FAYETTEVILLE, Ark. — A University of Arkansas economist studied performance data from hundreds of U.S. psychiatric hospitals and found that private, for-profit hospitals -- the most profitable psychiatric hospitals and most prevalent in terms of ownership type -- cycle patients through their doors faster than not-for-profit hospitals and public hospitals operated by federal, state and local government. Despite their higher profit margins, however, private, for-profit hospitals are not more efficient than private, not-for-profit psychiatric hospitals.

“Compared to other hospital ownership forms, for-profit hospitals in our sample provided patient care at a lower average cost and kept patients in the hospital for a shorter time on average,” said Gary Ferrier, professor of economics in the Sam M. Walton College of Business at the University of Arkansas. “The shorter length of stay and lower per-patient expenditures in the for-profit setting may indicate a 'cream-skimming’ for selective patients.”

Ferrier defined cream-skimming as a process in which hospitals keep the most profitable patients -- those with the best insurance coverage or those who are the least costly to care for.

Ferrier and Vivian Valdmanis, associate professor of health policy at the University of the Sciences in Philadelphia, used two data-analysis techniques -- data envelopment analysis and cluster analysis -- to examine the relative performance of 506 U.S. psychiatric hospitals in their use of resources. Cluster analysis is a research method used to group subjects into natural and meaningful categories. Data envelopment analysis helps researchers construct best-method models so that individual subjects -- in this case, psychiatric hospitals -- can be compared and measured based on an ideal of efficiency and productivity.

Ferrier and Valdmanis measured efficiency by hospitals’ management of resources in terms of various “inputs” and “outputs.” Inputs included resources that hospitals put into the system, such as the number of full-time physicians, nurses and other personnel, as well as the potential for capital based on the number of staffed and licensed beds. Outputs were the same as services provided. Specifically, they included the number of inpatient days defined by payer type -- Medicare, Medicaid and private. The researchers examined data based on ownership type, which were defined as non-federal state and local public hospitals, federal public hospitals, private for-profit hospitals and private not-for-profit hospitals.

The researchers then analyzed raw data and developed four efficiency rankings: “very efficient,” “moderately efficient,” “moderately inefficient” and “very inefficient.” Almost 40 percent of private, not-for-profit hospitals were considered very efficient. These hospitals were better at employing inputs to produce a given level of outputs. In other words, they did a better job of managing personnel and other resources relative to services provided. In contrast, private, for-profit hospitals, despite their higher profit margins, were not as efficient. Only 22.1 percent of all private, for-profit hospitals were very efficient. They failed to minimize input costs, which affected profit. Still, these hospitals made more money than their private, not-for-profit competitors because they charged more for services and avoided unprofitable patients.

  Fourteen percent of the non-federal, state and local public hospitals made the highest efficiency category, while no federal, public hospitals were classified as very efficient. Most (66.8 percent) private, for-profit hospitals were considered moderately efficient. Almost half (49.3 percent) of all private, not-for-profit hospitals achieved this efficiency ranking. Only one federal, public hospital was moderately efficient, and approximately one out of 10 non-federal, state and local public hospitals made this category.

The overwhelming majority (83.3 percent) of federal, public hospitals were moderately inefficient. Fourteen percent of non-federal, state and local public hospitals made this category, while 12.3 percent of private, not-for-profit hospitals and 9.9 percent of private, for-profit hospitals were moderately inefficient. Two out of three (67.3 percent) non-federal, state and local public hospitals were considered very inefficient, and, surprisingly, three, or 1.1 percent, of private, for-profit hospitals also received the poorest efficiency ranking. No federal, public hospitals and no private, not-for-profit hospitals were ranked as very inefficient.

De-institutionalization may explain why public hospitals at all governmental levels are inefficient, Ferrier said. Many of these hospitals have shut down, but those that remain open may not have adjusted their inputs (doctors, nurses, other staff and beds) according to decreased services. In other words, these hospitals may be open for reasons other than psychiatric care for patients.

“One reason for their inefficiency is that an objective function of public hospitals is to provide employment opportunities for the local population,” Ferrier said. “So, due to 'pork barrel’ politics, these hospitals remain open even though the apparent demand for services, particularly inpatient care, is low.”

Ferrier said that overall their findings support earlier studies reporting that not-for-profit psychiatric hospitals offer a greater number of services than for-profit hospitals, and that government hospitals tend to care for the chronically ill.

The researchers’ study, published in the Journal d’Economie Medicale, a leading European health economics journal, was unique because it addressed the use of resources by psychiatric hospitals. Previous studies examined efficiency and productivity based only on accounting ratios, such as expenditures per patient, staff per patient and profit margins.