Dr Michael Kolodziej Reflects on OCM and What Might Come After

The Oncology Care Model (OCM) is not a sustainable model, but while it finishes out the 5-year pilot, CMS will likely be fine-tuning bundled payments in medical oncology, said Michael Kolodziej, MD, vice president and chief innovation officer at ADVI Health, Inc.

Transcript

Based on what we’ve seen with the first 2 performance period results, does the Oncology Care Model look like a sustainable model? Do you think it will inform what comes next?

Is it a sustainable model? No, it’s not a sustainable model. There’s no way that even the majority of practices are going to choose to take 2-sided risk. Now, I would have said that no one’s going to take 2-sided risk, but it is clear that some practices are going to take 2-sided risk. But they’re the minority. And we already know that the practices participating in the OCM are highly selected. So, the concept that this model will be rolled out to everybody and everybody will take 2-sided risk in the model is I think just ridiculous, if you want to know the truth.

But let’s get to question 2. Question 2 is much more important, because it asks whether or not what we’ve learned from the model could inform the next payment proposal. And I say this with just a little asterisk—because so much stuff is coming out of DC right now that any one of those things could so remarkably reshape the way cancer care is delivered in this country—that it may become moot whether or not there is a successor.

But let’s say that the [International Pricing Index] is too complicated, and we never implement it—we just don’t do it. And let’s say nobody wants to be a CAP [Competitive Acquisition Program] vendor, because they can’t make money out of it—so, CAP goes away. So, in that case, this total cost of care model, coupled with all the information they’ve gotten from the registry reporting and what have you, makes it possible, I believe, for them to design clinically meaningful bundled reimbursement models and just say to a practice, “This is how much we pay for early-stage hormone receptor—negative breast cancer, take it or leave it.”

So, not unlike what they’re doing in radiation right now. The reason this has never been possible in medical oncology is because medical oncology is too clinically heterogeneous. You can’t just have a breast cancer model. Now, I don’t know if you need 50 breast cancer models or whether you need 5—I tend to believe you probably need closer to 5. That’s why the radiation episodes make sense. Because there’s a limited number of tools in their toolbox, and the choice of tool is driven by clinical features of the patient.

I think Medicare can do the same thing. And don’t think—I know—they could not have done it, even if they wanted to—and they did want to—before what they did with the OCM, but now they know all kinds of stuff. They know: what are the cost centers for a typical patient? They know what a reasonable [emergency room] utilization rate looks like. They know what a reasonable hospitalization rate looks like. They know about each of those things for a different subset of patients, including different age, different geography, different comorbidities. They’ve got all that stuff. I mean they’ve got so much stuff that they really could build this and execute it if they so choose.

We won’t see this now for a couple of years. I think because some practices will accept 2-sided risk. Let’s say, a third of practices accept 2-sided risk—I don’t think that’s very far from correct, I think that’s probably right—then those third will see the end of this OCM pilot, which is a 5-year project, but while they’re in year 4/5, they’re going to be fine-tuning and developing these bundles, I think. I think that’s what’s going to happen.