Gatekeeping

To the Editor

The Sounding Board by Franks et al. (Aug. 6 issue)^* was timely. As a general internist for the past 18 years, I have an additional observation regarding strategies to minimize overtreatment. There has been a deterioration in communication and medical etiquette between generalists and specialists. Too commonly, patients are referred to specialists and diagnostic and therapeutic plans are made without a timely discussion with the referring physician. Our patients would be better served if we discussed their consultation with them after the consultant had had an opportunity to communicate with us.

Over the past two decades, with the explosion in the number of specialists, the function of the primary physician as the overseer of patient management has been steadily eroded, and unfortunately, we generalists have acquiesced in this process.

Paul A.G. Cohen, M.D.
301 S. Eighth St., Philadelphia, PA 19106

To the Editor

I am deeply troubled by the focus of the Sounding Board article by Franks et al. on the role of the gatekeeper in preventing patients from receiving care. With reference to my specialty, there is a literature^{1 2 3 4 5} to show that dermatologists seeing patients directly are both cost efficient and more accurate and are able to care for patients with fewer visits.

Michael J. Franzblau, M.D.
1300 S. Eliseo Dr., Greenbrae, CA 94904

To the Editor

In health maintenance organizations (HMOs), the gatekeeper has a tremendous financial incentive to underuse resources. Perhaps another system that does not link the gatekeeper's income so closely with the deprivation of care might be more equitable.

Jerome S. Reich, M.D.
16800 N.W. 2nd Ave., North Miami Beach, FL 33169

To the Editor

If you believe that physicians who overuse resources to maximize fee-for-service profits will not underuse them to maximize profits in an HMO, then it is easy to believe that there is inadequate evidence that "financial incentives to limit services adversely affect patients' outcomes." The following experience might argue against the notion that merely converting a fee-for-service physician to an HMO physician will create a paragon of virtue.

A 25-year-old patient with florid hyperthyroidism was sent to me from a Flint, Michigan, HMO for a consultation, but no testing or treatment. She returned to the HMO with a request for permission to test and treat with iodine-131. One month later she came to me with an authorization to test and treat. A pregnancy test was positive. After an in-depth discussion of the risks of the treatment of hyperthyroidism during pregnancy, the patient asked me to treat her. She very much wanted the baby. She returned to the HMO with a request for permission to begin treatment with antithyroid drugs, with follow-up every three weeks until delivery. She came back six weeks later, having lost the baby, and requested treatment with iodine-131. When I asked her what treatment she had received in the interim, she replied "Nothing."

Joel I. Hamburger, M.D.
29877 Telegraph Rd., Southfield, MI 48034

To the Editor

I should like to call your attention to an error in the Sounding Board article by Franks et al. While discussing the risks of overtreatment that might be reduced by effective gatekeeping, the authors incorrectly report some findings of the Harvard Medical Practice Study^* of New York State hospital records. They state that in 3.7 percent of the hospitalizations there was evidence of adverse events due to "medical mismanagement." In fact, as indicated in that study, the figure for all adverse events was 3.7 percent, whereas the figure for events judged to be due to negligence was 1.0 percent.

This error does not invalidate the authors' point that admissions to a hospital are associated with a risk of misadventure, and that unnecessary hospitalization exposes patients to needless risk. Correction of the error will, however, keep the record straight.

Norton Spritz, M.D.
Veterans Affairs Medical Center, New York, NY 10010

To the Editor

The authors reply:

To the Editor: Dr. Cohen notes, and we agree, that one problem with the management role of primary care physicians is inadequate communication between primary care physicians and their consultants. We have argued elsewhere for the need for a better understanding of the elements involved in the referral process,^* and we view this as part of a comprehensive effort to improve the capacity of primary care physicians to ensure the optimal use of resources by their patients.

Dr. Franzblau is concerned that our focus is on preventing patients from receiving care. Our intent was to address the risk of overtreatment and to review the literature suggesting that the primary care physician, by virtue of training and clinical experience, is best able to balance the patient's needs with the appropriate mix of health services, thus avoiding both overtreatment and undertreatment. As we noted, most but not all studies indicate that for specific medical problems the quality of care provided by specialists is similar to that provided by primary care physicians, and most studies show that primary care physicians use fewer resources. Again, more research is needed on the impact on the overall care of patients who refer themselves to specialists, as compared with those who are referred by primary care physicians.

Finally, we agree with Dr. Reich that there is a risk of undertreatment in HMOs, and Dr. Hamburger provides an anecdote about undertreatment from his own practice. There are also many anecdotes suggesting the occurrence of overtreatment, and our review of more systematic studies suggests that the risk of overtreatment is a greater hazard than the risk of undertreatment and that the quality of care provided by HMOs is equal or superior to that provided by the fee-for-service sector.

The central purpose of our paper was not to argue for HMOs themselves, but to suggest that the quality of care provided to HMO patients is advantageous because of the emphasis in many HMOs on a strong base of primary care services. How primary care providers — and specialists —are paid for their services is a related, but separate, issue. Under any reimbursement system, primary care physicians have an important function in ensuring the delivery of high-quality care to all patients, one that has been undermined historically by fee-for-service practice. Ideally, reimbursement for physicians' services should be structured in a way that provides no strong incentives for either overtreatment or undertreatment.

These views do not necessarily reflect the views or official policy of the Public Health Service or the Department of Health and Human Services, but are solely those of the authors.

Peter Franks, M.D.
University of Rochester School of Medicine, Rochester, NY 14642

Carolyn M. Clancy, M.D.
Paul A. Nutting, M.D., M.S.P.H.
Agency for Health Care Policy and Research, Rockville, MD 20852