Reimbursement Remedy Backfires: Stress Tests Have Shifted From Private Practice to Hospitals, Driving Up Costs
Decade-old financial incentives created by CMS to cut down on self-referrals and inappropriate tests have not had their desired effect.
ANAHEIM, CA—Reimbursement changes instituted by the Centers for Medicare & Medicaid Services (CMS) 10 years ago to reduce self-referrals for stress testing appear to have backfired.
According to a new analysis, the overall rate of stress tests has decreased since CMS reduced Medicare fee-for-service (FFS) reimbursement levels for office-based cardiac testing, according to a new study covering the years between 2005 and 2014. However, the decline was greater for Medicare Advantage beneficiaries than for FFS beneficiaries, and most FFS testing simply switched from office- to hospital-based settings, likely due to the higher rate of reimbursement.
Beginning in 2007, CMS changed its reimbursement rates for certain cardiovascular imaging tests, whereby those in outpatient setting were cut while reimbursements for the same procedures done in-hospital went up. Interventional cardiologists have expressed concern over the years regarding the unintended consequences of these financial incentives, yet the CMS rules have remained.
“It was known at least anecdotally that there was a lot of movement of cardiovascular practice from these outpatient independent settings to owned practices within healthcare systems working within hospitals,” lead study investigator Frederick Masoudi, MD (University of Colorado Anschutz Medical Campus, Aurora), told TCTMD. “What wasn't as clear was the extent to which these changes in reimbursement affected practice, because ultimately . . . part of it was the hope that sort of the overall reimbursement cuts were made with the [intention] that less inappropriate testing would be done—sort of a blunt instrument.”
To look at this specifically, Masoudi and colleagues analyzed a random 5% sample of Medicare FFS patients as well as patients from three integrated health systems in the Medicare Advantage sector who were not subject to the reimbursement changes. Their results, presented in a poster session at the American Heart Association 2017 Scientific Sessions, showed that the overall stress-test rate declined more for the Medicare Advantage population than the Medicare FFS group (49% vs 30%) between 2005 and 2014. This suggests that there is “no evidence that the reimbursement cuts reduced testing rates,” investigators said. Notably, reimbursement levels since 2007 have declined for office-based testing, but they have climbed for hospital-based outpatient tests.
While there was a decline in office-based testing in both cohorts, this reduction was much more pronounced in the FFS arm than it was in the Medicare Advantage sector. Also, there was an increase of testing rates in the hospital-based setting for FFS but not for Medicare Advantage.
The findings would suggest that the “perturbation” of the payment structure in the FFS sector didn't induce greater declines in stress testing compared with the control group of Medicare Advantage patients, Masoudi said. “But at the same time, not only were there not additional declines that were induced by this change in reimbursement structure, there was a marked change in the venue in which these tests were performed—a shift toward a higher cost environment that was induced by these reimbursement changes.”
There was a marked change in the venue in which these tests were performed—a shift toward a higher cost environment that was induced by these reimbursement changes. Frederick Masoudi
“There's no reason to believe that a test that’s performed in an outpatient setting is somehow less valuable or less informative than one performed in a hospital,” he continued. “It's just more tests that were more expensive to the system with these changes in payment.”
Looking forward, Masoudi said this issue is “part of this larger dialogue about a transition from payment for volume to payment for quality. What I hope would happen is there would be in a way that's clinically sensible and accounts for the challenges of clinical practice that rewards the use of testing in appropriate situations and incentivizes the use of testing based on guidelines rather than again this FFS reimbursement.”
When systems are designed to “pay per service delivered, you get services delivered,” and “when you pay for the quality of services delivered, you get quality, provided that the measurements that are underlying that payment scheme really do reflect the quality of care that's delivered,” he observed.
‘Validation of What Everyone Told CMS Would Happen’
These findings are “no surprise,” Ralph Brindis, MD, MPH (University of California, San Francisco School of Medicine), who was not involved in the study, told TCTMD. “This is again validation of what everybody actually told CMS would happen.”
Brindis referenced a recent collaborative report released by Avalere Health and the Physician’s Advocacy Institute regarding the implications of hospital employment of physicians, which grew by 49% between 2012 and 2015, on both Medicare and its beneficiaries. Physician employment has resulted in more than $3.1 billion in increased costs between 2012 and 2015, according to the report, and Medicare and the program’s beneficiaries were responsible for $2.7 billion and $411 million, respectively.
“You could argue that maybe [hospital employment of physicians is] a methodology to have coordination of care or coordinated referral mechanisms” and that physicians have also benefitted in terms of having a reduced administrative burden, but the financial incentives cannot be ignored, Brindis said. “When you have physicians in an integrated system where all of a sudden the same office which was billing in a private physician's office a certain amount . . . [is] incorporated in a hospital-based system [and] is charging substantially more for the same service delivered in the same location, this represents huge increased costs.”
Over time, CMS might “begin to catch up” and start reimbursing less for hospital-based procedures, he commented, “but they have a long way to go in terms of dealing with this.”
A Call for ‘Better Stewardship’
Masoudi believes that CMS should “be very interested in understanding what the impact of these changes were on behavior [and seeing] whether or not this is something that reflects what they wanted, or whether or not this is an unintended consequence. My guess is it's more the latter—that maybe there wasn't an expectation that it would have such marked effects on the way the tests were ordered.”
Brindis, in turn, said that moving to an environment with more bundled payments related to disease management or a capitated system, “where there's internal motivation to deliver care, particularly if there's accountability related to quality,” is a “much better way of dealing with these things than the way that CMS has done in the past.”
The conversation also needs to revolve around physician quality. The study results are “important for practicing cardiologists to see and . . . reflect on the appropriateness of the tests that we order,” Masoudi said, noting that they weren't able to characterize the appropriateness of any of the tests they analyzed. “I do think they give us reason to take a pause and a step back and make sure we're doing the right thing for the right patient at the right time,” he said.
If future studies could accurately assess the appropriateness of tests being conducted, “that would certainly be of great interest,” he added. As chief science officer of the National Cardiovascular Data Registry, Masoudi said that “one of the things that we're doing with the registry programs is integrating appropriate use criteria into the programs, so that we're able to measure appropriateness and report it back at scale to large numbers of programs and get benchmarks nationwide in terms of what centers are doing.”
In addition, Brindis said that he hopes clinicians begin “utilizing some of the infrastructure tools, for example [like] the American College of Cardiology has put out, in terms appropriate use of imaging and help encourage CMS and payers to achieve value.”
We have to be better stewards of our healthcare dollar. Ralph Brindis
And they need to “be appropriate utilizers of imaging tests or angioplasty,” he added. “The professional societies and responsible clinicians can help lead the charge in increasing value in care and quality for our patients. We have to be better stewards of our healthcare dollar. This abstract is a perfect example of the perverse incentives that have led to poor value.”
Photo Credit: Extracted from: American Heart Association. Answers by Heart: What Is a Stress Test?
Masoudi F. Effect of financial incentives on cardiac testing rates and location. Presented at: American Heart Association 2017 Scientific Sessions. November 13, 2017. Anaheim, CA.
- Masoudi and Brindis report no relevant conflicts of interest.