Regional Variations in Health Care and the Dartmouth Research

A story in yesterday’s New York Times begins: “For years, health policy experts have said health care spending is much higher in New York City and Boston because doctors and hospitals there provide more services, practicing medicine in a more intensive way. But new government data show that Medicare costs per patient in those cities are slightly below the national average when the numbers are adjusted for the cost of living and other factors.”

The Times goes on to explain that President Obama “says that the nation could save huge sums if all doctors and hospitals were as efficient as those in lower-cost states like Iowa, Minnesota, Washington and Wisconsin. Lawmakers from those states have reached an agreement with House Democratic leaders that would increase federal Medicare payments to health providers in their states. Higher-cost states, which could see their Medicare payments reduced, are fighting back.”

The paper acknowledges that Dr. Denis A. Cortese, president of the Mayo Clinic, based in Rochester, Minn., agrees with the president, saying that Medicare wastes billions of dollars a year because it “pays the most to health care providers and geographic areas that provide the lowest-quality care at the highest costs.” In other words, Medicare is rewarding inefficient hospitals that keep patients longer because it takes them longer to diagnose the patient, because they run more tests, because more specialists are called in to consult, and --because they don’t always consult with each other--  patients suffer complications, or fall victim to medical error.

But, the Times notes, “Dr. Steven M. Safyer, president of Montefiore Medical Center in the Bronx, disagrees, saying : ‘Our Medicare expenditures reflect the low socioeconomic status of the population and the very high cost of doing business here. Many of our patients do not receive regular care before becoming eligible for Medicare and have no one to care for them after they leave the hospital.’”

Moreover—and here’s the eye-opening paragraph in the Times story: “The Medicare Payment Advisory Commission (MedPac), an independent federal panel that advises Congress, has found that geographic variation in Medicare spending is substantial. But it told Congress recently that much of the variation could be explained by local differences in the cost of providing care and in the health status of beneficiaries, as well as by extra payments, authorized by Congress, for hospitals that train doctors or treat large numbers of low-income patients.

After adjusting for those factors, the commission said, Medicare payments per beneficiary were 92 percent of the national average in New York City and 95 percent of the national average in Boston.

Wait a minute—is MedPac saying that more than two decades of Dartmouth research showing that patients receive far more intensive aggressive and expensive care in cities like New York and Boston, is wrong?

No. Looking at the dollars that Medicare spends is one thing. Looking at utilization of health care resources—what percent of patients see 10 or more specialists during their last six months of life, how many days they spend in the hospital, how many tests they undergo—is something else. “When you look at utilization, you don’t have to price-adjust,” Dr. Elliott Fisher a professor of medicine and at Dartmouth Medical School and Dartmouth’s Center for the Evaluative Clinical Sciences explained in a phone interview.  “If you look at hospitals in affluent parts of Manhattan, you’ll find that utilization is twice as high as in Rochester, New York.”

The table below compares spending and utilization in Rochester and in Manhattan both during the last six months of life and during the last two years for Medicare patients suffering from very similar chronic diseases. How sick were they? They all died within two years. (The data looks back on their care during those years.) The table also zeroes in on two hospitals—NYU Medical Center in Manhattan and Strong Memorial Hospital in Rochester.

*Part B spending for imaging per decedent  (click table to enlarge)

Source: The Dartmouth Atlas of Health Care 

While differences in the dollar amount that Medicare spends is affected by differences in local prices, that doesn’t explain the enormous disparities in the amount of time patients spent in the hospital, the number of physician visits, or the percent who were treated by 10 or more specialists.

Clearly, patients in Manhattan were receiving far more aggressive, intensive care, and inevitably, this winds up being more expensive care.  Meanwhile, reams of Dartmouth data show that outcomes are no better in regions where doctors and hospitals pull out all the stops. Sometimes they are worse. After all, every test, every treatment exposes a patient to some risk. If the test is redundant, if the extra days of hospitalization weren’t necessary, than the patient is, by definition, exposed to risk without benefit. “Hospitals are dangerous places,” says Fisher, “especially if you don’t need to be there.”

Other Dartmouth Studies Do Adjust for Differences in Local Prices, Race, and Socioeconomic Status

The table above does not factor in differences in local prices, race, age, sex, socioeconomic status and severity of illness. But in other studies, when comparing spending and outcomes at academic medical centers in different regions of the country, Dartmouth researchers do adjust for all of these factors.

And, after making these adjustments they find enormous differences regional differences, even at academic medical centers, where one might expect physicians would follow similar evidence- based guidelines. One would  be wrong.

Fisher explains: “In a study published in Health Affairs, we compared outcomes and quality of care for heart attack, colon cancer and hip fracture patients, adjusting for underlying differences in both health and socioeconomic status. We then compared costs of care, adjusting for price differences -- and found that at the academic medical centers where patients received more intense care, per-patient costs were still about 60% higher over the long term -- with no better outcomes or quality.”

Specifically, “Patients in the higher-intensity hospitals spent more time in the hospital and intensive care unit (ICU); had more frequent physician visits (especially in the inpatient setting); have more specialists involved in their care; and received more imaging services, diagnostic testing, and minor (but not major) procedures.”

When researches compared long term survival, it turned out that mortality rates were slightly higher for heart attack patients and cancer patients who received the  priciest, most intensive care.

In today’s article, the Times quoted Dr. Darrell G. Kirch, president of the Association of American Medical Colleges, saying that the proposed redistribution of Medicare money could have “catastrophic unintended consequences for teaching hospitals and their patients.”

 What he didn’t say is that, in fact not all teaching hospitals provide more costly care. The level of intensity tends to be tied to the region where the medical center is located. Secondly, proposals for reducing Medicare payments in some regions most often suggest cutting payments to hospitals (not individual doctors) that fail to approach benchmark standards in terms of cost and quality.

Why Do Patients In Some Regions Receive More Care?

Why is care so much more aggressive in some academic medical centers than in others?  In the past, some Manhattan hospitals have explained that their patients are simply more sophisticated, and more demanding than patients elsewhere. But researchers point out that in general, patients who want more aggressive management of their illnesses are likely to choose academic medical centers. “Therefore,” Fisher and colleagues note in the Health Affairs article, “we would expect patients’ preferences to differ relatively little on this dimension across these hospitals. Data from other studies also argue against patients’ preferences as a major determinant of the differences in practice we observed across hospitals.”

Another Dartmouth study that compares similar patients at Stanford University Medical Center, to Medicare enrollees who were cared for at New York University(NYU) Medical Center  finds that the NYU patients spent more than twice as many days in the hospital (twenty-seven versus ten) and saw more than three times as many physician visits (seventy-six versus twenty-three) during their last six months of life. Yet it’s hard to imagine the patients at NYU are significantly more knowledgeable, sophisticated  or demanding than patients at Stanford.

The single factor that does seem to explain differences in cost and intensity of care is the supply of specialists and hospital beds in a given region. In areas where there are more doctors and more beds per capita, patients are more likely to find themselves in one of those beds—and they are more likely to wind up with ten or more specialists consulting on their case. The researchers explain: “Given the current culture and financial incentives of U.S. medicine, which favors intervention among the seriously ill (even in situations where the scientific evidence is weak), physicians in a given care system are likely to draw upon all available resources to care for their patients.

“Moreover, it is often easier to manage patients’ care in the inpatient setting; the outpatient setting imposes a greater burden on the physician (for example, to arrange telephone, visiting nurse, or office visit follow- up to ensure that the patient is responding to therapy). The inpatient setting also lowers the threshold for further intervention: It is easier to obtain tests, perform minor discretionary procedures, or request a consultation. Studies of ICU bed availability reveal the same phenomenon: When more beds are available, the average severity of the patients cared for in the ICU is lower and the average length-of-stay is longer.”

By failing to note that you don’t have to adjust for prices when you’re looking at utilization, and by overlooking the studies where Dartmouth does adjust for local prices, income and demographics, the Times piece creates confusion. This is truly unfortunate, because the president is making the case that we can achieve significant savings in Medicare spending—by reducing the amount of needless spending that doesn’t improve outcomes and, in fact, often makes them worse.

“Next week, we have an article coming out in the New England Journal of Medicine addressing these issues, including differences in local prices,  says Fisher. "Factors like socioeconomic status do matter, he adds: “we don’t want to penalize a hospital in the Bronx that cares for much more vulnerable patients. But when you look at hospitals in the most affluent section of Manhattan, it’s hard to explain more aggressive, more intensive, more expensive care” compared to, say, Johns Hopkins in Baltimore. Especially because patients don’t benefit from pricier care.

One can only hope that both bloggers and reporters in the mainstream media will  read piece in NEJM and make a concerted effort to explain the waste in our health care system to their readers.

: