a new role for health insurers

Historically, health care has been organized largely around providers. Patients have needed access to providers to receive care and to arrange for tests, referrals, and procedures.

This need has resulted in a sense of paternalism in the development of healthcare delivery systems:" Patients see providers at the providers' convenience, obtain care wherever providers refer them, and are encouraged to follow providers' directions more or less passively.

This traditional approach in part was cultural. When modern healthcare delivery systems were being formed during the first three quarters of the 20th century, physicians were highly regarded and respected. The country as a whole seemed to be more accepting of directives from sources of authority such as the government, religious bodies, and employers.

However, since the 1970s, a cultural shift in America-undoubtedly accelerated by the Vietnam War and various social movements-has encouraged independent thinking and increased tolerance of different opinions. The notion of paternalism rapidly has been marginalized as a social construct; the ready availability of information on the Internet is likely to keep it that way.

But there is a considerable amount of misinformation on the Internet, and healthcare consumers cannot easily determine what medical information is accurate and useful. Amid ongoing healthcare reform, insurers can play an active role in filling this void.

Providing Actionable Information

Unfortunately, studies have shown that when patients are considering where to go for care, they rarely use accurate and reliable information provided by independent agencies. For instance, since the late 1970s, the New York State Department of Health's Statewide Planning and Research Cooperative System (SPARCS) has provided physician- and hospital-specific risk-adjusted mortality and morbidity rates for New York hospitals that perform coronary artery bypass graft (CABG) surgery and percutaneous coronary intervention (PCI). Despite evidence that providers have changed their behaviors in response to these data-with some poor performers going out of business or moving to a different state-there remains substantial variation in risk-adjusted rates. Hospitals and physicians with statistically significantly high risk-adjusted mortality rates are still reported by SPARCS each year.

Yet only rarely do New York residents-or California residents, who have a similar reporting system-use this information when considering CABG surgery or PCI; indeed, volume effects of such reports were found to be modest, transient, and largely limited to white Medicare patients in New York.b

Meanwhile, a substantial body of research has identified a strong "volume-outcomes" relationship wherein health outcomes are better when providers or hospitals perform a higher number of a particular procedure eachyear.c The association makes sense: Practice makes perfect. The Leapfrog Group, a not-for-profit membership organization composed of large employers, has tried to integrate adherence to minimum volume standards into hospital referral practices, albeit with an understanding that the relationship is imperfect. For instance, Leapfrog employers allow their employees to obtain care for CABG or pancreatic cancer only at hospitals that perform a minimum of 450 or 11 such surgeries per year, respectively.

Getting Through to Patients

When designing benefits, most insurers have tended not to use information on anticipated outcomes based on historical mortality rates or procedure volumes. Perhaps because of concerns about consumer backlash stemming from restriction of provider or hospital choice, insurers have been reluctant to try to steer patients to hospitals where outcomes are likely to be better.

To determine whether and where to obtain a procedure, patients should be provided with information about different treatment options and the estimated short- and long-term out-of-pocket costs of each option, along with different sites of care and the risk-adjusted outcomes of each.

But patients likely need help collating, interpreting, and validating the data. Insurers, including Medicare, should provide that assistance.

Instead of focusing on developing narrow networks that may be defined more by costs than by outcomes, insurers should design benefits in a way that steers patients toward providers that produce better outcomes. For example, copayments could be lower for patients who use hospitals with excellent risk-adjusted outcomes instead of hospitals with poorer outcomes.

The benefits of this approach could be far-reaching:

> Better outcomes and lower out-of-pocket costs for patients who choose a hospital with better risk-adjusted outcomes and higher volumes

> Lower costs for insurers stemming from better outcomes

> Increased volumes for high-quality hospitals and providers (potentially leading to continued improvement in the healthcare team's experience and outcomes and driving down the hospital's costs per unit, making it more profitable)

Insurers should aggregate, validate, and provide information on aspects of healthcare quality that patients regard highly (e.g., patient satisfaction, adherence to processes with demonstrated effectiveness, avoidance of patient harm, risk-adjusted patient outcomes) to help them make informed healthcare decisions and align those decisions with their own healthcare values. By motivating patients to make informed decisions, encouraging them to get only care that they want and need, and using behavioraleconomics techniques that encourage them to choose providers with the best outcomes, insurers can take steps to reduce healthcare costs and improve quality and outcomes.

If widespread, such information would drive patients to use higher-value care, thereby concentrating care among the highest-value providers-and improving the value of health care delivered in the United States.

a. Merriam-Webster defines paternalism as the attitude or actions of a person or organization that protects people and gives them what they need but does not give them any responsibility or freedom of choice.

b. Romano, RS. and Zhou, H., 'Do well-publicized risk-adjusted outcomes affect hospital volume?" Medical Care, April 2004.

c. Halm, E.A., Lee, C, and Chassin, M.R., "Is Volume Related to Outcome in Health Care? A Systematic Review and Methodologie Critique of the Literature," Annals of Internal Medicine, September 2002.

William B. Weeks, MD, PhD, MBA, is a professor, Geisel School of Medicine, Dartmouth College, Hanover, N.H., and a member of HFMA's New Hampshire-Vermont Chapter (wbw@dartmouth.edu).