Home Health & Hospice Week

MedPAC model riddled with problems, industry reps say.

You'd better brush up on your OASIS accuracy because your Medicare reimbursement could ride on it even more.

Medicare would base home health agency pay for performance payment rates on 20 OASIS indicators, if the Medicare Payment Advisory Commission gets its way. That's one of the P4P recommendations MedPAC is considering for its June report to Congress on the matter, according to the Commission's January 10 meeting.

HHAs would receive a quality score based on aggregating those 20 OASIS measures, which include toileting, ambulating and managing oral medications, notes the American Association for Homecare.

MedPAC staffer Sharon Cheng proposed setting a reward pool for the top 20 percent of agencies and a penalty pool for the bottom 20 percent. The Commis-sion could also recommend incentive payments for agencies that improve the most.

Dilemma: MedPAC would have to decide whether to recommend setting the reward and penalty thresholds ahead of time based on historical data so agencies would know what they would be shooting for. Otherwise agencies would find out the thresholds well after the time period because the 20 percent lines would be calculated after claims had processed.

Smaller HHAs could have a tough time being measured under this system, Cheng acknowledged. To combat having a few patients skew an agency's data, MedPAC is considering recommending:

• "confidence intervals" that would give agencies a range instead of a hard and fast outcome score, based on statistical significance.

• aggregating data over two years.

• allowing small agencies to form voluntary associations with other similarly situated providers to pool their outcome data.

P4P Proposal Hurts Small Providers

Industry representatives greeted the proposed P4P structure with dismay. "The model P4P system displayed by MedPAC staff fails on a number of fronts to meet the P4P principles that the National Association for Home Care & Hospice and other organizations in the home care community have established," NAHC blasts.

The proposal relies solely on OASIS data without any consideration of process measures. "Up until this meeting, it had been NAHC's impression that MedPAC recognized the serious problems attendant to relying simply on OASIS outcomes data," the trade group tells members in its newsletter.

"There are real issues about using OASIS outcome measures in a rewards and punishments context," warns Bob Wardwell with the Visiting Nurse Associations of America. There are no "good answers" yet to a number of disturbing questions about OASIS measures, says Wardwell, a former top Centers for Medicare & Medicaid Services official.

Confidence in the smaller number of measures chosen for CMS' proposed P4P demonstration project is hard enough, notes consultant Judy Adams with Charlotte, NC-based LarsonAllen. "Adding more outcomes under the MedPAC proposal is not reasonable or practical," Adams tells Eli.

And gaming is a significant problem--for example, scoring patients low on admission and high on discharge, Adams adds.

Other problems: Until CMS can "separate outcomes for Medicare recipients from Medicaid and come to grips with the influence of other health care provider practice patterns on home health outcomes, most small providers will have a difficult time qualifying for incentive payments," Adams believes.

CMS has done a better job addressing some of those issues in its P4P demo proposal, NAHC and Adams say.

Commissioners Voice Doubts

Commissioners were unsure about the proposed system as well. Foremost, they worried that a P4P payment pool of only 1 to 2 percent wouldn't influence a provider group with 17 percent average profit margins and 25 percent of agencies experiencing margins over 27 percent. "If you can get 27 [percent] by taking certain steps to control your population ... that's going to be much more powerful than a 1 or 2 percent pool," Commissioner Ralph Muller noted in the meeting.

CMS may have to fix the home health prospective payment system's inherent distributional flaws before a P4P incentive can be effective, several commissioners suggested.

Basing the P4P rewards and penalties on the 20 OASIS indicators also leaves chronic care patients unaccounted for. "When I look at the list of measures we have for home health performance, I feel like we've got a lot about recovery and rehabilitation and less about chronic care management," said Commissioner Bill Scanlon. "So if an agency is small and dealing more with chronic care management, then I think they are potentially disadvantaged by the system."

And the commissioners too worried about "gameability" of the data. "We need to be sure that the system has integrity in terms of the data that we're using to make rewards," Scanlon cautioned.

Measures questioned: Using measures based on OASIS data may just be too simple for the sophisticated P4P system proposed, commissioners suggested.

In light of all those problems, the home health industry may not be a good place for CMS to start with P4P, commissioners said. "I wouldn't make [home health] the mainstay of where [a P4P effort] begins," Muller said.

P4P may be a lower priority than agencies realize, says consultant Tom Boyd with Rohnert Park, CA-based Boyd & Nicholas. Hospitals and physicians, with their larger patient and reimbursement volumes, are likely to see P4P first, he predicts. That's especially true now that the party that controls Congress, the Democrats, seem less enthusiastic about the concept of P4P.

Stay Tuned For P4P Developments

"The MedPAC approach is not the silver bullet to make P4P a valid, fair and reliable value-based purchasing system," Wardwell sums up.

Unless the Commission changes its P4P structure significantly, "NAHC will not support its recommendations," the trade group says.

But given the commissioners' many reservations, significant changes could still be ahead. The commissioners "sure did tease out many of devils in the details that could make this system not work as intended," Wardwell observes.