The Measles, Mumps, and Rubella (MMR vaccine) is one of the most effective and well-studied vaccines in the world, with a firmly established safety profile. Yet, a fraudulent 1998 study ignited a persistent myth linking MMR to autism, fueling waves of disinformation that refuse to die. Today, false claims about MMR threaten public health and put millions at risk. Here’s what the science really says about MMR and Autism.
MMR: Decades of Testing, Data, and Debunking the Zombie Myths
The measles, mumps, and rubella (MMR) vaccine has been in use since 1971 and is one of the most widely studied vaccines. Over the decades, rigorous pre-licensure trials, large post-licensure studies, and systematic reviews from around the world have consistently affirmed its safety and effectiveness. Nonetheless, zombie myths persist – such as the baseless claim that MMR was “never tested against a placebo.”
This post, chronologically reviews some of the key evidence from each decade, highlighting the study designs (randomised trials, observational studies, reviews) and data quality. I also address the placebo question by examining how vaccines are tested, including the famous Peltola twin study, and why alternative trial designs are often used.
Pre-Licensure Testing (1960s–1970s)
Developing Separate Vaccines
Before the combined MMR vaccine existed, its components (measles, mumps, and rubella vaccines) underwent extensive testing on their own. For example, John Enders’ team tested the first measles vaccine in large trials from 1958 to 1960, involving small groups initially and then thousands of children in New York City and Nigeria
These trials demonstrated high efficacy – the measles vaccine was hailed as 100% effective and licensed in 1963.
Combining into MMR vaccine
In 1969–1971 legendary vaccinologist Dr. Maurice Hilleman at Merck pioneered combining the three vaccines into a single shot. After development work and clinical evaluations, the trivalent MMR vaccine was licensed by the FDA in 1971.
Pre-licensure testing at that time focused on ensuring that the combined vaccine induced immunity to all three diseases without unexpected side effects. Studies showed that giving the three components together did not weaken the immune response. In fact, antibody responses in people who received MMR were equivalent to those who received separate measles, mumps, and rubella shots.
Early controlled trials confirmed that one dose of MMR could safely produce protective antibodies for all three illnesses, laying the groundwork for its widespread adoption.
Initial Efficacy and Use
Following licensure, many countries began introducing MMR in routine immunisation. Data from the 1970s demonstrated strong real-world effectiveness. The combined vaccine simplified immunisation programmes (fewer injections) and by the late 1970s and early 1980s, mass vaccination programs were driving down measles cases dramatically. (In The Gambia, measles transmission was interrupted as early as 1967 through vaccination campaigns, and the WHO’s Expanded Programme on Immunization in 1974 made measles and later MMR a priority .
These early successes confirmed that the MMR vaccine worked as intended in larger populations.
1980s: Early Post-Licensure Safety Studies
By the 1980s, the MMR vaccine was part of childhood vaccine schedules in many countries, and researchers turned a keen eye to its safety in large populations. This decade saw both randomised controlled trials (RCTs) and observational studies that assessed MMR’s side effects and helped refine vaccine formulations.
The famous Peltola Study from Finland
Twin Study with Placebo (Peltola 1986): A landmark double-blind, placebo-controlled-crossover trial in twins was conducted in Finland by Peltola et al. in 1986 to quantify true vaccine reactions. In this study, 581 twin pairs were immunised in a crossover design – one twin received MMR vaccine while the other received a placebo, then they swapped in a later round.
This unique design allowed researchers to distinguish coincidental symptoms from actual vaccine effects. The findings were reassuring: the vast majority of reported symptoms after vaccination were temporally associated but not causally caused by MMR vaccine.
The true rate of side effects attributable to the vaccine was estimated at only 0.5–4.0%.
Moreover, some common complaints (e.g. mild respiratory symptoms, nausea) occurred more often in the placebo group than in the vaccinated group, underscoring that children get sick occasionally with or without vaccines.
This well-controlled trial provided high-quality evidence that reactions to MMR less common that many realise. The most severe symptoms that were more likely after vaccination was high fever. Here is the table from the paper that shows the different between the saline group and the vaccine group.
Click Here to go to the Peltola paper abstract. Below is a table from that study that shows the difference in commone reactions between vaccinated and unvaccinated.
| Symptom or Sign | Maximum difference in rate (%) | CI_95% | Peak frequency (days after vaccination) |
| Local erythema (>2 cm) | 0.8 | 0.1–1.4 | 2 |
| Other local reaction | 0.4 | 0–1.4 | 2 |
| Mild fever (≤38.5°C rectal) | 2.7 | 0–6.1 | 10 |
| Moderate fever (38.6–39.5°C) | 2.9 | 1.6–4.3 | 9 |
| High fever (>39.5°C) | 1.4 | 0.7–2.1 | 10 |
| Irritability | 4.1 | 2.1–6.1 | 10 |
| Drowsiness | 2.5 | 1.4–3.6 | 11 |
| Willingness to stay in bed | 1.4 | 0.5–2.3 | 11 |
| Generalised rash | 1.6 | 0–3.0 | 11 |
| Conjunctivitis | 2.1 | 0.9–3.2 | 10 |
| Arthropathy | 0.8 | 0.2–1.3 | 7–9 |
| Peripheral tremor | 0.4 | 0–0.9 | 9 |
| Cough and/or coryza | -1.5† | -4.6–1.6 | 9 |
| Nausea and/or vomiting | -0.8† | -1.6–0 | 7–8 |
| Diarrhoea | 0.7 | 0–1.7 | 11 |
| Symptoms and signs caused by MMR vaccination and peak day of occurence. Adapted from Peltola, Heikki et al. FREQUENCY OF TRUE ADVERSE REACTIONS TO MEASLES-MUMPS-RUBELLA VACCINE. The Lancet, Volume 327, Issue 8487, 939 – 942 | |||
Active Comparator Trials of MMR Vaccine
While the Finnish twin study used a saline placebo, most vaccine trials for ethical reasons use an active comparator (another vaccine) instead of a saline placebo. [See my blog on this topic] By the 1980s, it was no longer ethical to leave children completely unprotected against measles in a trial, since an effective single vaccine existed. For instance, in the UK, before introducing MMR vaccine nationally in 1988, researchers ran an RCT comparing MMR to the standard single measles shot. In this randomised single-blind trial (Edees et al., 1991), 420 children were assigned to receive either MMR or measles vaccine alone.
The results showed no significant difference in the rate of fevers, rashes, or other mild reactions between the MMR group and the measles-only group.
Both groups had the expected brief fever and rash in about a week after vaccination, indicating that MMR’s reactogenicity was on par with the already-accepted measles vaccine. Immune responses were also robust: over 92% developed immunity to measles in both groups (93% with MMR vs 97% with single measles), while the MMR group additionally achieved high immunity to mumps (88%) and rubella (96%).
Global Monitoring and Improvements
Post-licensure surveillance in the 1980s also uncovered rare side effects linked to specific vaccine strains, which led to improvements. For example, a particular mumps strain (Urabe) used in some MMR vaccine versions was found to carry a small risk of aseptic meningitis. Careful case monitoring showed that within 2–3 weeks of Urabe-containing MMR, there was an elevated risk of meningitis (RR ~14–22 in some studies).
This prompted many countries (e.g. Canada, UK) to switch to the safer Jeryl Lynn mumps strain by the early 1990s. Such findings highlight the global collaboration in vaccine safety – even infrequent adverse events (on the order of one-in-tens-of-thousands) were detected and addressed through post-marketing studies, even back in the 1980s.
Addressing the “no Placebo” claims
The Peltola twin study demonstrates that, contrary to some claims, MMR has been tested against a placebo in a rigorous trial.
However, it’s true that most MMR vaccine studies did not use inert placebos – and for good reason. Once standalone measles, mumps, and rubella vaccines were known to save lives, it became unethical to withhold them. Instead, new combinations like MMR were tested against the standard of care (e.g. comparing MMR to giving the three vaccines separately, or to an older vaccine formulation). This is a common ethical approach in vaccine research. What matters is that the MMR’s safety and efficacy were thoroughly vetted – through both placebo-controlled and active-controlled trials – before and after it was licensed.
1990s: UK Introduction and the Autism Controversy
The 1990s brought MMR vaccine into new national programs and unfortunately also saw the rise – and refutation – of a false autism scare. Research during this decade was largely observational, focusing on population health outcomes.
Real-World Effectiveness
By the early 1990s, many countries that introduced MMR observed dramatic declines in measles, mumps, and rubella cases. In the United States, measles was well on the way to elimination by 1993 thanks to high MMR uptake
The UK introduced MMR in 1988 as a single-dose schedule (adding a second dose in 1996), after local trials had confirmed its immunogenicity (as noted above). Following introduction, disease rates plummeted in the UK as well, confirming the vaccine’s effectiveness in a new population.
Wakefield’s fraudulent 1998 Paper
In 1998, a small British case series ignited public concern by suggesting a link between MMR and autism. This study by Andrew Wakefield and colleagues involved just 12 children and was not a controlled trial or epidemiological study. The authors also falsified data, lied about ethics, lied about where the particpants came from, and did not disclose massive conflicts of interest. Essentially the controversy rested on the claims of the parents of 8 children. Despite this massive fraud, the paper gained undue media attention. Years later it was exposed as one of the biggest scientific frauds in history. The paper was retracted, with the lead author found guilty of multiple counts of misconduct. [Wiki has a good summary and links].
The episode underscored the importance of robust data when evaluating vaccine safety claims, along with the responsibilities of health leaders and the media to fact check and address issues of public concern.
Epidemiological Studies Refute Autism Link with MMR Vaccine
In response to the autism allegations, scientists swiftly conducted large epidemiological studies to see if children who got MMR vaccine were any more likely to develop autism than those who didn’t. One of the first was a comprehensive time-trend analysis in the UK (Taylor et al., 1999). This study identified 498 children with autism in certain regions and examined whether autism rates changed after the 1988 introduction of MMR. The analysis found no “step-up” increase in autism incidence after MMR began and no difference in the average age of diagnosis before vs. after vaccination was in use.
They also looked for clustering of autism onset shortly after vaccination and found none. The authors’ interpretation was clear: there was no epidemiological evidence for a causal link between MMR vaccine and autism.
A natural experiment occurred in Japan. Concern over a particular MMR formulation led Japan to withdraw the MMR vaccine in 1993 (reverting to single vaccines). If MMR were causing autism, one would expect autism rates to drop after 1993 – but the opposite happened. A population study in Yokohama, Japan (Honda et al., 2005) found that even after MMR use stopped, autism diagnoses continued to rise in cohorts born 1993–96.
The authors concluded that MMR vaccine s very unlikely to be a main cause of autism and that stopping its use did not reduce autism rates.
Findings like these, from different countries, provided strong observational evidence through the 1990s that MMR vaccination was not linked to autism.
2000s: Large-Scale Studies and Systematic Reviews
During the 2000s, an even greater emphasis was placed on large cohort studies, case-control studies, and formal systematic reviews to further ensure MMR’s safety profile. With millions of children now having received MMR vaccine, coupled with increased computing power and advances in epidemiological methods, researchers could study rare outcomes with much more statistical power than before.
Massive Cohort Studies
In 2002, the New England Journal of Medicine published a landmark Danish study by Madsen et al. This was a population-based cohort study of over 537,000 children born in Denmark (1991–1998), of whom about 440,000 had received MMR.
Investigators linked vaccination records to the national autism registry. The results were unambiguous: autism rates were the same in vaccinated and unvaccinated children. The relative risk of autism in MMR-vaccinated kids was essentially 1.0 (specifically, RR = 0.92, 95% CI 0.68–1.24) indicating no increase in risk. There was also no association between autism and the timing of MMR or age at vaccination.
Finland contributed further robust data. When Finland added MMR vaccine to its schedule in 1982, health authorities set up an intensive 14-year surveillance project to catch any serious adverse events. By 1996, about 1.8 million Finnish children had received nearly 3 million doses of MMR.
Medical reports of potential adverse events were collected and investigated. The outcomes, published in 2000, were reassuring: only 173 serious events were reported in that entire period, and nearly half of those turned out to be caused by other factors upon closer review.
Importantly, not a single case of autism or inflammatory bowel disease was found to be caused by the vaccine in this cohort. The Finnish researchers concluded that serious vaccine events were exceedingly rare and greatly outweighed by the health risks of the diseases that MMR prevents.
Institute of Medicine & Expert Reviews
In 2004, the U.S. Institute of Medicine (IOM) convened an expert panel to review all available evidence on vaccines and autism. After examining epidemiological studies from multiple countries and possible biological mechanisms, the IOM concluded “the body of evidence favours rejection of a causal relationship between MMR and autism.”
In plain terms, the best science at that point 20-years ago (of which there was now a lot) did not support any link. The committee recommended no changes to vaccination policy and encouraged continued vaccine safety monitoring and research into autism’s real causes
Cochrane Systematic Reviews
The Cochrane Collaboration, an independent network known for rigorous systematic reviews, first reviewed MMR in 2005 and updated their review in 2012. The 2012 Cochrane review (Demicheli et al., updated by Di Pietrantonj et al.) is one of the most comprehensive looks at MMR. It analysed 67 studies including 5 RCTs, 1 controlled clinical trial, 27 cohort studies, 17 case-control studies, and other designs, covering a total of about 14.7 million children.
This enormous evidence base confirmed that one dose of MMR vaccine is ~95% effective at preventing measles (and a second dose ups effectiveness even further).
It also assessed safety across all these studies. The authors found no credible evidence that MMR is associated with autism or with any other pervasive developmental or chronic disorders in children.
They did note that the design of some older studies could have been better in reporting safety outcomes, but ultimately “existing evidence on the safety and effectiveness of MMR vaccine supports current policies of mass immunization”
In other words, the best available data continued to strongly endorse the vaccine’s use and its favourable benefit-risk profile.
2010s: Ongoing Surveillance and Continued Confirmation
Becasue vaccine safety science never sleeps… In the 2010s, research on MMR largely reinforced what was already known, often with even larger datasets. Meta-analyses combining results from multiple studies and new nationwide cohort studies provided further reassurance to the global community.
Meta-Analysis of Autism Studies
By 2014, enough large studies on MMR and autism had been published that researchers could statistically pool their results. A 2014 meta-analysis looked at 10 epidemiologic studies (both cohort and case-control) encompassing over 1.2 million children in total. It concluded that MMR vaccination is not associated with an increased risk of autism.
This meta-analysis essentially synthesised a decade’s worth of research into one clear message: children who receive MMR are no more likely to be diagnosed with autism than those who don’t. The finding has been echoed by other reviewers and national committees. In fact, across three separate committees of the U.S. National Academy of Medicine (formerly IOM) over the years, the evidence has been consistently judged to reject any causal link between MMR and autism
Modern Large Cohort (2019)
One of the newest high-quality studies came from Denmark again. In 2019, Hviid et al. extended the Danish analysis to a cohort of 657,461 children (born 1999–2010) with even longer follow-up. This nationwide cohort study examined overall autism risk as well as risk in subgroups (such as children with autistic siblings or other risk factors). The results: MMR vaccination did not increase autism risk (adjusted hazard ratio ~0.93, essentially no difference from 1).
There also was no evidence that MMR triggered autism in susceptible subgroups, and no clustering of autism cases shortly after vaccination.
Continued Safety Monitoring
Beyond autism, researchers in the 2010s have kept monitoring other potential associations. The Cochrane review mentioned above was updated again around 2020 (adding varicella vaccine in the analysis) and upheld the same conclusions. Vaccine safety surveillance systems (like VAERS in the U.S. and similar systems in Europe, Australia, NZ, and almost all other countires in the world) have not detected new serious adverse event patterns or signals from MMR in over 50 years of use. Some known rare side effects are tracked – for instance, febrile seizures occur in about 1 in 3,000–4,000 toddlers after MMR (usually benign and without lasting effects), and a temporary drop in platelet count occurs in roughly 1 in 30,000–40,000 doses.
These events are orders of magnitude less severe or less frequent than the complications of measles or rubella infection. Meanwhile, the vaccine’s benefits – preventing brain damage or death from measles, preventing congenital rubella syndrome in babies, etc. – are well documented. Decades of global research have led to improvements like the current two-dose schedule, which provides ~97% protection against measles
Take home message
From the small trials of the 1960s to the multi-million-child analyses of today, the MMR vaccine’s testing history is remarkably extensive. We have seen randomised controlled trials (including a placebo-controlled twin study) that established its safety and immunogenicity, observational studies across different countries that found no link to autism or other chronic conditions, and systematic reviews pooling huge amounts of data that reaffirm the vaccine’s safety and effectiveness.
After more than 50 years and billions of doses, the scientific consensus is clear: the MMR vaccine’s benefits far outweigh its risks. Its pre-licensure trials and extensive post-licensure evaluations make it one of the best-characterised medical products we have. This wealth of evidence should give providers and parents confidence that MMR is a safe, life-saving vaccine – a triumph of preventive medicine built on decades of global research and high-quality data.
If anyone tells you otherwise they are either lying or deeply misinformed.
Useful resource on Safety of MMR and other Measles-containing Vaccines
The WHO Rate Sheets summarise the safety profiles of vaccines. Here is a link to the MMR one. https://www.who.int/publications/m/item/mmr-vaccine-rates-information-sheet
Discover more from diplomatic immunity
Subscribe to get the latest posts sent to your email.