Design Studies that Industry Won't Do
Posted on July 22, 2009
Joseph R. Calabrese (bio) describes one way to sync academic research with industry studies.
When it comes to interfacing with the pharmaceutical industry, the first thing we should do is to design studies that the industry won't do. So our focus should be on subgroups of patients that normally don't do well and are normally excluded from the industry-sponsored multi-center trials.
So, as an academic, we have a commitment to the conduct of trials that nobody wants to do. So for me, these have been patients with rapid cycling bipolar disorder, patients with co-occurring substance use disorders, and more recently, patients with metabolic syndrome.
Now in addition, drug trials conducted by the pharmaceutical industry tend to focus on efficacy. And with efficacy, what happens is that you use exclusion criteria to improve assay sensitivity. So as a result, you get a good trial, you get a good finding, but the generalizability is limited.
So in academia, what we need to do is to complement this research. It's not that efficacy research is bad. Efficacy research is important, but it's just a first step.
As academicians, we have a responsibility to conduct trials that have fewer exclusion criteria. Now when you do that, what you find is that the trials we conduct are more likely to show modest effect sizes and more likely to show side effects. When the industry does trials, there tends to be a focus on efficacy, and the commitment to teasing out differences in side effect profile isn't always as rigorous.
So we have to be very rigorous about looking at side effects and teasing out relevant side effects. And I think, by doing so, that distinguishes our research, our commitment to drug trials from the commitment the industry has to drug trials.
Excerpted from interview with researcher at the 2009 Career Development Institute for Bipolar Disorder in Pittsburgh, PA.
Please note that the feedback is viewed only by 4researchers staff and is not intended for communication with individual contributors.
Use the form below to submit feedback about this article. If you would like a response, please be sure to include your e-mail address.