 We were able to actually do quantitative analysis to determine that the best timing of a post-hospitalization visit was at seven days. And by doing that, we affected the care of 70,000 members a year who are discharged from Kaiser Permanente hospitals and our readmission rates are coming down in response. So much of medicine historically is passed down from generation to generation. And we do things just because that's the way they were done before without really knowing if they improve the lives of people. And so I'm very data-driven and data-oriented and it just makes complete sense to me that we should use data to find out whether what we're doing actually works or not. And so we ask questions that were important to patients and the doctors who take care of them and to help to generate the body of knowledge that would allow other physicians to practice better medicine. The CERT or the Care Improvement Research Team is our research operations partnership and the overall goal is to improve quality and affordability. At any point in time we have somewhere in the neighborhood of 20 or 25 different projects that are ongoing and those results are most important to informed decision-making by local leaders but we also share those results with the rest of the world. Seven years ago there were virtually no programs in other hospitals or health systems around the country and the more typical situation is that research is separate and hospital operations are run completely independently. So the idea of bringing researchers and clinical leaders and administrators together was really novel and groundbreaking.