As efforts to contain rising health care costs intensify, a new Harvard study suggests that shifting costs onto chronically ill elderly patients can backfire and result in higher overall costs through increased hospitalizations.

The research, conducted by Amitabh Chandra, a professor of public policy at the Harvard Kennedy School, examined patients’ health care utilization after copayment increases for office visits and prescription drugs in the California Public Employees Retirement System (CalPERS), the program that covers state and local government retirees there.

Chandra, who conducted the study with Jonathan Gruber of the Massachusetts Institute of Technology and Robin McKnight of Wellesley College, said the study doesn’t simply say copayments are bad or good, but rather has a more complex message for those making health system changes. Though the copayment increases were counterproductive for elderly patients with a chronic disease like diabetes or hypertension, the study showed that copayments worked as desired for those not chronically ill. Those patients reduced office visits and prescription drug utilization with no negative effects on their health.

Those broader results indicate that copayments can be effective cost-sharing mechanisms that prompt patients to consider whether they really need care, Chandra said. And the results show that most patients do a good job of deciding what care to cut out and what to maintain.

More attention needs to be paid, however, to those who are chronically ill, Chandra said. For the subset of patients who are fighting diseases such as diabetes, high blood pressure, arthritis, and Alzheimer’s disease, the cost shift backfired. Patients with those conditions cut back on prescription drugs and delayed office visits enough to warrant increased hospitalization, more than offsetting any cost savings recognized from their copayments.

“That’s a disaster because not only is care more expensive, their health is much worse,” Chandra said.

For those patients, other interventions should be designed that encourage them to get the maintenance care critical to their health, Chandra said. Eliminating copayments, tying copayments to the therapeutic value of the drug, or even establishing a “negative copay” that pays them for taking their medications and making office visits could effectively keep them healthy and costs lower.

The key, Chandra said, is tailoring the health care system in a way that wrings out costly unneeded aid while encouraging care that is effective at improving patients’ health.

“In general, people get it right in cutting back,” Chandra said. “The question we’re all interested in is how do you design a system where patients don’t just get less care, but get more valuable care.”

Understanding health care utilization by the elderly is critical because people over age 65 use 36 percent of health care in the United States, although they make up just 13 percent of the population. In addition, with an aging population, health care costs for America’s elderly promise to rise.

The study, published in the March issue of the American Economic Review, fills a knowledge gap left by a seminal study conducted 30 years ago. That study, the RAND Health Insurance Experiment, excluded elderly patients and concluded that shifting costs to patients would reduce utilization without a corresponding decline in patient health. In the decades since, Chandra said, the U.S. health system has changed dramatically, and care for the elderly has become a major concern.

The work of Chandra and his colleagues also highlighted a quirk in the U.S. system that allocates overall savings differentially to insurers involved in patient care. Because Medicare pays for hospital stays and private supplemental insurers provide prescription drug coverage, all the financial benefits of reduced prescription drug use went to the supplemental insurer. Offsetting hospitalization costs from patients not taking their medication, however, were borne not by the supplemental insurer, but by Medicare, which pays for hospitalization. That shift not only costs the government money, but argues for a system where a single insurer pays for all of a patient’s care so that unintentionally created perverse incentives do not wind up undermining a patient’s health.

Bringing men’s lax back