As CMS prepares to move forward with new models of value-based cancer care, a radiation oncologist, speaking at the Institute for Value-Based Medicine®, discussed her findings regarding whether accountable care models had an effect on cancer care spending.
As CMS prepares to move forward with new models of value-based cancer care, a radiation oncologist, speaking at the Institute for Value-Based Medicine®, an initiative of The American Journal of Managed Care®, discussed her findings regarding whether accountable care models had an effect on cancer care spending.
Miranda Lam, MD, MBA, presented a 2018 study1 looking at Medicare accountable care organizations (ACOs) for patients with cancer; Lam, of Dana-Farber Cancer Institute/Brigham and Women's Hospital, Boston, Massachusetts, said the results, which found an $11 difference per beneficiary, indicate models created for primary care may not be applicable for oncology purposes.
Lam examined national Medicare claims from 2011 to 2015; practices that became part of ACOs were identified and matched to non-ACO practices within the same region. She and her coauthors looked at whether the introduction of ACOs led to lower overall spending, lower spending by service type, and spending on cancer-specific services (radiation therapy, chemotherapy, hospitalizations, emergency departments, and hospice).
A difference-in-difference analysis showed that the introduction of ACOs had no meaningful impact on overall spending (-$308 per beneficiary in ACOs vs -$319 in non-ACOs; difference, $11; 95% CI, -$275 to $297; P = .94).
Within the 11 different cancer types examined, there was no change in total spending. And changes in spending and utilization did not meaningfully differ between ACO and non-ACO patients within cancer-specific categories.
Lam used the study as a prelude to her thoughts about CMS’ Radiation Oncology (RO) Model, which CMS proposed in July. Unlike the ACOs she studied, participation in the 2-sided risk model will be mandatory and, as of this writing, is slated to begin January 1, 2020.2 Currently, the plan proposes to randomize 40% of radiation providers and hospitals into the mandatory model, based on zip code.
There are some other differences between this model and previous bundle-type models as well. For instance, the payments will be site neutral between freestanding and hospital outpatient departments, and payments will be prospective (payment will be up front once an episode begins, for 90-day episodes of care) rather than retrospective.
The 2-sided risk model will be risk-adjusted and differ by geographic location, for 17 types of cancer. Three percent of the payments will be earned based on patient satisfaction and quality measures, which will include a care plan for pain, depression screening and a follow-up plan, an advanced care plan, and a treatment summary communication.
Lam acknowledged why CMS was taking the road of mandatory participation, saying “we might even benefit from having an initial sort of voluntary pilot phase to work out the some of the kinks.” But, Lam said, she knew that so few providers and practices join voluntary models.
In addition, CMS wants quality metrics, which some radiation oncologists might hesitate to supply. But Lam said this is not only CMS’ fault, “This is also on us as a field, in that we don’t have great quality metrics.”
“I think quality metrics in oncology are difficult—and definitely in radiation oncology, we often don't have [them]. We use process measures or structural measures [as] their surrogates for actual quality measures. The things that probably people care about are really hard to measure—local control, overall survival, those take time,” Lam said, adding that “they don't fit into that bundle as well.”
In addition, she said she “worries a little bit about potential skimping on care, certain areas where you may not get paid more for something that’s much more resource intensive, but maybe better for the patient.”
Another area of concern: innovation and clinical trials, which are not included in bundling, with the exception of trials funded by the National Institutes of Health. And, she noted, “Our field is really based in technology,” saying that during the past “decade or two radiation oncology has really made great strides in changing the way we deliver care that I think is really good for patients.”
There are other worries as well: the risk that without adjustments, that the bundles might widen disparities, which are already known to affect cancer care and outcomes more broadly. The other is that bundled payment would not work in palliative care, where the 90-day timeframe might disincentivize a provider from treating some patients.
Lastly, she said, not all cancers are the same, even if they seem to be. Yet as she understood the model as proposed, CMS would pay one price for treating lung cancer. “What does it mean to treat lung cancer? Does that mean stage I lung cancer, or someone who is stage III or IV, or someone whose lung cancer has metastasized to the bone?”
However, she said, “I think many of us agree that the way we pay for healthcare is not sustainable,” adding that she hopes CMS listens to stakeholders who have provided feedback about the RO model.