To the Editors:

We appreciate Dr. Seaman's interest in our article. We anticipated that our article would engender some discussion and refer the reader to the editorial by Wortmann in this issue (pp. 643–645) commenting on our conclusions and the state of clinical academic rheumatology.

Dr. Seaman raises some important points for consideration. His first point is that ordering tests, procedures, and therapies could present a conflict of interest if a clinician-educator's academic rheumatology salary was dependent on this revenue. Although such is possible, we have more confidence in the integrity of our colleagues in academic medicine who are charged with teaching appropriate care and are the role models for fellows in training programs. God help our subspecialty if this is not the case. Additionally, our ordering of indicated tests, procedures, and therapies is no different from the cardiac catheterizations, colonoscopies, and other procedures that our academic colleagues perform. Do they perform unnecessary procedures? We hope not.

We do not propose to increase salary based on downstream income. Rather, we had hoped that our article would primarily provide data to practice administrators (who generate monthly reports pushing academic clinicians to “bill their salaries”) that rheumatologists do generate substantial revenue for the medical practice and the hospital. With this realized, administrators might be in a position to accept the shortfalls to “billing our salaries” that many of the nonprocedural subspecialties face because of uncompensated teaching time, poor insurance contracts, and the burden of academic taxes. The reality of today's environment is that when seeing patients becomes the clinical academic rheumatologist's sole function, he or she is unlikely to remain in academic medicine. Only in the unusual situation in which salaries are 100% covered regardless of clinical income, such as in certain government or military health care systems or when departments of medicine function as group practices, is this not the case. Indeed, over the past 10 years in which I have been at the University of Colorado, 3 academic rheumatologists who worked primarily as clinician educators have left for private practice. This attrition is not unique to the University of Colorado, which remains a terrific place to work. Academic rheumatology cannot afford this loss of talented clinician-educators simply because of inappropriate compensation models in academic practice plans.

Dr. Seaman points out that 40% of our downstream revenue was generated by prescriptions for IVIG, which he says “has limited therapeutic use in rheumatic diseases and carries considerable cost and some risk.” We would like to point out that this therapy was used in only 4 patients with rheumatic conditions (including dermatomyositis) that were refractory to all other immunosuppressive therapies. We would contend that most busy academic rheumatology practices at tertiary referral centers where the most challenging patients are referred have at least 4 patients with complex disease who are receiving IVIG. Furthermore, we think it unlikely that a clinical academic rheumatologist would have an incentive to order IVIG or other expensive therapies or procedures that were not indicated. As most clinical rheumatologists have experienced, all medical insurance carriers require prior authorization for these types of therapies and procedures. This process is quite time-consuming and likely to result in a denial unless the clinician can rigorously justify ordering the test or procedure.

Dr. Seaman takes issue with our recommendation that academic rheumatologists should see mostly patients with inflammatory disease and not those with fibromyalgia. He indicates that our conclusion that the limited downstream revenue generated from fibromyalgia patients (as compared with that generated from patients with inflammatory disease) is insufficient reason not to see patients with fibromyalgia, and that a better reason is that managing fibromyalgia does not best utilize the expertise of the rheumatologist. Although we agree with his statement, many of our colleagues who provide care for patients with fibromyalgia would disagree. Hospital administrators, however, look at this issue a different way. They would like to avoid complaints from patients (and referring physicians) who cannot get access to the subspecialty care that the patient thinks he or she needs. Consequently, some administrators have mandated that all patients be seen. We hope that our downstream revenue argument will be more persuasive so that patients with inflammatory disease do have earlier access to our expertise (because initiation of definitive therapy makes a clear difference in outcome).

Dr. Seaman asks whether other subspecialty patients should replace rheumatology patients if those patients generate more downstream income. From the hospital administration's standpoint, patient groups that generate the most income are more desirable for the practice plan and the hospital. This is evidenced by aggressive advertising campaigns by hospitals for cardiology, oncology, and gastroenterology (not rheumatology) patients. Furthermore, in many hospitals (including our own), subspecialty practices are encouraged to limit access to patients without insurance. This creates moral and ethical dilemmas for practicing clinicians but points out that the administration's interest in the bottom line is clear.

Dr. Seaman agrees that compensation is inadequate for teaching time that takes away from the ability of clinicians to see more patients. He points out that there are professional problems with applying downstream income to offset this loss of income but offers no solution other than to read Kassirer's book On the Take (which we have read). We agree that any compensation system taking into account downstream income must be managed and monitored to prevent potential abuses, especially if it includes bonus plans for increased revenues. However, if the current compensation systems at many academic centers continue to apply only steeply discounted billing towards a clinical academic rheumatologist's salary base, it is likely that clinical academic rheumatologists will be less likely to stay in academic medicine for their entire careers. This attrition will result in fewer experienced teachers, lower quality of fellowship training programs, and inferior patient care.

Sterling G. West MD, MACP, FACR*, Pendleton B. Wickersham MD*, * University of Colorado Health Sciences Center Denver, Colorado.