Tribune Exclusive Report: Medicare Spending
Is more health care better? The conclusion of a Dartmouth study is that extra care for seniors doesn't mean better outcomes

April 13, 2008
By Sarah Arnquist

Medicare spends nearly 75 percent less per capital on chronically ill seniors in San Luis Obispo County during their last two years of life than on their counterparts in Los Angeles.

That's because elderly patients dying here are less likely to be poked and tested in a hospital during their final months and are more likely to die at home.

Similar differences in the amount, intensity and cost of care at the end of life exist nationwide. They can't be explained away by variations in patient health status or prices, experts say.

The disparities are caused largely by variations in the amount of time patients spend in the

hospital and the volume of services they receive - in other words, by the differences in how doctors practice in different places, said Dr. David Goodman, co-author of the Dartmouth Atlas of Health Care.

Despite the extra care, Medicare patients in Los Angeles don't live any longer or have better outcomes, he said, than patients in areas with roughly half the medical spending, such as Minnesota's renowned Mayo Clinic.

If doctors nationwide practiced as efficiently as those at the Mayo Clinic, which means having excellent outcomes coupled with relatively low expenses, the nation could reap billions in health care savings, Goodman said.

The Dartmouth researchers say the common belief that more health care means better care is “under scientific assault.” Their research increasingly is influencing policymakers hoping to reduce the nation's $2 trillion annual health care tab.

Policy implications

For three decades, the Dartmouth researchers have compiled evidence into their Atlas showing that a patient's care largely depends on his or her zip code.

The Atlas is a database that analyzes Medicare spending and procedures nationwide. The Web site, www.dartmouthatlas.org , allows people to look up rates of Medicare spending at individual hospitals. Medicare patients in Los Angeles receive more medical care than almost all other regions of the country, but that doesn't mean they receive better care, Goodman said.

Health care, he said, is like sunshine. There are limits to how much is good for you. “Once you pass those limits, what once was good is no longer.”

On average, chronically ill Medicare patients cost $44,795 in San Luis Obispo County and $53,400 at the Mayo Clinic in their last two years of life. At UCLA Medical Center, they cost $93,842.

The California average was $57,900. Only three areas in the state had lower costs than San Luis Obispo. Santa Barbara County's costs were lowest at $41,595 per enrollee.

The figures are based on the records of 4.7 million patients nationwide who died between 2001 and 2005.

There's little evidence that the extra care is what people want during their final years, said Dr. Steven Schroeder, former president of the Robert Wood Johnson Foundation and a professor at UC San Francisco Medical School.

“By and large people don't want to die in the hospital,” Schroeder said. “It's a less efficient, less desirable and very costly way to practice medicine.”

Skeptics

Dartmouth Atlas skeptics say the research design is flawed because it looks only at people who died, and therefore doesn't take into account the people who may have lived because of the aggressive care.

It also doesn't sufficiently adjust for differences in health status or socioeconomic status, said Dr. Tom Rosenthal, chief medical officer at the UCLA hospital.

UCLA has complicated patients who seek out experimental treatments, Rosenthal said. To draw conclusions from only the patients who died and ignore the survival rates is not fair or logical, he said. “You can't ignore those factors and consider this a scientifically sound study.”

The Dartmouth research is designed to standardize differences in patient populations; there is no relationship between the age people die and the intensity of services they receive in the last two years, Goodman said.

Doctors at high-spending hospitals are not bad doctors, he said. They believe they are doing what's in the best interest of their patients. But their style of practice is not sustainable.

At current spending rates, Medicare will exhaust its trust fund by 2019. The group that advises Congress on Medicare payments frequently cites the Dartmouth research.

Medicare is the federal health program for people 65 and older. It accounts for 20 percent of the nation's health care spending. About $1 out of every $3 Medicare spends goes to enrollees with chronic conditions in their final two years of life.

While the Dartmouth research looks exclusively at Medicare data, Goodman said, the conclusions apply to the entire health sector.

Medical culture

Compared with similar patients in Los Angeles, Medicare patients in San Luis Obispo County spend half as much time in the hospital, see doctors half as often in their final months, and are nearly three times as likely to use hospice services, according to the Atlas.

In other words, local doctors tend to practice more conservatively- like doctors do at the Mayo Clinic - and less aggressively than the doctors at UCLA Medical Center.

The local care philosophy is different from that in urban areas out of necessity, said Pismo Beach internist Scott Robertson. Patients probably work more closely with primary care doctors and see specialists less often because there are fewer specialists here. However, he said, that doesn't necessarily translate into worse care.

Robertson's explanation underscores the Dartmouth research that shows the availability of medical resources, such as hospital beds, specialists and technology, is the most important factor affecting a doctor's decisions about patient care.

“What the Dartmouth studies have shown is that the more doctors and specialists an area has, the more stuff that gets done to the population,” UCSF's Schroeder said.

That makes sense, he said, because a doctor's job is to per-form tests and procedures, and they are paid based on how many they do - not on the quality of patient care.

Why? Different areas have different medical cultures that influence individual doctors' decisions, Schroeder said. Deciding when to send patients to specialists, which tests to perform, and whether to admit them to the hospital for a procedure often falls within the “gray area” of medicine.

“There's no hard and fast rule,” he said.

But the more hospital beds available, the more likely a patient will end up in one, he said.

Patients who want to spend their final years inside a hospital receiving tests and pingponging between specialists should seek out hospitals and doctors that provide that style of care, Goodman said.

Patients who prefer to work with a primary care doctor to stay out of the hospital and avoid extra tests and procedures, he said, should seek out that kind of care. Not only will it save them money, they will avoid the risks of hospitalization, such as infection, complication or error.

The less intense care may actually be harder to find. Fewer doctors are going into primary care nationwide, because they can make substantially more as specialists.

“Primary care doctors are in eclipse because of the way we pay for services,” Schroeder said.

Backward incentives

Another effect of the disparity in costs and services is that enrollees in low-cost areas, such as San Luis Obispo County, subsidize the higher spending regions, such as Los Angeles, New York and Boston.

The system also does not reward local doctors for keeping patients out of the hospital and costs down. Instead of incentivizing them to practice like doctors at the Mayo Clinic, the system rewards them for practicing like doctors in Los Angeles. “ The way things work is (doctors') costs are fixed, and the only way to make more money is to see more patients and to do more things to the patients you see,” Robertson said.

Patients often want more care, he said. It's difficult for patients to assess the quality of medical care so they perceive more care as better care, even though the extra care may provide no added benefit and could be harmful.

Reforming the system

Faced with unsustainable growth in health costs, policymakers at all levels are paying close attention to the Dart-mouth research. The Medicare Payment Advisory Commission cited the Dartmouth Atlas 50 times in its 236-page report last year to Congress on Medicare's payment policies.

The MedPAC report acknowledged inequities in the current system, which “reinforces a general style of medical practice beyond the financial means of an increasing number of Americans.”

UCLA's Rosenthal said the questions raised by the Atlas are worth further exploration, but more studies in peer-reviewed journals are necessary before policymakers act. UCLA has conducted one study, he said, and is seeking to publish it.

“A healthy scientific debate needs to occur before policy gets made,” he said.

The Dartmouth researchers have testified before Congress recommending solutions to lower health spending. Among them are increasing the supply of primary care doctors and improving coordination of care, Goodman said.

Ultimately, he said, changing the way physicians and hospitals are paid is a must so they are rewarded not for doing more but instead for the quality of care they provide.

Politically, that will be challenging, Robertson said. Every specialty and sector of health care will fight to protect its turf.

“You do what's right for your patient in the exam room, but sometimes the patients get forgotten in the policy,” he said.