Common Factors Among Successful Practices in the OCM

Practices participating in the Oncology Care Model have now received performance results for 2 periods, and many practices were surprised that they didn’t perform as well as expected. In general, some of the practices have found it difficult to predict success in the model, but there have been positive results from the OCM.

The US Oncology Network has 15 practices participating in OCM, and Marcus Neubauer, MD, chief medical officer, admitted that it was a little surprising to him to see which practices didn’t perform as well as others. And learning why some practices succeeded while others didn’t is important in order to make changes that will improve performance going forward.

“And I think it’s been somewhat difficult to figure out why practices get certain results, and, frankly, I think it’s been hard for [the Center for Medicare and Medicaid Innovation] to tell us,” he said.

However, the practices that performed well under the model often had some things in common to drive that success. Neubauer pointed to changes such as expanding clinic hours so people aren’t going to the emergency department, hiring more advanced practice providers, and incorporating clinical pathways.

Charles Saunders, chief executive officer of Integra Connect, added that OCM practices also seem to be performing better than other practices in the community that are not in the OCM, which is measured by the trend factor.

“I would say the increase in the positive trend factor adjustments suggests that they’re holding the line on prices a little bit more than the surrounding communities,” he said, attributing the difference to those practice changes that Neubauer highlighted.

While only a few practices were “in the money,” Saunders said, the difference between practices’ actual cost and the target price improved about 41% from performance period 1 to performance period 2.

“That is, they’re losing less money, and some practices have actually gone into positive territory with substantial performance-based bonus,” he explained.

Still, he believes practices can perform better in the OCM if CMS does a better job of predicting target price, which is constructed using a number of factors, including the number of comorbidities a patient has. Comorbidities are based on Hierarchical Condition Category coding, but oncologists do not do a very good job of documenting these comorbidities, Saunders said.

For example, of the practices Integra has looked at, almost 40% of patient cases had either no or only 1 comorbidity. For a patient with lung cancer between the ages of 65 and 85, it’s incredibly unlikely that is really the case.

“So, that means that people are undercoding, and the downside of that is that CMS then projects the risk adjustment lower than it should be and the target price you get is unrealistically low,” Saunders said.