A major problem that is now occurring is scientists manipulating data to achieve results that they want. So far the public still has a lot of faith in proclamations by scientists, but I think that will change unless a better review process can be implemented. The MMR vaccine given to babies is an example of how the current situation can go wrong. One study concluded that the MMR vaccine increased the incidence of autism. A not very smart politician named Robert Kennedy, Jr. picked up on this, and launched a major publicity campaign that convinced a lot of people to not vaccinate their children. Many additional studies were done that showed no correlation between the MMR vaccine and autism, but Kennedy and others continued their campaign. Now an investigation calls the initial study into question. (The initial study involved such a small sample that I can't see why anyone ever acted on the results without more study.)

http://www.timesonline.co.uk/tol/life_and_style/health/article5683671.ece

Several authors of the original report have now withdrawn their support.

http://query.nytimes.com/gst/fullpage.html?res=9F05E4DE133FF937A35750C0A9629C8B63