Vaccines for measles, mumps, rubella, and varicella in children

Carlo Di Pietrantonj; Alessandro Rivetti; Pasquale Marchione; Maria Grazia Debalini; Vittorio Demicheli

doi:10.1002/14651858.CD004407.pub5

. 2021 Nov 22;2021(11):CD004407. doi: 10.1002/14651858.CD004407.pub5

Vaccines for measles, mumps, rubella, and varicella in children

Carlo Di Pietrantonj ^1,^✉, Alessandro Rivetti ², Pasquale Marchione ³, Maria Grazia Debalini ⁴, Vittorio Demicheli ¹

Editor: Cochrane Acute Respiratory Infections Group

PMCID: PMC8607336 PMID: 34806766

Abstract

Background

Measles, mumps, rubella, and varicella (chickenpox) are serious diseases that can lead to serious complications, disability, and death. However, public debate over the safety of the trivalent MMR vaccine and the resultant drop in vaccination coverage in several countries persists, despite its almost universal use and accepted effectiveness. This is an update of a review published in 2005 and updated in 2012.

Objectives

To assess the effectiveness, safety, and long‐ and short‐term adverse effects associated with the trivalent vaccine, containing measles, rubella, mumps strains (MMR), or concurrent administration of MMR vaccine and varicella vaccine (MMR+V), or tetravalent vaccine containing measles, rubella, mumps, and varicella strains (MMRV), given to children aged up to 15 years.

Search methods

We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (the Cochrane Library 2019, Issue 5), which includes the Cochrane Acute Respiratory Infections Group's Specialised Register, MEDLINE (1966 to 2 May 2019), Embase (1974 to 2 May 2019), the WHO International Clinical Trials Registry Platform (2 May 2019), and ClinicalTrials.gov (2 May 2019).

Selection criteria

We included randomised controlled trials (RCTs), controlled clinical trials (CCTs), prospective and retrospective cohort studies (PCS/RCS), case‐control studies (CCS), interrupted time‐series (ITS) studies, case cross‐over (CCO) studies, case‐only ecological method (COEM) studies, self‐controlled case series (SCCS) studies, person‐time cohort (PTC) studies, and case‐coverage design/screening methods (CCD/SM) studies, assessing any combined MMR or MMRV / MMR+V vaccine given in any dose, preparation or time schedule compared with no intervention or placebo, on healthy children up to 15 years of age.

Data collection and analysis

Two review authors independently extracted data and assessed the methodological quality of the included studies. We grouped studies for quantitative analysis according to study design, vaccine type (MMR, MMRV, MMR+V), virus strain, and study settings. Outcomes of interest were cases of measles, mumps, rubella, and varicella, and harms. Certainty of evidence of was rated using GRADE.

Main results

We included 138 studies (23,480,668 participants). Fifty‐one studies (10,248,159 children) assessed vaccine effectiveness and 87 studies (13,232,509 children) assessed the association between vaccines and a variety of harms. We included 74 new studies to this 2019 version of the review. Effectiveness Vaccine effectiveness in preventing measles was 95% after one dose (relative risk (RR) 0.05, 95% CI 0.02 to 0.13; 7 cohort studies; 12,039 children; moderate certainty evidence) and 96% after two doses (RR 0.04, 95% CI 0.01 to 0.28; 5 cohort studies; 21,604 children; moderate certainty evidence). The effectiveness in preventing cases among household contacts or preventing transmission to others the children were in contact with after one dose was 81% (RR 0.19, 95% CI 0.04 to 0.89; 3 cohort studies; 151 children; low certainty evidence), after two doses 85% (RR 0.15, 95% CI 0.03 to 0.75; 3 cohort studies; 378 children; low certainty evidence), and after three doses was 96% (RR 0.04, 95% CI 0.01 to 0.23; 2 cohort studies; 151 children; low certainty evidence). The effectiveness (at least one dose) in preventing measles after exposure (post‐exposure prophylaxis) was 74% (RR 0.26, 95% CI 0.14 to 0.50; 2 cohort studies; 283 children; low certainty evidence). The effectiveness of Jeryl Lynn containing MMR vaccine in preventing mumps was 72% after one dose (RR 0.24, 95% CI 0.08 to 0.76; 6 cohort studies; 9915 children; moderate certainty evidence), 86% after two doses (RR 0.12, 95% CI 0.04 to 0.35; 5 cohort studies; 7792 children; moderate certainty evidence). Effectiveness in preventing cases among household contacts was 74% (RR 0.26, 95% CI 0.13 to 0.49; 3 cohort studies; 1036 children; moderate certainty evidence).  Vaccine effectiveness against rubella, using a vaccine with the BRD2 strain which is only used in China, is 89% (RR 0.11, 95% CI 0.03 to 0.42; 1 cohort study; 1621 children; moderate certainty evidence).  Vaccine effectiveness against varicella (any severity) after two doses in children aged 11 to 22 months is 95% in a 10 years follow‐up (rate ratio (rr) 0.05, 95% CI 0.03 to 0.08; 1 RCT; 2279 children; high certainty evidence). Safety There is evidence supporting an association between aseptic meningitis and MMR vaccines containing Urabe and Leningrad‐Zagreb mumps strains, but no evidence supporting this association for MMR vaccines containing Jeryl Lynn mumps strains (rr 1.30, 95% CI 0.66 to 2.56; low certainty evidence). The analyses provide evidence supporting an association between MMR/MMR+V/MMRV vaccines (Jeryl Lynn strain) and febrile seizures. Febrile seizures normally occur in 2% to 4% of healthy children at least once before the age of 5. The attributable risk febrile seizures vaccine‐induced is estimated to be from 1 per 1700 to 1 per 1150 administered doses. The analyses provide evidence supporting an association between MMR vaccination and idiopathic thrombocytopaenic purpura (ITP). However, the risk of ITP after vaccination is smaller than after natural infection with these viruses. Natural infection of ITP occur in 5 cases per 100,000 (1 case per 20,000) per year. The attributable risk is estimated about 1 case of ITP per 40,000 administered MMR doses. There is no evidence of an association between MMR immunisation and encephalitis or encephalopathy (rate ratio 0.90, 95% CI 0.50 to 1.61; 2 observational studies; 1,071,088 children; low certainty evidence), and autistic spectrum disorders (rate ratio 0.93, 95% CI 0.85 to 1.01; 2 observational studies; 1,194,764 children; moderate certainty). There is insufficient evidence to determine the association between MMR immunisation and inflammatory bowel disease (odds ratio 1.42, 95% CI 0.93 to 2.16; 3 observational studies; 409 cases and 1416 controls; moderate certainty evidence). Additionally, there is no evidence supporting an association between MMR immunisation and cognitive delay, type 1 diabetes, asthma, dermatitis/eczema, hay fever, leukaemia, multiple sclerosis, gait disturbance, and bacterial or viral infections.

Authors' conclusions

Existing evidence on the safety and effectiveness of MMR/MMRV vaccines support their use for mass immunisation. Campaigns aimed at global eradication should assess epidemiological and socioeconomic situations of the countries as well as the capacity to achieve high vaccination coverage. More evidence is needed to assess whether the protective effect of MMR/MMRV could wane with time since immunisation.

Plain language summary

Does the measles, mumps, rubella and varicella (MMRV) vaccine protect children, and does it cause harmful effects?

Background

Measles, mumps, rubella (German measles) and varicella (chickenpox) are infectious diseases caused by viruses. They are most common in children and young adults. They are not always serious, but can cause disability (such as deafness), complications and death. If pregnant women catch rubella, it may cause loss (miscarriage) of, or harm to, their unborn babies.  A vaccine is a medicine that prevents infection by a specific disease. The MMR (measles, mumps, rubella) vaccine protects people against all three of these infections (a combined vaccine). Doctors can vaccinate against chickenpox at the same time by mixing the chickenpox (varicella) vaccine with the MMR vaccine (MMRV) or giving it separately at the same time (MMR+V).  The MMR vaccine has reduced measles, mumps and rubella infections. However, some people think the MMR vaccine causes unwanted effects such as autism, swelling of the brain (encephalitis), meningitis, learning difficulties, type 1 diabetes, and other conditions. As a result, the number of children being vaccinated has fallen.  This is the 2019 update of a review first published in 2005 and previously updated in 2012.  Review question

We wanted to find out how effectively MMR, MMR+V and MMRV vaccines stop children (up to 15 years old) from catching measles, mumps, rubella and chickenpox. We also wanted to know if the vaccines cause unwanted effects.

Study characteristics

We looked for studies that assessed MMR, MMRV or MMR+V vaccines, given in any dose or time schedule, compared with not giving the vaccine, or giving a placebo vaccine (a sham treatment), to healthy children up to 15 years old. Studies needed to measure the number of cases of measles, mumps, rubella and chickenpox, and report whether children suffered any unwanted effects attributable to vaccination. We checked each study to make sure it used robust methods so that we could judge how reliable its results were.

Results We found 138 studies with more than 23 million children. Fifty‐one studies (10 million children) assessed how effective the vaccines were at preventing the diseases, and 87 studies (13 million children) assessed unwanted effects. In this 2020 update we have included 74 new studies published since 2012.  Measles: results from seven studies (12,000 children) showed that one dose of vaccine was 95% effective in preventing measles. Seven per cent of unvaccinated children would catch measles and this number would fall to less than 0.5% of children who receive one dose of vaccine.  Mumps: results from six studies (9915 children) showed that one dose of vaccine was 72% effective in preventing mumps. This rose to 86% after two doses, (3 studies, 7792 children). In unvaccinated children, 7.4% would catch mumps and this would fall to 1% if children were vaccinated with two doses.  The results for rubella (1 study, 1621 children) and chickenpox (one study, 2279 children) also showed that vaccines are effective. After one dose, vaccination was 89% effective in preventing rubella, using a vaccine with the BRD2 strain which is only used in China, and after 10 years the MMRV vaccine was 95% effective at preventing chickenpox infection.  Unwanted effects Overall, the studies found that MMR, MMRV and MMR+V vaccines did not cause autism (2 studies 1,194,764 children), encephalitis (2 studies 1,071,088 children) or any other suspected unwanted effect. Our analyses showed very small risks of fits due to high temperature or fever (febrile seizures) around two weeks after vaccination, and of a condition where blood does not clot normally (idiopathic thrombocytopenic purpura) in vaccinated children. 

Certainty of the evidence

Our certainty (confidence) in the evidence is slightly limited by the design of most of the studies. Nonetheless, we judged the certainty of the evidence for the effectiveness of the MMR vaccine to be moderate, and that for the varicella vaccine to be high. Our certainty in the evidence for autism and febrile seizures was also moderate.

Conclusions

Our review shows that MMR, MMRV and MMR+V vaccines are effective in preventing the infection of children by measles, mumps, rubella and chickenpox, with no evidence of an increased risk of autism or encephalitis and a small risk of febrile seizure.

Search date

This review includes evidence published up to 2 May 2019.

Summary of findings

Background

Description of the condition

Measles, mumps, and rubella (MMR) are serious diseases that can lead to potentially fatal illnesses, disabilities, and death. MMR is particularly prevalent in low‐income countries where vaccination programmes are inconsistent and mortality rates from disease are high. Large‐scale vaccination programmes have reduced MMR incidence, prevalence, and rates of complications in high‐income countries (Hambrosky 2015).

Measles is highly contagious with a case‐fatality rate ranging from 0.01% to 0.1% in high‐income countries to 3% to 30% in low‐income areas (Wolfson 2009). Otitis media (7% to 9%), pneumonia (8%), and diarrhoea (1% to 6%) are the most frequently reported complications of measles. These complications are responsible for the large proportion of measles‐related morbidity and mortality (Perry 2004). Pneumonia is the most common fatal complication of measles, occurring in 56% to 86% of measles‐related deaths (Bester 2016).

Rubella is an acute viral disease mostly affecting school‐aged children and young adults with high incidence and prevalence worldwide in the pre‐vaccine era (Lambert 2015). Women of childbearing age are susceptible to rubella infection before conception or during early pregnancy which can result in miscarriage, fetal death, or congenital rubella syndrome. These conditions are the most serious complications of rubella with incidence varying from fewer than 2 per 100,000 live births in the Americas and Europe to 121 per 100,000 live births in Africa and South East Asia (Vynnycky 2016).

Mumps is a viral infection that mostly affects children. Peak incidence occurs among those aged five to nine years (Hviid 2008). Annual incidence of 100 to 1000 cases/100,000 population was reported in the pre‐vaccine era with greater than 90% reduction after mumps vaccines were introduced (Hambrosky 2015). Orchitis (inflammation of the testicles) is the most common age‐related complication (12% to 66% of cases) (Yung 2011). The most serious complications are aseptic meningitis (1% to 10%) and deafness (4%) (Yung 2011).

Varicella (chickenpox) is a widespread and highly contagious infectious disease with peak incidence in children aged up to 15 years (Gershon 2015). Most epidemiological data are from high‐income countries and account for high pre‐vaccine incidence (from 320 to 1600 cases per 100,000) with case‐fatality rates of approximately 3 per 100,000 cases (Amjadi 2016; Helmuth 2015). Typically, varicella‐zoster virus (VZV) becomes latent in ganglionic neurons after primary infection, and reactivation may occur to cause zoster (shingles); risk increases with age (Gershon 2013).

Description of the intervention

The single‐component live attenuated vaccines of MMR were first licenced in the USA in the early 1960s (Plotkin 2017), and have been shown to be highly effective. Some combination vaccines were available from the early 1970s, including trivalent MMR vaccines; a combination of MMR with varicella (MMRV) was made available from 2005 (Plotkin 2017; WHO Position Paper 2017). At least two MMR vaccines are authorised worldwide and marketed widely:

MMR‐II or MMRVaxPro by Merck/MSD is a live‐virus vaccine. It is a sterile lyophilised preparation of 1000 TCID50 (50% tissue culture infectious doses) Enders' attenuated Edmonston measles strain propagated in chick embryo cell culture; mumps 20000 TCID50 Jeryl Lynn strain propagated in chick embryo cell culture; and rubella 1000 TCID50 Wistar RA 27/3 propagated on human diploid lung fibroblasts. The growth medium is medium 199 (5.7 mg) used with neomycin as stabiliser;
Priorix vaccine, Glaxo SmithKline Beecham (GSK), is a lyophilised mixed preparation of the attenuated Schwarz measles CCID50 (50% cell culture infective dose) strain; RIT 4385 mumps CCID50 (derived from Jeryl Lynn strain); and CCID50 Wistar RA 27/3 rubella strain of viruses. These are obtained separately by propagation either in chick embryo tissue cultures (mumps and measles) or MRC5 human diploid cells (rubella). The vaccine also contains residual amounts of neomycin (25 µg per dose).

A World Health Organization (WHO) pre qualified MMR vaccine has also been licenced by the Serum Institute of India/Masu Co Ltd for Asian markets. It is a sterile lyophilised preparation containing live attenuated Edmonston‐Zagreb measles virus (not less than 1000 CCID50), Leningrad‐Zagreb mumps virus (not less than 5000 CCID50), and Wistar RA 27/3 rubella virus (not less than 1000 CCID50).

Other commercial formulations of MMR vaccines have been used over the past 30 years, and to date are authorised in few countries, or have been withdrawn from marketing for commercial, safety, or both commercial and safety reasons:

Morupar by Chiron contains live attenuated Schwarz measles strain 1000 TCID50, propagated in chick embryo cell culture; Wistar RA 27/3 rubella strain 1000 TCID50, propagated on human diploid lung fibroblasts; and Urabe AM9 mumps 5000 TCID50, propagated in chick embryo cell culture, with neomycin as stabiliser (withdrawn globally because of increased allergic reactions due to the manufacturing process);
Trimovax by Pasteur‐Merieux Serums and Vaccines contains live attenuated Schwarz measles strain, 1000 CCID50; Urabe AM9 mumps strain, 5000 TCID50; and Wistar RA 27/3 rubella strain, 1000 TCID50;
Triviraten Berna contains live attenuated Edmonston‐Zagreb (EZ 19) measles strain, 1000 TCID50; Rubini mumps strain, 5000 TCID50; and Wistar RA 27/3 rubella strain, 1000 TCID50 propagated on human diploid cells. The product contains lactose (14 mg), human albumin (8.8 mg), sodium bicarbonate (0.3 mg), medium 199 (5.7 mg), and distilled water as solvent.

Two main MMRV combined vaccines are authorised for worldwide use and contain live attenuated Oka/Merck strain VZV:

ProQuad by Merck/MSD is a live‐virus vaccine with the same composition as MMR‐II/MMRVaxPro, including live attenuated Oka/Merck VZV strain, 3.99 log10 PFU (plaque forming units) propagated on MRC‐5 human diploid cells; and
Priorix Tetra by GSK is a live‐virus vaccine with the same composition as Priorix, including live attenuated Oka/Merck VZV strain, 103.3 PFU propagated on MRC‐5 human diploid cells.

The components of monovalent and subsequently combined MMR vaccine are described below (Plotkin 2017). Most attenuated measles vaccines currently produced worldwide are derived from the Edmonston strain. Vaccines containing non‐Edmonston‐derived strains are also in use, including Leningrad‐16, Shanghai‐191, CAM‐70, and TD97. In most cases the virus is cultured in chick embryo cells. However, a few vaccines are attenuated in human diploid cells. Most vaccines contain traces of antibiotics (e.g. 25 µg neomycin per dose), but some do not. Sorbitol and gelatine are used as stabilisers (Plotkin 2017; WHO Position Paper 2017).

More than 10 mumps vaccine strains (Jeryl Lynn, Urabe, Hoshino, Rubini, Leningrad‐3, L‐Zagreb, Miyahara, Torii, NK M‐46, S‐12, and RIT 4385) have been used throughout the world, but the Jeryl Lynn strain is the most widely used to date (Plotkin 2017). Although some manufacturers produce live mumps vaccines containing the Urabe AM9 virus strain, some countries have promptly stopped Urabe strain‐containing MMR vaccines because of concerns about vaccine‐associated meningitis. Viruses are often cultured in chick embryo fibroblasts (as with the Jeryl Lynn and Urabe strain‐containing vaccines), but quail and human embryo fibroblasts are also used. Most vaccines also contain neomycin (25 µg per dose) (WHO Position Paper 2017).

Most rubella vaccines used throughout the world contain the RA 27/3 virus strain. Exceptions are vaccines produced in Japan, which use different virus strains: Matsuba, DCRB 19, Takahashi, TO‐336 (cultured in rabbit kidney cells), and Matsuura (produced using quail embryo fibroblasts) (Plotkin 2017). The RA 27/3 strain is used most often because of consistent immunogenicity, induction of resistance to re‐infection, and low rate of adverse effects (WHO Position Paper 2017). The live virus produces viraemia and pharyngeal excretion, but both are of low magnitude and are non‐communicable (Plotkin 2017).

All available monovalent VZV vaccines consist of the Oka virus strain, which was subsequently attenuated by sequential passage in cultures of human embryonic lung cells, embryonic guinea pig cells, and the human diploid cell line WI‐38 or MCR‐5 (Plotkin 2017). The titre of VZV is around 14 times higher in the MMRV vaccines described than in the monovalent VZV vaccine (WHO Position Paper 2014).

How the intervention might work

Combined MMR (trivalent vaccine, containing measles, rubella, mumps strains), MMR+V (concurrent administration of MMR vaccine and varicella (chickenpox) vaccine), and MMRV (tetravalent vaccine containing measles, rubella, mumps, varicella strains) vaccines are widely recommended by health authorities and offer advantages over individual vaccines in the facilitation of current immunisation implementation strategies. Moreover, trivalent vaccines are included in the WHO Expanded Programme on Immunization, and are used in almost all European countries, the USA, Canada, Australia, New Zealand, and 100 other countries around the world (Orenstein 2018; WHO GVAP 2013). Quadrivalent MMRV vaccines are also recommended, but have to date been implemented in a limited number of countries where varicella vaccination is routinely recommended (WHO Immunization Monitoring 2019). According to accepted recommendations, the first dose of both MMR and MMRV should be administered on or after the child's first birthday (from 9 to 15 months of age), and the second dose at least 28 days later, or from 4 to 10 years of age (WHO Immunization Monitoring 2019; WHO Position Paper 2017). Combined vaccines provide a significant improvement in the efficiency of childhood immunisation, and a meaningful reduction in costs through increasing immunisation coverage against specific diseases with a single injection (Vesikari 2007).

Until 2011, single‐component measles vaccine was largely used in nearly all African and several Asian, and Western European WHO member states with different implementation strategies (single‐dose or second‐dose administration) (WHO GVAP 2013). A first dose of measles‐containing vaccine at nine months of age has been recommended in all countries with ongoing transmission and high risk of measles mortality among infants to ensure adequate protection. The introduction of a second measles‐containing vaccine dose at 15 to 18 months of age has been recommended when coverage of at least 80% for the first dose of measles‐containing vaccine has been reached for three consecutive years. By 2011, all 194 WHO member states had introduced or begun the process of introducing a two‐dose measles vaccination strategy through routine immunisation services, supplementary immunisation activity, or both (WHO Strategic Plan 2012). However, this policy was revised in April 2017, and recommended including the second measles vaccine dose in national vaccination schedules regardless of the coverage level (WHO Position Paper 2017). As of December 2010, 131 of the 194 WHO member states included MR or MMR combined vaccines in routine immunisation programmes (WHO Strategic Plan 2012). Relevant progress has been made toward the ambitious goals of the Global Measles and Rubella Strategic Plan 2012 to 2020 (WHO Strategic Plan 2012), with a further 23 of 194 WHO member states introducing a second dose of measles‐containing vaccine, and 17 countries introducing the rubella‐containing vaccine (Orenstein 2018).

Between 2000 and 2017, estimated measles vaccine coverage increased globally from 72% to 85%, with a reported 83% reduction of annual measles incidence and 80% reduction in estimated measles mortality (Dabbagh 2018). Estimated global rubella vaccine coverage increased from 39% to 46%, with high regional variability ranging from 12% in South East Asia to 94% in Europe (Orenstein 2018). According to Regional Verification Commissions in the American, European and Western Pacific Regions, the goal of measles elimination (end of endemic transmission for at least three years) had been reached by the end of 2015 in 61 member states (34/35, 21/53, and 6/27 member states respectively in the Americas, Europe, and western Pacific) and elimination of rubella in 55 member states (35/35 and 20/53 member states in the Americas and Europe, respectively) (Orenstein 2018; Perry 2015). However, measles elimination milestones have not been met in several countries in all WHO regions, and measles resurgence has been reported from 2017 to 2019 because of large outbreaks (Dabbagh 2018; Zimmerman 2019).

A global technical consultation requested by the WHO assessed the feasibility of measles elimination through mass immunisation and convened that eradication is biologically, technically, and operationally feasible (WHO 2011). MMR capability to eliminate the targeted diseases has been demonstrated in a number of countries and different scenarios.

The largest country to have ended endemic measles transmission is the USA, where the elimination of endemic measles had been previously verified in 2000 (CDC 2005; CDC 2012; Orenstein 2004). The interruption of indigenous transmission was first observed in 1993 after refining the elimination strategy to face the large resurgence of measles that occurred from 1989 to 1991 (CDC 1992; Watson 1998). Incidence has remained at less than 1 case per 1 million population continuously since 1997, with most measles cases from 2001 representing importations or import‐associated infections (CDC 2012; Fiebelkorn 2017). The elimination of rubella and congenital rubella syndrome was verified in 2004 by an external expert panel (CDC 2005). The incidence remained below 1 case per 10 million population with an annual median number of 10 cases (range 4 to 18 cases) (CDC 2012; Hinman 2011). Recent studies and reviews of USA measles and rubella outbreaks showed that most imported cases were unvaccinated people in areas with suboptima vaccination coverage and in regions where herd immunity threshold for first or second dose had not been reached, or both (Fiebelkorn 2017; Lee 2019; Papania 2014).

In Europe, measles and rubella outbreaks and endemic transmission persisted at regional levels due to suboptima vaccination coverage (Zimmerman 2019). Despite the substantial reduction of measles and rubella incidence, 21 of 53 countries in the European Union had interrupted the endemic transmission of measles, and 20 member states had interrupted endemic transmission of rubella (Muscat 2014; Orenstein 2018; WHO Regional Office for Europe 2016).

Finland was the first European country to end endemic measles transmission through a national vaccination programme as a two‐dose schedule launched in 1982, with an unremitting 95% coverage for both doses until 2017 (National Institute for Welfare and Health 2017; Peltola 2008). Incidence declined to 1 case per 1 million population for all MMR diseases in 1995, and in 1999 the country was documented as being free of indigenous measles, mumps, and rubella (Davidkin 2010). Since then, a few clusters of MMR imported cases have been observed annually without any outbreaks (WHO 2017).

After the introduction of MMR vaccine in 1988 for children aged 13 to 15 months with a catch‐up campaign for preschool‐aged children, the annual incidence of measles declined sharply in England and Wales, from 160/100,000 in 1989 to 17/100,000 in 1995 (Gay 1997; Ramsay 2003). The interruption of indigenous transmission was first observed in 1996 after a widespread vaccination campaign in 1994 and the introduction of the second MMR dose in 1995 (Vyse 2002). Nevertheless, endemic transmission in the UK re‐established in 2006 because of intense media coverage of the fraudulent Wakefield claim of a suspected link among MMR vaccines and autism (Public Health England 2019a). Moreover, an increased number of mumps‐confirmed cases were reported in England and Wales (Public Health England 2019b). However, after different nationwide vaccination campaigns, the UK had interrupted endemic transmission of measles and rubella by 2014, and elimination was certified in 2017 from the Regional Verification Commission for Measles and Rubella Elimination. Furthermore, a significant reduction of mumps cases in school‐aged children has been observed with persisting outbreaks in young adults (Public Health England 2019c).

Although varicella vaccines are licenced worldwide, a limited number of countries routinely recommend varicella vaccination with a one‐ or two‐dose programme (WHO Immunization Monitoring 2019). The USA was the first country to recommend a routine one‐dose programme in 1996, and an updated routine two‐dose programme in 2006 (Marin 2007). A progressive reduction of overall varicella incidence has been observed in target age groups, with more than 90% decrease in cases when maintaining coverage with two doses over 80%. Moreover, a significant reduction of zoster incidence has been observed in children and adolescents, but it is too early to observe the impact of childhood varicella vaccination in adults and the elderly (Harpaz 2019). Similar data have been reported in some European countries: Italy and Spain reported 75% and 89% reductions, respectively, despite lower rates of immunisation coverage (Bechini 2015; Garcia Cenoz 2013). No evidence suggested a shift of varicella disease burden to older age groups after the introduction of varicella vaccination, but significant reductions in hospitalisations, complications, and deaths have been reported globally (Wutzler 2017).

Why it is important to do this review

Despite its worldwide use, no systematic reviews studying the effectiveness and safety of MMR or MMRV vaccines are available.

Objectives

To assess the effectiveness, safety, and long‐ and short‐term adverse effects associated with the MMR (trivalent vaccine, containing measles, rubella, mumps strains), or MMR+V (concurrent administration of MMR vaccine and varicella vaccine), or MMRV (tetravalent vaccine containing measles, rubella, mumps, varicella strains), given to children aged up to 15 years.

Methods

Criteria for considering studies for this review

Types of studies

We included randomised controlled trials (RCTs), controlled clinical trials (CCTs), prospective and retrospective cohort studies (PCS/RCS), case‐control studies (CCS), interrupted time‐series (ITS) studies, case cross‐over (CCO) studies, case‐only ecological method (COEM) studies, self‐controlled case series (SCCS) studies, person‐time cohort (PTC) studies, and case‐coverage design/screening methods (CCD/SM) studies. See Appendix 1 for study design definitions (based on Farrington 2004; Harris 2006; Higgins 2011; Jefferson 1999; Last 2001; Maclure 1991; Morgenstern 1995). A study taxonomy is shown in Appendix 2.

Observational study design was crucial in this review because the main concern about MMR/V vaccination is in regard to safety. The cohort, case‐control, and case‐only studies are valid study designs to investigate the possible association between vaccination and rare adverse events (Farrington 2004).

Types of participants

Healthy children aged up to 15 years, or adults who received MMR or MMRV/MMR+V vaccination between 0 and 15 years of age. We included studies (or data sets) where participants received vaccination before 16 years of age. For studies conducted in the general population, only data regarding participants vaccinated under 15 years were included in analyses. Studies where most participants received vaccination when aged 16 years or older were excluded.

Types of interventions

Vaccination with any combined MMR or MMRV/MMR+V vaccine given in any dose, preparation, or time schedule compared with no intervention or placebo.

MMR (trivalent vaccine containing measles, rubella, mumps strains). MMR+V (concurrent administration of MMR vaccine and varicella vaccine). MMRV (tetravalent vaccine containing measles, rubella, mumps, varicella strains).

Types of outcome measures

Primary outcomes

Effectiveness: clinical and/or laboratory‐confirmed cases of measles, mumps, rubella, or varicella.
Safety: encephalitis or encephalopathy, aseptic meningitis, seizure (febrile/afebrile), autism spectrum disorders, inflammatory bowel disease, cognitive delay, developmental delay, idiopathic thrombocytopenic purpura, Henoch‐Schönlein purpura, type 1 diabetes, asthma, dermatitis or eczema, hay fever, rhinoconjunctivitis, hypersensitivity/allergy, acute leukaemia, demyelinating diseases, multiple sclerosis, encephalomyelitis, acute disseminated encephalomyelitis (ADEM), gait disturbances, bacterial or viral infections.

Secondary outcomes

Short‐term side effects: local reactions (e.g. soreness and redness at the site of inoculation) and systemic reactions (e.g. fever, rash, vomiting, and diarrhoea) following MMR or MMRV vaccination.

Search methods for identification of studies

Electronic searches

We searched the following databases up to 2 May 2019:

the Cochrane Central Register of Controlled Trials, which contains the Cochrane Acute Respiratory Infections Group's Specialised Register (CENTRAL; 2019, Issue 5) in the Cochrane Library using the strategy in Appendix 3;
MEDLINE via PubMed (from 1966 to 2 May 2019) using the strategy in Appendix 3; and
Embase via Elsevier (from 1974 to 2 May 2019) using the strategy in Appendix 3.

We searched the following trial registers on 2 May 2019:

US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (www.clinicaltrials.gov); and
World Health Organization International Clinical Trials Registry Platform (WHO ICTRP) (apps.who.int/trialsearch).

We used the strategies in Appendix 3 and did not restrict the results by language or publication status (published, unpublished, in press, or in progress).

Searching other resources

For effectiveness trials, we searched bibliographies of all relevant articles obtained and any published reviews for additional studies. We also searched trial registers (WHO ICTRP and ClinicalTrials.gov) for unpublished, prospectively registered trials. For safety trials, we assessed bibliographies of all relevant articles and any published reviews for additional studies. We imposed no language restrictions on all searches.

Data collection and analysis

Selection of studies

Two review authors (CDP, AR) independently applied the inclusion criteria to all identified and retrieved articles. A third review author (VD) arbitrated in case of disagreements about the eligibility of a study.

Data extraction and management

Two review authors (CDP, AR) independently performed data extraction using a data extraction form (Appendix 4). A third review author (VD) checked data extraction and arbitrated in case of disagreement. For each study, relevant information was summarised and reported by main outcomes in Additional tables and Characteristics of included studies.

We used a two‐letter prefix to distinguish types of study designs and whether these related to effectiveness/efficacy or safety (only). The first letter signifies the study design (a = RCT, b = case control, c = cohort, d = self‐controlled case series, e = case cross‐over, f = case‐coverage design, g = case‐only ecological method, h = interrupted time series), and the second letter signifies the endpoint (a = effectiveness/efficacy, b = safety only). See Appendix 2.

We classified the funding sources of included studies as follows.

Government or not‐for‐profit organisation: explicitly stated that funding sources were public institutions, not‐for‐profit organisations, health department, or other government institutions. All authors were affiliated with public institutions, and none were affiliated with the pharmaceutical industry. All critical aspects of the research (participant selection, outcome assessment, statistical analysis, vaccine supplies) were conducted without pharmaceutical industry support.
Pharmaceutical industry: explicitly declared that funding was provided by the pharmaceutical industry. All authors were affiliated with the pharmaceutical industry. All critical aspects of the research (participant selection, outcome assessment, statistical analysis, vaccine supplies) were conducted with pharmaceutical industry support.
Mixed (government and pharmaceutical industry): at least one author was affiliated with the pharmaceutical industry. Statistical analysis was conducted with pharmaceutical industry support. Study vaccines were supplied by the pharmaceutical industry.
Not stated or unclear: funding source was not declared, therefore it was not possible to apply the funding classification criteria.

Assessment of risk of bias in included studies

Two review authors (CDP, AR) independently assessed the methodological quality of the included studies (Appendix 5). We assessed the quality of RCTs and quasi‐RCTs using criteria adapted from the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We assessed the quality of non‐RCTs in relation to the presence of potential confounders that could make interpretation of the results difficult. We evaluated the quality of case‐control (prospective and retrospective) and cohort studies using the appropriate Newcastle‐Ottawa Scales (Stang 2010; Wells 2000). We applied quality control assessment grids based on those developed by the University of York, NHS Centre for Reviews and Dissemination (Appendix 5) to historical controlled trials (HCTs), interrupted time‐series (Khan 2001).

Experimental and quasi‐experimental studies

See Appendix 5.

Random sequence generation

Low risk of bias: e.g. a table of random numbers or computer‐generated random numbers.
High risk of bias: e.g. alternation, date of birth, day of the week, or case record number.
Unclear risk of bias: if insufficient information was provided.

Allocation concealment

Low risk of bias: e.g. numbered or coded identical containers were administered sequentially; an on‐site computer system that could only be accessed after entering the characteristics of an enrolled participant; or serially numbered, opaque, sealed envelopes, or sealed envelopes that were not sequentially numbered.
High risk of bias: e.g. an open table of random numbers.
Unclear risk of bias: if insufficient information was provided.

Blinding

Low risk of bias: if adequate double‐blinding (e.g. placebo vaccine) or single‐blinding (i.e. blinded outcome assessment) was used.
High risk of bias: if there was no blinding.
Unclear risk of bias: if insufficient information was provided.

Incomplete outcome data

Low risk of bias: no missing data, or the proportion of missing data compared with the observed event risk was not enough to have a clinically relevant impact on the intervention effect estimate.
High risk of bias: when the proportion of missing data compared with observed event risk was large enough to induce clinically relevant bias in the intervention effect estimate.
Unclear risk of bias: if insufficient information was provided.

Non‐experimental studies

See Appendix 5.

We used different methodological quality checklists (unpublished) for the different case‐only design studies for:

self‐controlled case series (SCCS) and person‐time cohort (PTC) checklist based on Farrington 2004 and Petersen 2016;
case cross‐over studies (CCO) checklist was based on Farrington 2004 and Maclure 1991; and
case‐coverage methods/screening method (CCM/SM); and for case‐only ecological method (COEM) studies checklist was based on Farrington 2004.

We assessed evidence quality as a component of interpreting the overall results. We assigned the following 'Risk of bias' categories (Higgins 2011):

low risk of bias: plausible bias unlikely to seriously alter the results;
unclear risk of bias: plausible bias that raises some doubt about the results; and
high risk of bias: plausible bias that seriously weakens confidence in the result.

Measures of treatment effect

We used risk ratio (RR) and its confidence interval (CI) as measures of effect for RCT and cohort studies. We used the odds ratio (OR) and its CI for case‐control studies. The usual effect measure for case‐only studies is the rate ratio (rr). We calculated vaccine efficacy (or effectiveness) as VE = (1 − effect estimate) x 100, expressed as a percentage. For cohort and RCT/CCT studies VE = (1 − RR) x 100. For case‐control studies VE = (1 − OR) x 100. For study designs adopting the rr as effect measure (rate = events/person‐time), the vaccine effectiveness is VE = (1 − rr) x 100.

The inclusion of different studies involved different estimation methods and statistical models, so we are dealing with different measures of effect. Cohort studies may use the RR to compare two groups, or more sophisticated statistical models such as the logistic regression model or the proportional hazard regression model, where the effect measures reported are OR or hazard ratio (HR), respectively. Case‐control studies adopt the logistic regression model, so the effect measure is the OR. Case‐only studies design (SCCS, person‐time cohort, case cross‐over studies) use the Poisson regression model. In this case the effect measure is rr. Consequently, in order to perform meta‐analysis in some cases we had to convert one measure of the effect into another using the formulae described in Higgins 2011.

We converted temperatures to degrees celsius (°C) using the formula °C = (Fahrenheit − 32)/1.8.

Unit of analysis issues

We considered analytical studies that provided data at the person‐level for this review. The only ecological design considered was case‐only ecological study (COES). The differences between ecological study design and case‐only ecological study are described in Appendix 1.

Where several vaccine arms from the same study design were included in the same analysis, we split the placebo group equally between the different arms, so that the total number of participants in a single analysis did not exceed the actual number in the study.

Dealing with missing data

For this update we wrote to study authors to request missing data or for clarification. The response was disappointing, and we desisted from further attempts. Our analysis relies on existing data. Whenever possible we used the intention‐to‐treat (ITT) population. When necessary and possible we used strategies described in Di Pietrantonj 2006 to impute missing outcome data.

Assessment of heterogeneity

We calculated the I² statistic for each pooled estimate to assess the impact of statistical heterogeneity. The I² statistic can be interpreted as the proportion of total variation amongst effect estimates due to heterogeneity rather than sampling error, and is intrinsically independent from the number of studies. When the I² statistic is less than 30%, there is little concern about statistical heterogeneity (Higgins 2011). We used random‐effects models throughout to take account of the between‐study variance in our findings (Higgins 2011). Not all studies reported detail sufficient to enable a full analysis of the sources of heterogeneity.

Assessment of reporting biases

A detailed description of the study quality is provided in the Risk of bias in included studies section. We assessed publication bias by inspecting the funnel plots and heterogeneity (I²) (see Assessment of heterogeneity). Due to the limited number of studies in each comparison, the assessment of publication bias was not applicable. Since the evidence presented in this review originated mainly from published data, we cannot be sure that our results are not affected by publication bias. We were unable to retrieve unpublished papers, thus our results could be affected by publication bias.

Data synthesis

We carried out quantitative and qualitative data syntheses separately for efficacy/effectiveness and safety. We grouped studies for quantitative analysis according to study design (see Types of studies), vaccine type (MMR, MMRV, MMR+V), virus strain, and study settings. We incorporated heterogeneity into the pooled estimates by using the DerSimonian Laird random‐effects model.

Most of the studies included in this review were observational studies, therefore quantitative synthesis is performed on adjusted estimates by multivariate models. The estimates are adjusted for age and gender. The multicentre studies also take into account the geographical area, address, school, paediatric practice, and health organisation/insurance. Some studies adjusted estimates for the health history and health status of the older siblings.

As explained in the Measures of treatment effect section, the different studies involved different statistical models and estimation methods, so we are dealing with different measures of effect. Consequently, in some cases, in order to perform the meta‐analysis, we converted one measure of effect into another using the formulae described in Higgins 2011.

The cohort studies on MMR vaccine effectiveness against measles and mumps present estimates not adjusted by multivariate models but report binary data (fourfold frequency table) stratified by doses. In this case, the quantitative synthesis is performed on binary data. If some studies reported adjusted estimates, we used the method described in Di Pietrantonj 2006 to convert adjusted effect estimates into adjusted binary data.

We used RR for comparisons between vaccine and placebo/control groups for RCTs and cohort studies. We used rr for cohort studies using Poisson regression or the proportional hazard regression model. We OR for case‐control studies and rr for case‐only study designs.

We classified and discussed included studies according to the type of outcomes for which they provided evidence, effectiveness, and possible association with harms or local and systemic adverse effects. We illustrated study characteristics, design, population, and outcomes definitions in Additional tables.

Subgroup analysis and investigation of heterogeneity

We planned to carry out subgroup analyses where data were available, as follows.

Age group
1. aged < 5 years, aged 5 to 10 years;
2. aged < 6 years, aged 11 to 16 years; and
3. aged < 1 year, aged 1 to 4 years, aged 5 to 14 years.
Number of doses administered
1. all doses, 1 dose, 2 doses, at least 1 dose (or any dose).
Length of follow‐up
1. < 5 years, 5 to 10 years.
Risk period (self‐controlled case series)
1. 0 to 30 days, 31 to 60 days, 61 to 90 days.
Disease severity
1. moderate, severe.

Sensitivity analysis

We had planned to perform a sensitivity analysis on results by applying fixed‐effect and random‐effects models to assess the impact of heterogeneity on our results. We performed a sensitivity analysis by excluding studies at high risk of bias to assess the robustness of our conclusions.

Summary of findings and assessment of the certainty of the evidence

We created 21 'Summary of findings' tables using the outcomes listed in Appendix 6.

Effectiveness against measles
Effectiveness against mumps
Effectiveness against rubella
Effectiveness against varicella
Safety ‐ short‐term side effects
Safety ‐ encephalitis or encephalopathy
Safety ‐ aseptic meningitis
Safety ‐ seizures (febrile/afebrile)
Safety ‐ autism spectrum disorders
Safety ‐ inflammatory bowel disease
Safety ‐ cognitive/developmental delay
Safety ‐ idiopathic thrombocytopenic purpura
Safety ‐ Henoch‐Schönlein purpura
Safety ‐ type 1 diabetes
Safety ‐ asthma
Safety ‐ eczema/dermatitis
Safety ‐ hay fever, rhinoconjunctivitis, hypersensitivity/allergy
Safety ‐ acute leukaemia
Safety ‐ demyelinating diseases ‐ multiple sclerosis ‐ acute disseminated encephalomyelitis (ADEM)
Safety ‐ gait disturbances
Safety ‐ bacterial or viral infections, immune overload

We used the five GRADE considerations (study limitations, consistency of effect, imprecision, indirectness, and publication bias) to assess the quality of a body of evidence as it relates to the studies that contributed data to the meta‐analyses for the prespecified outcomes (Atkins 2004). We used the methods and recommendations described in Section 8.5 and Chapter 12 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), employing GRADEpro GDT software (GRADEpro GDT 2015). We justified all decisions to down‐ or upgrade the quality of studies using footnotes, and made comments to aid readers’ understanding of the review where necessary.

Results

Description of studies

Results of the search

We updated searches on 2 May 2019 and identified 13,196 records for screening. We retrieved 101 papers after reviewing titles and abstracts, 74 of which we considered for this 2019 update. We also evaluated 16 studies identified as awaiting classification in our previous update (Demicheli 2012), of which we considered 12 studies. We included a total of 74 new studies, plus 12 studies from our previous update, for a total of 86 new included studies for this 2019 update. This review includes a total of 138 studies (see Figure 1; Figure 2).

Included studies

We included nine randomised controlled trials (RCTs) (aa‐Henry 2018; aa‐Povey 2019; aa‐Prymula 2014; ab‐Bloom 1975; ab‐Edees 1991; ab‐Freeman 1993; ab‐Lerman 1981; ab‐Peltola 1986; ab‐Schwarz 1975); one controlled clinical trial (CCT) (ab‐Ceyhan 2001); 63 cohort studies (PCS/RCS) (ca‐Arciuolo 2017; ca‐Arenz 2005; ca‐Barrabeig 2011a; ca‐Barrabeig 2011b; ca‐Bhuniya 2013; ca‐Chamot 1998; ca‐Chang 2015; ca‐Choe 2017; ca‐Compés‐Dea 2014; ca‐Giaquinto 2018; ca‐Greenland 2012; ca‐Hales 2016; ca‐La Torre 2017; ca‐Livingston 2013; ca‐Lopez Hernandez 2000; ca‐Ma 2018; ca‐Marin 2006; ca‐Marolla 1998; ca‐Musa 2018; ca‐Nelson 2013; ca‐Ogbuanu 2012; ca‐Ong 2005; ca‐Ong 2007; ca‐Rieck 2017; ca‐Schlegel 1999; ca‐Snijders 2012; ca‐Spackova 2010; ca‐Tafuri 2013; ca‐Takla 2014; ca‐Wichmann 2007; ca‐Woudenberg 2017; cb‐Ahlgren 2009; cb‐Barlow 2001; cb‐Beck 1989; cb‐Benjamin 1992; cb‐Benke 2004; cb‐Beyerlein 2017; cb‐DeStefano 2002; cb‐Dunlop 1989; cb‐Gavrielov‐Yusim 2014; cb‐Hviid 2004; cb‐Hviid 2008; cb‐Hviid 2019; cb‐Jacobsen 2009; cb‐Jain 2015; cb‐Klein 2010; cb‐Klein 2012; cb‐Klein 2017; cb‐Madsen 2002; cb‐Makino 1990; cb‐McKeever 2004; cb‐Miller 1989; cb‐Mrozek‐Budzyn 2013; cb‐Robertson 1988; cb‐Rowhani‐Rahbar 2013; cb‐Schink 2014; cb‐Sharma 2010; cb‐Stokes 1971; cb‐Swartz 1974; cb‐Timmermann 2015; cb‐Uchiyama 2007; cb‐Vestergaard 2004; cb‐Weibel 1980); 35 case‐control studies (CCS) (ba‐Andrade 2018; ba‐Castilla 2009; ba‐Cenoz 2013; ba‐Defay 2013; ba‐Fu 2013; ba‐Giovanetti 2002; ba‐Goncalves 1998; ba‐Harling 2005; ba‐Hungerford 2014; ba‐Jick 2010; ba‐Kim 2012; ba‐Liese 2013; ba‐Mackenzie 2006; ba‐Vazquez 2001; bb‐Ahlgren 2009; bb‐Baron 2005; bb‐Bertuola 2010; bb‐Black 1997; bb‐Black 2003; bb‐Bremner 2005; bb‐Bremner 2007; bb‐Chen 2018; bb‐Da Dalt 2016; bb‐Davis 2001; bb‐De Stefano 2004; bb‐Dockerty 1999; bb‐Groves 1999; bb‐Ma 2005; bb‐Mallol‐Mesnard 2007; bb‐Mrozek‐Budzyn 2010; bb‐Ray 2006; bb‐Shaw 2015; bb‐Smeeth 2004; bb‐Uno 2012; bb‐Vcev 2015); 16 self‐controlled case series/person‐time cohort studies (SCCS/PTC) (db‐Andrews 2012; db‐Dourado 2000; db‐Farrington 1995; db‐France 2008; db‐Macartney 2017; db‐MacDonald 2014; db‐Makela 2002; db‐McClure 2019; db‐Miller 2003; db‐Miller 2005; db‐Miller 2007; db‐O'Leary 2012; db‐Perez‐Vilar 2018; db‐Stowe 2009; db‐Taylor 1999; db‐Ward 2007); 3 case cross‐over studies (CCO) (eb‐Ki 2003; eb‐Lafaurie 2018; eb‐Park 2004); and 11 case‐only ecological method studies (COEM) (ga‐Boccalini 2015; ga‐Pozza 2011; ga‐Tafuri 2015; gb‐da Cunha 2002; gb‐da Silveira 2002; gb‐Fombonne 2001; gb‐Fombonne 2006; gb‐Honda 2005; gb‐Jonville‐Bera 1996; gb‐Seagroatt 2005; gb‐Taylor 2002).

We classified studies reported as field trials or controlled trials as cohort studies when the allocation procedure was not mentioned.

Vaccine effectiveness

We included 51 studies on MMR/MMRV effectiveness with the following study designs: 3 RCTs/CCTs, 31 cohorts, 14 case‐control, and 3 COEM. Two studies reported vaccine efficacy data against two diseases (measles and mumps) and were thus included in two different comparisons (ca‐La Torre 2017; ca‐Marolla 1998). We presented studies evaluating effectiveness in four main comparisons, as follows.

Measles: 17 studies included effectiveness data: 14 cohort studies, ca‐Arciuolo 2017; ca‐Arenz 2005; ca‐Barrabeig 2011a; ca‐Barrabeig 2011b; ca‐Bhuniya 2013; ca‐Choe 2017; ca‐Hales 2016; ca‐La Torre 2017; ca‐Marin 2006; ca‐Marolla 1998; ca‐Musa 2018; ca‐Ong 2007; ca‐Wichmann 2007; ca‐Woudenberg 2017, and 3 CCS (ba‐Defay 2013; ba‐Hungerford 2014; ba‐Jick 2010). See also Table 22 and Table 23.
Mumps: 21 studies included effectiveness data: 14 cohort studies, ca‐Chamot 1998; ca‐Compés‐Dea 2014; ca‐Greenland 2012; ca‐La Torre 2017; ca‐Livingston 2013; ca‐Lopez Hernandez 2000; ca‐Ma 2018; ca‐Marolla 1998; ca‐Nelson 2013; ca‐Ogbuanu 2012; ca‐Ong 2005; ca‐Schlegel 1999; ca‐Snijders 2012; ca‐Takla 2014, and 7 CCS (ba‐Castilla 2009; ba‐Fu 2013; ba‐Giovanetti 2002; ba‐Goncalves 1998; ba‐Harling 2005; ba‐Kim 2012; ba‐Mackenzie 2006). See also Table 24 and Table 25.
Rubella: 1 cohort study included effectiveness data (ca‐Chang 2015). See also Table 26.
Varicella: 14 studies included effectiveness data: 3 RCTs (aa‐Henry 2018; aa‐Povey 2019; aa‐Prymula 2014), 4 cohort studies (ca‐Giaquinto 2018; ca‐Rieck 2017; ca‐Spackova 2010; ca‐Tafuri 2013), 4 CCS (ba‐Andrade 2018; ba‐Cenoz 2013; ba‐Liese 2013; ba‐Vazquez 2001), and 3 COEM (ga‐Boccalini 2015; ga‐Pozza 2011; ga‐Tafuri 2015). See also Table 27, Table 28, Table 29, and Table 30.

1. Measles: effectiveness ‐ cohort studies.

Study	Population characteristics	Case definition	Vaccine/strain	N vaccinated sample size (dose)	N control	N events in exposed/ N total exposed or person‐time versus N events in non‐exposed/ N total non‐exposed or person‐time	Vaccine effectiveness  VE% (95% CI)
ca‐Barrabeig 2011b	Children attending day‐care and preschool centres (a) ≥ 15 months (all ages) (b) 15 to 23 months (c) 24 to 35 months (d) ≥ 36 months ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (e) Indirect effectiveness (e1) 12 to 23 months (e2) 24 to 35 months (e3) ≥ 36 months	Confirmed measles was defined as laboratory‐confirmed case or met the WHO clinical case definition and was epidemiologically linked to laboratory‐confirmed case.	Priorix/Schwarz or MDS/Enders dose 1 at 9 to 12 months dose 2 at 15 months	(a) N = 1027 (any dose) (a1) N = 830 (1 dose) (a2) N = 197 (2 doses) (b) N = 269 (any doses) (c) N = 384 (any doses) (d) N = 374 (any doses)	(a) n = 94 (b) n = 57 (c) n = 20 (d) n = 17 unvaccinated	(a) 5/1027 versus 12/94 (a1) 5/830 versus 12/94 (a2) 0/197 versus 12/94 (b) 3/296 versus 6/57 (c) 1/384 versus 4/20 (d) 1/374 versus 2/17	(a) 96.2% (89.4% to 98.6%) (a1) 95.3% (86.9% to 98.%) (a2) 100% (‐% to ‐%) (b) 89.4% (58.9% to 97.3%) (c) 98.7% (88.9% to 99.8%) (d) 97.7% (76.1% to 99.8%) VE = (1 − RR) x 100 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (e1) 71.1% (63.5% to 78.8%) (e2) 80.0% (56.3% to 94.3%) (e3) 88.2% (63.6% to 98.5%) VE = (ARU − ARV)/ARU x 100 Orenstein 1985
ca‐Bhuniya 2013	Children aged 9 to 59 months (at 30 June 2011) (a) 9 to 59 months (b) 9 to 12 months (c) > 12 months	A clinical case of measles is defined as fever with maculopapular rash and either conjunctivitis or cough or coryza (catarrhal inflammation of the mucous membrane in the nose). A confirmed case of measles is defined as a clinical case who is positive for anti‐measles virus nucleoprotein immunoglobulin M antibodies in serological tests but has not been vaccinated against measles during last 1 month.	MMR vaccine not described	(a) N = 50 (1 dose)	(a) N = 18	(a) 15/50 versus 16/18	(a) 66.3% (46.9% to 78.6%) (b) 66.6%() (c) 65.4%() (*) no statistical evidence VE = (1 − RR) x 100
ca‐Choe 2017	Outbreak at a university in 2014 Students born between 1984 and 1993. N = 14,465 VE > 10 years after vaccination	The definition of suspected measles case was individuals with following features: fever and rash and at least 1 of cough, coryza, or conjunctivitis. All suspected cases were quarantined and were interviewed using standardised questionnaire, and physical examinations were performed by trained physicians. Presence of symptoms (fever, rash, cough, coryza, or conjunctivitis), travel history, and days of illnesses were assessed.	MMR/not stated 2 doses	N = 11448	N = 3017	52/11448 versus 33/3017	60% (38.2% to 74.1%) VE = (1 − RR) x 100
ca‐La Torre 2017	N = 11,004 children born between 2008 and 2010 who underwent vaccination in 2009 to 2011. Follow‐up = 24 months	Hospitalisation for (a) measles (b) mumps (see also Table 24) (c) measles and mumps (d) all infectious diseases (e) all respiratory diseases The effectiveness of MMR vaccine in reducing hospitalisations for any infection was assessed by analysing 2 distinct databases (vaccination record) and (hospital discharge): Hospital discharge diagnosis which contained the following ICD‐9 codes in primary or secondary diagnosis: 001 to 139 for infectious and parasitic diseases; 460 to 519 for respiratory diseases	MMR not described the vaccination records of the database of the Roma Local Health Unit from which relevant data were extracted, such as date of birth; MMR vaccination (yes/no); MMR dose (only for vaccinated); personal tax code. The cohort was recomposed through record linkage of the 2 archives, registration and vaccination of hospital discharge records, using personal tax codes as a common identification in both archives.	(1) 1 dose N = 5392 (2) 2 doses N = 3310 (3) any dose N = 8702	Unvaccinated N = 2302	(a1) 3/5392 versus 9/2302 (a2) 0/3310 versus 9/2302 (a3) 3/8702 versus 9/2302 (b1) 1/5392 versus 1/2302 (b2) 0/3310 versus 1/2302 (b3) 1/8702 versus 1/2302 (c1) 4/5392 versus 10/2302 (c2) 0/3310 versus 10/2302 (c3) 4/8702 versus 10/2302 (d1) 82/5392 versus 262/2302 (d2) 70/3310 versus 262/2302 (d3) 414/8702 versus 262/2302 (e1) 202/5392 versus 424/2302 (e2) 183/3310 versus 424/2302 (e3) 809/8702 versus 424/2302	Unadjusted estimates (a1) 85.8% (47.5% to 96.1%) (a2) 96.3% (37.1% to 99.8%) (a3) 91.2% (67.5% to 97.6%) (b1) 57.3% (−582% to 97.3%)* (b2) 76.8% (−468% to 99.1%)* (b3) 73.5% (−322% to 98.3%)* (c1) 82.9% (45.6% to 94.6%) (c2) 96.7% (43.5% to 99.8%) (c3) 89.4% (66.3% to 96.7%) (d1) 86.6% (83% to 89.5%) (d2) 81.4% (75.9% to 85.6%) (d3) 84.7% (81.4% to 87.4%) (e1) 79.7% (76.1% to 82.7%) (e2) 70% (64.6% to 74.5%) (e3) 76% (72.6% to 78.9%) () no statistical evidence VE = (1 − RR) x 100 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Adjusted estimates any doses (a) 91% (68% to 99%) (b) not reported (c) 90% (66% to 97%) (d) 71% (66% to 75%) (e) 82% (52% to 93%) VE = (1 − HR)100
ca‐Marolla 1998	Children (19 to 67 months) whose parent required a paediatrician visit during a measles outbreak peak	Clinical diagnosis patient records and parent interviews	(a) Pluserix Schwarz (b) Morupar Schwarz (c) Triviraten Edmonston‐Zagreb vaccination records	(a) N = 329 (1 dose) (b) N = 747 (1 dose) (c) N = 1023 (1dose)	N = 646 unvaccinated	(a) 0/329 versus 114/646 (b) 2/747 versus 114/646 (c) 8/1023 versus 114/646 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (a) 0/ 19,836 PT (b) 2/ 12,906 PT (c) 8/ 31,329 PT (control) 114/22,188 PT = person‐time in months	(a) 100% (‐% to ‐%) (b) 97% (88% to 99%) (c) 95% (90% to 98%) VE = (ARU − ARV)/ARU x 100 Orenstein 1985
ca‐Musa 2018	Children aged up to 14 years. N = 2784 (children aged > 14 years, N = 2300). Data were presented by age group. The study included all students in 40 classes with 1 or more registered measles cases in the period February 2014 to September 2015. VE ≤ 5 years since vaccination 6 to 14 years since vaccination	Measles diagnosis was confirmed according to WHO guidelines. The clinical criteria for measles were fever, maculopapular rash (i.e. non‐vesicular rash), and cough or coryza (i.e. runny nose) or conjunctivitis (i.e. red eyes). The laboratory criteria for measles surveillance case confirmation were measles IgM antibody detection, or measles virus isolation, or measles viral RNA detection by RT‐PCR, or a significant rise in measles IgG antibody in paired sera. All suspected cases were investigated and classified based on clinical, laboratory, and epidemiological data, based on the WHO case definition.	MMR/not stated (a) 1 dose (b) 2 doses (c) ≤ 5 years since vaccination (d) 6 to 14 years since vaccination	(a) N = 100 (b) N = 606 (c) N = 20 (d) N = 76	N = 95	(a) 3/100 versus 35/95 (b) 6/606 versus 35/95 (c) 1/20 versus 35/95 (d) 2/76 versus 35/95	(a) 91.9% (74.4% to 97.4%) (b) 97.3% (93.8% to 98.8%) (c) 86.4% (6.6% to 98.0%) (d) 92.9% (71.2% to 98.2%) VE = (1 − RR) x 100
ca‐Ong 2007	Children from primary school in Singapore (aged 8 to 14 years, > 5 years since vaccination) during a measles outbreak	Clinical with laboratory confirmation. Active survey and serological confirmation	MMR vaccine not described Vaccination status was ascertained from health booklet.	N = 171 (1 dose)	N = 13 unvaccinated	2/171 versus 7 /13	97.8% (90.6% to 99.5%) VE = (1 − RR) x 100
ca‐Wichmann 2007	School outbreak 2006. Students aged 10 to 15 years (N = 875) 16 to 21 years (N = 139) VE < 10 years after vaccination > 10 years after vaccination	Clinical or laboratory	MMR/not stated (a) 1 dose (b) 2 doses (c) unknown vaccination status ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐	All ages (a) N = 199 (b) N = 561 (c) N = 218 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ 10 to 15 years (a) N =196 (b) N = 502 (c) N = 144 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ 16 to 21 years (a) N = 3 (b) N = 59 (c) N = 74	All ages N = 36 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ 10 to 15 years N = 33 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ 16 to 21 years N = 3	All ages (a) 2/199 versus 19/36 (b) 2/5611 versus 19/36 (c) 30/218 versus 19/36 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ 10 to 15 years (a) 2/196 versus 18/33 (b) 2/502 versus 18/33 (c) 25/144 versus 18/33 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ 16 to 21 years (a) 0/3 versus 1/3 (b) 0/59 versus 1/3 (c) 5/74 versus 1/3	All ages (a) 98.1% (92.2% to 99.5%) (b) 99.3% (97.2% to 99.8%) (c) 73.9% (59.0% to 83.4%) VE = (1 − RR) x 100 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ 10 to 15 years (a) 98.1% (92.3% to 99.5%) (b) 99.3% (97.0% to 99.8%) (c) 68.2% (48.9% to 80.2%) ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ 16 to 21 years (a) 66.7% () (b) 97.8% (53.7% to 99.9%) (c) 79.7% () VE = (1 − RR) x 100 (*) no evidence
ca‐Woudenberg 2017	Infants aged 6 to 14 months living in municipalities where coverage with the first dose of MMR vaccine was < 90%. Infants aged 6 to 11 months were offered an extra vaccination (and would thus still be eligible for their second MMR vaccination at the age of 14 months). Infants aged 12 to 14 months were offered an early MMR vaccination as an alternative to the regular time point at 14 months of age. All infants were eligible for another dose of MMR scheduled at 9 years of age.	Laboratory‐confirmed measles N = 1080 infants eligible for analysis laboratory‐confirmed	MMR vaccine: (M‐M‐RVAXPRO; Sanofi Pasteur MSD). This vaccine contains measles virus Enders’ Edmonston strain. Vaccination status was checked in the national vaccination register. Parents were asked whether their infant(s) had had measles in the preceding 3 months.	N = 919	N = 311	3/106,631 (PT‐days) versus 10/23,769 (PT‐days)	HR (95% CI)() 0.29 (0.05 to 1.72) () adjusted estimates Cox proportional hazard model VE = 1 − HR
ca‐Arenz 2005	Household contacts 55 families, 43 children (a) 1 dose (b) 2 doses (c) any dose	Clinical	MMR/strain not stated	(a) N = 13 (b) N = 4	N = 26	(a) 1/13 versus 19/26 (b) 0/4 versus 19/26 (c) 1/20 versus 19/26	(a) 96.9% (71.8% to 99.7%) (b) 95.7% (10.6% to 99.8%) (c) 97.7% (79.3% to 99.7%) VE = (1 − RR) x 100 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (a) 90% (35% to 97%) (b) not reported (c) 92% (48% to 98%) VE = (ARU − ARV)/ARU x 100 Orenstein 1985
ca‐Hales 2016	Household contacts adolescents and young adults (10 to 29 years) (a) any dose (b) 1 dose (c) 2 doses (d) 3 doses	Clinical or laboratory confirmation, or both	MMR vaccine not described	(a) N = 302 (b) N = 27 (c) N = 205 (d) N = 70	(a) N = 16	Pre‐campaign MMR doses (a) 16/302 versus 2/16 (b) 3/27 versus 2/16 (c) 13/205 versus 2/16 (d) 0/70 versus 2/16 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐	Pre‐campaign MMR doses (a) (No data) (b) 23.1% (−425.0% to 87.3%)* (c) 63.4% (−103.0% to 90.6%)* (d) 95.9% (45% to 100%) ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Campaign MMR doses: 78.7% (10.1% to 97.7%) for pre‐exposure doses ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ 50.4% () for postexposure doses () no statistical evidence ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ VE = (1 − OR) x 100 from logistic regression
ca‐Marin 2006	Household contacts (6 months to 14 years) of primary measles cases	Secondary cases Clinical (WHO definition) or IgM positive antibody of secondary cases Standardised questionnaires	MMR vaccine not described Vaccination records	(a1) N = 48 (1 dose) (a2) N = 106 (2 doses) (b) N = 44 (> 2 doses) (c) N = 219 any doses contacts	N = 21 unvaccinated	(a1) 2/48 versus 11/21 (a2) 3/106 versus 11/21 (b) 1/44 versus 11/21 (c) 17/219 versus 11/21	(a1) 92.0% (67.2% to 98.1%) (a2) 94.6% (82.3% to 98.4%) (b) 95.7% (68.6% to 99.4%) (c) 85.2% (72.7% to 92.0%) VE = (1 − RR) x 100
ca‐Arciuolo 2017	Postexposure prophylaxis Childrena aged < 19 years N = 208	All who subsequently developed measles were considered as contacts.	MMR not described MMR PEP administered within 72 hours of initial exposure.	N = 44	N = 164	(a) 2/44 versus 45/164	(a) 83.4% (34.4% to 95.8%) VE = (1 − RR) x 100
ca‐Barrabeig 2011a	Postexposure prophylaxis N = 166 children with median age of 16.5 months (range 6 to 47 months) Candidates for the intervention were susceptible contacts who had not received either measles‐containing vaccine or had not suffered measles.	Clinical and laboratory	MMR not stated (a) at least 1 dose (b) vaccinated ≤ 3 days (c) vaccinated 4 to 5 days (d) vaccinated 6 to 7 days (e) vaccinated 8 to 9 days (f) vaccinated 10 to 12 days	(a) N = 54 (b) N = 17 (c) N = 14 (d) N = 14 (e) N = 8 (f) N = 1	N = 21	(a) 12/54 versus 13/21 (b) 1/17 versus 13/21 (c) 4/14 versus 13/21 (d) 5/14 versus 13/21 (e) 1/8 versus 13/21 (f) 1/1 versus 13/21	(a) 64.1% (34.5% to 80.3%) (b) 90.5% (34.5% to 98.6%) (c) 53.8% (0.0% to 81.1%) (d) 42.3% (0.0% to 81.1%) (e) 79.8% (0.0% to 73.5%) (f) not reported VE = (1 − RR) x 100

Study	Population characteristics	Case definition	Controls/ selection	MMR strain/exposure	N cases vaccinated/N cases versus N controls vaccinated/N controls	OR (95% CI)	VE% (95% CI)
ba‐Defay 2013	Children aged 5 to 17 years (a) outside of outbreak school (b) all participants	(a) N = 61 (b) N = 102 confirmed by laboratory testing or epidemiologic link is notifiables by both physicians and laboratories in Quebec	(a) N = 305 (b) N = 510 Controls were matched for date of birth (± 6 months) and school attended in 2010 to 2011.	MMR‐II (Merck Canada, Montreal, Quebec) Cases and controls received 2 doses of measles‐containing vaccine.	No data reported amongst unvaccinated.	‐	‐
ba‐Hungerford 2014	Participants (median age 16 years, upper quartile age 76 years) living in Merseyside (UK)	N = 42 microbiological confirmation: oral fluid/blood test IgM positive or PCR positive	N = 42 Control group participants were selected at random, matched 1:1 by general medical practice and aged within 1 year.	MMR vaccine not described (a) vaccinated appropriately for age (b) under age for vaccination (< 14 months) (c) all ‐ vaccinated Unvaccinated: incompletely or partially vaccinated for age (> 13 months)	(a) 5/27 versus 23/29 (b) 15/37 versus 12/18 (c) 20/42 versus 35/42	Risk factors for measles infection (univariate analysis) age > 13 months and incomplete vaccination 6.3 (1.9 to 33.4) ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (Multivariate analysis) under age for routine vaccination 20.4 (2.0 to 300) incomplete/partial vaccination for age > 13 months 22.1 (3.8 to 300) (**) adjusted for confounders	Risk factors for measles infection (univariate analysis) age > 13 months and incomplete vaccination 84.1% (47.4% to 97.0%) ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (Multivariate analysis) under age for routine vaccination 95.1% (50.0% to 100%) incomplete/partial vaccination for age > 13 months 95.5% (73.7% to 100%) (**) adjusted for confounders VE = (1 − OR) x 100
ba‐Jick 2010	Participants aged 1 to 19 years	N = 1261 clinical definition	N = 4996 randomly selected, matched for year of birth, gender, general practice attended, index date	MMR or MR not described (a) 1 dose (b) > 1 dose	(a) 409/1221 versus 2012/4750 (b) 40/852 versus 246/2984	(a) 0.49 (0.41 to 0.58)* (b) 0.39 (0.26 to 0.58)* *adjusted estimates, conditional logistic regression	(a) 51.0% (42.0% to 59.0%) (b) 61.0% (42.0% to 74.0%) VE = (1 − OR) x 100

Study	Population characteristics	Case definition	Controls/selection	MMR strain/exposure	N cases vaccinated/ N cases versus N controls vaccinated/ N controls	OR (95% CI)	VE% (95% CI)
ba‐Castilla 2009	Children aged between 15 months and 10 years from Navarre region (Northern Spain) at the time a mumps outbreak occurred (between August 2006 and June 2008)	(a) N = 181 (b) N = 72 (c) N = 241 Laboratory or epidemiological confirmation of clinical cases: swelling of 1 of more salivary glands for at least 2 days with either laboratory (PCR or IgM positive) or epidemiological confirmation (i.e. epidemiological relation with other laboratory confirmed or clinical mumps cases). Obtained from cases notified to the regional health authority	(a) N = 875 (b) N = 353 (c) N = 1205 matched for sex, municipality, district of residence, and paediatrician	(a) 1 dose (b) 2 doses (c) any dose MMR/Jeryl Lynn doses received at least 30 days before symptom disease onset. Blinded review of primary care vaccination registry	(a) 169/181 versus 852/875 (b) 59/72 versus 330/353 (c) 228/241 versus 1182/1205	‐	(a) 66% (25% to 85%) (b) 83% (54% to 94%) (c) 72% (39% to 87%) adjusted for confounders
ba‐Fu 2013	Children in Guangzhou aged 8 months to 12 years during 2006 to 2012	N = 1983 randomly selected clinical definition	N = 1983 matched 1:1 by birth date, gender, residence not reported breakdown by type of vaccine administrated	(a) MMR/Jeryl Lynn RIT4385 (b) measles‐mumps (c) missing (vaccine type) (d) any vaccine 1 dose	(a) 112 versus 145 (b) 242 versus 261 (c) 620 versus 837 (d) 974/1983 versus 1243/1983	(a) OR extracted from VE reported 0.49 (0.26 to 0.93)	(a) 51.3% (7.2% to 95.0%)
ba‐Giovanetti 2002	Children and adolescents aged 14 months to 15 years from urban area of Alba and Bra and 10 rural towns (n = 12,800 residents from 0 to 15 years) during 2000 to 2001 epidemic	Clinical diagnosis (cases notified by national infectious diseases surveillance system) N = 139 notified mumps cases	N = 139 randomly selected from immunisation registry, matched for birth year and address. (controls received at least 1 MMR dose)	MMR vaccine not specified. Vaccination registry and phone interviews, immunisation should have been received at least 30 days before disease onset.	90/139 versus 111/139	0.46 (0.27 to 0.80)	53.7% (20.4% to 73.0%)
ba‐Goncalves 1998	Children and adolescents (15 months to 16 years) from Oporto (Portugal)	Clinical diagnosis Cases reported by GPs or hospital doctors, occurred during the 1995 to 1996 mumps outbreak (a) N = 73 (b) N = 133 (c) N = 189	2 consecutive vaccination records of the same sex, month and birth year as the case were selected. (a) N = 169 (b) N = 236 (c) N = 378 Controls received at least 1 MMR dose.	Assuming that before 1 November 1992 MMR mumps Urabe strain was administered, subsequently the Rubini strain (a) Urabe (b) Rubini (c) all at least 1 MMR dose	(a) 56/73 versus 142/169 (b) 116/133 versus 209/236 (c) 172/189 versus 351/378	‐	(a) 70% (25% to 88%) (b) 1% (−108% to 53%) adjusted for confounders
ba‐Harling 2005	Children and adolescents aged between 1 and 18 years from religious community in Northeast London. Mumps outbreak	Clinical diagnosis N = 156 (GP notification to the local CCDC, mumps diagnoses from electronic practice list, verbal reports by community members) ‐‐‐‐‐‐‐‐‐‐‐‐ Laboratory confirmation of clinical diagnosis N = 43 GP notification to the local CCDC of notified cases, IgM and mumps RNA testing was offered	N = 175 randomly selected and stratified for age and sex from practice list	Jeryl Lynn 1 or 2 MMR doses received at least 1 month before index date (a) at least 1 dose (b) 1 dose (c) 2 doses	79/156 versus 134/175	(a) 0.31 (0.20 to 0.50)	(a) 69% (50% to 80%) (crude) (a) 69% (41% to 84%) adjusted for age, sex, practice ‐‐‐‐‐‐‐‐‐‐ Laboratory‐confirmed cases (a) 65% (25% to 84%) (b) 64% (40% to 78%) (c) 88% (62% to 96%) All adjusted for age, sex, practice. Proportion of vaccinated in cases and controls not provided.
ba‐Kim 2012	Children (a) prospective case‐control study from March 2010 to October 2011 (b) retrospective case‐control study 2008 to 2009 in western Seoul, Incheon, and Goyang (c) total	(a) N = 55 (a1) 1 dose (a2) 2 doses (a3) any dose ‐‐‐‐‐‐‐‐‐‐‐‐ (b) N = 122 (b1) 1 dose (b2) 2 doses (b3) any dose ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (c) N = 177 (c1) 1 dose (c2) 2 doses (c3) any dose	(a) N = 165 (a1) 1 dose (a2) 2 doses (a3) any dose ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (b) N = 449 (b1) 1 dose (b2) 2 doses (b3) any dose ‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (c) N = 614 (c1) 1 dose (c2) 2 doses (c3) any dose	MMR vaccine not described (assumed to be Jeryl Lynn following Park 2015) For (a) and (b): data about demographic characteristics and MMR vaccination status were collected from cases and controls.		‐‐‐‐‐‐‐‐‐‐(a)‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (a1) 0.58 (0.05 to 6.90) (a2) 1.1 (0.09 to 13.3) (a3) 0.67 (0.06 to 7.35) ‐‐‐‐‐‐‐‐‐‐(b)‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (b1) 0.33 (0.02 to 5.33) (b2) 0.11 (0.01 to 2.12) (b3) 0.33 (0.02 to 5.33) ‐‐‐‐‐‐‐‐‐(c)‐‐‐‐‐‐‐‐‐‐‐ (c1) 0.58 (0.10 to 3.56) (c2) 0.42 (0.06 to 2.81) (c3) 0.50 (0.08 to 2.99)	‐‐‐‐‐‐‐‐(a)‐‐‐‐‐‐‐‐‐ (a1) 42.0%* (a2) −10.0%* (a3) 33.0%* ‐‐‐‐‐‐‐‐‐(b)‐‐‐‐‐‐‐‐‐‐‐ (b1) 67.0%* (b2) 89.0%* (b3) 67.0%* ‐‐‐‐‐‐‐‐‐(c)‐‐‐‐‐‐‐‐‐‐‐ (c1) 42.0%* (c2) 58.0%* (c3) 50.0%* *no statistical evidence
ba‐Mackenzie 2006	About 600 pupils attending a boarding school in Scotland during a mumps outbreak that peaked between October and November 2004	Virological confirmation of clinical diagnosis N = 20 (aged 13 to 17 years). Cases notified to consultant in public health medicine. Acute cases with virological positive test	N = 40 matched for age, sex, residential status, UK or international students	MMR vaccine not described (a) 1 dose (b) 2 doses (c) any dose Not specified. Pre‐outbreak vaccination status obtained by medical notes held in the school, communication with parents, and from Scottish Immunisation Recall System.	(a) 9/18 versus 20/34 (b) 2/11 versus 6/20 (c) 11/20 versus 26/40	(a) 0.7 (0.22 to 2.21) (b) 0.52 (0.09 to 3.16) (c) 0.66 (0.22 to 1.97)	(a) 30.0%* (b) 48.1%* (c) 32.4%* *no statistical evidence

Study ID and design	Population enrolled	Outcome	Vaccine arms n = sample size	Comparator arm n = sample size	Vaccine arm events/n	Comparator arm events/n	VE% (95% CI)
aa‐Prymula 2014 RCT	This study is the first phase (1 September 2005 to 29 June 2009) of an RCT. The study was done in 111 study centres in Europe: Czech Republic (22), Greece (11), Italy (9), Lithuania (9), Norway (5), Poland (10), Romania (9), Russia (14), Slovakia (17), and Sweden (5). An eligible participant was a healthy child aged 12 to 22 months at the time of the first vaccination; had a negative history of varicella, mumps, measles, and rubella diseases and vaccinations; and was one of the following: (1) at home with at least 1 sibling (with negative history of varicella disease and vaccination), (2) attending a child minder (where at least 1 child was without a known positive history of varicella disease and vaccination), (3) playing for more than 5 min weekly with children without a known positive history of varicella disease and vaccination, (4) registered to attend a day‐care centre from 24 months of age. An eligible participant’s parents or guardians had direct access to a telephone and were deemed by the investigator of being capable of complying with the requirements of the trial protocol.	The primary efficacy endpoint was the occurrence of confirmed varicella from 42 days after the second vaccine dose to the end of the first phase of the trial. The secondary efficacy endpoint was the occurrence of confirmed varicella graded by severity over the same time period. Varicella cases (a) All (b) Moderate/severe Follow‐up = 3 years	MMRV group: 2 doses of MMRV (Priorix‐Tetra, GSK) N = 2279 MMR+V group: 1 dose MMR (Priorix, GSK) and monovalent varicella vaccine (Varilrix, GSK) at dose 2 N = 2263	MMR group (control): 2 doses of MMR (Priorix, GSK) N = 743	MMRV (a) 37/2279 (b) 2/2279 MMR+V (a) 243/2263 (b) 37/2263	MMR (a) 201/743 (b) 117/743	MMRV (a) 94.9% (92.4% to 96.6%) (b) 99.5% (97.5% to 99.9%) MMR+V (a) 65.4% (57.2% to 72.1%) (b) 90.7% (85.9% to 93.9%) VE = (1 − HR) x 100
aa‐Henry 2018 RCT linked to aa‐Prymula 2014	Healthy children aged 12 to 22 months. n = 5803 children enrolled and vaccinated (TVC) in phase A, n = 4580 in the TVC in phase B, n = 3829 completed the study up to Year 6; n = 5289 ATP cohort for efficacy in phase A + B, n = 3791 in the ATP cohort for efficacy in phase B	Varicella cases (a) All (b) Moderate/severe (c) Severe Follow‐up = 6 years	ATP cohort for efficacy phase A + B MMRV n = 2279 MMR+V n = 2266 Phase B MMRV n = 1802 MMR+V n = 1593 MMRV group 2 doses of MMRV (Priorix‐Tetra, GSK) at Day 0 and Day 42 MMR+V group 1 dose of MMR (Priorix, GSK) at Day 0 and 1 dose of monovalent varicella vaccine (Varilrix, GSK) at Day 42	ATP cohort for efficacy phase A + B MMR n = 744 Phase B MMR n = 396 MMR group 2 doses of the MMR (Priorix, GSK) vaccine at Day 0 and Day 42	Phase A + B MMRV (a) 71/2279 (b) 6/2279 (c) 0/2270 MMR+V (a) 419/2266 (b) 58/2266 (c) 1/2266 Phase B MMRV (a) 33/1800 (b) 4/1800 (c) 0/1800 MMR+V (a) 176/1592 (b) 18/1592 (c) 0/1592	Phase A + B MMR (a) 325 /744 (b) not reported Phase B (a) 125/396 (b) not reported	Phase A + B MMRV (a) 95.0% (93.6% to 96.2%) (b) 99.0% (97.7% to 99.6%) (c) undefined MMR+V (a) 67.0% (61.8% to 71.4%) (b) 90.3% (86.9% to 92.8%) (c) 94.6% (55.3% to 99.4%) Phase B MMRV (a) 95.3% (93.1% to 96.8%) (b) 98.4% (95.5% to 99.4%) (c) undefined MMR+V (a) 69.5% (61.5% to 75.8%) (b) 91.8% (85.9% to 95.2%) (c) undefined VE = (1 − HR) x 100
aa‐Povey 2019 RCT linked to aa‐Prymula 2014	Children aged 12 to 22 months were eligible for inclusion if: had not received MMR or varicella vaccines, or both, or had measles‐mumps‐rubella or varicella zoster or herpes zoster diseases, or both, and were at home with at least 1 sibling with negative history of varicella disease and vaccination, at a child‐minders where at least 1 child was without a known positive history of varicella disease and vaccination, playing for more than 5 min/week with children without a known positive history of varicella disease and vaccination, or registered to attend day care from 24 months.	Varicella cases (a) All (b) Moderate/Severe Follow‐up = 10 years	Phase A + B MMRV n = 2279 MMR+V n = 2266 Phase B MMRV n = 1800 MMR+V n = 1591 MMRV group 2 doses of MMRV (Priorix‐Tetra, GSK) at Day 0 and Day 42 MMR+V group 1 dose of MMR (Priorix, GSK) at Day 0 and 1 dose of monovalent varicella vaccine (Varilrix, GSK) at Day 42	Phase A + B MMR n = 744 Phase B MMR n = 396 MMR group 2 doses of the MMR (Priorix, GSK) vaccine at Day 0 and Day 42	Phase A + B MMRV (a) 71/2279 (b) 6/2279 MMR+V (a) 469/2266 (b) 67/2266 Phase B MMRV (a) 33/1800 (b) 4/1800 MMR+V (a) 176/1592 (b) 18/1592	Phase A + B MMR (a) 352/744 (b) 176/744 Phase B (a) 149/396 (b) 59/396	Phase A + B MMRV (a) 95.4% (94.0% to 96.4%) (b) 99.1% (97.7% to 99.6%) MMR+V (a) 67.2% (62.3% to 71.5%) (b) 89.5% (86.1% to 92.1%) Phase B MMRV (a) 95.9% (94.1% to 97.1%) (b) 98.7% (96.4% to 99.5%) MMR+V (a) 69.8% (62.8% to 75.5%) (b) 90.0% (84.2% to 93.7%) VE = (1 − HR) x 100

Study ID and design	Population enrolled	Vaccine arm n = sample size	Comparator arm n = sample size	Outcome	MMR vaccine arm events/n	Other vaccine arms events/n	Comparator arm events/n
ab‐Bloom 1975; RCT	Children aged 11 months to 4 years Observation period 21 days	MMR vaccine Measles Schwarz Mumps Jeryl Lynn Rubella Cendehill n = 183 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Temperature above normal sample size n = 160 Normal temperature rectal 99.6 °F (37.5 °C) (163 children) Oral 98.6 °F (37 °C) (6 children) Axillary 97.6 °F (36.4 °C) (26 children)	Placebo n = 40 ‐‐‐‐‐‐‐‐‐‐ Temperature sample size n = 35	Reactions (a) Rash (b) Lymphadenopathy (c) Coryza (d) Rhinitis (e) Cough (f) Other total ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Temperature above normal (a) 1.5 to 2.4 °F (b) 2.5 to 3.4 °F (c) 3.5 to 4.4 °F (d) 4.5 to 4.9 °F (e) ≥ (normal + 1.5) °F	MMR vaccine (a) 22/183 (b) 2/183 (c) 4/183 (d) 2/183 (e) 5/183 (f) 35/183 total 70/183 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Temperature above normal (a) 17/160 (b) 1/160 (c) 5/160 (d) 2/160 (e) 25/160	‐	Placebo arm (a) 2/40 (b) 1/40 (c) 4/40 (d) 4/40 (e) 1/40 (f) 8/40 total 20/40 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Temperature above normal (a) 2/35 (b) 2/35 (c) 0/35 (d) 0/35 (e) 4/35
ab‐Ceyhan 2001; CCT	Infants aged 38 to 40 months Observation period 28 days	Arm A: n = 442 (1) MV/Rouvax Measles Schwarz at 9 months; and (2) MMR/Trimovax Measles Schwarz Mumps Urabe AM9 Rubella Wistar RA 27/3 at 15 months ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Arm B: n = 495 (3) MMR/Trimovax Measles Schwarz Mumps Urabe AM9 Rubella Wistar RA 27/3 at 12 months	No placebo arm	Systemic reactions (a) Fever (b) Runny nose (c) Cough (d) Rash (e) Diarrhoea ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Local (f) Redness (g) Swelling ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Total events (x) Fever (y) Systemic (z) Local	MMR vaccine (2)15 months; (3)12 months (a) 40/442; 55/495 (b) 7/442; 22/495 (c) 36/442; 34/495 (d) 16/442; 19/495 (e) 2/442; 5/495 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Local (2)15 months; (3)12 months (f) 14/442; 19/495 (g) 2/442; 3/495 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Total events (2)15 months; (3)12 months (x) 40/442; 55/495 (y) 61/442; 80/495 (z) 16/442; 22/495	MV vaccine (1) 9 months (a) 38/442 (b) 19/442 (c) 28/442 (d) 2/442 (e) 5/442 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Local (f) 7/442 (g) 2/442 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Total events (x) 38/442 (y) 54/442 (z) 9/442
ab‐Edees 1991; RCT	Children aged 12 to 18 months. Observation period 21 days	Arm A: n = 196 MV/Rouvax Measles Schwarz ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Arm B: n = 198 MMR/Trimovax Measles Schwarz Mumps Urabe AM9 Rubella Wistar RA 27/3	No placebo arm	Local symptoms  (a) Erythema (b) Induration (c) Pain ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Specific systemic  (a) Rash (b) Parotitis (c) Conjuntivitis (d) Testicular swelling (e) Arthralgia (f) Arthritis (g) Convulsion ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Non‐specific systemic  (a) Fever (b) Adenopathy (c) Nasopharyngeal disorders (d) Gastrointestinal disorders (e) Restlessness Restlessness: used to describe a non‐specifically unwell child; it covers terms such as irritable miserable tearful clingy not sleeping.	MMR vaccine (Arm B) Local (a) 18/198 (b) 1/198 (c) 9/198 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Specific systemic (a) 87/198 (b) 5/198 (c) 17/198 (d) 0/198 (e) 0/198 (f) 0/198 (g) 0/198 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Non‐specific systemic (a) 76/198 (b) 2/198 (c) 113/198 (d) 83/198 (e) 124/198	MV vaccine (Arm A) Local (a) 16/196 (b) 0/196 (c) 14/196 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Specific systemic (a) 100/196 (b) 0/196 (c) 21/196 (d) 0/196 (e) 0/196 (f) 0/196 (g) 0/196 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Non‐specific systemic (a) 74/196 (b) 3/196 (c) 115/196 (d) 74/196 (e) 147/196
ab‐Lerman 1981; RCT	Children aged 15 months to 5 years Observation period 42 days	Arm(1): n = 43: Measles (MSD) Arm(2): n = 41: Mumps (MSD) Jeryl Lynn Arm(3): n = 47: Rubella HPV‐77:CE‐5 Arm(4): n = 142 MMR (MSD) with Rubella HPV‐77:DE‐5 Arm(5): n = 46: Rubella/Wistar RA27/3 Arm(6): n = 141: MMRII (MSD) with Rubella Wistar RA27/3	Placebo arm n = 42 (vaccine diluent) 1 dose subcutaneously	Reactions (a) Local reaction (b) Fever 101 to 102.9 °F (fever 38.3 to 39.4 °C) (c) Fever 103 to 104.9 °F (fever 39.4 to 40.5 °C) (d) Respiratory symptoms (e) Rash (f) Lymphadenopathy (g) Sore eyes (h) Joint symptoms	MMR vaccine  Arms: (4); (6) (a) 7/142; 11/141 (b) 31/142; 35/141 (c) 11/142; 16/141 (d) 97/142; 102/141 (e) 24/142; 28/141 (f) 6/142; 11/141 (g) 24/142; 23/141 (h) 1/142; 1/141	Other vaccine arms: (1); (2); (3); (5) (a) 1/43; 6/41; 3/47; 2/46 (b) 12/43; 6/41; 6/47; 11/46 (c) 2/43; 3/41; 3/47; 2/46 (d) 34/43; 26/41; 31/47; 31/46 (e) 5/43; 1/41; 6/47; 5/46 (f) 1/43; 2/41; 2/47; 2/46 (g) 6/43; 8/41; 8/47; 8/46 (h) 0/43; 0/41; 0/47; 0/46	Placebo arm (a) 3/42 (b) 10/42 (c) 0/42 (d) 31/42 (e) 4/42 (f) 0/42 (g) 4/42 (h) 0/42
ab‐Peltola 1986; RCT	Pairs of twins aged (a) 14 to 18 months (first dose) (b) 6 years (second dose) Observation period 21 days	MMR vaccine Vivirac (MSD) 2 doses n = 581	Placebo arm n = 581		No data available for quantitative synthesis
ab‐Schwarz 1975; RCT	Children aged 10 months to 8 years Observation period 21 days	MMR vaccine Measles Schwarz Mumps Jeryl Lynn Rubella Cendehill n = 403	Placebo arm n = 205	Temperature (1) Axillary (2) Rectal ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (a) < 37.0 °C (b) 37.0 to 37.4 °C (c) < 37.5 °C (d) 37.5 to 37.9 °C (e) 38.0 to 38.4 °C (f) 38.5 to 38.9 °C (g) 39.0 to 39.4 °C (h) 39.5 to 39.9 °C (i) 40.0 to 40.4 C° ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Reactions (s1) Rash (s2) Lymphadenopathy (s3) Conjunctivitis (s4) Otitis media (s5) Coryza (s6) Rhinitis (s7) Pharyngitis (s8) Cough (s9) Headache (s10) Parotitis (s11) Orchitis (s12) Arthralgia (s13) Paraesthesia	MMR vaccine (1) Temperature axillary (a) 56/244 (b) 154/244 (c) 210/244 (d) 21/244 (e) 6/244 (f) 2/244 (g) 3/244 (h) 2/244 (i) 0/244 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (2) Temperature rectal (a) not reported (b) not reported (c) 48/142 (d) 51/142 (e) 30/142 (f) 8/142 (g) 1/142 (h) 1/142 (i) 3/142 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Reactions (s1) 36/403 (s2) 4/403 (s3) 8/403 (s4) 4/403 (s5) 8/403 (s6) 69/403 (s7) 2/403 (s8) 7/403 (s9) 1/403 (s10) 0/403 (s11) 0/403 (s12) 1/403 (s13) 0/403		Placebo arm (1) Axillary temperature (a) 32/176 (b) 132/176 (c) 164/176 (d) 9/176 (e) 2/176 (f) 1/176 (g) 0/176 (h) 0/176 (i) 0/176 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (2) Rectal temperature (a) Not reported (b) Not reported (c) 6/28 (d) 13/28 (e) 6/28 (f) 1/28 (g) 2/28 (h) 0/28 (i) 0/28 ‐‐‐‐‐‐‐‐‐‐‐‐‐‐ Reactions (s1) 9/205 (s2) 4/205 (s3) 5/205 (s4) 1/205 (s5) 5/205 (s6) 59/205 (s7) 2/205 (s8) 1/205 (s9) 1/205 (s10) 0/205 (s11) 0/205 (s12) 0/205 (s13) 0/205
ab‐Freeman 1993; Cluster‐RCT	Children aged 13 to 15 months Observation period 30 days	MMR vaccine MMRII (MSD) n = 253	No placebo arm	Reactions (a) Lymphadenopathy (b) Nasal discharge (c) Rash (d) Otitis media (e) Conjunctival abnormality (f) Abnormal tonsils	Reactions (a) 57/240 (b) 15/240 (c) 11/240 (d) 8/240 (e) 8/240 (f) 2/240

Study ID and design	Population	Outcome definition	Exposure MMR/MMRV vaccine	Findings	Crude data	Estimate (95% CI)
bb‐Ray 2006 Case‐control	Cases: (n = 452) children aged 0 to 6 years with outcome of interest. Controls: (n = 1280) matching for HMO, location, age within 7 days, sex, and length of enrolment in health plan	1. Encephalopathy: acute generalised disturbance of brain function requiring hospitalisation and consisting of coma or stupor that cannot be attributed to medication or postictal state. Such cases must have altered consciousness, delirium, obtundation and/or confusion. 2. Reyes syndrome: clinical symptoms of acute encephalopathy with altered level of consciousness as well as: absence of inflammatory changes in cerebrospinal fluid as indicated by 5 white blood cells/mm³ or brain histology showing cerebral oedema without perivascular or meningeal inflammation, plus evidence of hepatitis or liver failure documented by a 3‐fold or greater elevation in serum glutamic oxaloacetic transaminase, serum glutamate pyruvate transaminase or serum ammonia or fatty changes of hepatocytes on liver biopsy or autopsy, plus absence of other aetiologies for cerebral or hepatic abnormalities. 3. Encephalitis/encephalomyelitis: evidence of acute neurologic disease presenting with non‐specific signs such as fever, seizures, altered consciousness, headache, vomiting, meningismus, or anorexia. Multifocal involvement of the central nervous system and evidence of cerebrospinal fluid inflammation (7 white blood cells/mm³) were required. Diseases with other known aetiologies were excluded. For data analysis, all cases were stratified on the basis of their aetiology: known, unknown, suspected but unconfirmed (this last when a diagnosis was not confirmed by a diagnostic test). Hospitalisation cases for encephalopathy, Reyes syndrome, or encephalitis (primary or secondary diagnosis) in children aged 0 to 6 years, members of the health plan of 4 HMOs in the USA, and occurred between 1 January 1981 and 31 December 1995, were considered as possible cases. Hospital charts were reviewed by abstracter (not blind to vaccination status of the cases) who included in first instance encephalitis diagnoses by a neurologist with clear aetiology and excluded all cases with a condition other than encephalopathy. All other neurologic cases were reviewed by a neurologist (blind to vaccination status of the cases) and included as cases if they met case definition (see column on the right).	Vaccine exposure time interval relative to onset of encephalopathy (a) 7 to 14 days (b) 0 to 14 days (c) 0 to 30 days (d) 0 to 60 days (e) 0 to 90 days MMR type not reported. Vaccination status of both cases and controls was ascertained from medical records.	The findings do not support a conclusion that there is an increased risk of encephalitis or encephalopathy after MMR vaccination. Although this study is large, encephalopathy is rare and thus it is not possible to exclude completely a small increase in the risk of encephalopathy after MMR vaccination. However, if such an increased risk exists, the absolute risk is extremely small and it is much lower after vaccination than after measles. This corresponds roughly to an all‐cause incidence (not an attributable risk) of 1 in 200,000 after MMR, a rate that is not statistically different from background. Consequently, our results support the continued use of DTP and MMR vaccines.	N cases vaccinated/ N cases versus N controls vaccinated/ N controls (a) 1/452 versus 6/1280 (b) 1/452 versus 7/1280 (c) 4/452 versus 13/1280 (d) 8/452 versus 33/1280 (e) 15/452 versus 44/1280	OR (95% CI) ‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐‐ (a) 0.40 (0.05 to 3.46) (b) 0.35 (0.04 to 2.95) (c) 0.85 (0.27 to 2.68) (d) 0.64 (0.27 to 1.50) (e) 0.98 (0.47 to 2.01) adjusted estimates
db‐Makela 2002 Person‐time cohort	Children immunised aged 1 to 7 years old. Between November 1982 and September 1986 n = 535,544 n = 119 children hospitalised for encephalitis (MMR vaccine was administered before the disease), and only 97 between 0 and 24 months after MMR vaccination.	Encephalitis: acute or subacute onset of neurologic symptoms. Presence of neurologic symptoms or findings (clinical or laboratory, e.g. microbiological, electroencephalographic, computed tomographic) indicative of involvement of the brain parenchyma, such as coma, seizures, focal neurologic findings, or mental function impairment. Absence of evidence of other diagnoses, including non‐inflammatory conditions, and no microbiological or other laboratory findings suggestive of a non‐viral infection. When pleocytosis in CSF is present, the term encephalitis is used, implying an inflammatory response within the brain. The presence of normal CSF findings does not preclude the diagnosis if the other criteria are satisfied. Encephalopathy: clinically resembles encephalitis but no inflammatory response is evident. Chronic encephalopathy: persistence of acute findings usually over several months. The National Hospital Discharge Register was consulted by using the following ICD‐8 codes: 065.99, 066.01, 066.02, 072.01, 292.20, 292.38, 292.39, 323.00, 323.01, 323.08, 323.09, 781.70, 999, 999.10. Medical records of hospitalised participants were reviewed (in order to evaluate possible other causes of the event) and their correspondence to diagnostic criteria (see column on the right) examined.	Exposure risk period: (a) 0 to 3 months after vaccination Control period: (b) 4 to 24 months Observation period: (c) 0 to 24 months MMR II vaccine (Merck & Co, West Point, PA) measles: Enders‐Edmonston mumps: Jeryl Lynn rubella: Wistar RA 27/3 Vaccination data were assessed through vaccination register.	Not significant excess of hospitalisation within 3 months of vaccination (P = 0.28) Incidence of encephalitis of undefined cause amongst 1‐ to 7‐year‐old children decreased from 19.9 per 100,000 in 1983 to 13.0 per 100,000 in 1985.	(a) 9 cases (3 months) (b) 88 cases (21 months) (c) 97 cases (24 months)	rr (95% CI)* 0.72 (0.36 to 1.42) (*)rate ratio amongst risk period (b) and control period (a)
db‐Ward 2007 Self‐controlled case series	Children aged 2 to 35 months (immunised with MMR; NK) with outcome of interest diagnosed between October 1998 and September 2001 (n = 107)	Onset of illness: day of hospital admission Fever: temperature of 37.5 °C; the questionnaire asked whether there was a fever and also for the maximum temperature recorded at any site by any method Encephalopathy: a depressed or altered level of consciousness Case definition of serious neurologic disease: any child 2 to 35 months old with a severe illness with fever and convulsions (see Table 35) and/or encephalitis was included Encephalitis: encephalopathy for at least 24 hours and at least 2 of the following: fever, convulsions, focal neurologic findings (≥ 24 h), pleocytosis (> 5 leukocytes per μL CSF), characteristic abnormal results of neuroimaging (computerised tomography or MRI), herpes simplex virus nucleic acid (or nucleic acid of any other virus proven to cause encephalitis) in CSF; or postmortem histologic evidence of encephalitis Exclude: viral (aseptic) meningitis without encephalopathy the following confirmed causes were excluded: hypoxic/ischaemic; vascular; toxic; metabolic, neoplastic, traumatic and pyogenic infections uncomplicated convulsions or a series of convulsions lasting < 30 min immunocompromised children Cases of suspected encephalitis and/or severe illness with fever and convulsion occurring in children aged between age 2 and 35 months through Britain and Ireland were identified by consultant paediatricians taking part in a survey (October 1998 to September 2001) and notified to the British Paediatric Surveillance Unit. Details about neurologic illnesses were collected by reporting paediatricians by means of a detailed questionnaire. For diagnostic purposes, saliva, blood, and cerebrospinal samples were also collected. Questionnaires were reviewed by study investigators in order to assess whether reported cases corresponded to an analytical case definition taking into account severe illness with fever and convulsion and encephalitis (see column on the right).	Exposure risk period: 15 to 35 days after immunisation, because this is the incubation period for postinfectious encephalitis induced by wild‐type measles and for aseptic meningitis induced by the Urabe vaccine strain mumps MMR vaccine type, not reported. Immunisation history of cases was obtained by the Immunisation Department of the Health Protection Agency (other than MMR vaccine, the study also considers DTP, Hib, and MenC vaccines). Only cases with known vaccination history were included in the analysis.	Regarding MMR vaccine, there was no evidence of a raised relative incidence of serious neurologic disease 15 to 35 days after immunisation.	Within 15 to 35 days with concurrent primary HHV‐6 or HHV‐7 infection (a) all (5 cases) (b) no (4 cases) (c) yes (1 case)	rr (95% CI) (a) 1.34 (0.52 to 3.47) (b) 1.52 (0.52 to 4.41) (c) 0.86 (0.10 to 7.23)

	Study design	Low risk of bias		Unclear risk of bias		High risk of bias
	Study design	n	Row %	n	Row %	n	Row %	n total
Effectiveness studies	RCT/CCT	3	100%					3
	Case‐control	8	57.1%	4	28.6%	2	14.3%	14
	Prospective/retrospective cohort	4	13.0%	21	67.7%	6	19.4%	31
	Case‐only ecological method			2	66.7%	1	33.3%	3
	Subtotal	15	29.4%	27	53.0%	9	17.6%	51

	Study design	Low risk of bias		Unclear risk of bias		High risk of bias
	Study design	n	Row %	n	Row %	n	Row %	n total
Safety studies	RCT/CCT	2	28.6%	2	28.6%	3	42.9%	7
	Case‐control	8	38.1%	11	52.4%	2	9.5%	21
	Prospective/retrospective cohort	14	43.8%	4	12.5%	14	43.8%	32
	Self‐controlled case series/person‐time cohort	11	68.8%	5	31.2%			16
	Case cross‐over	1	33.3%	2	66.7%			3
	Case‐only ecological method	2	25.0%	4	50.0%	2	25.0%	8
	Subtotal	38	43.7%	28	32.2%	21	24.1%	87
Total (all studies)		53	38.4%	55	39.9%	30	21.7%	138

	Study design	Low risk of bias		Unclear risk of bias		High risk of bias
	Study design	n	Row %	n	Row %	n	Row %	n total
Safety studies (excluding short‐term side effects studies)	Case‐control	8	38%	11	52%	2	10%	21
	Prospective/retrospective cohort	14	64%	4	18%	4	18%	22
	Self‐controlled case series/person‐time cohort	11	69%	5	31%			16
	Case cross‐over	1	33%	2	67%			3
	Case‐only ecological method	2	25%	4	50%	2	25%	8
	Total	36	51%	26	37%	8	11%	70

All studies included	Low risk of bias		Unclear risk of bias		High risk of bias		Total
Publication year	N	Row %	N	Row %	N	Row %	Total
1971 to 1980	0	0%	1	20%	4	80%	5
1981 to 1990	2	29%	0	0%	5	71%	7
1991 to 2000	3	20%	6	40%	5	40%	15
2001 to 2010	21	39%	23	43%	10	18%	54
2011 to 2019	27	47%	24	42%	6	11%	57
Total	53	36%	54	42%	30	22%	138

Only safety studies	Low risk of bias		Unclear risk of bias		High risk of bias		Total
Publication year	N	Row %	N	Row %	N	Row %	Total
1971 to 1980			1	20%	4	80%	5
1981 to 1990	2	29%			5	71%	7
1991 to 2000	2	20%	4	40%	4	40%	10
2001 to 2010	17	40%	17	41%	8	19%	43
2011 to 2019	17	74%	5	22%	1	4%	22
Total	38	39%	27	37%	22	24%	87

Effectiveness against measles
Patient or population: children 9 months to 15 years old Setting: general population or school or day‐care centre or general practitioner or households Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of measles amongst unvaccinated	Risk of measles amongst vaccinated
Cohort studies ‐ 1 dose	Study population		RR 0.05 (0.02 to 0.13)	12,039 (7 observational studies)	⊕⊕⊕⊝ MODERATE¹
	66 per 1000	3 per 1000 (1 to 9)
Cohort studies ‐ 2 doses	Study population		RR 0.04 (0.01 to 0.28)	21,604 (5 observational studies)	⊕⊕⊕⊝ MODERATE¹
	19 per 1000	1 per 1000 (0 to 5)
Cohort studies household contacts ‐ 1 dose	Study population		RR 0.19 (0.04 to 0.89)	151 (3 observational studies)	⊕⊕⊝⊝ LOW
	508 per 1000	97 per 1000 (20 to 452)
Cohort studies household contacts ‐ 2 doses	Study population		RR 0.15 (0.03 to 0.75)	378 (3 observational studies)	⊕⊕⊝⊝ LOW
	508 per 1000	76 per 1000 (15 to 381)
Cohort studies household contacts ‐ 3 doses	Study population		RR 0.04 (0.01 to 0.23)	151 (2 observational studies)	⊕⊕⊝⊝ LOW
	351 per 1000	14 per 1000 (4 to 81)
Cohort studies postexposure prophylaxis	Study population		RR 0.26 (0.14 to 0.50)	283 (2 observational studies)	⊕⊕⊝⊝ LOW
	314 per 1000	82 per 1000 (44 to 157)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MMR: measles, mumps, rubella vaccine; RR: risk ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Effectiveness against mumps
Patient or population: children 9 months to 15 years old Setting: general population or school or day‐care centre or general practitioner or households Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of mumps amongst unvaccinated	Risk of mumps amongst vaccinated
Cohort studies ‐ Jeryl Lynn strain ‐ 1 dose	Study population		RR 0.24 (0.08 to 0.76)	9915 (6 observational studies)	⊕⊕⊕⊝ MODERATE¹
	91 per 1000	22 per 1000 (7 to 69)
Cohort studies ‐ Jeryl Lynn strain ‐ 2 doses	Study population		RR 0.12 (0.04 to 0.35)	7792 (5 observational studies)	⊕⊕⊕⊝ MODERATE²
	74 per 1000	9 per 1000 (3 to 26)
Cohort studies ‐ Jeryl Lynn strain ‐ unspecified number of doses	Study population		RR 0.23 (0.14 to 0.35)	2011 (4 observational studies)	⊕⊕⊝⊝ LOW
	97 per 1000	22 per 1000 (14 to 34)
Cohort studies ‐ Jeryl Lynn strain ‐ household contacts	Study population		RR 0.26 (0.13 to 0.49)	1036 (3 observational studies)	⊕⊕⊕⊝ MODERATE²
	408 per 1000	106 per 1000 (53 to 200)
Cohort studies ‐ Urabe strain ‐ unspecified numbers or at least 1 dose	Study population		RR 0.23 (0.12 to 0.44)	2721 (4 observational studies)	⊕⊕⊝⊝ LOW
	202 per 1000	47 per 1000 (24 to 89)
Cohort studies ‐ Rubini strain ‐ unspecified numbers or at least 1 dose	Study population		RR 0.96 (0.55 to 1.65)	4219 (4 observational studies)	⊕⊕⊝⊝ LOW
	202 per 1000	194 per 1000 (111 to 334)
Cohort studies ‐ mumps strain not reported or any strain	Study population		RR 0.52 (0.29 to 0.94)	769 (2 observational studies)	⊕⊕⊝⊝ LOW
	225 per 1000	117 per 1000 (65 to 212)
Cohort studies ‐ third dose versus 2 doses	Study population		RR 0.59 (0.33 to 1.05)	5417 (2 observational studies)	⊕⊕⊝⊝ LOW
	7 per 1000	4 per 1000 (2 to 8)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MMR: measles, mumps, rubella vaccine; RR: risk ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Effectiveness against rubella
Patient or population: children 9 months to 15 years old Setting: school Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of rubella amongst unvaccinated	Risk of rubella amongst vaccinated
Cohort studies secondary cases ‐ any strain	Study population		RR 0.11 (0.03 to 0.42)¹	1621 (1 observational study)	⊕⊕⊕⊝ MODERATE²
	0 per 1000	0 per 1000 (0 to 0)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MMR: measles, mumps, rubella vaccine; RR: risk ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Effectiveness against varicella
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMRV or MMR+V vaccine Comparison: MMR vaccine (RCTs), unvaccinated (cohort studies)
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of varicella amongst unvaccinated with MMR vaccine	Risk of varicella amongst vaccinated with MMRV vaccine
MMRV randomised controlled trial ‐ any severity ‐ 2 doses ‐ follow‐up at 5 years	Study population		Rate ratio 0.05 (0.03 to 0.08)	3022 (1 RCT)	⊕⊕⊕⊕ HIGH
	271 per 1000	14 per 1000 (8 to 22)
MMRV randomised controlled trial ‐ any severity ‐ 2 doses ‐ follow‐up between 5 and 10 years	Study population		Rate ratio 0.05 (0.04 to 0.06)	3023 (1 RCT)	⊕⊕⊕⊕ HIGH
	437 per 1000	22 per 1000 (17 to 26)
MMRV randomised controlled trial ‐ any severity ‐ 2 doses ‐ follow‐up at 10 years	Study population		Rate ratio 0.05 (0.04 to 0.06)	3023 (1 RCT)	⊕⊕⊕⊕ HIGH
	473 per 1000	24 per 1000 (19 to 28)
MMRV randomised controlled trial ‐ moderate/severe cases ‐ 2 doses ‐ follow‐up at 5 years	Study population		Rate ratio 0.00 (0.00 to 0.02)	3022 (1 RCT)	⊕⊕⊕⊕ HIGH
	157 per 1000	0 per 1000 (0 to 3)
MMRV randomised controlled trial ‐ moderate/severe cases ‐ 2 doses ‐ follow‐up between 5 and 10 years	Study population		Rate ratio 0.01 (0.00 to 0.02)	3023 (1 RCT)	⊕⊕⊕⊕ HIGH
	237 per 1000	2 per 1000 (0 to 5)
MMRV randomised controlled trial ‐ moderate/severe cases ‐ 2 doses ‐ follow‐up at 10 years	Study population		Rate ratio 0.01 (0.00 to 0.02)	3023 (1 RCT)	⊕⊕⊕⊕ HIGH
	237 per 1000	2 per 1000 (0 to 5)
MMR+V randomised controlled trial ‐ any severity ‐ 2 doses ‐ follow‐up at 5 years	Study population		Rate ratio 0.35 (0.28 to 0.43)	3006 (1 RCT)	⊕⊕⊕⊕ HIGH
	271 per 1000	95 per 1000 (76 to 116)
MMR+V randomised controlled trial ‐ any severity ‐ 2 doses ‐ follow‐up between 5 and 10 years	Study population		Rate ratio 0.33 (0.29 to 0.38)	3010 (1 RCT)	⊕⊕⊕⊕ HIGH
	437 per 1000	144 per 1000 (127 to 166)
MMR+V randomised controlled trial ‐ any severity ‐ 2 doses ‐ follow‐up at 10 years	Study population		Rate ratio 0.33 (0.29 to 0.38)	3010 (1 RCT)	⊕⊕⊕⊕ HIGH
	473 per 1000	156 per 1000 (137 to 180)
MMR+V randomised controlled trial ‐ moderate/severe cases ‐ 2 doses ‐ follow‐up at 5 years	Study population		Rate ratio 0.09 (0.06 to 0.14)	3006 (1 RCT)	⊕⊕⊕⊕ HIGH
	157 per 1000	14 per 1000 (9 to 22)
MMR+V randomised controlled trial ‐ moderate/severe cases ‐ 2 doses ‐ follow‐up between 5 and 10 years	Study population		Rate ratio 0.10 (0.07 to 0.13)	3010 (1 RCT)	⊕⊕⊕⊕ HIGH
	237 per 1000	24 per 1000 (17 to 31)
MMR+V randomised controlled trial ‐ moderate/severe cases ‐ 2 doses ‐ follow‐up at 10 years	Study population		RR 0.10 (0.08 to 0.14)	3010 (1 RCT)	⊕⊕⊕⊕ HIGH
	237 per 1000	24 per 1000 (19 to 33)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MMR: measles, mumps, rubella vaccine; MMRV: measles, mumps, rubella, and varicella vaccine; MMR+V: concurrent administration of MMR vaccine and varicella vaccine; RCT: randomised controlled trial; RR: risk ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: short‐term side effects (local or systemic reactions)
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Short‐term side effects amongst unvaccinated	Short‐term side effects amongst vaccinated
Temperature ‐ RCT/CCT axillary	Study population		RR 2.04 (1.09 to 3.83)	420 (1 RCT)	⊕⊕⊝⊝ LOW ¹
	68 per 1000	139 per 1000 (74 to 261)
Temperature ‐ RCT/CCT rectal	Study population		RR 0.84 (0.67 to 1.06)	170 (1 RCT)	⊕⊕⊝⊝ LOW ¹
	786 per 1000	660 per 1000 (526 to 833)
Temperature ‐ RCT/CCT measurement site not reported	Study population		RR 1.36 (0.83 to 2.23)	520 (2 RCTs)	⊕⊕⊕⊕ HIGH
	182 per 1000	247 per 1000 (151 to 405)
Temperature ‐ cohort studies orally	Study population		RR 1.37 (1.04 to 1.81)	334 (1 observational study)	⊕⊝⊝⊝ VERY LOW ²
	377 per 1000	517 per 1000 (392 to 683)
Temperature ‐ cohort studies measurement site not reported	Study population		RR 1.12 (0.84 to 1.49)	457,123 (4 observational studies)	⊕⊝⊝⊝ VERY LOW ²
	31 per 1000	35 per 1000 (26 to 46)
Rash ‐ cohort studies	Study population		RR 1.49 (0.73 to 3.04)	457,261 (3 observational studies)	⊕⊝⊝⊝ VERY LOW ²
	4 per 1000	6 per 1000 (3 to 13)
Lymphadenopathy ‐ RCT/CCT	Study population		RR 1.32 (0.52 to 3.33)	1156 (3 RCTs)	⊕⊕⊕⊝ MODERATE ²
	21 per 1000	28 per 1000 (11 to 70)
Lymphadenopathy ‐ cohort studies	Study population		RR 1.98 (0.19 to 20.97)	454,085 (2 observational studies)	⊕⊝⊝⊝ VERY LOW ²
	0 per 1000	1 per 1000 (0 to 6)
Coryza ‐ RCT/CCT	Study population		RR 0.45 (0.12 to 1.63)	831 (2 RCTs)	⊕⊕⊝⊝ MODERATE ¹
	37 per 1000	17 per 1000 (4 to 60)
Coryza ‐ cohort studies	Study population		RR 1.13 (1.05 to 1.20)	3176 (1 observational study)	⊕⊕⊝⊝ LOW
	502 per 1000	567 per 1000 (527 to 602)
URTI (rhinitis pharyngitis) ‐ RCT/CCT	Study population		RR 0.31 (0.06 to 1.56)	831 (2 RCTs)	⊕⊕⊝⊝ LOW ¹
	265 per 1000	82 per 1000 (16 to 414)
URTI (rhinitis pharyngitis) ‐ cohort studies	Study population		RR 1.44 (1.26 to 1.64)	966 (1 observational study)	⊕⊝⊝⊝ VERY LOW ²
	484 per 1000	697 per 1000 (610 to 794)
Cough ‐ RCT/CCT	Study population		RR 1.99 (0.45 to 8.81)	831 (2 RCTs)	⊕⊕⊝⊝ LOW ^{1, 2}
	8 per 1000	16 per 1000 (4 to 72)
Rash ‐ RCT/CCT	Study population		RR 2.05 (1.21 to 3.48)	1156 (4 RCTs)	⊕⊕⊕⊕ HIGH
	52 per 1000	107 per 1000 (63 to 182)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; CCT: controlled clinical trial; MMR: measles, mumps, rubella vaccine; RCT: randomised controlled trial; RR: risk ratio; URTI: upper respiratory tract infection
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: encephalitis or encephalopathy
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of encephalitis or encephalopathy amongst unvaccinated	Risk of encephalitis or encephalopathy amongst vaccinated
Case‐control: MMR (risk interval from 0 to 90 days)	Study population		OR 0.98 (0.64 to 1.50)	452 cases, 1280 controls (1 observational study)	⊕⊕⊝⊝ LOW
	34 per 1000	34 per 1000 (22 to 51)
Self‐controlled case series/person‐time cohort	Study population		Rate ratio 0.90 (0.50 to 1.61)	1,071,088 (2 observational studies)	⊕⊕⊝⊝ LOW
	22 per 100,000	20 per 100,000 (11 to 36)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MMR: measles, mumps, rubella vaccine; OR: odds ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: aseptic meningitis
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of aseptic meningitis amongst unvaccinated	Risk of aseptic meningitis amongst vaccinated
Case‐control ‐ Jeryl Lynn ‐ risk interval 0 to 30 days	Study population		OR 0.85 (0.21 to 3.41)	59 cases, 118 controls (1 observational study)	⊕⊕⊝⊝ LOW
	59 per 1000	51 per 1000 (13 to 177)
Case cross‐over ‐ Urabe or Hoshino	Study population		OR 4.00 (2.23 to 7.20)	(2 observational studies)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Case cross‐over ‐ Jeryl Lynn or Rubini	Study population		OR 0.60 (0.18 to 1.99)	(1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series ‐ any strain	Study population		Rate ratio 12.40 (3.12 to 49.35)	(1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series ‐ Urabe	Study population		Rate ratio 30.71 (13.45 to 70.10)	564,635 (3 observational studies)	⊕⊕⊝⊝ LOW
	16 per 100,000	490 per 100,000 (214 to 1.117)
Self controlled case series ‐ Leningrad‐Zagreb	Study population		Rate ratio 6.40 (0.78 to 52.47)	(1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Person‐time cohort ‐ Jeryl Lynn	Study population		Rate ratio 1.30 (0.66 to 2.56)	1,071,088 (1 observational study)	⊕⊕⊝⊝ LOW
	30 per 100,000	39 per 100,000 (20 to 77)
Case‐only ecological method ‐ Urabe	Study population		Rate ratio 9.12 (5.73 to 14.52)	1,054,305 (1 observational study)	⊕⊕⊝⊝ LOW
	9 per 100,000	80 per 100,000 (51 to 128)
Case‐only ecological method ‐ Leningrad‐Zagreb	Study population		Rate ratio 18.56 (12.09 to 28.51)	1,164,964 (3 observational studies)	⊕⊕⊝⊝ LOW
	0 per 100,000	0 per 100,000 (0 to 0)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MMR: measles, mumps, rubella vaccine; OR: odds ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: seizures (febrile/afebrile)
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of seizures  (febrile/afebrile) amongst unvaccinated	Risk of seizures  (febrile/afebrile) amongst vaccinated
Cohort studies ‐ within 1 week after MMR vaccination	Study population		Rate ratio 2.45 (2.21 to 2.71)	1,451,990 (2 observational studies)	⊕⊕⊕⊝ MODERATE¹
	108 per 1000	264 per 1000 (238 to 292)
Cohort studies ‐ between 1 and 2 weeks after MMR vaccination	Study population		Rate ratio 3.16 (2.89 to 3.46)	2,147,638 (2 observational studies)	⊕⊕⊕⊝ MODERATE¹
	13 per 1000	42 per 1000 (38 to 46)
Cohort studies ‐ > 2 weeks after MMR vaccination	Study population		Rate ratio 0.97 (0.49 to 1.94)	1,018,998 (1 observational study)	⊕⊕⊝⊝ LOW
	3 per 1000	3 per 1000 (1 to 5)
Self‐controlled case series/person‐time ‐ between 1 and 2 weeks after MMR vaccination	Study population		Rate ratio 3.36 (2.65 to 4.24)	505,493 (5 observational studies)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series/person‐time ‐ > 2 weeks after MMR vaccination	Study population		Rate ratio 1.18 (0.93 to 1.50)	102,099 (3 observational studies)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series/person‐time ‐ between 1 and 2 weeks after vaccination; MMRV	Study population		Rate ratio 6.08 (4.95 to 7.47)	180,480 (2 observational studies)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series/person‐time ‐ between 1 and 2 weeks after MMR+V vaccination	Study population		Rate ratio 3.13 (2.38 to 4.10)	181,088 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
MMRV vs MMR+V ‐ by brand ‐ from 0 to 42 days after vaccination (Priorix‐Tetra)	Study population		RR 1.95 (0.85 to 4.48)	115,022 (1 observational study)	⊕⊕⊝⊝ LOW
	1 per 1000	1 per 1000 (0 to 2)
MMRV vs MMR+V ‐ by brand ‐ from 7 to 10 days after vaccination (Priorix‐Tetra)	Study population		RR 1.69 (0.93 to 3.07)	114,922 (1 observational study)	⊕⊕⊝⊝ LOW
	1 per 1000	1 per 1000 (0 to 2)
MMRV vs MMR+V ‐ by brand ‐ from 0 to 42 days after vaccination (ProQuad)	Study population		RR 1.30 (1.17 to 1.44)	1,381,609 (4 observational studies)	⊕⊕⊝⊝ LOW
	2 per 1000	2 per 1000 (2 to 3)
MMRV vs MMR+V ‐ by brand ‐ from 7 to 10 days after vaccination (ProQuad)	Study population		RR 2.01 (1.70 to 2.38)	1,381,609 (4 observational studies)	⊕⊕⊝⊝ LOW
	2 per 1000	4 per 1000 (3 to 4)
MMRV vs MMR ‐ by brand ‐ from 0 to 42 days after vaccination (Priorix‐Tetra)	Study population		RR 1.28 (1.00 to 1.64)	292,535 (2 observational studies)	⊕⊕⊝⊝ LOW
	1 per 1000	2 per 1000 (1 to 2)
MMRV vs MMR ‐ by brand ‐ from 7 to 10 days after vaccination (Priorix‐Tetra)	Study population		RR 2.49 (1.66 to 3.74)	292,535 (2 observational studies)	⊕⊕⊝⊝ LOW
	1 per 1000	3 per 1000 (2 to 5)
MMRV vs MMR ‐ by brand ‐ from 0 to 42 days after vaccination (ProQuad)	Study population		RR 1.60 (1.42 to 1.82)	1,049,831 (3 observational studies)	⊕⊕⊝⊝ LOW
	43 per 100,000	69 per 100,000 (61 to 78)
MMRV vs MMR ‐ by brand ‐ from 7 to 10 days after vaccination (ProQuad)	Study population		RR 1.46 (1.32 to 1.61)	1,989,157 (4 observational studies)	⊕⊕⊝⊝ LOW
	21 per 100,000	30 per 100,000 (28 to 34)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MMR: measles, mumps, rubella vaccine; MMRV: measles, mumps, rubella, and varicella vaccine; MMR+V: concurrent administration of MMR vaccine and varicella vaccine; RR: risk ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: autistic spectrum disorders
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk of ASD amongst unvaccinated	Risk of ASD amongst vaccinated
Cohort studies ‐ all children, MMR	Study population		Rate ratio 0.93 (0.85 to 1.01)	1,194,764 (2 observational studies)	⊕⊕⊕⊝ MODERATE¹
	451 per 100,000	419 per 100,000 (383 to 455)
Cohort studies ‐ autism risk (low), MMR	Study population		Rate ratio 1.00 (0.89 to 1.14)	93,071 (1 observational study)	⊕⊕⊕⊝ MODERATE¹
	85 per 100,000	85 per 100,000 (76 to 97)
Cohort studies ‐ autism risk (moderate/high), MMR	Study population		Rate ratio 0.80 (0.64 to 0.98)	1914 (1 observational study)	⊕⊕⊝⊝ LOW	The apparent protective effect is due to indication bias.
	12 per 1000	9 per 1000 (7 to 11)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). ASD: autism spectrum disorders; CI: confidence interval; MMR: measles, mumps, rubella vaccine
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: inflammatory bowel disease
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of IBD amongst unvaccinated	Risk of IBD amongst vaccinated
Case control ‐ all IBD, MMR	Study population		OR 1.42 (0.93 to 2.16)	409 cases, 1416 controls (3 observational studies)	⊕⊕⊕⊝ MODERATE¹
	0 per 1000	0 per 1000 (0 to 0)
Case control ‐ ulcerative colitis, MMR	Study population		OR 1.35 (0.81 to 2.23)	292 cases, 582 controls (2 observational studies)	⊕⊕⊕⊝ MODERATE¹
	0 per 1000	0 per 1000 (0 to 0)
Case control ‐ Crohn's disease, MMR	Study population		OR 0.64 (0.42 to 0.98)	514 cases, 804 controls (3 observational studies)	⊕⊕⊕⊝ MODERATE¹
	0 per 1000	0 per 1000 (0 to 0)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; IBD: inflammatory bowel disease; MMR: measles, mumps, rubella vaccine; OR: odds ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: cognitive delay ‐ developmental delay
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of cognitive delay ‐  developmental delay amongst unvaccinated	Risk of cognitive delay ‐  developmental delay amongst vaccinated
Cohort study ‐ MDI‐BSID II 24th month, MMR	Study population		OR 1.35 (0.15 to 12.07)	337 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Cohort study ‐ MDI‐BSID II 36th month, MMR	Study population		OR 0.37 (0.03 to 4.28)	337 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Cohort study ‐ Raven 5th year, MMR	Study population		OR 1.22 (0.23 to 6.51)	337 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Cohort study ‐ WISC‐R verbal 6th year, MMR	Study population		OR 1.23 (0.09 to 16.92)	337 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MDI‐BSID II: Mental Development Index of Bayley Scales of Infant Development, second edition; MMR: measles, mumps, rubella vaccine; OR: odds ratio; WISC‐R: Wechsler Intelligence Scale for Children, Revised Form
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: idiopathic thrombocytopenic purpura
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of ITP  amongst unvaccinated	Risk of ITP  amongst vaccinated
Case‐control ‐ case cross‐over ‐ case controls MMR	Study population		OR 2.80 (1.50 to 5.23)	410 cases, 2040 controls (2 observational studies)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series ‐ MMR vaccine ‐ age from 9 to 23 months	Study population		Rate ratio 4.21 (2.28 to 7.78)	3,723,677 (5 observational studies)	⊕⊕⊕⊝ MODERATE¹
	17 per 100,000	72 per 100,000 (39 to 132)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; ITP: idiopathic thrombocytopenic purpura; MMR: measles, mumps, rubella vaccine; OR: odds ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: Henoch‐Schönlein purpura
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of HSP  amongst unvaccinated	Risk of HSP amongst  vaccinated
Case‐control ‐ MMR vaccine	Study population		OR 3.40 (1.18 to 9.81)	288 cases, 617 controls (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; HSP: Henoch‐Schönlein purpura; MMR: measles, mumps, rubella vaccine; OR: odds ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

PERMALINK

Vaccines for measles, mumps, rubella, and varicella in children

Carlo Di Pietrantonj

Alessandro Rivetti

Pasquale Marchione

Maria Grazia Debalini

Vittorio Demicheli

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Summary of findings

Background

Description of the condition

Description of the intervention

How the intervention might work

Why it is important to do this review

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Primary outcomes

Secondary outcomes

Search methods for identification of studies

Electronic searches

Searching other resources

Data collection and analysis

Selection of studies

Data extraction and management

Assessment of risk of bias in included studies

Experimental and quasi‐experimental studies

Random sequence generation

Allocation concealment

Blinding

Incomplete outcome data

Non‐experimental studies

Measures of treatment effect

Unit of analysis issues

Dealing with missing data

Assessment of heterogeneity

Assessment of reporting biases

Data synthesis

Subgroup analysis and investigation of heterogeneity

Sensitivity analysis

Summary of findings and assessment of the certainty of the evidence

Results

Description of studies

Results of the search

1.

2.

Included studies

Vaccine effectiveness

1. Measles: effectiveness ‐ cohort studies.

2. Measles: effectiveness ‐ case‐control studies.

3. Mumps: effectiveness ‐ cohort studies.

4. Mumps: effectiveness ‐ case‐control studies.

5. Rubella: effectiveness ‐ cohort studies.

6. Varicella: effectiveness ‐ RCTs/CCTs.

7. Varicella: effectiveness ‐ cohort studies.

8. Varicella: effectiveness ‐ case‐control studies.

9. Varicella: effectiveness ‐ case‐only ecological method studies.

Vaccine safety‐harms

10. Safety: short‐term side effects (local or systemic reactions) ‐ RCTs/CCTs.

11. Safety: short‐term side effects (local or systemic reactions) ‐ non‐RCT study designs.

12. Safety: encephalitis or encephalopathy.

13. Safety: aseptic meningitis.

14. Safety: seizure (febrile/afebrile).

15. Safety: MMRV versus MMR/MMR+V ‐ febrile seizures.

16. Safety: autistic spectrum disorders.

17. Safety: inflammatory bowel disease.

18. Safety: cognitive delay, developmental delay.

19. Safety: idiopathic thrombocytopenic purpura.

Safety: type 1 diabetes
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of type 1 diabetes amongst unvaccinated	Risk of type 1 diabetes amongst vaccinated
Cohort study MMR ‐ all children	Study population		Rate ratio 1.09 (0.98 to 1.21)	1,666,829 (2 observational studies)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Cohort study MMR ‐ children with at least 1 sibling with type 1 diabetes	Study population		Rate ratio 0.86 (0.34 to 2.16)	3848 (1 observational study)	⊕⊕⊝⊝ LOW
	6 per 1000	5 per 1000 (2 to 12)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MMR: measles, mumps, rubella vaccine
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: asthma
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of asthma  amongst  unvaccinated	Risk of asthma  amongst  vaccinated
Cohort study (rate ratio) ‐ all ages	Study population		Rate ratio 1.05 (0.80 to 1.39)	1,067,712 (3 observational studies)	⊕⊕⊝⊝ LOW
	32 per 1000	33 per 1000 (25 to 44)
Cohort studies (risk ratio) ‐ all ages	Study population		RR 0.63 (0.24 to 1.63)	886 (3 observational studies)	⊕⊕⊝⊝ LOW
	414 per 1000	261 per 1000 (99 to 674)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MMR: measles, mumps, rubella vaccine; RR: risk ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: eczema ‐ dermatitis
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: vaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of eczema ‐ dermatitis amongst unvaccinated	Risk of eczema ‐ dermatitis amongst vaccinated
Cohort study (rate ratio)	Study population		Rate ratio 3.50 (2.38 to 5.15)	14,353 (1 observational study)	⊕⊝⊝⊝ VERY LOW ¹
	0 per 1000	0 per 1000 (0 to 0)
Cohort study (risk ratio)	Study population		RR 0.75 (0.29 to 1.94)	555 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MMR: measles, mumps, rubella vaccine; RR: risk ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: hay fever, rhinoconjunctivitis, hypersensitivity/allergy
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of hay fever, rhinoconjunctivitis, hypersensitivity/allergy amongst unvaccinated	Risk of hay fever, rhinoconjunctivitis, hypersensitivity/allergy amongst vaccinated
Cohort study ‐ rhinoconjunctivitis	Study population		OR 0.64 (0.19 to 2.11)	489 (1 observational study)	⊕⊕⊝⊝ LOW
	211 per 1000	146 per 1000 (48 to 360)
Cohort study ‐ hypersensitivity/allergy	Study population		OR 0.63 (0.14 to 2.77)	544 (1 observational study)	⊕⊕⊝⊝ LOW
	429 per 1000	321 per 1000 (95 to 675)
Case control ‐ hay fever	Study population		OR 1.16 (0.92 to 1.45)	0 cases, 0 controls (2 observational studies)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MMR: measles, mumps, rubella vaccine; OR: odds ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: acute leukaemia
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of acute leukaemia  amongst unvaccinated	Risk of acute leukaemia  amongst vaccinated
Case‐control ‐ acute leukaemia	Study population		OR 0.97 (0.76 to 1.24)	941 cases, 1667 controls (2 observational studies)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Case‐control ‐ acute lymphoblastic leukaemia	Study population		OR 0.91 (0.72 to 1.14)	1375 cases, 2316 controls (4 observational studies)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Case‐control ‐ acute myeloblastic leukaemia	Study population		OR 0.56 (0.29 to 1.07)	62 cases, 1258 controls (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MMR: measles, mumps, rubella vaccine; OR: odds ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: demyelinating diseases ‐ multiple sclerosis ‐ acute disseminated encephalomyelitis
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of demyelinating diseases ‐ multiple sclerosis ‐ ADEM amongst unvaccinated	Risk of demyelinating diseases ‐ multiple sclerosis ‐ ADEM amongst vaccinated
Case‐control ‐ multiple sclerosis	Study population		OR 1.13 (0.62 to 2.05)	206 cases, 888 controls (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Case‐control ‐ ADEM	Study population		OR 1.03 (0.44 to 2.42)	272 cases, 1096 controls (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). ADEM: acute disseminated encephalomyelitis; CI: confidence interval; MMR: measles, mumps, rubella vaccine; OR: odds ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: gait disturbances
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of gait disturbances amongst unvaccinated	Risk of gait disturbances amongst vaccinated
Self‐controlled case series (hospitalisations) ‐ hospitalisations ‐ risk period: 0 to 60 days	Study population		Rate ratio 0.46 (0.16 to 1.34)	127 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series (GP visits) ‐ GP visit ‐ risk period: 0 to 5 days	Study population		Rate ratio 1.88 (1.30 to 2.72)	1398 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series (GP visits) ‐ GP visit ‐ risk period: 6 to 60 days	Study population		Rate ratio 0.93 (0.78 to 1.11)	1398 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; GP: general practitioner; MMR: measles, mumps, rubella vaccine
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Safety: bacterial or viral infections, immune overload
Patient or population: children 9 months to 15 years old Setting: general population Intervention: MMR vaccine Comparison: unvaccinated
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)
	Risk of bacterial or viral infections, immune overload amongst unvaccinated	Risk of bacterial or viral infections, immune overload amongst vaccinated
Self‐controlled case series ‐ lobar pneumonia ‐ lobar pneumonia risk period (0 to 90 days)	Study population		Rate ratio 0.75 (0.64 to 0.89)	2412 (2 observational studies)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series ‐ invasive bacterial infections ‐ invasive bacterial infections risk period (0 to 90 days)	Study population		Rate ratio 0.90 (0.71 to 1.13)	2412 (2 observational studies)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series ‐ encephalitis meningitis ‐ encephalitis meningitis risk period (0 to 90 days)	Study population		Rate ratio 0.84 (0.20 to 3.51)	2025 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series ‐ herpes ‐ herpes risk period (0 to 90 days)	Study population		Rate ratio 1.17 (0.56 to 2.46)	2025 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series ‐ pneumonia ‐ pneumonia risk period (0 to 90 days)	Study population		Rate ratio 0.72 (0.32 to 1.60)	2025 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series ‐ varicella zoster ‐ varicella zoster risk period (0 to 90 days)	Study population		Rate ratio 0.93 (0.68 to 1.27)	2025 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
Self‐controlled case series ‐ miscellaneous viral infections ‐ miscellaneous viral infections risk period (0 to 90 days)	Study population		Rate ratio 0.68 (0.43 to 1.08)	2025 (1 observational study)	⊕⊕⊝⊝ LOW
	0 per 1000	0 per 1000 (0 to 0)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MMR: measles, mumps, rubella vaccine
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Date	Event	Description
18 November 2021	Amended	In the previous publication of this review, Analysis 3.1 and Additional Table 5 reported that vaccine effectiveness from one cohort study that attempted to estimate MMR vaccine effectiveness in a population who received two rubella strain‐based MMR vaccines (ca‐Chang 2015): MMR containing the BRD‐II rubella strain, or MMR containing the RA27/3 rubella strain. The vaccine efficacy was 89% (95% confidence interval 56% to 95%). In response to a feedback comment, we specified in the Abstract, Discussion, and summary of findings table 3, that the vaccine type BRD2 against rubella is only used in China.
18 November 2021	New citation required but conclusions have not changed	Our conclusions remain unchanged.

Date	Event	Description
8 July 2020	Amended	The NIHR disclaimer and funding stream detail have been added to the Sources of support section.
2 May 2019	New search has been performed	A new author joined the team to update this review. We included new vaccines MMRV and MMR+V in an updated search from 4 October 2016 to 2 May 2019. We included 34 studies on safety and 40 studies on effectiveness. We included 4 studies on safety and 8 studies on effectiveness that were previously awaiting classification in our 2012 review update.
2 May 2019	New citation required but conclusions have not changed	Our conclusions remain unchanged.
4 October 2016	Feedback has been incorporated	Feedback comment inserted.
12 May 2011	New search has been performed	We updated the searches and included 33 new trials in the review, including one previously excluded trial (ca‐Marolla 1998). We excluded 50 new trials, and 13 new trials are awaiting classification. The conclusions remain unchanged.
1 February 2011	New citation required but conclusions have not changed	A new author joined the team to update the review.
6 May 2008	Amended	Converted to new review format.
8 August 2006	Feedback has been incorporated	Feedback comment and reply added to review.
18 December 2004	New search has been performed	Searches conducted.