Rx for Reform
I live in Massachusetts, where 97 percent of the population now has health insurance. Yet state leaders are scrambling to cut costs in the face of...
I live in Massachusetts, where 97 percent of the population now has health insurance. Yet state leaders are scrambling to cut costs in the face of huge budget deficits. As we seek to create universal health insurance nationwide, Massachusetts can teach us two lessons: Universal coverage is possible, yes, but without serious cost control, it is not sustainable.
The best way to slow health care inflation is to change the way we pay for care. The financial incentives in our current system are backwards. Under fee-for-service payments, providers make more money by performing more services. If they figure out economical ways to keep their patients healthy, they go broke.
John Wennberg at Dartmouth University has studied Medicare data for decades. His research shows that in regions with more doctors per capita, more procedures get done, since most doctors get paid per procedure. Yet this higher spending yields worse outcomes.
Wennberg's research also shows that integrated, managed health care providers--such as the Mayo Clinic, Kaiser Permanente and Intermountain Health Care--are more efficient. Regions dominated by such providers have costs that are nearly one-third lower than in other areas because those providers use electronic health records, evidence-based standards of care, a more cost-effective mix of personnel, and other ways to keep costs down and quality up.
The conclusion is obvious: To slow health-care inflation, replace fee-for-service systems with competition between integrated health plans that charge annual per-patient fees.
How do we get from here to there? Federal reform could offer states incentives to experiment, so we can learn what models are effective in what types of markets.
States could create purchasing pools involving all their programs--Medicaid, the Children's Health Insurance Program, state employees' and retirees' plans--and invite health insurance companies to compete for this market based on both price (of annual per-patient fees) and quality. States could partner with private and nonprofit employers willing to join the pool as well.
Wisconsin's insurance program for state employees offers a good example of how such price competition could work. The program defines a standard benefit package, asks health plans to submit bids specifying what they would charge for this package, then ranks the bids.
Wisconsin uses price and quality measures to define three tiers. Plans in tier one, which are low in price but still high in quality, cost the least for state employees. If employees prefer more expensive plans in either tier two or three--because their physician is not part of a tier-one plan, for instance--they pay part of the difference in premium cost. The vast majority choose tier-one plans.
Wisconsin put this three-tier approach into effect in 2003. In Dane County, which includes the state capital of Madison, the state employees' plan covers nearly 30 percent of the non-Medicare and non-Medicaid market. By 2006, state employees' costs for individual and family plans in Dane County had fallen 14 percent below the statewide average.
Both patients and doctors can be satisfied in such a system, as long as they have choices. Wisconsin's experience demonstrates that a government can restrain costs by forcing price competition in just 30 percent of the market. This is true because most health plans compete for that 30 percent, and when they strive to become more efficient to capture a bigger share, the changes they make affect the rest of their business.
On average, government employees, Medicaid patients and other state-funded recipients total 21 percent of the market. If states could recruit some private employers into the pool and cover at least some of the uninsured, they could reach 30 percent.
Competition for prepaid plans would reward the most cost-effective providers. Integrated, managed care systems would flourish. Innovation would spread.