June 12, 2019
Waste and inefficiency in drug development is a big problem. It is estimated that as much as 85% of biomedical research may be wasteful—because of biases in study design, lack of publication, unnecessary duplication, or investigating questions of little importance. It is also estimated that only about 1 (or maybe 2) out of every 10 drugs that enters into clinical testing will turn out to be effective.
This rather depression picture of productivity in pharmaceutical research has led to much head-scratching among scientists and policy researchers who are interested in making the enterprise more efficient. For example, a few years ago, the medical journal The Lancet ran an entire series of articles devoted to discussions of how to reduce waste and improve the value of research. This series included suggestions on how to improve prioritization, study design, regulation, access, and the quality of scientific reporting.
But as valuable as these statistics and articles are for illuminating the 100,000 foot view of the research and development enterprise, there may be important patterns or properties of the research enterprise that we can only see if we “zoom in” a little bit more. In what follows, I attempt to provide this “bird’s eye view”—i.e., the 30,000 foot view, where we can look at particular trials of particular sponsors, and begin to ask questions about whether the patters or trends that we see represent an industry working as we think it should; or where there may be particular opportunities to make this system more efficient.
The figure below is a proof-of-concept for this way of looking at the research enterprise—it depicts the clinical trials from 10 large pharmaceutical companies (AbbVie, Bayer, Gilead, GSK, Johnson & Johnson, Merck, Novartis, Pfizer, Roche, and Sanofi) for the past 20 years or so. The underlying data source is ClinicalTrials.gov, which is the national registry for clinical trials in the United States. Every registered trial from each of these companies is represented as a node/bubble. The trials are organized by their starting date (on the x-axis) and patient population/disease under study (on the y-axis). The disease areas are clustered according to the National Library of Medicine’s Medical Subject Heading (MeSH) tree. The color of a node indicates the company (e.g., Pfizer is darker blue, GlaxoSmithKline is pink); the shape indicates the trial’s status (e.g., completed studies are circles, recruiting studies are triangles); and size indicates the number of patients enrolled.
The figure is also interactive. You can mouse-over any node to see more info or click the node to open the trial’s registration page at ClinicalTrials.gov.
In total, this graph shows 13,749 trials, which included more than 6 million patients. This represents an enormous social and scientific investment (likely costing several hundred billion dollars) and this is only the activities of 10 companies—a small piece of the total clinical research activity.
I’ll offer some take-away thoughts at the bottom of the page (note: it’s a long scroll to get down there), but for those interested in the efficiency of the pharmaceutical research enterprise, I think it is worth taking some time to just scroll up and down and explore this picture. There are thousands of stories about science and medicine represented here, raising thousands of questions about what we’ve learned and what we should test next—questions that we wouldn’t even think to ask until we’ve seen the data presented this way.