This Spring, Caravan Health hosted a panel of experts to discuss the future of accountable care. This was a chance to dig deeper into the what makes a successful accountable care model as highlighted in the recent Health Affairs
For the first time, we heard from Colleen Norris, an actuary who analyzed the ACO data, as well as Tim Putnam, a hospital CEO with long experience in the Caravan Health collaborative ACO model, and Stephen Shortell, a UC Berkeley Dean Emeritus who has conducted research on ACO leadership. The panel was moderated by Caravan Health Senior Vice President Tim Gronniger.
Caravan Health CEO and Founder Lynn Barr started off the webinar by with a recap of how Caravan Health developed the collaborative ACO model
to coordinate efforts across multiple independent hospitals and physician groups. The company got its start working with rural hospitals and created a collaborative model not by choice, but by necessity. There were simply not enough hospitals and physicians serving enough lives in these sparsely populated areas to start an ACO on their own. Lynn and other early pioneers brought those hospitals together to transform care and work toward value-based payment. That model of collaborating across otherwise unrelated health systems became an area of expertise and paved the way for future success.
Today, Caravan Health has helped form and manage 38 ACOs. As described in Health Affairs
and discussed on the webinar, we have learned that ACOs must be at least 100,000 lives to predictably start showing positive results. As a result, several Caravan Health ACOs will merge into larger, collaborative ACOs starting next year. This timeline aligns with the changes in risk coming to all participants in the ACO programs. In track 1, without downside risk, ACOs could afford to be a bit forgiving of those providers who couldn’t manage quality improvements. In the risk-bearing model required of almost everyone in the next few years, ACOs must raise their game – the collaborative model facilitates that transformation.
Colleen Norris, an actuary with Milliman, gave a more detailed picture of the data showing the effect on size on ACO outcomes
. She explained that ACOs face a similar dynamic to health insurers. Just like health insurers, ACOs find more predictable results with larger populations. Both ACOs and health insurers thrive on stability, on having enough lives so that a few high-cost patients don’t put the whole organization in danger.
Colleen showed a graph illustrating that an ACO of 10,000 lives with expected savings of 1 percent has a 1 in 3 chance of performance year costs exceeding benchmark year costs. This creates an expected loss of 0.5 percent of the benchmark – a very high rate of loss for an organization that is doing everything right. Colleen said that one notable finding is that at for an ACO of 100,000 lives, the chances of experiencing losses greater than 2 percent is “vanishingly small.” That means there is a very slim chance of having to write a check for losses.
This graphic from Milliman shows how the outcomes for large ACOs fall in a much narrower range than smaller ACOS. That narrow range is a great illustration of the stability and predictability that ACOs need.
This graph shows the range of expected outcomes for ACOs of different sizes. The lightest colors in the center of each bar illustrate a 50 percent level of confidence achieving that outcome. The next lightest colors show 75 percent confidence and the darkest colors show 90 percent confidence.
Tim Putnam, the CEO and President of Margaret Mary Health in Batesville, Indiana, is a longtime client of Caravan Health. Tim offered the perspective of a small hospital working to collaborate with other small hospitals, all of which want to remain independent but are counting on each other to do the hard work of practice transformation. He talked about how, in the community of small hospital CEOs, it was easy to tell who was serious about what they call the “volume to value transition”. This close-knit group of providers learned quickly that the key is the nurses and others delivering care
One conclusion was unmistakable – even when all the providers were doing the hard work of quality improvement, results were variable. It was difficult to accept that working hard wasn’t consistently creating results. The data from Colleen and the team at Milliman explains it – the population is simply too small. Random and local factors – like a worse than average flu season – were impacting the results.
Webinar participants also heard an academic perspective – Dean Emeritus of the UC Berkeley School of Public Health, Stephen Shortell, talked a bit about the role of population health management in successful accountable care arrangements. One question is who should run ACOs. Should it be hospitals, physicians, or a hybrid? Steve’s research shows that it may not matter – there are not significant differences between those three types of ACO leadership. There is some variation within the groups of hospital, physician, or hybrid models. The most important factor seems to be culture. Those that are prepared for a change of culture from fee-for-service to value-based payment
, and those who are ready for data-driven decision making, are succeeding regardless of who is running the ACO.
Thanks to all who participated in our panel webinar.