option or by contracting directly with many or most providers in a community. As a result of this inclusiveness, plans have less ability to select carefully from among the larger pool of physicians those who provide higher quality care in their particular fields. It also prevents plans from creating a small group of practitioners who could easily work together on mutually agreed upon goals.

As pressure to increase the size of networks increases, the ability of each plan to affect practice decreases. For example, if a given health plan provides only 10 percent of a physician's business, no single plan has enough data, influence, or incentive to work with those physicians to improve quality. For one thing, they lose economies of scale. For example, obtaining data about the performance of 1,000 physicians for whom the health plan provides 10 percent of their patients is much more costly and less useful than obtaining data from 100 physicians for whom the plan provides 100 percent of their patients.

In addition, when plans deal with physicians and hospitals that are also in most other plans, there is a "free-rider" concern. In such an open-panel model, any serious investment in improving quality in one plan also benefits all of the plan's competitors. For example, sending a primary care physician to a course on dermatology so that she is better able to identify and treat skin lesions benefits not only the patients in that plan but those of its competitors as well.

In short, the market imperative for plans to engage most of the physicians in an area of nonexclusive relationships creates networks that are too large and too broad to address improvement efforts efficiently, even if plans are willing to accept the free-rider burden. One way out of this dilemma would be for physicians to organize themselves into manageable units, to assume financial risk, and to be accountable for the quality of the services that they provide. Collaboration is another strategy that plans might use to achieve economies of scale and avoid free-rider problems.

Potential Areas for Collaboration

Within the reasonable constraints of antitrust laws, how might plans collaborate on clinical initiatives that could lead to greater quality and efficiency? One possibility would be to carry out the objectives of the Health Care Quality Improvement Act of 1986. Currently, when plans identify possibly serious quality deficiencies in the care provided by contract physicians, they find that it is easier to terminate those physicians "without cause" than to follow the requirements of the Health Care Quality Improvement Act. These requirements include a due process hearing to identify grossly substandard care provided by the physician and the reporting of quality deficiencies to the National Practitioner Data Bank.

Similarly, many physicians identified as providing substandard care would prefer being terminated without cause and foregoing due process to avoid being reported to the Data Bank. By not reporting a physician, the plan can protect its own patients and the integrity of its own operations, but that does not help the



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement