Study designs of evaluations included in the review
Controlled epidemiological studies were eligible for inclusion. The included studies were case series, time-series analyses, cross-sectional studies, self-matched case-series and retrospective cohort studies.
Specific interventions included in the review
Studies of MMR vaccination were eligible for the review.
Participants included in the review
The participants in any study that investigated the possible association between ASD and MMR vaccination and which described the methods for the diagnosis of ASD were eligible for the review. The studies had to have reported a systematic method of identifying a sample. All of the included studies were of children: those diagnosed with ASD or with some developmental delay disorder (with possible temporal association with MMR); those who had developed possibly significant symptoms after MMR vaccination; those who had all received the MMR vaccine; or inclusive or random samples of a general population of children. Two studies included a control group: one of matched controls, and one comprising a clinical sample from before the introduction of the MMR vaccine. Overall, the included studies were based on nine distinct sources of data.
Outcomes assessed in the review
Studies that reported outcomes relevant to the testing of the association between the development of ASD and MMR vaccination were eligible for the review. The studies had to describe methods used for the diagnosis of ASD. The included studies compared rates of ASD in children vaccinated with MMR and unvaccinated children; examined the change in rates of ASD associated with change in MMR vaccination coverage; and examined the temporal association between ASD or a variant of ASD and MMR. The included studies defined variant ASD as developmental regression or gastrointestinal symptoms.
How were decisions on the relevance of primary studies made?
The authors stated that they selected papers for the review independently and in duplicate. Any disagreements were resolved by consensus, with the help of a third reviewer if required.