We’re told science is reliable because of the hypothetical experimental method. Everything can be objectively tested, measured, reported, and—this is key—replicated by other scientists. But what if you can’t trust all the scientists? What if some scientists are “behaving badly”?

Every medication sold in the United States must first undergo clinical trials to test for safety, efficacy, and proper dosage. In this first installment of Scientists Behaving Badly,[1] let’s look at two clinical trial scenarios.

Imagine you are a scientist conducting a clinical trial. You find that a particular arthritis drug is showing some promising results. A few of your test subjects have reported some heart issues, but it may not be related to the drug. A lot of people would benefit from having this drug on the market right now, so you decide to petition the FDA to accelerate the clinical trial process. They approve it, and the arthritis drug goes to market.

Everything is going great, your pharmaceutical company is making a tidy profit, and your drug is helping people. And then the reports start coming in. Your company has been conducting another trial testing whether the arthritis drug can also help decrease the chances of getting colon cancer. Several people in that trial have also reported heart problems. Conducting a post-market trial would be incredibly expensive, and pulling the drug from the shelves would cost millions. Is it really worth it to do another clinical trial?[2]

Scenario two: imagine that you are conducting phase 3 clinical trials for a breast cancer drug that will help women. Phase 3 trials find out if the new drug works better than the standard treatment. You and your team have been working on this drug for several years and are very close to making it widely available. Just like the arthritis drug, you are seeing promising benefits.

But, there are some side effects showing up in this trial that you did not see before. Aside from the normal fatigue and nausea, some of the patients seem to be having other, more serious issues. You aren’t sure if these issues are noteworthy or not. If you report them, then you’ll have to do more clinical trials. You may even have to go back to square one. You’ve worked so hard, and come this far with a promising drug to help women with breast cancer. Is it really necessary to report these observations?[3]

Scientists can “behave badly.” These scenarios are based on actual cases where the scientists ended up under-reporting negative results. The lesson for us is to realize that there are lots of pressures on researchers to come up with new drugs, and to avoid additional costly trials. Keep that in mind when you read the next glowing news report of a drug breakthrough.

[1] This series does not assume that ALL scientists behave badly or is anti-science in anyway.

[2] This is based on the controversy surrounding Vioxx, an anti-inflammatory drug that went on the market in 1999. After conducting a second clinical trial in 2005 Merk discontinued the production of Vioxx. See http://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatientsandProviders/ucm106290.htm. For the post-market trial publication, see “Pooled analysis of Rofecoxib Placebo-Controlled Clinical Trial Data” Joseph S. Ross, MD, MHS; David Madigan, PhD; Kevin P. Hill, MD, MHS; David S. Egilman, MD, MPH; Yongfei Wang, MS; Harlan M. Krumholz, MD, SM  Arch Intern Med. 2009;169(21):1976-1985. doi:10.1001/archinternmed.2009.394; http://archinte.jamanetwork.com/article.aspx?articleid=1108579 (accessed 01/15/13).  For the New York Times article, see “Despite Warnings, Drug Giant Took Long Path to Vioxx Recall.”  The New York Times, November, 2004; http://www.nytimes.com/2004/11/14/business/14merck.html?_r=0 (accessed 01/15/13).

[3] A report come out this past November showing that there is a significant amount of spin and bias in reporting clinical trial results with breast cancer drugs, particularly drugs that show promising results. For the research article, see “Bias in reporting of end points of efficacy and toxicity in randomized, clinical trials for women with breast cancer” F.E. Vera-Badillo, R. Shapiro, A. Ocana, E. Amir, and I.F. Tannock Ann Oncol 24, 1, 2013; http://annonc.oxfordjournals.org/content/early/2013/01/04/annonc.mds636.full (accessed 01/15/13).  For the news brief, see Science Daily: http://www.sciencedaily.com/releases/2013/01/130109215234.htm (accessed 01/15/13).