Health-risk adjustment helps ensure consumer choice in markets
Health-risk adjustment of premiums not only promotes fairer compensation among health plans but it also can help ensure that consumers have adequate choice of providers in their markets, experts say.
The Minneapolis-based employer coalition, Buyers Health Care Action Group (BHCAG), is risk-adjusting payments to providers based on the health status of their patients partly because it wants to attract niche players, the specialty providers, disease management programs and centers of excellence that typically serve high-risk, high-cost patients.
If we weren't doing risk adjustment, we wouldn't have some of these care systems at the table, said Ann L. Robinow, executive director of care systems and finance for BHCAG. Ms. Robinow spoke at a Jan. 29 meeting on health risk adjustment in Washington, D.C. The meeting, which included reports on health risk adjustment efforts in four states Minnesota, California, Colorado and Washington was sponsored by the Robert Wood Johnson Foundation and conducted by the Alpha Center.
Premiums traditionally have been adjusted for age, gender and geographic region, but theres growing interest in adjusting payments for the health status of enrollees, too. The idea is that this more refined risk adjustment will better compensate plans that enroll higher risk or sicker populations and reduce incentives for cherry-picking healthy enrollees.
Jan Malcolm, system vice president, public affairs, for Allina Health System, said her plan used to offer an open-access product to state employees. While the product was very popular, it became clear that it also attracted higher-risk, more costly enrollees. Because the state did not have a risk adjustment system that compensated Allina for those differences, the health plan was forced to discontinue the product, Ms. Malcolm said.
Better risk adjustment, Ms. Malcolm said, is needed to allay public mistrust of payment incentives under managed care. Whatever we can do to rebuild those incentives is critical, she said. We have to demonstrate that were willing to do risk-adjusted payments.
But there are many obstacles to widespread adoption of health risk adjustment. Many purchasers find the concept of health risk adjustment too difficult to understand and too costly to support. There also are serious data collection issues. Reliable information about patient diagnoses is difficult to get, particularly in outpatient settings, where it frequently isn't recorded in the claim at all. There is also the daunting statistical challenge of trying to reliably risk-adjust for small populations as well as accounting for comorbid diseases and geographic variations.
The next generation of risk adjustment will measure declines and improvements in health outcomes and adjust payments accordingly, said Richard V. Anderson, vice-president of health policy for Kaiser Foundation Health Plan in Oakland, CA. BHCAG hopes to create incentives for providers to improve their management of enrollees with chronic disease, said Ms. Robinow.
Before risk adjustment can be used in this way, though, coding of chronic diseases in outpatient settings will have to improve, said Craig Christianson, MD, medical director of U-Care Minnesota.
Here are brief summaries of experience with health risk adjustment in four states:
The California Health Insurance Purchasing Collective (HIPC), a state-run small group purchasing alliance that includes 20 insurance carriers enrolling 120,000 beneficiaries, uses only certain high-cost inpatient diagnoses to risk-adjust, partly because of data limitations. The HIPC set a tolerance band of 0.95 to 1.05, so that risk adjustment is triggered only when a carriers risk assessment is 5% above or below the average for the HIPC.
The California HIPC reported that 1% of total premiums was transferred during the first year of health risk adjustment and 0.15% of total premiums was shifted the second year. The small shift in the second year may be due to a more even mix of risk among insurers or from better data reporting. Two HMOs that withdrew after implementation scored below 0.95.
In Minneapolis, BHCAG, which contracts directly with 25 provider care systems for health care coverage of 115,000 beneficiaries, implemented health risk adjustment in January 1997. BHCAG makes fee-for-service (FFS) payments to care systems which are adjusted quarterly, based on how actual payments to each system compare with a risk-adjusted, per-member-per-month (PMPM) claim target. Daniel Dunn of Integrated Healthcare Information Services, Inc. noted that the largest adjusted increase was about $22 PMPM and the largest adjusted decrease was $20 PMPM. Relative risk among care systems varied from 0.78 to 1.25 for the first quarter of 1997.
Colorado's Medicaid program moved toward risk adjustment in two phases because of data issues. In 1997, Colorado adjusted premiums based on prior fee-for-service costs for a large sampling of Medicaid recipients. This year, it will use a diagnosis-based grouper system to predict costs and adjust premiums.
Colorado Medicaid contracts with five managed care plans that enroll 70,000 Medicaid recipients (about 28% of the total enrollment). Interestingly, the state has found that the population in HMOs is sicker and costlier than the population in fee-for-service. About 2.7% of total premiums was shifted in the first year of health risk adjustment, and the bulk of that transfer involved a 2.15% increase in HMO payments relative to FFS.
The Washington State Health Care Authority (HCA), which administers benefits for 240,000 state employees and dependents in 17 health plans, is starting a phased-in approach to health risk adjustment this year. The HCA ran a hypothetical simulation of how its risk adjustment models would have affected 1997 premiums. Under full-fledged health risk adjustment, 6.5% of total premiums would have been shifted. About 5% of total premiums would have been transferred, under the phase-in approach being used in 1998, which will be based on enhanced demographic data and health status.
For more information, contact the Alpha Center, which conducted the conference, at (202)296-1818.