Is there growing evidence that vaccines cause autism and other illnesses, once you factor out pharma-funded studies?

The MMR-autism link is the most studied question in vaccine safety, and the large, well-designed studies consistently find no connection. The evidence against a link is strong, but it is not complete.

The strongest evidence comes from a Japanese study that is a natural experiment [1]. MMR vaccination rates in one city dropped from about 70% to zero between 1988 and 1996. Yet autism diagnoses at age 7 kept rising, from about 48 to 117 per 10,000 children [1]. If MMR caused autism, removing it should have stopped or slowed the rise. It did not. A 2014 meta-analysis pooled over 1.2 million children from cohort studies and found no increased risk of autism from any childhood vaccine, with an odds ratio of 0.99 [4]. A separate analysis of over 500,000 Finnish children found no clustering of autism diagnoses after MMR vaccination [3].

The thimerosal question is less settled. One VAERS analysis found a higher rate of autism reports after thimerosal-containing DTaP vaccines compared to thimerosal-free ones [6]. Another study found similar signals for autism and other neurodevelopmental disorders [5]. But these are passive reporting systems, which detect only a fraction of real events and are prone to bias. A Danish ecological study found no correlation between thimerosal use and autism rates [2]. The overall picture is mixed, with the strongest studies finding no link and weaker signal-detection studies finding associations that could be due to reporting bias.

My call: the evidence that MMR causes autism is very weak and contradicted by the best-designed studies. The evidence on thimerosal is less clear, with passive surveillance suggesting a signal that population-level studies do not confirm. Confidence in the MMR conclusion is high; confidence on thimerosal is moderate.