BACKGROUND: The relation between the measles-mumps-rubella (MMR) vaccine and autism has been a controversial topic for a long time. Numerous studies have evaluated the possible relationship between MMR and autism, but individual results are relatively powerless to identify differences. In addition, the existing meta-analyses are imperfect and out of date. Therefore, a new meta-analysis is required.

MATERIAL AND METHODS: We searched for relevant studies in PubMed, Medline, Embase, Web of Knowledge, and Google Scholar until May 2015 using defined selection criteria. Heterogeneity was measured by the Q-test and I2 statistic. Funnel plots were used to explore publication bias. Data were analyzed with the random-effects model. The overall odds ratio (OR) with 95% confidence intervals (CI) was calculated.

RESULTS: Four cohort studies involving 633 756 participants and 5 case-control studies involving 8975 participants were included. The pooled data show no association between MMR vaccination and autism (OR: 0.88, 95% CI: 0.67 to 1.14). A significant heterogeneity was found between the studies without potential publication bias. Therefore, subgroup analyses and sensitivity analyses were performed.

CONCLUSIONS: This meta-analysis indicates that the MMR vaccine has no relationship with autism, but convincing the public of this remains a major challenge.

Keywords: Autistic Disorder, Measles-Mumps-Rubella Vaccine