I think the problem is that the public ends up reading abstracts and papers without the proper education necessary to adequately critique scientific methods. The vast majority of the conclusions written in discussions and abstracts are conjecture that is merely hinted to by the data to make studies sound more important than they are. They also fail to account for causal relationships and do not adequately address confounding bias. These issues are further compounded by sampling issues and sample size issues. To put it more simply, the quality of the science in papers reporting no association between vaccines and autism is for the most part far superior then those that claim the opposite.
The Wakefield paper is a perfect case as it was used as ammunition for a vaccine-autism link when any scientist could clearly see that the paper was a small case series with no controls, linking three common conditions and relying on parental recall. Each of these factors making all observations for the Wakefield paper inconclusive. We don't even have to take into account that the paper was a complete fraud. Then contrast his paper (or other papers such as the one linking acetaminophen use, vaccines and autism) with papers like so:
Which are done on large populations, are controlled for confounding bias and rely on reliable medical records and you can see the case for vaccines causing the majority of autism cases is very thin.
Comment Reply
I think the problem is that the public ends up reading abstracts and papers without the proper education necessary to adequately critique scientific methods. The vast majority of the conclusions written in discussions and abstracts are conjecture that is merely hinted to by the data to make studies sound more important than they are. They also fail to account for causal relationships and do not adequately address confounding bias. These issues are further compounded by sampling issues and sample size issues. To put it more simply, the quality of the science in papers reporting no association between vaccines and autism is for the most part far superior then those that claim the opposite.
The Wakefield paper is a perfect case as it was used as ammunition for a vaccine-autism link when any scientist could clearly see that the paper was a small case series with no controls, linking three common conditions and relying on parental recall. Each of these factors making all observations for the Wakefield paper inconclusive. We don't even have to take into account that the paper was a complete fraud. Then contrast his paper (or other papers such as the one linking acetaminophen use, vaccines and autism) with papers like so:
http://www.nejm.org/doi/full/10.1056/NEJMoa021134
http://jama.ama-assn.org/content/285/9/1183.short
http://onlinelibrary.wiley.com/doi/10.1111/j.1469-7610.2005.01425.x/full
Which are done on large populations, are controlled for confounding bias and rely on reliable medical records and you can see the case for vaccines causing the majority of autism cases is very thin.
September 21, 2011 - 3:26pmThis Comment
Reply