For-profit Hospital Ownership Means Higher Costs, Dartmouth-VA Study Finds

HANOVER, NH � Medicare costs are higher in areas served by for-profit hospitals than in those served by non-profit hospitals, according to a national study by researchers at Dartmouth and White River Junction (Vermont) Veterans Affairs Medical Center.

The comprehensive analysis, reported in the August 5 New England Journal of Medicine, is the first to examine the association between for-profit hospital ownership and health care spending for the community served, say the investigators. It offers insights for health care policy and cost containment at a time when more hospitals are converting to for-profit status, they note.

Comparing spending for Medicare beneficiaries over a six-year period, the Dartmouth researchers found that per capita spending was greater and rose faster in areas where for-profit hospitals predominated. Furthermore, spending in areas where hospitals converted to for-profit ownership grew faster than in stable not-for-profit areas.

Authors of the study are Drs. Elaine Silverman, general medicine fellow, and Elliott Fisher, professor of medicine and of community and family medicine, at Dartmouth Medical School and the VA Medical Center, and Jonathan Skinner, John French Professor of Economics at Dartmouth College.

"Advocates of for-profit hospitals believe that they are more likely to respond to market forces to produce high quality services at lower costs," says Silverman, also an instructor at DMS.

"Previous work has focused on the costs of individual services. We have examined the impact of hospital ownership on the Medicare population as a whole and found costs in areas of for-profit hospitals higher. There is no evidence that for-profit hospitals provide a higher quality of service.

"Based on our results, if all Medicare enrollees had lived in areas served by not-for-profit hospitals, we estimate that Medicare would have saved over $5 billion in 1995 alone," Fisher says. "If the trends we observed continue through the year 2010, the excess Medicare costs associated with for-profit hospital service areas would equal the estimated cost of the President's proposed Medicare drug benefit."

Among the Dartmouth findings:

• Per capita Medicare spending was higher in areas served by for-profit hospitals than those served by not-for-profit hospitals in 1989, 1992, 1995.
• Spending grew more rapidly in for-profit areas over the six-year period examined.
• For-profit spending was greater not only in hospital care, but also in physician services, home health and other long-term care and nursing facilities.
• Spending increases were significantly higher in areas where hospitals converted from not-for-profit to for-profit compared to stable non-profits.

Between 1990 and 1996, 192 non-profit hospitals converted to for-profit status, with half of the conversions occurring in 1995-1996, the investigators point out. They say it is increasingly important to examine how this trend will affect our health care system.

"Our study reveals that we have a health care system that continues to reward those who do more. And we can't assume that more is better. We need a health care system that rewards those who do better," Fisher adds.

The investigators looked at spending in 1989, 1992 and 1995, using information from the groundbreaking Dartmouth Atlas of Health Care, as well as national Medicare data. They drew on American Hospital Association surveys to categorize hospital service areas according to those served by for-profits, not-for-profits or a combination.

Thirty-one states had at least one for-profit hospital service area, but these areas were concentrated in the southeastern U.S. and California. For-profit areas had higher acute hospital admission rates, were more likely to be affiliated with multi-hospital systems, and less likely to be associated with medical schools than non-profit hospital service areas.

The Dartmouth analysis controlled for these and other factors that might influence health care costs, including age, sex, race, regional price differences, location, number of physicians, medical school affiliation, multi-hospital system membership and mortality rates.

Total per capita Medicare spending in 208 stable for-profit hospital areas that remained under for-profit ownership over the study period was greater than in the 2,860 stable not-for-profit areas, the researchers found. The absolute difference was $452 in 1989, $402 in 1992 and $732 in 1995. Spending was higher in each category of service reviewed; the greatest increases were in hospital spending and home health care.

"Our results suggest that organizational behavior of for-profit hospitals leads to increased health care spending in the Medicare population," says Silverman. She stresses that further work is needed to determine precisely how for-profits treat patients differently and whether patients benefit from more costly care.

The study was supported by grants from the Robert Wood Johnson Foundation and the National Bureau of Economic Research.

For additional information, contact Elaine Silverman at 802-295-9363 extension 5545 or Elliott Fisher at 603-650-1822.

Hali Wickner

Return to News Archive for 1999