Relations between hospitals and physicians are broken. Can they be improved by lodging accountability for cost and quality in delivery systems consisting of local hospitals and the physicians who work within and around them?

Extended hospital medical staff. The prospects for these still mostly virtual systems –- dubbed “extended hospital medical staffs” by Dartmouth professor of medicine Elliott Fisher –- was one of the topics of discussion in a Dec. 6 conference call involving Fisher, Urban Institute senior fellow Bob Berenson, and Project HOPE senior fellow Gail Wilensky. (The full transcript is posted in the Contributing Voices section of our blog.) These analysts authored the three main papers in a Dec. 5 Health Affairs Web Exclusive package on hospital-physician relations [2-week free access], and we asked them to come together and react to each other’s ideas.

In his paper, Fisher and his coauthors wrote: “Virtually all physicians are either directly or indirectly affiliated with a local acute care hospital, whether through their own inpatient work or through the care patterns of the patients they serve.” And “because most patients receive their care within the context of” these networks, “the hospital and its extended medical staff provide a natural organizational setting within which to improve the overall experience of care.”

Some reservations. However, during the call, both Berenson and Wilensky expressed reservations about this approach. Berenson, who worked in the Health Care Financing Administration (which is now the Centers for Medicare and Medicaid Services) under President Clinton, noted that medical staff organizations at hospitals have proven “dysfunctional. . . . So the question I would have for Elliott is: How do you lodge accountability without control?” There “would be many members of the medical staff who simply don’t have any common business relationship to each other and would not want to participate,” he observed.

Wilensky warned that if Fisher’s new entities were imposed on hospitals and physicians by payers, rather than entered into voluntarily, “it might solve some problems, but it would exacerbate the issues of conflict between hospitals and physicians because of the change in power structure implied by moving to this model.”

In their paper, Wilensky, a HCFA administrator under President George H.W. Bush, and coauthors suggested allowing physicians and hospitals to enter into voluntary, comprehensive, “gain-sharing” arrangements, as a transitional strategy towards more integrated forms of care. They said this could overcome the mismatch between Medicare’s hospital payment system, which encourages efficiency through flat per discharge payments, and the program’s physician payment system, which offers a separate fee for each service.

Fisher was candid about the difficulties his idea would face, and but he argued that it was still worth pursuing to improve coordination across different sites of care. “I think just starting . . . with performance measurement at this level, and then thinking about how pay-for-performance could reward performance across hospitals and their staffs, would be a logical first step,” he said.

Medicare physician payment. Fisher also suggested that the extended hospital medical staff could function as a useful unit of volume measurement in Medicare’s physician-payment “sustainable growth rate” (SGR) formula. “There could be a mixed model of voluntary if people want to, or we can assign you to a group if you’re not willing to. Moving volume constraints such as the SGR down to a level where all the physicians can easily know who each other is” could change the “current financial incentives to put in a new MRI or to start a new ambulatory care surgery center,” Fisher said.

Berenson offered another idea for SGR reform, suggesting that large multispecialty physician practice groups could be allowed to “opt out” of the SGR formula: “If they’re going to be held accountable for cost through a different mechanism, and that mechanism is a sort of a shared savings review, then they opt out of the SGR, and the SGR is the default that the physicians who haven’t opted out are left in,” Berenson explained.

Gain sharing. In her paper and during the conversation, Wilensky emphasized that she was advocating broad-based gain-sharing initiatives involving disease management programs, quality improvement efforts, and the like, not the more narrow sorts of gain sharing in which hospitals and physicians agree on which brand of medical device to use in order to get a volume discount. She called this these narrow initiatives “beside the point and trivial; . . . you take something that could be an important transitional strategy and give it a bad name because it doesn’t do anything.”

Berenson praised Wilensky’s focus on “gain sharing with a big ‘G’.” He expressed concern, “based on Jeff Goldsmith’s views about ‘consulting fees’ — as he puts in quotes — from device manufacturers to use their products exclusively, whether some forms of [narrower] gain sharing might actually just exacerbate that kind of behavior, of physicians and hospitals being bought off by device manufacturers.”

In the end, though, Berenson was skeptical that comprehensive gain sharing — or indeed any strategy — could restore the central role the hospital has traditionally played for physicians: “I think we have to assume . . . that there’s going to be continued migration of services just due to new technology, and then combined with convenience to the patient, etc., out of the hospital into ambulatory settings,” he said. “And so we need, I think, to also address basically how we’re paying for these services, and the distortions in payment both in the hospital payment system and in the physician payment system that just overly reward certain kinds of activities and under-reward other activities.”

Click here for a full transcript.