The effects of predetermined payment rates for Medicare home healthcare

Health Services Research, Oct, 1997 by Randall Brown, Barbara Phillips, Christine Bishop, Craig Thornton, Grant Ritter, Amy Klein, Peter Schochet, Kathleen Skwara

Between 1983 and 1994, total Medicare outlays for home healthcare increased from $1.6 billion to $13 billion, raising home health's share of total Medicare costs from 2.8 percent to 7.8 percent. Costs per beneficiary increased from $78 to $352 in 1993 dollars during this period. This rapid increase warrants exploration of alternatives to Medicare's current cost-based method of reimbursing home health agencies.

Little incentive exists under cost-based reimbursement for agencies to provide Medicare beneficiaries with services in a cost-effective manner. Although the current system imposes a limit on the amount Medicare will pay per visit, agencies below the limit have no incentive to reduce their costs. While controlling home health costs today probably will require a payment methodology that also encourages fewer visits, controlling growth in the average cost of a visit will be important under any payment method. For example, the Medicare Preservation Act of 1995 (HR2425), passed by the U.S. House of Representatives, called for paying agencies a fixed price per visit, putting a ceiling on the average cost per episode of care, and sharing savings with agencies that hold costs below the ceiling. Agencies' ability to control cost per visit will influence whether they survive under such a ceiling and the amount of the savings to be shared with the Health Care Financing Administration (HCFA).

In this article we report the findings from a five-year study of a demonstration program designed to test an alternative method of paying home health agencies that does provide an incentive for efficient production of services to Medicare beneficiaries.

THE DEMONSTRATION

DEMONSTRATION DESIGN

Under HCFA's per-visit rate-setting demonstration, which began in fall 1990, home health agencies were paid prospectively determined rates for each of the six types of home health visits (nursing; physical, occupational, and speech therapy; home health aides; and medical social workers), with adjustments for sizable changes in volume. Profits and losses were shared with HCFA. For each type of visit, HCFA set the prospective rate for an agency participating in the demonstration at the agency's cost per visit of that type in the base year (the agency's fiscal year preceding entry into the demonstration), adjusted for expected inflation (rates were increased by the same amount for all agencies, about 5 percent each demonstration year). If the agency's total base-year cost exceeded the legislatively mandated cost limit for that year, the base-year rates for each type of visit were reduced by the ratio of the agency's base-year limit to its actual base-year costs.

Payments were retroactively adjusted at the end of each year for demonstration agencies that experienced (1) sizable changes in the total number of Medicare visits rendered relative to their base year, or (2) large profits or losses. Per-visit reimbursement rates were reduced 1 percent for agencies whose total number of Medicare visits increased by 10 to 20 percent. Rates were increased by 1 percent for agencies whose volume decreased by 10 to 20 percent. Each additional 10 percentage point change in volume added an additional 1 percent to the change in the reimbursement rate, up to a maximum 5 percent change for agencies whose volume increased or decreased by more than 50 percent. These adjustments for volume were intended to reflect the effects of economies of scale on agencies' average costs and to discourage agencies from increasing their volume. The profit- and loss-sharing provision required agencies to return a specific percentage of their profits on Medicare visits, if profits exceeded 5 percent of Medicare-allowable costs.

This payment methodology differs substantially from the current cost-based reimbursement system. The current system pays agencies as claims are submitted, then performs an end-of-year reconciliation between the agency's costs of providing care to Medicare patients and the amount already paid by Medicare.

The demonstration, which was open to all nongovernmental home health agencies that had been in operation for at least three years in five states (California, Florida, Illinois, Massachusetts, and Texas), attracted only 47 agencies, far fewer than the 67 that HCFA sought. In each state, the participating urban agencies within each of three strata (freestanding proprietary, freestanding voluntary or nonprofit, and facility-based agencies) were randomly assigned to treatment or control status as they entered the demonstration. Rural agencies formed a separate stratum. Treatment group agencies were paid according to the demonstration rules described earlier; control group agencies were paid under the usual cost-based method. Twenty-six of the participating agencies were assigned to the treatment group, 21 to the control group. (The unequal numbers were due to an odd number of agencies in some strata at the time of randomization.) Each agency participated in the demonstration for three years, with participation beginning at the start of the agency's next fiscal year after application. The first agencies began demonstration operations in October 1990; the last began in October 1991.