The Utah Primary Care Experience

INTRODUCTION

I appreciate the opportunity to comment on the article by Magill and colleagues,¹ which describes a 7-year saga to operate a primary care network for an academic health science center. There are a number of notable things to comment on as one looks from afar. In doing so, I will compare their experience in Utah with ours in Colorado. Our approach, which I will describe, differed for several important reasons (most of them financial).

The early 1990s was a period of challenge and change for most academic health centers (AHCs). The emergence and rapid growth of managed care plans with their requirements that specialist services be preceded by a primary care physician (PCP) referral and an approved preauthorization from the health plan put these centers at a marked disadvantage when competing for insured patients. These managed care plans had the power, through contracts, to channel their members to specific physicians and hospitals. This power combined with these plans’ general belief that AHCs exemplified the phrase “go in for a cold, come out with a transplant” led to a general unwillingness to contract with AHCs.

In reality, in many places, they were right. For decades, academic PCPs had 2 obligations—to train new physicians and to provide the highest quality of medical care. These missions resulted in excessive use of diagnostic facilities and specialist referrals, and higher costs for patients when cared for by an AHC—with serious questions as to the true medical necessity of many of the services. To make matters worse, the primary care faculty of an AHC was remarkably small when compared with its specialist faculty. As a result, health plans had minimal interest in contracting with academically based PCPs. And without access to those PCPs, referrals of insured patients to academic specialists often declined.

It quickly became apparent that AHCs would have to expand their PCP base, modify their clinical practice habits, and capture patient populations that would sustain the costs associated with these changes. The approach described by Magill and colleagues was one way to try to attain these goals.

THE UTAH EXPERIENCE

I offer 3 observations on the Utah experience. First, the decision of the University of Utah to purchase practices was similar to actions taken by other health systems in the 1990s. The recognition that with the advent of managed care, practice plans of medical school faculty were not structured to compete effectively for patients because they were heavily populated with specialists and had few PCPs on their faculty is what led many schools to either add primary care faculty or buy primary care practices. The experience reported in this paper of staggering losses when the practices became liabilities instead of assets paralleled that in Philadelphia and possibly at other sites.

Second, the decision to try to turn around the operations after 2 years and more than $40 million in losses was obviously a difficult option, but likely the only politically viable one. To spend $37.4 million (one half of the reserves of the departments) and have nothing to show for it and then just close the operation and write it off without giving the departments any hope of being repaid was clearly not as good as trying to turn the situation around and do what was necessary to try to make the original idea work.

Third, the integrated governance of the school, practice plan, and hospital with a single balance sheet made it easier to determine success and to allow all the players to share in the turnaround and see the benefit of the $5 million a month in referral income and the $150 million net cumulative benefit. If the hospital, school, and practice plan were all different legal entities (as is true in Colorado), this outcome might have been more difficult to accomplish.

THE COLORADO EXPERIENCE

The Utah scenario could not have happened in Colorado. To put our situation into context, our University of Colorado Hospital (UCH) was spun off as a separate public authority with its own board of directors in 1991. The practice plan, University Physicians, Inc (UPI), is a 501(c)(3) organization, closely aligned with the school of medicine. As dean, I am president of the board of UPI, and my senior associate dean for administration and finance is also executive director of the practice plan. The school of medicine is a unit of a public health sciences center.

In 1993, UPI and UCH engaged in a joint strategic planning effort. We knew we needed a primary care network if we were to grow our clinical enterprise. Neither our Department of Medicine nor our Department of Family Medicine had robust clinical enterprises. We rejected the notion of buying primary care practices for a very simple reason: we did not have the money. As a result, we undertook a different strategy: we decided that just expanding primary care was not enough to do well in a competitive managed care environment. We needed to assure ourselves that we could expand our clinic population to more than 250,000 patients, so UPI and UCH started 3 health plans. We decided that rather than whine about managed care, we would practice it the best we could. Over the next several years, we started a capitated health plan for the faculty and staff of the University of Colorado system and became owners of the successful Colorado section of the 16-state TriCare Health Plan for military retirees and beneficiaries and part owners of a Medicaid health maintenance organization (Colorado Access) for the state of Colorado. Together, our ownership, administration, or both brought us the 250,000 patients we thought we needed.

The next step was to develop the primary care network to see these populations. The Colorado University Health Plan had 15,000 patients spread from Boulder through the Denver Metropolitan area to Colorado Springs (a 4,000–square mile area); TriCare and Colorado Access were statewide programs. We needed a primary care network that was statewide in nature. Rather than purchase one, we contracted for one. We were fortunate that at the same time as we were developing these programs, we were also expanding our undergraduate medical education program—sending students to PCP offices for half-day weekly experiences for 3 years. These physicians asked for reimbursement for their teaching time, something we had no ability to do within our resources. Alternatively, they were very willing to have a contract to see our insured patients and, under the contract, were willing to refer all their patients to us for specialty care and hospitalization. This strategy has worked successfully for more than a decade.

Unlike the Utah PCPs, many of our physicians are involved in teaching our students and residents. Most hold volunteer (clinical) faculty appointments and are not employees of the university. We also have relatively small full-time faculty practices in general internal medicine and family medicine. These practices have required a $1.4 million subsidy from both UPI and UCH for the last several years (50% from each unit). The downstream revenue to the hospital and the specialists in the practice plan has far exceeded this figure.

LEGAL CONSIDERATIONS

One of the potential barriers to these types of arrangements between separate practice plans and hospitals is the Stark laws. These laws prohibit a physician from referring patients covered by Medicare or Medicaid for laboratory services or certain other “designated health services” to an entity if the physician has a financial interest in that entity. When the school, practice plan, and hospital are all within or under 1 legal entity, there are no Stark issues. When the practice plan, hospital, or both are separate entities, then Stark could apply—except that there has been an academic safe harbor created to allow these entities to support each other. As long as there are written affiliation agreements between the major teaching institutions, the school of medicine, and the practice plan, most (but not all) of the revenue transfers between teaching hospitals and medical school practice plans are within the safe harbor.

In conclusion, it is obvious that medical schools and teaching hospitals have an inexorable drive to grow and succeed. Achieving an efficient and profitable clinical enterprise is not easy. No one has ever accused school of medicine faculty of being able to practice as efficiently as physicians in private practice in the community. But few schools have the resources to not rely on practice income. For schools relying on such income and situated in very competitive managed care environments, having examples such as the Utah and Colorado experiences to learn from could be helpful in trying to maintain the clinical revenue stream.