As the health care economic framework shifts from paying for volume to paying for value, the ability to quickly vet and test clinical interventions becomes increasingly important.1Within a value-based framework, it is no longer enough to understand that an intervention improves quality. Instead,(FHP) is compelled to focus on the financially efficient delivery of quality. Perhaps, this is best expressed in the abstract through the simple equation:

As a company, Ģý North America (FMCNA) has a long and robust history of generatingĢýand, in many cases, publishingĢýinsights garnered from the companyĢýs massive collection of clinical data. More recently, FHP risk arrangements have provided the organization with access to medical claims data. Combining claims data with clinical data creates an opportunity to generate valuable insights with respect to the financially efficient delivery of quality. To formalize an approach to this opportunity, FHP has created the Clinical Interventions Lab (the Lab).
The Lab is made up of individuals from across the enterprise, collectively working together to rapidly vet, prioritize, and pilot clinical interventions within the risk populations FHP serves. The Lab's foundational principles include:

At a high level, the process in place involves five steps (Figure 1):

Figure 1| Clinical Interventions Lab Framework
During its first six months of operation, the Lab has generated several valuable insights. The first is the value of data-driven decisions. One of the early interventions FHP considered was the deployment of intensive diabetes education and management.
A third party shared convincing evidence that poorly controlled diabetic patients were more expensive to care for than well- controlled diabetic patients. They were able to demonstrate that their diabetes management services not only improved control, but also substantially reduced the total cost of care. The FHP team was intrigued by the possibility of deploying that service to the at-risk ESRD diabetic population.
Much to the teamĢýs surprise, the sniff test yielded no difference in total cost of care among well-controlled and poorly controlled ESRD diabetic patients (Figure 2). The results identified in the general population were not present in the ESRD population.
ThatĢýs not to say glycemic control should be ignored in ESRD, but the Lab could not justify piloting an expensive service to advance the value equation. This is an example of using data to drive decisions, and it supports the Ģýfail early and oftenĢý principle as well.
A second valuable insight relates to the LabĢýs approach to pilots. Each pilot has a champion who is committed to testing the hypothesis that the Lab is examining with a specific clinical intervention. The champion works closely with a project manager, paying attention to the following operational pilot touch points:

Figure 2| Total cost of care per month per member per month (PMPM)versus average HgbA1c
One final insight. In order to move quickly, the LabĢýs approach to pilots more closely resembles an agile methodology as opposed to the classic waterfall approach to project management.2
FHPĢýs intent is to avoid investing substantial initial capital in developing a complex project plan that requires rigid adherence. Instead, the intent is to plunge in quickly, making adjustments to the pilot based on what is discovered along the way. FHPĢýs preference is to fail early rather than late and to iterate as the project progresses based on what is learned in the field.

Dr. Terry Ketchersid has clinical oversight over Ģý North America's Integrated Care Group, leading value-based care initiatives. He received his BA in chemistry from Austin College, his executive MBA from Duke UniversityĢýs Fuqua School of Business and his MD degree from the University of Texas Southwestern Medical School.
Rapid Testing of Clinical Interventions: How to Be ĢýClinically NimbleĢý
by Terry Ketchersid, MD, MBA