WASHINGTON -- Medicare should work harder on ferreting out the services it shouldn't be paying for because they either don't help patients or actually do them harm, several members of the Medicare Payment Advisory Commission (MedPAC) said.
"This is really important work because it's win/win," said commission member Rita Redberg, MD, of the University of California San Francisco. "Besides the billions that people are spending, people are being hurt. There is nothing good about getting tests you don't need and aren't going to help you feel better."
In 2014, Medicare spent anywhere from $2.4 billion to $6.5 billion on these types of services, which are known as "low-value care," MedPAC staff member Ariel Winter said during a commission meeting on Thursday. To arrive at those numbers, commission members analyzed Medicare claims data from 2012 to 2014, using 31 measures of low-value care developed by Aaron Schwartz of Harvard University and colleagues. The range in the estimate is because they looked at both a broad and a narrow version of each measure.
Approximately 37% of beneficiaries received at least one low-value service in 2014 if the broader measures were included, Winter said. In terms of volume, most of those services fell under two categories -- imaging and cancer screening -- while most of the spending on low-value services came in the areas of cardiovascular tests and procedures, and other surgical procedures, he said.
Medicare spending on imaging for nonspecific low-back pain -- one of the low-value procedures looked at -- amounted to $232 million in 2014, and occurred at a rate of 12 such procedures per 100 patients, Winter continued. Other low-value procedures included prostate-specific antigen (PSA) screening for men ages 75 or older ($79 million, 9 procedures per 100 patients), colon cancer screening for older adults ($405 million, 8 per 100 patients), and spinal injections for low-back pain ($1.26 billion, 6.6 per 100 patients). Use of low-value care varied greatly by geographic area, the researchers found, with five out of the top 10-highest areas being in Florida.
The researchers also looked to see whether some of the newer payment models -- such as accountable care organizations (ACOs) -- might result in less use of low-value care. Winter cited a 2015 study by Schwartz and colleagues showing that Medicare's Pioneer ACOs spent 4.5% less on low-value care compared with a control group of providers.
Reforming Medicare's payment and delivery system -- such as by encouraging formation of more ACOs -- is one idea commissioners could consider recommending as a way to reduce low-value service use, Winter said. Other possibilities include changes to quality measurement, changes to Medicare coverage policies, and increasing patients' stake in the game by using cost-sharing and conferring with their providers about which services are necessary.
Of those four options, "probably coverage policy is not a strength of Medicare, given the political environment and the difficulty of using well the recommendations coming from the U.S. Preventive Services Task Force," said commissioner Paul Ginsburg, PhD, of the Brookings Institution, a left-leaning think tank here. "To me, payment/delivery system reform gets at this indirectly by providing incentives for the organization to act in [the right] way ... so the dollar potential in that one is actually strongest."
Before anything is recommended for payment or delivery system reform, MedPAC should wait and see what the results are from the payment reforms being implemented through the Medicare Access and CHIP Reauthorization Act (MACRA), said commissioner Alice Coombs, MD, a physician at South Shore Hospital in Weymouth, Mass. "That's looking at costs, looking at resource utilization -- why layer another layer on top of providers? Why not wait and see how we change patterns of low-value services through whatever is already on the table?"
Commission member Craig Samitt, MD, of the health insurance firm Anthem in Indianapolis, Ind., said he would focus first on payment and delivery system reform. "I'm not sure the other three will be powerful or effective enough."
The reasons for using low-value services are multi-factorial, "but one thing that comes out of this [analysis] is that there has got to be some effective consumer education; I don't really know how that's going to happen," said commission member William Hall, MD, of the University of Rochester, adding that the problem "isn't going to go away."