Cost-cutting strategy employed by HMOs for 15 years is simply not working

Sep 19 2005

With rising concern over the cost of the new Medicare prescription drug benefit program – going into effect January, 2006 and estimated to cost $593 billion over the next decade – a new UCSF study reveals that a key cost-cutting strategy employed by HMOs for 15 years is simply not working.

Health insurance companies have increasingly sought to limit the amount of expensive drugs doctors prescribe to patients in order to keep drug costs form spiraling, according to the study authors. A major strategy has been to restrain drug costs by assuring that a medical group will make money if member doctors prescribe within the drug budget set by the insurance company, and will lose money if member doctors over-prescribe.

The underlying assumption is that placing doctors at financial risk for their drug prescribing practices will lead them to adopt new practices to control drug costs, the authors explain. These practices include hiring pharmacists for expert advice, using "physician profiling" to compare doctors' prescribing patterns, and adhering to professional protocols that specify what each drug should be prescribed for, at what dose and for how long.

The new study shows that providing financial incentives for doctors to rein in their prescription practices has not led to cost-cutting innovations.

The study is being published in the August issue of the Journal of Health Policy, Politics and the Law, available mid-September. It is based on a survey of executives in more than three dozen physician groups and HMOs in four large U.S. cities.

The survey found widespread dissatisfaction with the HMO strategy of making doctors financially liable for prescription drug costs above a certain limit. It also found doctors were often confused or unaware of the incentives, either because of unclear contracts with HMOs or failure of HMOs to share drug cost information with doctors appropriately. In addition, frequent changes in the contract made it hard for physicians to know if an investment in innovation made under today's deal would still pay off under tomorrow's, the survey shows.

"The problem lies in the contract terms and information-sharing between the HMO and doctors. Fix that, and you'll go a long way to controlling drug costs," says study co-author R. Adams Dudley, MD, UCSF associate professor of medicine and health policy.

HMOs negotiated their own prices for drugs with the manufacturers via pharmacy benefit managers but typically failed to disclose that information to physicians, the survey revealed.

"Profitably managing something as complex as prescription drug risk requires accurate, timely information, but most of the surveyed physician groups couldn't even calculate their total drug costs," says Jonathan D. Agnew, PhD, a study co-author and senior policy consultant at the British Columbia Medical Association.

Even when HMOs were willing to share price information, the study found that the data usually got to the doctors too late for them to determine if they were over or under budget.

"On the one hand, doctors get too much information from working with so many HMOs, each with its own array of prescription drug benefits and policies. On the other hand, the information they do get from the HMO is often incomplete or not timely enough to help them make good management decisions," says Helene Levens Lipton, PhD, professor of health policy and pharmacy at UCSF and lead author of the study.

The survey also found that HMO contracts with physician groups focused far more on efforts to contain costs than to maintain or improve the quality of care, Dudley said.

The researchers propose three major changes:

• Contracts between physician groups and HMOs should be clearer and more accurate, and information about changing drug-cost incentives should be provided to physicians in a more timely way.

• Doctors should not be held liable for the costs of doing the "right thing": If giving a prescription is the right thing to do, the costs should not be included in a budget target. Rather, the expenditure targets should be limited to situations in which medication use is more discretionary, such as whether or not a patient needs to continue on anti-ulcer drugs after an ulcer has healed.
• When HMOs evaluate a group's performance, the assessment should be fair. For example, in the late 1990s, doctors should not have been penalized for rising prescription costs related to the introduction of new drug therapies they could not have budgeted for, such as the first drug therapies for attention deficit disorder.

As another example of a drug-risk contract considered unfair by physician groups, the authors report that one physician group was penalized for not controlling its costs as well as another group in the HMO's network, even though the other group was in a different health care market.

Under the new Medicare drug benefit, reimbursement for HMOs depends, in part, on their establishing sound drug use management programs to contain drug costs, the authors point out.

"As HMOs establish and modify these programs, we hope the data presented here can help them avoid some of the major pitfalls we found in several large cities across the country," says study co-author Marilyn Stebbins, PharmD, UCSF professor of clinical pharmacy. "And doctors can learn to request fairer contracts and better information sharing."

http://www.ucsf.edu/