Cleaveland: Health costs can differ by region

Cleaveland: Health costs can differ by region

July 2nd, 2009 in Health

Unless the cost of health care is controlled, any reform passed by Congress will ultimately fail.

Most reform proposals are based on the twin goals of universal coverage and containing costs. For decades, medical costs have exceeded both the rate of inflation and growth of gross domestic product.

In this and later columns I will examine cost from four perspectives: regional variation, the use of technology, health care organization and consumer demand.

For more than 20 years, investigators at the Dartmouth Institute for Health Policy and Clinical Practice have studied regional variations in Medicare spending and health care facilities and manpower. A greater than three-fold difference in Medicare spending per person exists among hospital referral regions.

The differences are due to the volume of medical services delivered. Higher cost areas include Louisiana, eastern Texas, part of southern California extending into southern Nevada, San Francisco, New York City and the Baltimore-Washington area. Medicare spending is substantially lower in Washington State, Oregon, Idaho, Montana, the Dakotas, Minnesota, and Wisconsin. Health care spending tends to be above average in most Southern states.

High expenditure areas are characterized by more frequent hospitalizations, more days spent in intensive care units, and more frequent diagnostic tests and physician visits. Health care outcomes are essentially the same in high cost and low cost areas. More is not better.

Striking differences are also seen in numbers of hospital beds designated for acute care. Data from 2006 show a greater than two-fold variation even after adjusting for age and gender. Jackson, Mississippi, for example, had 4.44 acute beds per 1000 residents. Sacramento, California had 1.52.

In centers with high bed capacity, hospitalization is more likely for conditions in which out-patient care is an acceptable option. If a bed is available, it will be filled. Attempts to regulate expansion of hospitals and diagnostic facilities have generally failed because of local political and economic pressures.

Similar differences exist in the physician workforce. A 2.5 fold difference is seen in primary care providers. San Francisco is at the high end with 117 doctors per 100,000 residents. El Paso has 47.2. A like variation exists for specialty physicians: White Plains, New York has 215 doper 100,000; Sioux City, Iowa has 84.1. As carefully documented by the Dartmouth group, having more hospital beds and more physicians does not translate into better health, easier access, or better customer satisfaction.

An article by Atul Gawande recently published in the New Yorker analyzed health care in McAllen, Texas, one of the most expensive health care centers in America. Yearly Medicare costs per enrollee in 2006 averaged more than $15,000, twice the national average. By contrast, annual costs per Medicare enrollee at Rochester, Minnesota's Mayo Clinic averaged $6.600. This important article analyzes the causes of McAllen's high health costs: more testing, more surgeries, more physician-investors in health care facilities. Competition among providers led to expensive marketing and public relations ploys. The sad conclusion is that the drive for profits led to the sustained rise in costs. I urge you to read this article.

How can we address such wide regional variations?

* Market forces alone will not succeed. Health care providers can no more regulate themselves than banks and auto-makers. Too much money is at stake. Even so, providers take a lead in determining which tests and treatments are essential to good care.

* Medicare and private insurers could pay for services based on a national average of health care costs, adjusted for cost-of-living.

* Insurers could regulate volume of medical services by using tighter, evidence-based standards.

* A national commission with licensing and funding authority could determine needs for hospitals, diagnostic facilities and physicians.

* Integrated health care systems could lead to better quality of health care at lower costs. (This approach will be the subject of a later column.)

* The problems associated with regional variation are too complex to be solved on a local or statewide basis. An informed Congress, acting across political party lines, must address this vital issue.

Contact Clif Cleaveland at