eight months, and then further limiting it to infants two to six months and two to four months. A study in England¹⁶⁸ that examined admission rates for intussusception before and after introduction of rotavirus vaccine found no overall increase in hospital admission rate or disease severity, but found an increase in the admission rate in the 8–16 weeks age group (RR 1.46; CI 1.12, 1.91) that was compensated for by decreases in the 17 to 24 weeks (RR 0.77; CI 0.63, 0.94), 25 to 32 weeks (RR 0.71; CI 0.59, 0.86) and 41 to 52 weeks (RR 0.80; CI 0.66, 0.98) age groups. A case-control study¹⁷⁵ from Germany found that age at the start of rotavirus vaccine series did not modify the risk of intussusception.

In studies looking at the association of rotavirus vaccination and risk of type 1 diabetes, one study¹⁸⁹ found no statistically significant effect of year of birth and gender. A cohort study¹¹⁵ that assessed the association between rotavirus vaccine and type 1 diabetes reported no increased risk among children partially exposed to rotavirus vaccine (aHR 1.03; CI 0.62, 1.72) or children fully exposed to all recommended doses of rotavirus vaccine (aHR 1.50; CI 0.81, 2.77) compared with children unexposed to rotavirus vaccination; there were also no differences by age and vaccine brand. A study¹³³ that found no risk of convulsions among children who received rotavirus vaccine (Rotarix) and IPV compared to those who received IPV alone showed no difference in the lack of risk when assessed by any dose, after the first dose, or after the second dose. Finally, in a study¹⁵⁵ looking at a number of adverse events including otitis media, there was an association between otitis media and the second dose of rotavirus vaccine (HR 1.11; CI 1.08, 1.15), which was consistent across sex, region of the United States, and time period; this association was less pronounced with RotaTeq compared to Rotarix (HR 0.92; CI 0.89, 0.95).

Other Vaccines

A number of studies examined potential risk factors for adverse events associated with other vaccines.

One study²³⁵ that assessed the risk of seizures with varicella vaccine given concomitantly with other vaccines, including MMR, found that the risk seven to ten days after vaccination increased with age, from an IRR of 2.75 (CI 2.05, 3.70) at age 12 to 15 months to 3.64 (CI 1.86, 7.12) when administered at 16 to 23 months of age. In the same study, both MMR (IRR 2.65; CI 1.99, 3.55 at 12–15 months vs IRR 6.35; CI 3.15, 13.53 at 16–23 months) and MMR-V (IRR 4.95; CI 3.68, 6.66 at 12–15 months vs IRR 9.80; CI 4.35, 22.06 at 16–23 months) were associated with increasing risk of seizures by age. A study³⁵³ of measles-containing vaccines (including MMR-V) found lower increased risk of seizures when administered at 12–15 months of age compared to 16–23 months of age.

A study¹⁶⁷ examining the risk of seizures found no significant difference in the risk of seizures for children who had been born pre-term compared to full-term for MMR (IRR 1.2; CI 0.70, 2.0) or MMR-V (IRR 1.4; CI 0.51, 3.8).

Two studies examined risk factors for MMR and autism. A cohort study¹³⁸ in Denmark found no increased risk for autism, and this finding was consistently observed in subgroups of children defined according to sibling history of autism, autism risk factors based on a disease risk score, or receipt of other childhood vaccinations. Another study¹⁴¹ found that receipt of the MMR was not associated with increased risk of autism spectrum disorder (ASD), regardless of whether older siblings had ASD. A case-control study²²⁴ of MMR and autism found no increased risk of autism and no difference in the risk of autism following MMR at any age.

A study¹⁷⁴ of DTaP-IPV/Hib that found no increased risk of medically attended fever, seizure, meningitis/encephalitis/myelitis, non-anaphylactic serious allergic reaction, anaphylaxis, Guillain-Barré syndrome, or invasive Hib disease did identify an increased risk of medically attended fever among 1- to 2-year-olds who received DTaP-IPV/Hib vaccine versus historical comparators (RR 1.83; CI 1.34, 2.50) but not among infants under 1 year old (RR 0.83; CI 0.73, 0.94).

Five studies examined risk factors associated with PCV13 and adverse events. Two RCTs⁵⁵^, ²¹⁷ examined safety after both the infant and toddler doses, and found that PCV13 has an acceptable safety profile in both toddlers and infants, including one study²¹⁷ in which PCV13 was co-administered with DTaP. A pre-post study²²¹ of PCV13 compared to PCV7 that assessed pre-specified adverse events found no risk associated with PCV13, with no difference among age-stratified analyses. In a study²⁰⁶ comparing fever in children who received PCV13 and trivalent IIV versus trivalent IIV alone, the finding that PCV13 and trivalent IIV were associated with higher rates of fever did not change when excluding children whose families reported antipyretic use on days 0 to 1 or when analyses were limited to first enrollments (children could enroll for more than one vaccination event; this analysis was restricted to their first enrollment in the study). Another study of PCV13⁵⁹ found higher risk of febrile seizures when administered with IIV (IRR 2.80; CI 1.63, 4.83) than when given alone (IRR 1.54; CI 1.04, 2.28). The study included a mix of trivalent and quadrivalent IIV.

Two studies assess risk factors for MenB and adverse events in children. One RCT¹⁸⁶ examined different doses of the vaccine, including the approved dose, and found all doses to be well tolerated. Another RCT¹⁹⁶ found that local reactions and systemic events did not increase with concomitant administration of other vaccines. A self-controlled cases series study¹²⁷ of MenACWY-D (Menactra) that assessed multiple outcomes stratified by age groups (2–10 years and 9–23 months). The only significantly elevated outcome was among 2–10-year-olds (two cases of cellulitis/abscess occurred during the risk interval versus none during comparison interval; after medical record review, the cases were considered unrelated to vaccination).

Four studies of influenza vaccines in children looked at risk factors for particular subgroups. A study¹²⁹ of quadrivalent IIV (Flucelvax Quadrivalent) stratified children into age groups of 4 to 5 years, 6 to 8 years, and 9 to 17 years, and found no major differences in non-serious and non-severe adverse events by age. A study¹⁶⁰ of all quadrivalent IIV (brand not specified) found an increased risk of febrile seizures when administering quadrivalent IIV with PCV 13 and other vaccines (RR 12.3; CI 2.50, 58.90), but not for quadrivalent IIV without PCV13 (RR 0.6; CI 0.07, 4.85) or quadrivalent IIV alone (RR 1.2; CI 0.12, 11.20). A study⁷⁶ of LAIV in children with asthma and high-risk conditions found no increased risk of hospitalization. A second study²⁰⁷ of quadrivalent LAIV in children found no difference in fever when looking at those who received quadrivalent LAIV alone, when assessing only children aged 3 years and older, or when race/ethnicity or the presence of a high-risk medical condition were added to the model.

Among studies that included both adults and children, three reported sub-group analyses. In a study⁶¹ of quadrivalent LAIV versus IIV, the authors analyzed risk for different age groups, separately as well as by history of wheezing. Compared to other age groups, children aged two to four years had significantly lower risk of wheezing (aHR 0.67; CI 0.48, 0.92), lower respiratory tract infection (aHR 0.69; CI 0.51, 0.93), and hospitalization (aHR 0.39; CI 0.18, 0.87) with LAIV than those who received IIV. This effect was not observed for the within-cohort analysis that looked at these outcomes pre- and post-vaccine, nor when comparing LAIV to no vaccine (both aHR <1). Children aged five to nine years who received LAIV had lower risk of lower respiratory tract infection (aHR 0.55; CI 0.39, 0.79) and hospitalization (aHR 0.12; CI 0.03, 0.54) than those who received IIV and lower risk of hospitalization (aHR 0.14; CI 0.03, 0.63) than those who received no vaccine. Adolescents aged 9 to 17 years had lower risk of wheezing (aHR 0.51; CI 0.27, 0.96) and lower respiratory tract infection (0.57; CI 0.35, 0.93) when comparing LAIV to IIV. Among adults aged 18 to 49 years, LAIV was associated with significant decrease in risk of hospitalization when compared to no vaccine (aHR 0.33; CI 0.13, 0.87) and compared to IIV (aHR 0.13; CI 0.06, 0.30), except for the within cohort analysis. Among children with a history of wheezing or asthma, LAIV was associated with significantly less wheezing than was IIV (aHR 0.25; CI 0.04, 0.96), but the result was not significant when LAIV was compared to no vaccine (aHR 0.45; CI 0.06, 2.12). When looking specifically at wheezing among children aged 2–4 years of age with a history of wheezing or asthma, there was no difference in effects associated with LAIV versus IIV. In this study, quadrivalent LAIV was not associated with any increased risk of the adverse events examined.

The recent studies of HPV9 resulted in a number of additional publications that pooled together multiple RCTs to assess safety by geographic region and gender. Two studies that were secondary analyses of one of the RCTs¹³⁶ looked specifically at participants in Asian countries²⁵⁶ and in Latin American countries³⁵⁴ and identified no new safety concerns.

In a study²²⁰ of MenACWY-CRM (Menveo), the only significant finding was a reported association with Bell’s palsy (relative incidence 2.9; CI 1.1, 7.5). Stratified analyses demonstrated the increased risk only in participants receiving concomitant vaccines (RI 5.0; CI 1.4, 17.8), and no increased risk for those without concomitant vaccine (RI 1.1; CI 0.2, 5.5).

KQ2. Summary of Findings for Safety of Vaccines in Children

This section summarizes the findings across studies and safety of vaccines for children or children and adults for the pre-defined key adverse events. The section is stratified by vaccine type. The summary of findings tables document the results across studies grouped by vaccines. These tables show the number of RCTs, the number of other studies, the number of participants across pooled analyses, the studies contributing to the risk estimate, findings for the outcomes of interest, the criteria used to downgrade the SoE, and the SoE summary statement. The relative risk of an adverse event was derived by comparing the reported event rates in vaccinated participants compared to a control group across all studies that reported the data for that outcome. The absolute rates of adverse events (number of events, number of assessed participants) for the vaccine and the control group are also shown. In many instances, results were based on single occurrences of a specific adverse event. Where studies reported insufficient detail and did not contribute to the effect size estimates, the tables report the results as reported by the study authors.

Diphtheria, Tetanus, and Acellular Pertussis Vaccines

The summary of findings Table 11 documents the findings across studies of vaccines for diphtheria, tetanus, and pertussis (including DTaP and Tdap).

Table 11

KQ2: Update summary of findings and SoE for safety of diphtheria, tetanus, pertussis vaccines in children.

The 2014 AHRQ report concluded moderate SoE of no association of type 1 diabetes with DTaP/Tdap/Td (based on studies reviewed in the IOM report of vaccines containing diphtheria toxoid, tetanus toxoid, and acellular pertussis antigens either alone or in combination) and high SoE of no association with childhood leukemia with DTaP/Td (based on multiple large epidemiological studies). These findings remain unchanged with no new findings from this report (childhood leukemia was not examined as an outcome, and there were no studies on type 1 diabetes and DTaP or Tdap; Td no longer in use in children). The prior 2014 report found insufficient evidence for an association between acute disseminated encephalomyelitis, autism, cerebellar ataxia, idiopathic thrombocytopenic purpura, infantile spasms, multiple sclerosis relapse, seizures, serum sickness, sudden infant death syndrome, and transverse myelitis and DTaP/Tdap/Td based on the IOM report alone; these findings remain unchanged as the update either identified no new studies or the evidence continued to be insufficient.

For DTaP, the update found no evidence of increased risk of asthma or death (low SoE). The evidence was graded as insufficient or there were no new studies for several other key adverse events. For Tdap, the update found no evidence of increased risk of cardiovascular events or death (low SoE). There was either insufficient evidence or no new studies for several other key adverse events.

Table 11a summarizes the findings across the prior 2014 report and the update.

Table 11a

KQ2: Safety of DTaP and Tdap in children.

Haemophilus influenzae Type b Vaccines

Table 12 documents the evidence for the identified studies of Hib.

Table 12

KQ2: Update summary of findings and SoE for safety of Hib in children.

The prior 2014 report found high SoE for no association of Hib and childhood leukemia based on multiple large epidemiological studies; childhood leukemia was not examined as a key adverse event in the current report and the finding remains unchanged. The prior 2014 report also found moderate SoE for no association with short term serious adverse events following Hib across three high-quality RCTs; these vaccines had not been covered in the IOM report. This finding remains unchanged as there were either no studies in the update examining key adverse events or evidence was graded as insufficient due to study limitations.

Table 12a summarizes the findings across the prior 2014 report and the update.

Table 12a

KQ2: Safety of Hib in children.

Hepatitis Vaccines

Table 13 documents the evidence for the identified hepatitis vaccine studies.

Table 13

KQ2: Update summary of findings and SoE for safety of hepatitis vaccines in children.

HepA. The prior 2014 report found moderate SoE for idiopathic thrombocytopenic purpura in association with HepA among children 7 to 17 years of age based on a large post-licensure study (cases were rare and the majority were mild; no summary risk estimate could be calculated). There were no studies of this outcome in the update, and the finding remains unchanged

The evidence was either graded as insufficient or there were no new studies for several other key adverse events.

HepB. The prior 2014 report found high SoE for no association of hepatitis B vaccine and childhood leukemia based on multiple large epidemiological studies; this outcome was not examined as a key adverse event in the current report and the finding remains unchanged. The prior 2014 report concluded insufficient SoE for statements regarding food allergy; this outcome was also not examined as a key adverse event in the current report. The prior 2014 report also concluded that there was moderate SoE for no association with multiple sclerosis; there were no studies examining this outcome in the current report, and the finding remains unchanged.

Evidence in the update for HepB was either graded as insufficient or there were no new studies.

Table 13a summarizes the findings across the prior 2014 report and the update.

Table 13a

KQ2: Safety of hepatitis vaccines in children.

9-Valent Human Papillomavirus Vaccine

Table 14 documents the evidence for the identified studies of HPV9 in children.

Table 14

KQ2: Update summary of findings and SoE for safety of HPV9 in children.

The prior 2014 report found high SoE for increased risk of pain at injection site with HPV vaccines. The prior 2014 report also found moderate SoE for increased risk of anaphylaxis in persons with allergies, fever, headache, mild gastrointestinal adverse events, and skin infection with prior HPV vaccines. There was moderate SoE for no association with juvenile rheumatoid arthritis, type 1 diabetes, appendicitis, Guillain-Barré syndrome, seizures, stroke, syncope, or venous thromboembolism with prior HPV vaccines. The prior 2014 report indicated insufficient SoE for evidence statements regarding the presence or absence of acute disseminated encephalomyelitis, transverse myelitis, neuromyelitis optica, multiple sclerosis, onset of Hashimoto’s disease, chronic inflammatory demyelinating polyneuropathy, brachial neuritis, amyotrophic lateral sclerosis, transient arthralgia, pancreatitis, thromboembolic events, spontaneous abortion, and hypercoagulable states. However, all of the evidence review was for HPV2 and HPV4, which are no longer in use.

The current report assessed only HPV9 and found low SoE for no evidence of increased risk for autoimmune disease, birth defects, death, reproductive system events, seizures, or spontaneous abortion, largely compared to prior vaccines (HPV2 or HPV4). The evidence was graded as insufficient or there were no studies for several other key adverse events. Table 14a summarizes the findings across the prior 2014 report and the update.

Table 14a

KQ2: Safety of HPV9 in children.

Inactivated Poliovirus Vaccine

Table 15 documents the evidence for the identified studies of IPV.

Table 15

KQ2: Update summary of findings and SoE for safety of IPV in children.

The prior 2014 report concluded that there was insufficient evidence for food allergy related reactions based on one post-licensure study in newborns. This finding remains unchanged as the current report did not examine this outcome. The prior 2014 report found high SoE for no association with childhood leukemia based on multiple large epidemiological studies; childhood leukemia was not examined as a key adverse event in the current report and the finding remains unchanged.

The update either found insufficient evidence or no studies reporting on other key adverse events.

Table 15a summarizes the findings across the prior 2014 report and the update.

Table 15a

KQ2: Safety of IPV in children.

Influenza Vaccines

Table 16 documents the evidence for the identified studies of influenza vaccine in children.

Table 16

KQ2: Update summary of findings and SoE for safety of influenza vaccines in children.

IIV. The prior 2014 report only assessed trivalent IIV or monovalent H1N1 influenza vaccine. There was moderate strength evidence of an association of trivalent IIV with mild gastrointestinal events, which were not examined as a key adverse event in the current report. The prior report also found increased risk of febrile seizures with trivalent IIV (based on a self-controlled case series). The prior 2014 report found low SoE for no association between any serious adverse events in the short term with trivalent IIV in children who have received organ transplants and . The prior 2014 report also found low SoE for an association between influenza-type symptoms in children and influenza vaccines (which included both trivalent IIV and LAIV). There was insufficient evidence for acute disseminated encephalomyelitis and transverse myelitis based on the IOM report (trivalent IIV and LAIV assessed together). None of the influenza vaccines examined in the prior 2014 report are still in use.

The current report examines quadrivalent IIV only and found moderate SoE for no evidence of death. The current report also found low SoE for no evidence of increased risk for anaphylaxis or systemic allergic reaction, asthma, autoimmune disease, cardiovascular events, febrile seizures, or seizures. The evidence was graded as insufficient for several key adverse events, with no studies for others.

LAIV. The prior 2014 report combined findings across LAIV and IIV, and quadrivalent LAIV was not reviewed in the prior 2014 report (only trivalent LAIV). The prior 2014 report showed low SoE for increased risk of influenza-like symptoms with trivalent LAIV. There was also low SoE for no association with any serious adverse events in the short term in children with cancer (not examined as a key adverse event in current report). There was insufficient evidence for asthma exacerbation (LAIV specifically) and acute disseminated encephalomyelitis or transverse myelitis (all influenza vaccines assessed together). Trivalent LAIV, which was the subject of the prior 2014 report, is no longer in use.

The current report found low SoE for no evidence of increased risk for death and seizures with quadrivalent LAIV compared to IIV. Evidence was graded as insufficient for several key adverse events; there were no studies for other key adverse events. Table 16a summarizes the findings across the prior 2014 report and the update.

Table 16a

KQ2: Safety of influenza vaccines in children.

Measles, Mumps, and Rubella Vaccine

Table 17 documents the evidence for the identified studies of MMR in children.

Table 17

KQ2: Update summary of findings and SoE for safety of MMR in children.

The prior 2014 report found high SoE for no association with autism spectrum disorders based on the IOM report and an additional case-control study; this finding remains unchanged as the update found no evidence of increased risk based on two studies.

The prior 2014 report also found high SoE supporting a causal relationship between MMR and anaphylaxis in children with allergies based on IOM report, which was based on mechanistic evidence; the estimate of the magnitude of increased risk was not determined. This finding remains unchanged as there were no studies in the update. The prior 2014 report found an increased risk of febrile seizures, which was based on epidemiologic evidence from the IOM report as well as additional evidence from the prior 2014 report; no summary risk estimate could be calculated. This finding remains unchanged as there was insufficient evidence in the update for febrile seizures.

The prior 2014 report found moderate SoE supporting a causal relationship between MMR and transient arthralgia, which was based on mechanistic evidence from the IOM report; the estimate of the magnitude of increased risk was not determined. This finding remains unchanged as it was not examined as a key adverse event in the current report. The prior 2014 report found moderate SoE for a rare association with idiopathic thrombocytopenic purpura among children based on post-marketing studies (no summary risk estimate could be calculated); this finding also remains unchanged as there were no studies of this outcome in the update.

The update found low SoE for no evidence of increased risk of asthma.

The prior 2014 report found insufficient evidence for evidence statements regarding encephalitis, afebrile seizures, meningitis, cerebellar ataxia, acute disseminated encephalomyelitis, transverse myelitis, optic neuritis, neuromyelitis optica, multiple sclerosis onset, and chronic arthropathy. These findings remain unchanged as the update either did not find studies examining these outcomes, did not include these outcomes as key adverse events, or the new evidence was graded as insufficient (for acute disseminated encephalomyelitis, optic neuritis, and transverse myelitis). There were also several key adverse events for which there were no studies. Finally, the prior 2014 report noted that the IOM report found that evidence convincingly supported a causal relationship between MMR and measles inclusion body encephalitis in immunocompromised patients; the SoE for this finding was not graded and was not carried forward into this report as MMR would not be administered to immunocompromised patients.

Table 17a summarizes the findings across the prior 2014 report and the update.

Table 17a

KQ2: Safety of MMR in children.

Meningococcal Vaccines

Table 18 documents the evidence for the identified studies of meningococcal vaccine in children.

Table 18

KQ2: Update summary of findings and SoE for safety of meningococcal vaccines in children.

MenACWY. The prior 2014 report found moderate SoE supporting a causal relationship between MenACWY and anaphylaxis in children with allergies, based on mechanistic evidence from the IOM report; the estimate of the magnitude of increased risk was not determined. In the update, there was low SoE for no evidence of increased risk of anaphylaxis or systemic allergic reaction among all children (not restricted to those with allergies to the vaccine) based on two studies.

The prior 2014 report found insufficient evidence for encephalitis, acute disseminated encephalomyelitis, transverse myelitis, multiple sclerosis, Guillain-Barré syndrome, chronic inflammatory demyelinating polyneuropathy, and chronic headache. The following outcomes were upgraded to low SoE for no evidence of increased risk based on new studies identified in the update: acute disseminated encephalomyelitis, encephalitis/encephalopathy, multiple sclerosis, and transverse myelitis. Evidence for Guillain-Barré syndrome, chronic inflammatory demyelinating polyneuropathy, and chronic headache remained insufficient, either because the evidence was graded as insufficient (as for Guillain-Barré syndrome) or because the outcomes were not included as key adverse events.

The update also found moderate SoE for no evidence of increased risk for cardiovascular events, diabetes, febrile seizures, intussusception, idiopathic thrombocytopenic purpura, Kawasaki disease, or seizures. The update found low SoE for no evidence of increased risk of asthma, autoimmune disease, death, meningitis, or reproductive system events. The evidence was graded as insufficient evidence or there were no studies for several other key adverse events.

MenB. MenB was not in use at the time of the prior 2014 report. The current report found moderate SoE for no evidence of increased risk for anaphylaxis or systemic allergic reaction or reproductive system events, and low SoE for no evidence of increased risk for asthma, death, or seizures. The evidence was graded as insufficient or there were no studies for several other key adverse events. Table 18a summarizes the findings across the prior 2014 report and the update.

Table 18a

KQ2: Safety of meningococcal vaccines in children.

Pneumococcal Vaccines

Table 19 documents the evidence for the identified studies of pneumococcal vaccines in children.

Table 19

KQ2: Update summary of findings and SoE for safety of pneumococcal vaccines in children.

PCV13. The prior 2014 report found moderate strength evidence of an increased risk of febrile seizures with PCV13 among children based on one very large epidemiological study; no summary risk estimate could be calculated. The update identified six studies with conflicting evidence around the risk of febrile seizures (graded as low SoE for no evidence of increased risk). Thus, the original finding was downgraded from moderate to low SoE for an increased risk of febrile seizures.

The update found moderate SoE for no evidence of increased risk for death (most studies compared PCV13 to PCV7). The update also found low SoE for no evidence of increased risk for asthma, cardiovascular events, intussusception, meningitis, reproductive system events, or seizures (all compared PCV13 to PCV7). The update found insufficient evidence for several other key adverse events, and no studies for others.

PPSV23. There were no studies of PPSV23 in children in the prior 2014 report. In the update, the evidence was either graded as insufficient or there were no studies; thus, there was insufficient evidence to draw conclusions. Table 19a summarizes the findings across the prior 2014 report and the update.

Table 19a

KQ2: Safety of pneumococcal vaccines in children.

Rotavirus Vaccines

The large number of identified studies evaluating rotavirus vaccines is documented in Table 20.

Table 20

KQ2: Update summary of findings and SoE for safety of rotavirus vaccines in children.

The prior 2014 report found no other associations other than moderate SoE for increased risk of intussusception. The update found that the moderate SoE for risk from the last report was not confirmed when combining all studies that report sufficient detail at the latest time of follow-up. Some, but not all, observational studies showed increased risk of intussusception, particularly following the first dose of rotavirus vaccine. Taking all of the evidence together, the current report found moderate SoE for no risk of intussusception with rotavirus vaccine at the time of latest follow-up (SoE downgraded due to mixed results from observational studies that could not be combined).

The update found high SoE for no evidence of increased risk of diabetes, and moderate SoE for no evidence of increased risk for asthma, autoimmune disease, death, encephalitis/encephalopathy, febrile seizures, idiopathic thrombocytopenic purpura, seizures, or stroke. The update found low SoE for no evidence of increased risk of anaphylaxis or systemic allergic reaction, autoimmune thyroiditis (Hashimoto’s disease), Kawasaki disease, meningitis, or reproductive system events. The current report found insufficient evidence or no studies for some key adverse events. Table 20a summarizes the findings across the prior 2014 report and the update.

Table 20a

KQ2: Safety of rotavirus vaccines in children.

Varicella Vaccine

Table 21 documents the evidence for the identified varicella studies.

Table 21

KQ2: Update summary of findings and SoE for safety of varicella vaccines in children.

The prior 2014 report found high SoE supporting a causal relationship between varicella vaccine and the following, based on mechanistic evidence from the IOM report: anaphylaxis; disseminated Oka varicella zoster virus without other organ involvement; disseminated Oka attenuated varicella zoster virus with subsequent infection resulting in pneumonia, meningitis, or hepatitis in individuals with demonstrated immunodeficiencies; vaccine strain viral reactivation without other organ involvement (herpes zoster); and vaccine strain viral reactivation with subsequent infection resulting in meningitis or encephalitis; estimates of the magnitude of increased risk was not determined. All of these findings remain unchanged, as there are no studies addressing these outcomes in the update.

The prior 2014 report also found moderate strength evidence of a rare association with thrombocytopenic purpura among children aged 11 to 17 years based on one large epidemiological study (no summary risk estimate could be calculated); this finding remains unchanged given that no new studies of this outcome were identified in this report.

At the time of the prior 2014 report, there was insufficient evidence regarding seizures, acute disseminated encephalomyelitis, transverse myelitis, Guillain-Barré syndrome, small fiber neuropathy, onset or exacerbation of arthropathy, and thrombocytopenia. SoE for these findings remains insufficient as new evidence was either graded as insufficient (for seizures, acute disseminated encephalomyelitis, and transverse myelitis), there were no studies, or the outcome was not included as a key adverse event. There was insufficient evidence or no studies for several other key adverse events.

Table 21a summarizes the findings across the prior 2014 report and the update.

Table 21a

KQ2: Safety of varicella vaccines in children.

Combination Vaccines

Table 22 documents the evidence for the identified studies evaluating combination vaccines in children.

Table 22

KQ2: Update summary of findings and SoE for safety of combination vaccines in children.

DTaP-IPV-Hib-HepB. The current report examines DTaP-IPV-Hib-HepB, which was not in use at the time of the prior 2014 report. The current report found low SoE for no evidence of increased risk for asthma, death, febrile seizures, or meningitis, and evidence was graded insufficient for other outcomes. There were also no studies for several key adverse events.

DTaP-HepB-IPV. The prior 2014 report did not identify studies of DTaP-HepB-IPV and the update either found insufficient evidence or no new studies; thus, there was insufficient evidence to draw conclusions.

DTaP-IPV/Hib. The prior 2014 report found moderate strength evidence of an association of DTaP-IPV/Hib with febrile seizures based on one large, high quality post licensure study. This finding has been downgraded to insufficient taking into account both available studies (as there was a second study identified in this report with no evidence of increased risk). The update found low SoE for no evidence of increased risk for anaphylaxis or systemic allergic reaction.

DTaP-IPV. The prior 2014 report did not report findings related to DTaP-IPV. The update found low SoE for no evidence of increased risk for encephalitis/encephalopathy. Evidence was graded as insufficient or there were no studies for other key adverse events.

MMR-V. The prior 2014 report did not report findings related to MMR-V. The update found low SoE of no evidence of increased risk of encephalitis/encephalopathy and death. Evidence was graded as insufficient or there were no studies for other key adverse events.

Table 22a summarizes the findings across the prior 2014 report and the update.

Table 22a

KQ2: Safety of combination vaccines in children.

Key Question 3. What is the evidence that vaccines recommended for pregnant women are safe in the short term or long term for both the woman and her fetus/infant?

This section describes the evidence for the safety of vaccines routinely recommended for use in pregnant women, both in the pregnant women and their fetuses/infants.

Key Points

Insufficient evidence to permit evidence statements on hepatitis B vaccine, inactivated influenza vaccines, or recombinant influenza vaccine in pregnant women.
Tetanus, diphtheria, and acellular pertussis vaccine: No evidence of increased risk^d for maternal cardiovascular events, maternal death, maternal diabetes, eclampsia/pre-eclampsia, preterm labor, maternal reproductive system events, stillbirth, cardiovascular events in infants, death in infants, encephalitis/encephalopathy in infants, or seizures in infants (moderate SoE). No evidence of increased risk of maternal encephalitis/encephalopathy, autism in infants, birth defects in infants, or febrile seizures in infants (low SoE).

KQ3a. What adverse events are collected in clinical studies and in observational studies containing a control/comparison group?

Given the large number of adverse events that have been collected in the included studies, the table of collected adverse events is shown in Appendix C. The table lists both prespecified and incidentally collected, adverse events. The information comes from a variety of experimental and observational designs. Details of the studies, including the method used to assess adverse events, are documented in the evidence tables (Appendix D) and the risk of bias tables (Appendix C). Adverse events are shown in alphabetical order to facilitate the comparison across vaccines.

Studies collected adverse events for pregnant women and their infants/fetuses peri-and post-partum. The table lists the adverse events verbatim as reported in the original studies. As shown, the assessed adverse events varied greatly in detail as well as content. The collected data encompass the range of mild and transient events to serious adverse events with permanent consequences.

KQ3b. What adverse events are reported in clinical studies and in observational studies containing a control/comparison group?

The evidence tables in Appendix D contain all serious and severe adverse events reported in studies of pregnant women and their fetuses/infants.

KQ3c. What adverse events are associated with these vaccines in pregnant women?

This section further characterizes the risk of adverse events identified in studies of pregnant women who received routinely recommended vaccines.

KQ3c1. For each adverse event associated with a particular vaccine, what is the average severity and frequency?

No studies of key adverse events in pregnant women using HepB, quadrivalent IIV or quadrivalent RIV were found in this update. The prior 2014 report identified only evaluations of trivalent IIV and monovalent H1N1 influenza vaccine, which are no longer in use.

We did not detect any significant associations with adverse events following Tdap in pregnant women.

KQ3c2. For adverse events without statistically significant associations with a particular vaccine, what is the range of possible effects?

Below we report all pooled risk estimates for adverse events that were reported on by more than one study. We note any instances where individual studies contributing to a pooled risk estimate were statistically significant on their own, indicating increased risk of an adverse event. We also report on studies with risk estimates that could not be computed and pooled statistically. Finally, as for KQ1 and KQ2, we summarize the range of possible effects by vaccine in brief below for all adverse events for which the primary relative risk estimate did not favor the intervention (i.e., RR>1), but the association was not statistically significant. Where appropriate, we contextualize the finding to understand the range of possible effects.

Hepatitis B Vaccines in Pregnant Women

We did not identify any studies that assessed adverse events associated with HepB, in pregnant women in this update or in our original report.

Influenza Vaccines in Pregnant Women

We did not identify any studies that assessed key adverse events associated with quadrivalent IIV or any studies of quadrivalent RIV in pregnant women in this update.

One study of active surveillance used automated text messaging to identify adverse events following Tdap or IIV (of which >95% were quadrivalent IIV) among pregnant women in Australia.¹¹⁶ When comparing IIV plus Tdap to Tdap alone, there was no significant difference in women seeking medical advice via telephone for adverse events (RR 1.38; CI 0.60, 3.20) or being medically attended for adverse events (RR 1.79; CI 0.79, 4.04).

Tdap in Pregnant Women

We identified four RCTs¹²⁵^, ¹⁷²^, ¹⁷⁹^, ¹⁸¹ and 13 cohort studies⁵²^, ⁶⁴^, ⁶⁵^, ⁶⁷^, ⁹¹^, ¹²¹^, ¹²³^, ¹⁴⁴^, ¹⁵⁴^, ¹⁷⁰^, ¹⁸²^, ¹⁹¹^, ¹⁹⁸ that assessed adverse events in pregnant women and/or their offspring among women who received Tdap. RCTs assessed Adacel¹²⁵^, ¹⁷²^, ¹⁷⁹ or Boostrix;¹⁸¹ in most cases, the vaccine(s) used for women followed in the cohort studies were not specified. Three of the RCTs¹⁷²^, ¹⁷⁹^, ¹⁸¹ were placebo controlled, and one RCT¹²⁵ compared Tdap to Td. We did not separate out the one study that compared Tdap and Td in reporting results, as Tdap is sufficiently different from Td given the acellular pertussis component. The other studies used participants not vaccinated with Tdap as control groups. Two studies⁵²^, ¹⁷⁹ did not contribute to the meta-analyses as key adverse events were not reported; however, neither study found significant difference in maternal adverse events overall and one study⁵² that also looked at infants found no difference in adverse events.

Across ten studies,⁶⁷^, ¹²¹^, ¹²³^, ¹²⁵^, ¹⁷⁰^, ¹⁷²^, ¹⁸¹^, ¹⁸²^, ¹⁹¹^, ¹⁹⁸ we found no evidence of increased risk of preterm labor; women who received Tdap were at significantly lower risk for preterm labor (RR 0.62; CI 0.46, 0.82). However, the preterm labor analysis detected heterogeneity (I² 85%). Across six studies,¹²¹^, ¹²³^, ¹²⁵^, ¹⁵⁴^, ¹⁸¹^, ¹⁹¹ we found no evidence of increased risk of eclampsia/pre-eclampsia (RR 0.96; CI 0.92, 1.01); there was some evidence of publication bias (Egger p=0.063, Begg p=0.272). Across another six studies,⁶⁷^, ¹⁷⁰^, ¹⁷²^, ¹⁸¹^, ¹⁹¹^, ¹⁹⁸ there was no evidence of increased risk of stillbirth (RR 0.44; CI 0.11, 1.80). Across another six studies,¹²¹^, ¹²⁵^, ¹⁴⁴^, ¹⁷²^, ¹⁸¹^, ¹⁹¹ there was no evidence of increased risk of cardiovascular disorders (RR 0.86; CI 0.41, 1.84); there was considerable heterogeneity (I² 85%). Of note, one of the studies¹²¹ reported an increased risk of cardiovascular events (gestational hypertension; RR 1.30; CI 1.15, 1.48; events 262/8178 vs 1484/60372); however, the authors noted that the adjusted hazard ratio (aHR) was no longer significant (aHR 1.02; CI 0.88, 1.19).

Across four studies,¹²¹^, ¹²⁵^, ¹⁸¹^, ¹⁹¹ we found no evidence of increased risk for maternal deaths (RR 1.52; CI 0.07, 32.35). Although the RR was greater than 1, it is important to note that no deaths occurred in the intervention groups—one death occurred in a control group and the RR was inflated due to an imbalance in the sample sizes after adding a constant for computational purposes.

Across another four studies,¹²¹^, ¹⁴⁴^, ¹⁸¹^, ¹⁹¹ we found no evidence of increased risk of diabetes (RR 0.98; CI 0.88, 1.10). Across three studies,¹²⁵^, ¹⁸¹^, ¹⁹¹ we found no evidence of increased risk of reproductive system events (RR 0.52 CI 0.05, 5.91); we detected statistical heterogeneity (I² 53%). Across two studies,¹²³^, ¹⁹⁸ we found no evidence of increased risk of spontaneous abortion (RR 0.66; CI 0.00, 1004); we detected statistical heterogeneity despite the small number of studies (I² 81%). The risk estimate was imprecise as it was based on two studies of disparate size (4/138 vs 49/552 and 233/1252 vs 22931/130289).

Among other maternal outcomes, a cohort study reported that women who received Tdap had a decreased risk (adjusted) for caesarean delivery (aOR 0.78; CI 0.63, 0.98).⁶⁷ One cohort study¹²¹ identified an increased risk of lactation disorders (aHR 1.63; CI 1.15, 2.33), but significantly decreased risk of antenatal bleeding (aHR 0.61; CI 0.49, 0.78) as well as pre-eclampsia with severe features and preterm labor/delivery (already captured in meta-analyses). In another cohort study,¹⁵⁴ there was increased risk of post-partum hemorrhage both among those with optimally timed Tdap (aHR 1.23; CI 1.18, 1.28) and early Tdap (aHR 1.34; CI 1.25, 1.44). However, in another study¹⁸¹ which was an RCT, there was no increased risk of vaginal or intrauterine hemorrhage following Tdap (2.6%; CI 1.2, 5.0 vs 2.9%; CI 1.4, 5.3).

Among five cohort studies that assessed the risk for chorioamnionitis, two reported a slightly increased risk—one study⁹¹ among 197,564 women (adjusted rate ratio 1.23; CI 1.17, 1.28) and one study¹⁵⁴ among 1,079,034 women (aHR 1.11; CI 1.07, 1.15) who received optimally timed Tdap. One study¹²¹ of 68,550 women noted no difference in risk (aHR 1.10; CI 0.70, 1.75), as did a second study¹⁷⁰ that looked at 7,378 women who had received one dose of Tdap (6% vs 4%, p=0.31). Finally, one study⁶⁷ reported no difference in risk when evaluated in multiple ways among 1,759 women.

One study¹²⁵ assessed maternal autoimmune disease following Tdap and found a RR greater than 1 (RR 1.02; CI 0.06, 16.18) due to one event in each group of either Crohn’s disease or ulcerative colitis. Another study¹⁴⁴ assessed maternal seizures, finding an RR of 2.07 (CI 0.12, 32.41) due to seizures occurring in one subject in each of the intervention and control groups.

One study¹¹⁶ of active surveillance used automated text messaging to identify adverse events following Tdap or IIV (of which >95% were quadrivalent IIV) among pregnant women in Australia. When comparing Tdap plus IIV to IIV alone, there was no significant difference in women seeking medical advice via telephone for adverse events (RR 2.53; CI 0.89, 7.2) or being medically attended for adverse events (RR 1.82; CI 0.7, 4.51).

KQ3c3. For each adverse event associated with a particular vaccine, what are the risk factors for the adverse event?

As outlined, we did not find any adverse events associated with vaccines in pregnant women, but we discuss any risk factors that were examined for adverse events below.

Few studies addressed risk factors for adverse events, though three studies examined adverse events based on timing of Tdap. In one cohort study,¹⁵⁴ adverse events were stratified by timing of receipt of Tdap. Optimally timed vaccine was defined as 27 weeks onward, and early was defined as less than 27 weeks. Optimally timed Tdap was associated with significantly lower risk of preeclampsia/eclampsia (aHR 0.96: 0.94, 0.99), but early Tdap was not (aHR 1.05; CI 0.99, 1.12). Both optimally timed and early Tdap were associated with small increased relative risks of chorioamnionitis (aHR 1.11; CI 1.07, 1.15 for optimal and 1.19; CI 1.11, 1.28 for early) and postpartum hemorrhage (aHR 1.23; CI 1.18, 1.28 for optimal and 1.34; CI 1.25, 1.44 for early). However, these relative increases corresponded to low absolute risk increases. Early Tdap was associated with significantly increased risk of premature rupture of membranes (aHR 1.08; CI 1.02, 1.15), but optimally timed Tdap was not (aHR 1.03; CI 1.00, 1.06). However, it is possible that the study failed to adjust for all residual confounding, and that receipt of the Tdap prior to the recommended timing might have been a proxy for atypical care or anticipated premature birth. In addition, there is no clear biologically plausible mechanism whereby greater time since having received Tdap is more highly associated with a mechanical rupture of membranes than more temporally associated Tdap administration. Two other studies of timing of Tdap administration raised no concerns for early administration of Tdap. In one cohort study,¹⁴⁴ the subset of women receiving Tdap at or after 20 weeks’ gestation, as compared to their unvaccinated matches had no increased risk for incident gestational diabetes, thrombocytopenia, venous thromboembolism, or cardiac events (myocarditis, pericarditis, cardiomyopathy, or heart failure). In another study¹²³ comparing receipt of Tdap at 0 to 13 weeks’ gestation with receipt at 27 to 36 weeks’ gestation, earlier administration was not associated with increased risk of adverse events.

In a cohort study⁶⁵ of pregnant mothers in Australia who received Tdap compared to those who did, there was no increased risk of ADHD in their offspring compared to those who did not receive Tdap, including when looking by birth year and solely among nulliparous women.

KQ3d. What adverse events are associated with these vaccines in the fetus/infant?

KQ3d addresses the severity, range of possible effects, and risk factors for the adverse events in the fetus/infant of pregnant women.

KQ3d1. For each adverse event associated with a particular vaccine, what is the average severity and frequency?

We did not detect associations of in utero vaccine exposure with adverse events for the fetus or infant of vaccinated women. There were no studies in fetuses/infants of pregnant women using HepB, quadrivalent IIV or quadrivalent RIV. The prior 2014 report identified only evaluations of trivalent IIV and monovalent H1N1 influenza vaccine, which are no longer in use.

Among infants of women vaccinated for Tdap during pregnancy, we did not detect any statistically significant associations with adverse events.

KQ3d2. For adverse events without statistically significant associations with a particular vaccine, what is the range of possible effects?

Below we report all pooled risk estimates for adverse events that were reported on by more than one study. We note any instances where individual studies contributing to a pooled risk estimate were statistically significant on their own, indicating increased risk of an adverse event. We also report on studies with risk estimates that could not be computed and pooled statistically. Finally, as for KQ1 and KQ2, we summarize the range of possible effects by vaccine in brief below for all adverse events for which the relative risk estimate did not favor the intervention (i.e., RR>1), but the association was not statistically significant. Where appropriate, we contextualize the finding to understand the range of possible effects.

Hepatitis B Vaccines in the Fetus/Infant

We did not identify any studies that assessed adverse events associated with HepB, in the fetus/infant in this update or in our original report.

Influenza Vaccines in the Fetus/Infant

We did not identify any studies that assessed key adverse events associated with quadrivalent IIV or any studies of quadrivalent RIV in the fetus/infant in this update. The prior 2014 report identified only evaluations of trivalent IIV and monovalent H1N1 influenza vaccine, which are no longer in use.

Tdap in the Fetus/Infant

We identified four RCTs¹²⁵^, ¹⁷²^, ¹⁷⁹^, ¹⁸¹ and 13 cohort studies⁵²^, ⁶⁴^, ⁶⁵^, ⁶⁷^, ⁹¹^, ¹²¹^, ¹²³^, ¹⁴⁴^, ¹⁵⁴^, ¹⁷⁰^, ¹⁸²^, ¹⁹¹^, ¹⁹⁸ that assessed adverse events in pregnant women and/or their offspring among women who received Tdap. RCTs assessed Adacel¹²⁵^, ¹⁷²^, ¹⁷⁹ or Boostrix;¹⁸¹ in most cases, the vaccine(s) used for women followed in the cohort studies were not specified. Three of the RCTs¹⁷²^, ¹⁷⁹^, ¹⁸¹ were placebo controlled, and one RCT¹²⁵ compared Tdap to Td. As for outcomes in pregnant women, we did not separate out the one study that compared Tdap and Td, as Tdap is sufficiently different from Td given the acellular pertussis component. The other studies used participants not vaccinated with Tdap as control groups.

Fetal or infant outcomes of primary interest that were assessed in more than one study were birth defects, cardiovascular outcomes, death, encephalitis/encephalopathy, and seizures. We found no significant differences in fetal and infant adverse events across these studies.

Across eight studies,⁶⁷^, ¹²³^, ¹²⁵^, ¹⁷²^, ¹⁸¹^, ¹⁸²^, ¹⁹¹^, ¹⁹⁸ we found no evidence of increased risk of birth defects (RR 0.77; CI 0.43, 1.38), but we detected heterogeneity (I² 81%) and some evidence of publication bias (Egger test p=0.007, Begg test p=0.399). Of note, one of the studies¹⁸² reported an increased risk of birth defects (ankyloglossia [tongue tie]; RR 1.53; CI 1.33, 1.77; events 221/8299 vs 1063/61090); the next most common was congenital anomalies of the feet, which was less common in infants of mothers vaccinated with Tdap.

Across four studies,¹⁵⁴^, ¹⁸¹^, ¹⁸²^, ¹⁹¹ we found no evidence of increased risk of encephalitis/encephalopathy (RR 1.23; CI 0.60, 2.54). One of the studies¹⁵⁴ reported an increased risk of infant encephalopathy (RR 1.46; 1.23, 1.73; events 165/113094 vs 577/577000); however, this was based on crude rates, and in adjusted analyses there was no increased risk. Two¹⁸¹^, ¹⁹¹ of the remaining three studies showed either no cases of the adverse event (0/341 vs 0/346; 0/1199 vs 0/1259) and the remaining study¹⁸² showed a lower rate of the adverse event in the intervention group (12/8299 [0.14%] vs 101/61090 [0.16%]).

Across three studies,¹⁷⁰^, ¹⁸¹^, ¹⁹¹ we found no evidence of increased risk of cardiovascular events in infants of vaccinated mothers (RR 0.77; CI 0.50, 1.20). Across another three studies,¹⁸¹^, ¹⁹¹^, ¹⁹⁸ we detected no evidence of increased risk of deaths in infants (RR 0.15; CI 0.00, 8.88). Across yet another set of three studies,⁹¹^, ¹⁵⁴^, ¹⁸² we found no evidence of increased risk of seizures in infants (RR 1.02; CI 0.76, 1.35). While the RR was technically greater than 1, the percentages in each group experiencing seizures were essentially the same (0.25% in both groups when combining across all three studies).

We found no evidence of increased risk of febrile seizures, intussusception, meningitis, or stroke based on single studies¹²⁵^, ¹⁷⁰^, ¹⁷² reporting on the outcome. The study assessing meningitis was an RCT¹²⁵ with an RR of 1.03 (CI 0.07, 16.3). There was one case of infant meningitis in each of the two groups (N=134 versus N=138). No conclusions about the range of possible effects can be drawn based on the extremely wide confidence intervals.

Among other fetal or infant outcomes, one cohort study⁶⁴ of 81,993 infants in the United States found no difference in the risk of autism spectrum disorder between those exposed to Tdap in utero and those not exposed, with most who received a positive initial diagnosis receiving a confirmatory diagnosis after the age of two. A cohort study⁹¹ that followed 324,463 mother-infant pairs reported no difference in the risk for neonatal microcephaly between infants exposed to Tdap in utero and those not exposed.

One cohort study¹⁷⁰ found that infants exposed to Tdap in utero had a significantly higher gestational age at birth; a significantly decreased risk for preterm birth and small size for gestational age; and fewer days of hospitalization compared with those not exposed to Tdap. Likewise, another observational study¹⁸² reported a decreased risk for preterm birth, low birth weight, small for gestational age, large for gestational age, respiratory distress syndrome, transient tachypnea of newborn, tachycardia or bradycardia, hemolytic diseases, other neonatal jaundice, anemia, syndrome of infant of mother with gestational diabetes, and hypoglycemia in infants born to mothers who received Tdap during pregnancy.

A cohort study⁶⁵ of pregnant mothers in Australia who received Tdap compared to those who did not showed no association between prenatal Tdap and ADHD in offspring (HR 1.00; CI 0.88, 1.14).

KQ3d3. For each adverse event associated with a particular vaccine, what are the risk factors for the adverse event?

As outlined, we did not find any association of adverse events with vaccines in fetuses/infants exposed in utero, but we discuss any risk factors that were examined for adverse events below.

Of the three studies¹²³^, ¹⁴⁴^, ¹⁵⁴ that compared early versus optimal timing of Tdap, none identified any increased risk of adverse events among infants of pregnant women by timing of vaccine.

One cohort study¹⁷⁰ examined adverse events among infants of women who received one dose of Tdap versus two or more (from prior pregnancies). No difference in postnatal outcomes was noted between women who were administered Tdap at least twice in the past 5 years and those who received only a single dose.

KQ3. Summary of Findings for Safety of Vaccines in Pregnant Women and Their Fetuses/Infants

Table 23 documents the evidence for all identified studies evaluating the effects of vaccines in pregnant women for the prespecified key adverse events. The summary of findings table documents the results across studies grouped by vaccines. The table shows the number of RCTs, the number of other studies, the number of participants across pooled analyses, the studies contributing to the risk estimate, findings for the outcomes of interest, the criteria used to downgrade the SoE, and the SoE summary statement. The relative risk of an adverse event was derived by comparing the reported event rates in vaccinated participants compared to a control group across all studies that reported the data for that outcome. The absolute rates of adverse events (number of events, number of assessed participants) for the vaccine and the control group are also shown. In many instances, results were based on single occurrences of a specific adverse event. Where studies reported insufficient detail and did not contribute to the effect size estimates, the table reports the results as reported by the study authors.

Table 23

KQ3: Update summary of findings and SoE for safety of vaccines in pregnant women and their fetuses/infants.

Neither the prior 2014 report nor the update identified studies of HepB in pregnant women. The prior report found moderate SoE for no association of trivalent IIV and monovalent H1N1 influenza vaccine with serious adverse events in pregnant women. The update identified one study of quadrivalent IIV, but this study did not report on key adverse events. The update identified no studies of the quadrivalent RIV currently in use in pregnant women.

For Tdap, the update found no evidence of increased risk for maternal cardiovascular events, maternal death, maternal diabetes, eclampsia/pre-eclampsia, preterm labor, maternal reproductive system events, stillbirth, cardiovascular events in infants, death in infants, encephalitis/encephalopathy in infants, or seizures in infants (moderate SoE). The update found no evidence of increased risk of maternal encephalitis/encephalopathy, autism in infants, birth defects in infants, or febrile seizures in infants (low SoE). There was insufficient evidence due to the studies being graded as insufficient or no studies for some key adverse events.

Table 23a summarizes the findings across the prior 2014 report and the update.

Table 23a

KQ3: Safety of vaccines in pregnant women and their fetuses/infants.

Footnotes

b: “No evidence of increased risk” indicates that the outcome was studied and that the findings of the studies did not constitute evidence of increased risk of the adverse event following administration of that vaccine (because the risk either was not statistically significantly increased or was reduced).
c: “No evidence of increased risk” indicates that the outcome was studied and that the findings of the studies did not constitute evidence of increased risk of the adverse event following administration of that vaccine (because the risk either was not statistically significantly increased or was reduced).
d: “No evidence of increased risk” indicates that the outcome was studied and that the findings of the studies did not constitute evidence of increased risk of the adverse event following administration of that vaccine (because the risk either was not statistically significantly increased or was reduced).

Bookshelf ID: NBK572043

Contents

< Prev Next >

PubReader
Print View
Cite this Page
Gidengil C, Goetz MB, Maglione M, et al. Safety of Vaccines Used for Routine Immunization in the United States: An Update [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2021 May. (Comparative Effectiveness Review, No. 244.) Results.
PDF version of this title (9.4M)
Disable Glossary Links

Results - Safety of Vaccines Used for Routine Immunization in the United States:...
Results - Safety of Vaccines Used for Routine Immunization in the United States: An Update

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on