A recent article published in the New England Journal of Medicine (October 23, 2003) deserves the attention of both policy makers and a public suffocating under the expense of health care. It reports that during the last decade, the Department of Veterans Affairs (VA) undertook a structural reform designed to reduce its emphasis on hospitals in favor of providing medical care through a comprehensive, high-quality primary-care system. The VA oversees the largest health-care system in the United States.

Analyzing the effect of this change in the VA’s medical delivery, researchers found that the number of days spent in the hospital fell by a remarkable 50 percent, and that there was only a moderate increase in outpatient utilization when compared to the period prior to the reform. On top of that, urgent-care services fell by 35 percent, and still more impressive, survival rates were not affected. Those who conducted the study came to the conclusion that the decrease in hospital use did not diminish access to necessary medical care, and that this shift caused no adverse consequences for the health of VA beneficiaries. These findings, coupled with the study’s conclusion that more efficient care does not automatically equate with poorer care, bring to mind related research on how the American health-care system fails to utilize its vast resources properly.

Health economists have long noted variations in healthcare utilization by comparing geographic areas. In 2000, Medicare expenditures (a good measure of the utilization of medical services) were $10,500 per enrollee in Manhattan but only $4,823 per enrollee in Portland, Oregon. This difference, it turns out, was not caused by a greater burden of illness in New York, or even to its higher cost of living. Using statistical methods to control for variables, researchers eliminated possibly misleading comparisons, such as using dissimilar populations and/or variations in the cost of doing business, and they were able to document that the differences were real. They found, for example, that the gap in expenditure persisted even after population characteristics (age, sex, race, and burden of illness) and the price of medical services had been taken into account.