Scientific Fraud: Autism and Child Vaccines

CNN is reporting that researchers are labeling academic research that linked autism to childhood vaccines as "fraudulent." Here is a quote:

A now-retracted British study that linked autism to childhood vaccines was an "elaborate fraud" that has done long-lasting damage to public health, a leading medical publication reported Wednesday.

An investigation published by the British medical journal BMJ concludes the study's author, Dr. Andrew Wakefield, misrepresented or altered the medical histories of all 12 of the patients whose cases formed the basis of the 1998 study -- and that there was "no doubt" Wakefield was responsible.

"It's one thing to have a bad study, a study full of error, and for the authors then to admit that they made errors," Fiona Godlee, BMJ's editor-in-chief, told CNN. "But in this case, we have a very different picture of what seems to be a deliberate attempt to create an impression that there was a link by falsifying the data."

Britain stripped Wakefield of his medical license in May 2010. Efforts to reach him for comment were unsuccessful Wednesday.

"Meanwhile, the damage to public health continues, fueled by unbalanced media reporting and an ineffective response from government, researchers, journals and the medical profession," BMJ states.

The now-discredited paper panicked many parents and led to a sharp drop in the number of children getting the vaccine that prevents measles, mumps and rubella. Vaccination rates dropped sharply in Britain after its publication, falling as low as 80 percent by 2004. Measles cases have gone up sharply in the ensuing years.

Fraud in scientific research is something I am very concerned about. I find it reprehensible that researchers will lie in order to get their results published. Meanwhile, people rely on their studies and, in this case, people took medical risks that were unnecessary. Deadly diseases that could be avoided have been contracted because these researchers falsified their data.

In economic research such as in my field, the consequences of falsifying results may not seem to be as severe as normally business research won't lead people to take medical risks as it did in the case above. As such, fraud may be easier to rationalize in my area of research than it would be in a field such as medical science.

I am personally aware of falsified results that were nearly published in a top academic business journal. In the case I am aware of, the person was caught when the results didn't make sense to the co-authors. Unfortunately, I am sure that a careful and competent fraudster could have gotten away with fraud. As such, when I see someone who regularly publishes papers with wonderful data that is collected without help from his co-authors, I have to wonder if it's fraudulent.

Another case that troubles me is when I heard a comment from a co-author who was presenting a working paper with data that was collected by his very successful coauthor.  The first coauthor said something to the effect of: "We don't know where he gets his data--he won't tell us!" I think this is like a CEO who says: "I don't know where the CFO gets the revenues he's booking--he won't tell us!" In my mind, the CEO and the co-author are at least negligent if not co-conspirators in the fraud!

I believe that the problem of fraudulent research getting published is probably more widespread than we think. This is based on a simple analysis of the fraud triangle. First, tremendous pressure exists to publish in top journals. Second, opportunities are ripe for doing so as few controls exist to prevent or detect it. One control could come through co-authors who watch over each other. I've already established that that control is probably not very effective. Another control could come in the way of replication studies. This was how the cold fusion research was revealed to be bogus. However, in many areas of research including some business and economics, replicating a finding is often not publishable and the researcher in charge of collecting the data may do so without any help from his co-authors. Thus, opportunities to falsify data are probably ripe.

If pressure is high and controls are weak, we have to rely on the integrity of the researchers. I believe it's naive to think there aren't some fraudsters out there taking advantage of the system...