|Ragu Bharadwaj, SDM ’07|
Editor’s note: This is the third in a series of articles that follow Ragu Bharadwaj’s progress through the System Design and Management Program. Bharadwaj holds a PhD in biochemistry and is a research scientist at a local biotech firm.
In the first piece, Bharadwaj introduced the problems inherent to the drug development activities of today’s pharmaceutical industry and posited ways to improve these processes through the strategies and techniques taught in SDM. In the second, he discussed problems inherent in the drug development process. In this installment, he describes how systems thinking and lean techniques can improve the drug development process.
As I move well into my second year in the SDM program, I find that SDM courses are enlarging my thinking about how to address the vastly complex and enormously expensive problems of new drug discovery and development.
Today’s pharma industry is a classic systems problem waiting to be solved. The entire system from drug concept development through drug delivery and utilization is rife with problems, which has brought the industry to a crossroad.
In senior lecturer J. Bradley Morrison’s system dynamics class, I learned how the tools of systems thinking can—and must—be applied to optimize processes during drug discovery in order to determine where to extract more value, and consequently save money and time.
Morrison’s class opened my eyes to the multidisciplinary nature of this incredibly complex challenge. One of the main reasons is cultural. The pharmaceutical sector is well over 100 years old and many of the legacy structures and norms no longer serve today’s high-speed, high-tech global environment.
Functional silos are a major culprit, as they were in another “old” industry—manufacturing. However, while manufacturing has evolved, pharma still has a way to go—scientists remain organized by departments, communicating more within these departments than with the outside. Over time, as the softer sciences and disciplines comprising pharma research have matured, attributes such as throughput lead time and cycle times in iterative functions in research have become more important. Unfortunately, the industry has remained unchanged, and functional silos persist in, for example, development. These must be integrated in order for scientists to share information and work together more efficiently. This problem extends beyond the company’s walls to include outsourcers and vendors as well.
In Morrison’s class we learned to apply systems thinking and lean techniques to help make these changes happen. For example, taking a systems view of the cycle time and documenting the steps in the drug discovery process can help provide quantitative data needed to build a case for optimizing systems. Systems thinking can unequivocally show where cycle times can be reduced across multiple functions.
The harder challenge will be getting scientists to want to change how they work. Most are used to focusing in on one research problem rather than thinking about the big picture. In addition, most come from established academic labs where they train not to work in teams as much as with unparalleled excellence on a single problem. Furthermore, most of them have jobs that incorporate research and production components. Changing the work culture of these scientists is necessarily harder than doing so for other fields, mainly because we are attempting to change some portions of a scientist’s job without affecting other portions.
SDM’s course in systems dynamics has helped me see that there are opportunities for pharma companies to find additional revenue streams. For example, much of the data that is gathered in clinical trials is not used for FDA documentation submission. That data, however, still resides in the company’s computers and can be sold. Selling it would help reduce the cycle time in development of other drugs, potentially benefiting the system as a whole and ultimately, consumers and the health-care system at large. Right now, the old model simply involves selling the intellectual property in the form of patents, so this is an area that could profitably be optimized. Perhaps other assets of programs—such as aggregated structure-activity data, information showstoppers in the projects, could be examined for sale.
I plan to take Professor of the Practice Deborah Nightingale’s class, Integrating the Lean Enterprise, to learn tools and techniques to address the challenges of changing the culture. Applying lean optimization in the discovery process will help identify areas to increase speed and reduce iterations. For instance, if there is a two-to-four-week cycle time for a team of chemists to test new variants of a molecule, lean principles can be employed to determine the real time of the value-added portion of the cycle—which might be just one to two days for each week. Therefore if you double the number of cycles in each time frame, you will be making fewer molecules and utilize a smaller amount of the overall cycle time while producing richer information content (see diagram).
I believe that all of this can ultimately impact health-care policy. At present Institute Professor Robert Langer at Harvard-MIT Division of Health Sciences and Technology (HST) is looking at renewable organs and tissues. This raises huge social issues because as people begin to live longer there will be a significant strain on resources. There should be someone with a systems perspective involving engineering, management, and social science at a very high level looking at the potential problems.
I’m planning to take more classes at HST on the economics of health-care industries, new disruptive technologies, and how to evaluate biotech companies from a venture capital perspective. I also want to take a class at Harvard School of Public Health, where they are using systems thinking as the backdrop for looking at the current problems in Medicare, Medicaid, long-term care, health care, and insurance. (Publishers note: Cross-registration between MIT and Harvard at the graduate level can enhance the overall SDM educational experience.)
I’m now looking at almost everything with a systems lens and I believe I will apply my SDM learnings for years to come.