Flatlined

Chronic Illnesses

The evidence to support the assertion that at least one-third of health care services are unnecessary is from the work of the investigators at the Dartmouth Institute for Health Policy and Clinical Practice in Hanover, New Hampshire. The center's founder Dr. John Wennberg devised a technique of determining rates of usage of specific Medicare services in various regions of the US.

Wennberg and his colleagues, Dr. Elliott Fisher, an internist, and Dr. Jonathan Skinner, a health economist, have had access to the Medicare database for years. The investigators divided the US into 306 hospital referral regions where patients live and receive their medical care. Medicare pays a uniform rate across the US, adjusted for regional cost differences, so the study investigators adjusted for price differences and then examined the outcomes and care of all Medicare patients admitted with hip fractures, colorectal cancer, and heart attacks from 1993 to 1995 in all 306 hospital referral regions. When Medicare spending in a region is high, it is because of increased use of services not an increased price of services.

In the top twenty-five percent of regions Medicare spent sixty percent more per capita on patients with these three conditions than in the bottom twenty-five percent of areas. The range of spending between individual regions was much wider. For example, for the same condition Medicare paid for two and one-half times more tests, procedures, and admissions per Medicare beneficiary in Miami than in Minneapolis.(1) Doctors in high spending areas hospitalized more often, saw patients in the clinic more often, ordered more tests, and performed more minor procedures. At the end of life, patients in low-spending areas spent an average of six days in the hospital whereas patients in high-spending areas spent twenty days. (2)

The result of the increased use of medical services in high-spending areas was not improved outcomes, as might be expected. Patients cared for in those regions were at a significantly increased risk of dying over the five years after the initial hospital admission. More medical care not only worsened outcome, it also failed to increase the patients' satisfaction with their care or to improve their functional status.

It might be thought that academic medical centers that are supposed to be the best were more uniform in their care. But in high-spending areas, academic doctors behaved just like their peers in the same region. In the first six months after a hip fracture, patients cared for in academic medical centers in high-spending areas visited their doctors eighty-two percent more often, underwent twenty-six percent more imaging studies, ninety percent more diagnostic tests, and forty-six percent more minor surgery than those in academic medical centers in low-spending regions. (3)

What explains this astonishing regional variation in care? The only variables that predicted high-spending versus low-spending regions were an increased concentration of doctors and hospitals. (4)(5) Wennberg stated, “High-rate regions had thirty-two percent more hospital beds per capita, thirty-one percent more physicians, sixty-five percent more medical specialists, seventy-five percent more general internists, and thirty-seven percent more surgeons. Low-rate regions had twenty-five percent more family practice physicians than high-rate regions.”(6) Variations in the degree of illness among regions explained only twenty-seven percent of the variation whereas the local supply of hospital beds and specialists accounted for forty-two percent. (7)

When this group examined the overall picture that emerged from their findings, they concluded that low-spending areas should be the benchmark since patient satisfaction and patient functional status were the same as in high-spending areas while mortality was improved.

If doctors in high-spending areas practiced like doctors in low-spending areas, Medicare costs would be reduced by 28.9 percent with improved quality. (8) These savings are without rationing any needed care; without creating waiting lists or delaying anyone's access to a doctor, a test, or a procedure; and they are savings that would improve the quality of medical care.

It is not as if the doctors sit in a room and decide to build more hospital beds so they can admit more patients, collude to order more diagnostic tests, agree to perform more minor procedures so they can make more money, or keep patients in the hospital longer when they are dying. It does not work that way. In fact, practicing physicians are unaware of these regional differences and of their role in producing them.

This wastage reflects the unrestricted application of medical services in the absence of public reporting, in the absence of standards of medical practice, and in the absence of a full understanding of which patients benefit from which tests and procedures. The differences are local custom. The habits and circumstances of practice of Minnesota physicians are not the same as Florida physicians.

For instance, there are no standards for how often a patient should be seen after a heart attack or for heart trouble. If a town is home to a lot of cardiologists, they naturally want to be busy. The cardiologists all book their clinics to be full, and patients with heart trouble in that community are seen more often. A doctor that is liberal with patient visits is liberal with tests--and the more a doctor looks for, the more a doctor finds. Thus, excessive testing leads to unnecessary procedures. Half of the variation among regions in the number of visits to cardiologists is explained by the number of cardiologists in the community. (9)

The Dartmouth group attributes variations in care around the US to lack of coordinated care of patients, falsely optimistic assumptions about the benefits of aggressive treatment of the severely ill, and limited evidence about the kinds of care that benefit the chronically ill. (10)

Practicing Medicine Backwards

There is strong evidence that good primary care is the way to improve the management of chronic diseases. A higher concentration of primary care doctors in a community lowers the mortality rate from chronic diseases and helps prevent illness, but a higher concentration of specialists has no effect on mortality rates. (11)(12) Primary care doctors prevent disease when they can and manage it to avoid flare-ups when they cannot. These results are corroborated by state-to-state comparisons of treatments for pneumonia, diabetes, heart attacks, breast cancer, heart failure, and stroke.

A study of Medicare beneficiaries for the years 2000-2001 examined twenty-four measures that either prevent or treat these diseases. (13) The measures included immunizing elderly patients against pneumonia before they leave the hospital, the administration of drugs within twenty-four hours of a heart attack to decrease risk of death, eye examinations for diabetics every two years to detect eye disease before it becomes untreatable, and the discontinuation upon hospital admission of certain heart medications that can make a stroke worse or fatal. All twenty-four measures are inexpensive, safe, highly effective, and require a doctor to order them. In some states ninety-five percent of patients received them and in others only eleven percent. Low population states and the northeast ranked the best; high population states and the southeast consistently ranked the worst. What can possibly explain why only fifty-eight percent of Alabamans but eighty-six percent of New Hampshire's residents ever received a drug that reduces the mortality from a heart attack by fifteen percent?

A second group of investigators decided to find out. They designated states with high quality of care as those in which patients received a high percentage of the twenty-four life-saving treatments. Figure 1 gives the results. A ranking of one designates a state in which a high percentage of patients received the twenty-four measures and fifty-one is where this measure of quality is the lowest. The relationship between quality and spending was a straight line---in the wrong direction. The highest spending states provided the poorest care; low spending states the best care.

Figure 1 Relationship Between Quality and Medicare Spending, As Expressed by Overall Quality Ranking, 2000-2001

Source: Copyrighted and published by Project HOPE/Health Affairs as Katherine Baiker and Amitabh Chandra, “Medicare Spending, The Physician Workforce, And Beneficiaries' Quality of Care.” Health Affairs Web Exclusive, W-184-W197, April 7, 2004. The published article is archived and available online at www.healthaffairs.org.

The composition of the physician work force explained forty-two percent of the variation among states. The quality of care was better and the cost of care cheaper in states where more generalists practiced. High-spending states were home to more specialists who ordered more tests and treatments at more expense but failed to provide simple measures to prevent disease. Increasing the number of generalists in a state by one per ten thousand residents increased a state's overall rank by ten places; the same increase in specialists dropped the state's ranking by nine places. The specialists provided procedures and the generalists provided prevention. (14) In the medical industry more spending does not buy better health care, just more health care.

A Few Sick People

US medicine has mal-apportioned its resources, and it should turn its attention to high-cost patients. Health spending is concentrated in a few people--five percent of patients account for fifty percent of health spending; ten percent account for seventy percent of expenditures. (15)

Doctors can readily predict who these big spenders are. They're not necessarily the elderly; they are the chronically ill. In fact, fewer of the really elderly are disabled than they were in the past, and their dying process costs less than that of sixty-five-year-olds; they do not linger. (16)(17) While seventy percent of those over sixty-five have more than one chronic condition, fifty percent of middle-aged adults do also. (18)

The cost of care skyrockets as the number of chronic conditions increases. In 1999 the per capita cost of care of Medicare patients without chronic conditions was only $211 while for those with four chronic conditions it was $13,973. (19) Lung diseases, heart disease, and diabetes account for the majority of Medicare spending yet their management is grossly uncoordinated in the United States. Without the reconstitution of primary care, there is little possibility of improving.(20)

  1. John E. Wennberg, Elliott S. Fisher, and Jonathan S. Skinner, “Geography and the Debate Over Medicare Reform,” Health Affairs Web Exclusive, February 13, 2002.
  2. John E. Wennberg, “Variation in Use of Medicare Services Among Regions and Selected Academic Medical Centers: Is More Better?” Duncan W. Clark Lecture, New York Academy of Medicine, January 24, 2005
  3. Elliott S. Fisher, David E. Wennberg, Therese A. Stukel, et al., “Variations in the Longitudinal Efficiency of Academic Medical Centers,” Health Affairs Web Exclusive, October 7, 2004.
  4. Elliott S. Fisher, David E. Wennberg, Therese A. Stukel, et al., “Implications of Regional Variations in Medicare Spending: Part 1: The Content, Quality, and Accessibility of Care,” Annals of Internal Medicine,”138, no. 4 (2003): 273-287.
  5. Elliott S. Fisher, David E. Wennberg, Therese A. Stukel , et al., “Implications of Regional Variations in Medicare Spending: Part 2: Health Outcomes and Satisfaction with Care,” Annals of Internal Medicine 138, no. 4 (2003): 288-298.
  6. Wennberg, “Variation in Use of Medicare Services Among Regions and Selected Academic Medical Centers: Is More Better?”
  7. Wennberg, et al., “Geography and the Debate Over Medicare Reform.”
  8. Wennberg, et al., “Geography and the Debate Over Medicare Reform.”
  9. Wennberg, “Variation in Use of Medicare Services Among Regions and Selected Academic Medical Centers: Is More Better?”
  10. “New Study Shows Need for a Major Overhaul of how United States Manages Chronic Illness,” Robert Wood Johnson Foundation news release, May 16, 2006.
  11. Barbara Starfield, Leiyu Shi, Atul Grover, et al., “The Effects of Specialist Supply on Population's Health: Assessing the Evidence,” Health Affairs Web Exclusive, March 15, 2005.
  12. Barbara Starfield, Leiyu Shi, James Macinko, “Contribution of Primary Care to Health Systems and Health,” The Milbank Quarterly 83 no. 3 (2005): 457-502).
  13. 42. Stephen Jencks Stephen, Edwin Huff, Timothy Cuerdon, “Change in the Quality of Care Delivered to Medicare Beneficiaries: 1998-1999 to 2000-2001, Journal of the American Medical Association, January 15, 2003, 289(3), 305-312.
  14. Katherine Baiker and Amitabh Chandra, “Medicare Spending, The Physician Workforce, and Beneficiaries' Quality of Care,” Health Affairs Web Exclusive, April 7, 2004.
  15. Mark W. Stanton, “The High Concentration of U.S. Health Care Expenditures,” Agency for Health Care Research and Quality, Issue #19, June 2006.
  16. Dorothy P. Rice and Norman Fineman, “Economic Implications of Increased Longevity in the United States,” Annual Review of Public Health 25 (2004): 457-453.
  17. Andrew J. Rettenaier and Zijun Wang, “Explaining the Growth of Medicare: Part II,” Brief Analysis no. 408, National Center for Policy Analysis, August 6, 2002.
  18. Christine Hoffman, Dorothy Rice, Hai-Yen Sung, “Persons with Chronic Conditions. Their Prevalence and Costs,” JAMA 276, no. 18 (November 13, 1997): 1473-1479.
  19. Jennifer L. Wolff, Barbara Starfield, Gerald Anderson, “Prevalence, Expenditures, and Complications of Multiple Chronic Conditions in the Elderly,” Archives of Internal Medicine 162, no. 20 (November 11, 2002): 2269-2276.
  20. “High-Cost Medicare Beneficiaries,” A CBO Paper, Congressional Budget Office. May 2005.