At a June 13 congressional hearing, Elizabeth Fowler, Ph.D., J.D., director of the Center for Medicare and Medicaid Innovation (CMMI), was pressed to explain why so few of CMMI’s alternative payment models have produced cost savings.
In a meeting of the Health Subcommittee of the the U.S. House Energy and Commerce Committee, Cathy McMorris Rodgers (R-Wash.), chair of the House Energy and Commerce Committee, began by outlining her concerns.
CMMI was created to help improve how Medicare and Medicaid pay for healthcare and to be an engine in our drive towards value-based care, Rogers said. “CMMI was given a 10-year, $10 billion budget and extremely wide-ranging authorities with limited built-in congressional oversight. The only directives Congress gave CMMI were to achieve two goals: lowering the cost of delivering care and improved patient outcomes.”
Over the last decade and a half, CMMI has tested over 50 models to accomplish both those goals. When CMMI was created, the savings it was projected to generate were to be used to offset spending by the Affordable Care Act, Rogers continued. Originally, CBO estimated that CMMI would save $1.3 billion over its first decade of operation. That same model also projected CMMI would save as much as $77.5 billion in its second decade from 2020 to 2023. “However, when CBO looked at the actual results in a September 2023 report, the disparity between those expectations and the reality proved to be staggering. Instead of reducing spending by $1.3 billion in the first decade, CMMI increased spending by $5.4 billion. For the second decade, instead of saving $77.5 billion, CBO is now projecting CMMI to increase spending by $1.3 billion. I have a hard time believing any objective observer could look at the results thus far and describe CMMI as a success. So how do we move forward?”
Rep. Anna Eshoo (D.-Calif.) asked Fowler directly why CMMI is not generating savings. “Could you share with us what metrics CMMI uses to determine whether a model is successful and not just limited to savings?"
Fowler responded by saying: “I would say first of all that we have learned something from every model that we have tested. And I would also say that the innovation process itself is sometimes unpredictable. One factor that has affected our ability to generate savings is the fact that most of our models are voluntary, and this was an area that the Congressional Budget Office also pointed out. And when you have a voluntary model where providers can come in, if they think the terms look favorable or they can exit if they think the terms may turn against them, or they weren't performing as well as they thought they can, they can drop out of the model. And so that has led to risk selection, which has undermined our ability.”
Fowler said CMMI sees all of its models as successful in one sense in that CMMI is learning something regardless of the ultimate assessment. “We've also been spending a lot of time thinking about the impact on quality, and we've laid out a whole quality pathway. The statute gives us authority to examine both the quality of care provided and the impact on savings, and we are really leaning into the quality improvement angle, particularly when it comes to patient-reported outcomes and the patient care experience. And we're trying to learn more from this approach to thinking more broadly about quality.”
When it was her turn to ask questions, Rogers noted that in her opening statement, she discussed how far off the projected impact of CMMI has been compared to reality in terms of saving our healthcare systems money.
Here is a summary of the exchange between Rogers and Fowler.
Rogers: “Dr. Fowler, do you agree that reducing cost is fundamental to CMMI’s mission?
Fowler: That is a statutory mission outlined in in the statute that created the Innovation Center.
Rogers: Thank you. Despite having increased spending over $5 billion across the first decade of existence, in the CMMI 2021 strategic refresh and recent articles and op-eds and especially the recent launch of the so-called quality pathway, it seems there's an explicit shift in focus away from trying to reduce spending. Where does reducing spending rank in terms of your priority list?
Fowler: Reducing spending is our statutory mission, and we remain committed to that goal. Every model that we test goes through a rigorous review and evaluation and clearance process where our actuaries look very closely at what we're proposing. We wouldn't test anything that didn't have the potential to save money at the outset, as we make those announcements and as we implement the model.
Rogers: Many have suggested that most, if not all, of the new models CMMI has rolled out since 2021 are unlikely to save money based upon their designs. Are you willing to commit to shutting down any model that is failing to show signs of net savings after its first two years so we can correct CMMI’s failing fiscal trajectory?
Fowler: It's a great question. And when we do get information about a model's performance, sort of mid-year, we make adjustments accordingly. And in a couple of occasions since I've taken on this role, we have had to shut down models early. The emergency triage, transport and treat model is one where we didn't have enough participants and it didn't look like it would be generating those savings. Another model was the Medicare Part D modernization model, which also, unfortunately, we had to end early.
U.S. Rep. Michael Burgess, M.D. (R-Texas) said that he also sits on the Budget Committee, and works with CBO to understand how the mark was missed so dramatically. He asked Fowler: “From your perspective, what would be a reasonable savings target for CMMI over this next decade of CMMI’s existence?
Fowler responded that she is not sure she could come up with an estimate. “I think it depends on the rollout of the models and who comes in. And as I mentioned earlier, I think the nature of innovation can sometimes be unpredictable. So we go in with the optimistic assumption that we will generate savings, and fully expect that the models that we've announced will do that.”
Burgess replied: “Everybody knows what your funding is going to be. It's mandatory funding. Even if we don't pass an appropriations bill, you still get your $10 billion for the decade. What do you think is a reasonable return on CMMI’s next $10 billion investment?”
Fowler said that, again, “this goes back to everything that we're learning from the models and that we go into the field with the expectation of savings. We have a plan to spend the money that's been allocated to CMMI, and we fully expect to see positive results from from the models that we're testing.”
