Analyzing the data on ClinicalTrials.gov (CT.g) to uncover industry trends isn't easy at first. There are, after all, over 215,000 studies registered on the site. And before importing the data from these 215,000 studies, analysts will want to scrub the data for errors before doing any serious mining. Thanks to our Data Science team we have such a dataset that we use it not only to provide analytics for our clients, but to uncover trends we think our blog readers will want to know about.

In this post we zero-in on the "Funder Type" (aka agency class) field that is associated with each trial on CT.g. When registering a trial on the site, a trial manager must classify the trial as belonging to one of four funding types:

  • NIH
  • Industry
  • Other U.S. Federal Agency
  • All others (universities, organizations)

The last "all others" class is almost entirely comprised of trials funded by academic and medical centers. Pharmaceutical business, however, is mostly conducted with Industry trials. 

Because there is an multi-billion dollar ecosystem built around Industry-funded trials, our hypothesis was that that there are more industry-funded trials than any other type. However, take a look at the data of all clinical trials registered in the last 4 years, by funding type:

Out of approximately 20,000 studies registered each year, almost two thirds of these studies come out of academia and medical centers. This is not what we expected to see. When we dug further, and looked at sponsors for all trials in 2015 ranked by number of studies, this is what we found:

Of the 18 leading sponsors of trials in 2015 registered on ClinicalTrials.gov, only 4 are pharmaceutical companies; the rest are all academic or medical centers. Our hypothesis that industry-funded trials should make up most of the new trials registered in any given year wasn't even close. When we look at the data from the past 4 years, academia and medical centers have registered more trials than the industry at a rate of 2:1.

Questions or comments about our analysis? Join the discussion by leaving a comment.

6 Comments