Generalists or specialists--who does it better? (Value-Based Health Care).

THE U.S. HEALTH CARE SYSTEM IS UNDERGOING A historic transformation. Unbridled increases in the cost of health care during the past two decades, the development of a highly competitive global economy, and a burgeoning federal deficit have forced the major purchasers of heath care, businesses, and the U.S. government to focus on cost containment. (1)

Unable to control the cost of care for their employees and beneficiaries, the purchasers of health care are increasingly turning to managed care companies for help. Managed care companies use a variety of strategies to hold down costs, including restricting access to specialists and the services they provide. A method favored by many managed care companies is to require beneficiaries to select a primary care provider, who is almost always a generalist (family practitioner, general internist, or pediatrician), to guide them through the health care system, the so-called gatekeeper model.

The primary care physician often receives a financial incentive to limit referrals to specialists and/or is required to get approval before referring a patient. Moreover, primary care physicians are encouraged to provide as much of the needed services themselves. The rationale for the gatekeeper model is that primary care providers use less clinical resources than specialists, thus holding down the costs of health care. Recently, some managed care companies have begun to allow patients direct access to specialists, but they still rely on the primary care provider to coordinate and provide the bulk of the health care services.

Evidence clearly demonstrates the value of a primary care provider. Absence of a regular source of health care results in worse outcomes and lower patient satisfaction and patients who have a primary care provider use less medical resources than those who don't. (2) Thus, the concept of a primary care provider guiding a patient through the health care system and serving as the coordinator of care not only makes sense intuitively, but is supported by evidence.

The role of the generalist, (a.k.a. family doctor) is not new. The nobel laureate, Alexander Solzhenitsyn, wrote in his novel Cancer Ward, "generally speaking, the family doctor is the most comforting figure in our lives...he knows the needs of each member of the family...there is no shame in taking to him some trivial complaint you would never take to the outpatient clinic...and yet all neglected illnesses arise out of these trivial complaints.

The value of primary care notwithstanding, questions have been raised about whether generalist physicians, who are the primary care providers for most patients, have the skills to take on these new responsibilities and whether providing incentives to restrict access to specialists and/or provide care themselves compromises the quality of care. (3, 4, 5, 6)

The new role of the generalist in managed care has had a profound effect on the relationship between generalist and specialist physicians. The generalists, feeling a new-found sense of respect and power, are taking back some of the care that they once delegated to specialists and receiving a financial incentive to do so. In effect, generalists are saying, "We can do It just as well for lower cost." In contrast, the specialists, whose incomes are threatened because of lower payments for services rendered, loss of business, and an oversupply of physicians in the medical specialties, believe that their greater knowledge In their specialty area means that patients have better outcomes and satisfaction. Specialists and generalists are now in direct competition with one another.

We sought to determine whether the care provided by generalist and specialist physicians is different in terms of quality and cost. To address this question, a literature search was performed, using Medline and Healthstar from 1992-1997, to find articles comparing the health care provided by generalist and specialist physicians. This article summarizes the results of this search and provides a commentary on the role of physicians in assuring quality of care for their patients.

Literature review

The literature review yielded 21 articles that compared quality, patient satisfaction, and/or cost of care delivered by generalists, primary care providers, and specialist physicians. Articles comparing specialists to generalists in surgical specialties or specialists to subspecialists were excluded. The remaining articles could be effectively grouped according to which issues they assessed: medical knowledge, practice patterns, health outcomes, resource utilization and costs of treatment, and patient satisfaction. Several of the articles assessed multiple outcomes.

1. Medical knowledge

Five articles addressed the hypothesis that specialists have more complete knowledge within their field than generalists do. Ayanian et al surveyed cardiologists and generalists regarding treating acute myocardial infarction. (7) Their results demonstrated that cardiologists more often knew which medications had been proven to reduce mortality in acute myocardial infarction and were more likely to use them. For example, cardiologists were significantly more likely to answer that they would prescribe thrombolytics than internists or family practitioners did (94 percent versus 82 percent and 77 percent, respectively).

Mazzuca et at assessed generalist and rheumatologist knowledge in a case scenario for the treatment of hip osteoarthritis in patients with and without contraindications and whether they selected non-steroidal anti-inflammatory drugs. (8) The results of this survey, like Ayanian's, bespeak to the fact that specialists have more current and complete knowledge regarding treating disease within their specialty. Solomon et al compared the ability of dermatologists and generalists to diagnose dermatologic conditions via slides and a survey. (9) When the results were tallied, family practice residents correctly identified significantly fewer slides than did dermatology residents (48 versus 93 percent).

Shao et al surveyed specialists', generalists', and psychiatrists' attitudes towards and knowledge of depression. Psychiatrists possessed a more favorable attitude and demonstrated greater knowledge about treating depression--61 percent of generalists knew the efficacy of antidepressant medication compared to 81 percent of psychiatrists. (10) Finally, Hnutiak et al showed that generalists did not interpret pulmonary function tests correctly one third of the time. (11) The rate of correct pulmonary function test interpretation by pulmonary specialists was not measured.

Specialists have more complete knowledge, both with regard to diagnosing and treating diseases within their specialty, than generalists in these studies assessing physician knowledge.

2. Practice patterns

Five of the articles examined practice patterns of specialists and generalists. The objective of these studies was to determine which group of physicians was more likely to be using the most up-to-date medical knowledge and/or using resources appropriately. Similar to the studies comparing physician knowledge, the assumption underlying these studies was that using the most up-to-date and appropriate resources translates into better outcomes. Most of these studies come to the same conclusion: generalists lagged behind specialists in using new diagnostic and treatment modalities.

Markson et al compared using Zidovidine by primary care providers and AIDS specialists. (12) The latter were more likely to be using this therapy soon after it was recommended by an expert panel, although over time both groups of providers were prescribing this medication. Similarly, Hirth et al found that gastroenterologists adopted antibiotic therapy for peptic ulcer disease more quickly than generalists. (13) Nearly all gastroenterologists (99 percent) used H. Pylon eradication therapy immediately after guidelines were released, compared to only two-thirds of generalists. In a recent study of patients with asthma in a large HMO, Vollmer et al, found that those cared for by allergists were more likely to conform to national asthma management guidelines and reported better quality of life than patients of generalist physicians. (14)

Schieber et al found that cardiologists were more likely to prescribe proven therapies for unstable angina than generalists. (15) The cardiologists used aspirin (78 versus 68 percent), heparin (84 versus 67 percent) and beta blockers (30 versus 18 percent) more often in their initial hospital management of unstable angina than general internists. In at least one study comparing practice patterns, specialists misused resources less often than generalists. Stein et al showed that both cardiologists and non-cardiologists were using radio nuclide stress tests "inappropriately," but cardiologists did so less often. (16) Thus, in these studies, specialists used up-to-date, recommended treatments in their specialty area more often than generalists.

3. Treatment outcomes, cost of care, and resource utilization

The third group of articles dealt with the issue of treatment outcomes, cost, and resource utilization when generalists and specialists treated patients with similar conditions. Schreiber's study of a cohort of patients admitted to a large community hospital with E unstable angina found that cardiologists treated a cohort with more severe disease and had a trend towards lower mortality (4 versus 1.8 percent), while incurring only marginally greater costs. (12)

Similarly, James Jollis and his colleagues used data from the Cooperative Cardiovascular Project to analyze outcomes of elderly patients admitted to either generalists or cardiologists with acute myocardial infarction. (17) They found that patients admitted to cardiologists received recommended drug therapies more often than those admitted to other specialties and had longer hospital lengths of stay. In addition, after adjusting for baseline differences in patient profiles, patients admitted to cardiologists were 12 percent less likely to die within one year than those admitted to generalists.

Jollis' findings are corroborated by a study by the Pennsylvania Health Care Cost Containment Council (PHC4) that compared the mortality of patients suffering acute myocardial infarction treated by either cardiologists, internists, or family physicians. (18) The PHC4 data demonstrated significant differences in the risk-adjusted mortality among the three physician groups (cardiologists: 8.6 percent; internists: 10.8 percent: and family practitioners: 11.1 percent), with cardiologist's patients having the lowest mortality and shortest hospital stay.

In a similar study, Mitchell et al examined the outcomes of patients who were hospitalized with acute stroke and cared for by neurologists, generalists, or other specialists. (19) The patients of the neurologists had a lower 90-day mortality (16.1 percent) than generalists (24 percent) or other specialists (25 percent), but the neurologists used more resources and generated higher charges. Finally, Zarling et al compared a large number of patients admitted to the hospital with acute diverticulitis under the care of internists, family practitioners, and gastroenterologists. (20) There were no differences among the three groups in mortality, but the gastroenterologists' patients had a shorter length of stay (despite more diagnostic tests) and an approximately 50 percent lower 30-day readmission rate.

Turning to the outpatient setting, the results of comparative studies on outcomes are somewhat different than those in the inpatient setting. Murphy studied outcomes of patients with subacute illnesses presenting to a busy emergency department in Britain who were randomly triaged to either a general practitioner or an emergency department physician. (21) There were no significant differences in morbidity and mortality or subsequent return to the emergency department within the next 30 days for the patients seen by the general practitioners or emergency physicians. Furthermore the general practitioners performed fewer investigative studies, referred patients less often, and saved an average of [pounds]60 per episode of care when compared to the emergency department staff.

Two studies addressing cost and outcome of back pain were conducted by Carey et al 22 and Shekelle et al. (23) Cary compared outcomes of care for episodes of back pain provided by generalists, chiropractors, and orthopedic surgeons. The time to functional recovery, return to work, and complete recovery from back pain were similar in all groups, while generalists used significantly fewer resources. Shekelle's study quantified resource utilization and cost of episodes of care for acute back pain delivered by different types of providers. Their analysis demonstrated that while chiropractors billed the least per office visit, because the number of visits was significantly higher than for all other providers, the mean cost per episode of outpatient back pain was nearly three times greater for chiropractors than for the general practitioners and twice as much than general internists. Orthopaedic surgeons also were significantly more expensive than generalists.

Greenfield et al reported on the largest and most carefully done outcomes study to date. (24) Their group examined data from the Medical Outcomes Study and compared patients with adult onset diabetes or hypertension treated either by general internists, specialists (cardiologists and endocrinologists), or family practitioners. They documented no significant differences in any of several measured outcomes in either the hypertensive or the diabetic population at two and four years, regardless of who delivered their care, with the exception of diabetic foot ulcers, which were significantly improved under treatment by endocrinologists. Of note, Greenfield et al had previously reported that specialists used resources at a higher rate than generalists in treating these types of patients. (26)

Finally, Scott et al did a randomized trial of care of depression by psychiatrists, social workers, or generalists. After 16 weeks there was no significant difference in outcomes among the patients who were cared for by different providers. (26)

In summary, patients who have self limited or common chronic conditions appear to have the same outcomes, regardless of the specialty of their provider.

4. Patient satisfaction

Four articles also reported on patient satisfaction with medical care, As noted, Carey et al conducted a prospective observational study of randomly selected chiropractors, surgeons, and primary care providers to focus on their care of acute back pain. While the charge per episode of back pain was highest when chiropractors provided the care, patient satisfaction also was significantly greater. (22) Murphy et al also examined satisfaction of patients in the emergency room and found no significant difference in the satisfaction between patients treated by general practitioners and emergency medicine physicians. (21)

Scott et al's study on the treatment of depression revealed that satisfaction was significantly greater for those who saw social workers than with psychologists, psychiatrists, or general practitioners. (26) Finally, Owen et al found that patients in a dermatology clinic were more satisfied with the care of the dermatologist than with their general practitioner for the same condition. The latter is not surprising given that they were referred in the first place. (27)

The articles reviewed did not clearly resolve the issue of whose patients have greater satisfaction with their care; patient satisfaction seems to correlate with the frequency of contact, if outcomes are the same.

Who does it better?

Overall, the results of the literature review were not surprising. Generalists appear to be able to discern minor from more serious illness in otherwise healthy people, and they have similar outcomes to specialists in managing both self limited and common chronic conditions, such as hypertension. In contrast, specialists tend to have more knowledge in their specialty area, are more likely to use up-to-date and effective treatments, and often have better outcomes when taking care of patients within their specialty, particularly those who have acute, serious illnesses. The issues of which group uses clinical resources most appropriately and whose services result in greater patient satisfaction were not clearly answered by these studies.

Despite these general findings, the value of these studies is limited in determining who delivers the most effective or efficient care. The overall quality of the studies was mediocre. First, most contained significant limitations or flaws in design. Indeed, only one study was a randomized clinical trial, in which the patients were comparable in the different treatment groups. Moreover, most focused on a single episode of illness or care, examined only a few possible outcomes, did not define benchmarks for outcomes, nor measure long term outcomes or total health care costs.

A second problem is that the results did not always justify the conclusions. Better knowledge does not necessarily lead to better outcomes as demonstrated by comparing the studies of Shoa et al (10) and Scott et al (26) on depression. Moreover, the outcomes of patients may have been determined by factors other than the physician's specialty. For example, patients with diseases such as stroke or acute myocardial infarction may have better outcomes because they were cared for on special units or by nurses and other staff who are trained to deal with their unique problems. (28) The type of hospital is also important in determining outcomes--patients admitted to major teaching hospitals have better outcomes. (29) This and other potential confounders of the relationships between physician specialty and health outcomes were not measured or analyzed in most of the studies.

Another concern is that the differences in outcomes and cost between specialists and generalists were often modest and potentially could be rectified by implementing quality improvement programs. Also, there were large within group variations suggesting that the performance of individual physicians, regardless of specialty, was an important determinant of outcomes. Perhaps a physician's performance should determine whether he or she should care for patients with a specific illness, rather than specialty training.

Finally, there may be a publication bias in this literature. Indeed, the results could often be predicted on the basis of whether a generalist or specialist conducted the study. Both groups used their studies in a self-serving way, while attempting to expand or preserve their niche in the patient care marketplace. The common refrain was "we do it better and therefore the patient should be under our control. In fact, many of the studies contained distinct undertones of hostility, quite possibly due to the tension generalists and specialists feel as the health care market increasingly ratchets down their reimbursement.

As we pondered the results of these studies, we realized that asking who does it better, the generalists or the specialists, was the wrong question. Managed care companies and other health care purchasers have been successful, in part, because of the oversupply of physicians and the ability to pit one group against another in bidding for contracts. While competition is inevitable and there will be ever increasing pressure to do more with less, physicians must step back and ask the question, as professionals, how can we best create value for our patients?

The concept of consumer perceived value is complex. In an increasingly competitive environment, health care consumers will have an array of choices in terms of providers and price--they will chose the providers that offer the greatest perceived value. Of course, judgments of value are limited by available information, mobility, and income. Nonetheless, consumers will look at a provider's attributes (outcomes, service, provider skill and empathy, physical environment, location, etc.) and weigh these against the costs of acquiring the service (price, time, energy, and psychic costs) to determine who offers the greatest value. It is likely that consumers have different concepts of value and that they focus on the sum of their experience, rather than on an episode of care. Thus, health care providers must know what's important to their customers and how to create lasting value for them.

The literature reviewed addressed who provided better episodic care in a highly fragmented system in which work is designed around the specialization of physicians. It is a system that encourages separateness and rewards each individual providing service, whether or not it is in the best interest of the patient. This traditional, highly compartmentalized approach does not maximize consumer value, but rather provider value.

The approach providers should be taking is to design systems of care that maximize the long term health of patients and deliver service in a coordinated, efficient manner. The emphasis should be on care across the continuum of service providers and through time, rather than on individual performance, (i.e., total quality and cost). Physicians must organize themselves to deliver health care in a way that emphasizes quality, service, and efficiency realizing that there will be tradeoffs in these three parameters.

Thus, generalist and specialist physicians, in collaboration with other providers, should design systems of care where consumer value, either for specific conditions or global health, is defined and maximized over the long term. The roles of generalists and specialists can then be determined. Some generalizations can be made, however. Primary care providers, including nurse practitioners and physician assistants, can play an important role as coordinators of care. Primary care providers should serve as an entry point for patients in deciding what types of services may be needed to Improve their long term health and they should treat minor or self limited and most chronic illnesses. They should act as an advisor, a source of information, and help patients make rational decisions, weighing costs and outcomes of treatments that are recommended.

However, primary care providers must be taught to appropriately screen and refer patients to specialists when needed and should have incentives to restrict access to care. (30) Griffith et al provide a refreshing reminder of what is possible when physicians of different specialties subjugate their pride and focus on how to deliver effective health care. (31) They conducted a study comparing diabetic patients screened for retinopathy via direct opthalmoscopy by generalists and compared the results to ophthalmologists screening retinal photographs. Prior to study Initiation, all primary care physicians had received instruction and education in direct opthalmoscopy from the specialist. The direct examination screening process had a sensitivity of 100 percent and a specificity of 93 percent when compared to retinal photography. This study provides a rare example of cooperation between specialists and generalists and focuses on what's good for the patient.

Specialists will continue to play an important role in our health care system. They must teach primary care providers how to best manage illnesses within their specialty domain and when and to whom to refer patients for further treatment. A recent article by Leveton et al showed that consultation by a diabetes care team, led by an endocrinologist, shortened the length of stay by 56 percent over patients managed by internists alone. (33) Moreover, specialists can generate new knowledge and make advances in technology within their specialty.

The evidence does suggest that specialists are more skilled at treating acutely ill, complex patients, as well as those with unusual diseases. Specialists may play the role of primary care provider for the latter and for patients who have severe chronic illnesses. On the other hand, effective and efficient primary care requires broad knowledge. (33) Specialists may do a better job in their area of expertise but have deficiencies in other areas outside of their specialty. These will need to be rectified if specialists are to serve the role of primary care provider for selected groups of patients.

Importantly, the role of each provider must be defined in the context of a system developed and managed by physicians that emphasizes outcomes, quality, and efficiency. These health systems should be designed to receive capitation or prospective payment to eliminate the perverse incentives that are dividing the physician community. Moreover, it will give physicians control of the system and allow them to be creative in how services are delivered, ultimately focusing on the relationship with and duty to the patient. They can compete against managed care companies and other organizations whose primary strategy is to lower health care costs. We urge physicians to stop conducting studies that compare one professional to another, and work together to create systems of care that focus on consumer perceived value.

References

(1.) Sheeby B, Bolch E. Rosengart C. America's Health Care: The Big Squeeze. The Governance Institute. 1996.

(2.) Gabriel SE. "Primary Care: Specialists or Generalists." Mayo Clinic Proceedings 1996:71:415-418.

(3.) Brook R, McGlynn E, Cleary P. Quality of Health Care Part 2: Measuring Quality of Care. The New England Journal of Medicine. 1996:335:996-970.

(4.) Neilson E. When Does A Generalist Need a Specialist? Journal of General internal Medicine. 1996:11:247-248.

(5.) Greenfield S. Dividing Up the Turf: Generalists versus Specialists. Journal of General internal Medicine. 1996:11:245-246.

(6.) Ringel S. The Neurologist's Role in Stroke Management. Stroke. 1996:27:1935-1936.

(7.) Ayanian JZ. Hauptman PJ, Guadagnoli E, Antman EM, Pashos GL, McNeil BJ. Knowledge and Practices of Generalist and Specialist Physicians Regarding Drug Therapy for Acute Myocardial Infarction. New England Journal of Medicine. 1994:331:1136-1142.

(8.) Mazzuca SA, Brandt KD, Anderson SL, Musick BS, Katz BP. The Therapeutic Approaches of Community Based Primary Care Practitioners to Osteoarthritis of the Hip In an Elderly Patient. The Journal of Rheumatology. 1991:18:1593-1600.

(9.) Solomon BA. Collins R. Silverberg NB, Glass AT. Quality of care: Issue or oversight in health care reform? Journal of the American Academy of Dermatology. 1996;34:601-607

(10.) Shoa WA, Williams JW, Shuko L, Badgatt RG, Aaronson B, Cornell JE. Knowledge and Attitudes About Depression Among Non-generalists and Generalists. The Journal of Family Practice. 1997;44:161-168.

(11.) Hanatluk O, Moores L, Loughney T, Torrington K. Evaluation of Internists' spirometric interpretations. Journal of General Internal Medicine. 1996;11:204-208.

(12.) Markson LE, Cosler LE, Turner BJ. Implications of generalists' slow adoption of zidovudine in clinical practice. Archives of Internal Medicine. 1994:154:1497-1504.

(13.) Hirth RA, Fendrick AM, Chernew ME. Specialist and generalist physicians' adoption of antibiotic therapy to eradicate Helicobacter pylon infection. Medical Care. 1996;34:1199-1204.

(14.) Vollmer WM, O'Hollaren M, Ettinger KM, Stibolt T, Wilkins J, Buist AS, Linton KL, Osborne ML. Specialty differences in the management of asthma. A cross-sectional assessment of allergists' patients and generalists' patients In a large HMO. Archives of Internal Medicine. 1997:157:1201-1208.

(15.) Schreiber TL, Elkhatib A, Grimes CL, O'Neill WW. Cardiologist versus Internist (Management of Patients with Unstable Angina: Treatment Patterns and Outcomes. Journal of the American College of Cardiology. 1995:26:577-582.

(16.) Stein JH, Uretz EF, Parrillo JE, Barron JT. Cost and appropriateness of radio nuclide exercise stress testing by cardiologists and non-cardiologists. American Journal of Cardiology. 1996:77:139-142.

(17.) Jollis JG, DeLong ER, Peterson ED, Muhlbaier LH, Fortin DF, Califf RM, Mark DB. Outcome of Acute Myocardial Infarction According to the Specialty of the Admitting Physician. New England Journal of Medicine, 1996:335:18801887.

(18.) Nash IS, Nash DB, Fuster V. Do Cardiologists Do it Better? Journal of the American College of Cardiology, 1997:29:475-478.

(19.) Mitchell JB, Ballard DJ, Whisnant JP, Ammering CJ, Samsa GP, Matchar DB. What role do neurologists play in determining the costs and outcomes of stroke patients. Stroke. 1996:27:1937-1943.

(20.) Zarling EJ, Piontek F, Klemka-Walden L, Inczauskis D. The effect of gastroenterology training on the efficiency and cost of care provided to patients with diverticulitis. Gastroenterology. 1997;112:1859-1862

(21.) Murphy AW, Bury G. Plunkett PK, Gibney D, Smith M, Mullan E, Johnson Z. Randomized controlled trial of general practitioner versus usual medical care in an urban accident and emergency department. British Medical Journal. 1996:312:1135-1142.

(22.) Carey TS, Garrett J, Jackman A, McLaughlIn C, Fryer J, Smucker DR, and the North Carolina Back Pain Project. The outcomes and costs of care for acute lower back pain among patients seen by primary care practitioners, chiropractors and orthopedic surgeons. New England Journal of Medicine. 1995:333:913-917.

(23.) Shekella PG, Markovich M, Louie R. Comparing the Costs Between Provider Types of Episodes of Back Pain Care. Spine. 1995:20:221-226.

(24.) Greenfield S, Rogers W, Mangotich M, Camey MF, Tarlov AR. Outcomes of Patients with Hypertension and Non-insulin dependent Diabetes Mellitus Treated by Different Systems and Specialties. JAMA. 1995:274:1436-1444.

(25.) Greenfield S. Nelson EC, Zubkoff M, Manning W, Rogers W, Kravitz RL, Keller A, Tarlov AR, Ware JE Jr. Variations in resource utilization among medical specialties and systems of care. Results from the medical outcomes study. JAMA. 1992:267:1624-1630.

(26.) Scott AIF, Freeman CPL. Edinburgh primary care depression study: treatment outcome. patient satisfaction, and cost after 16 weeks. British Medical Journal. 1992:304:883-887.

(27.) Owen SA, Maayens E Jr. Weary PE. Patients' opinions regarding direct access to dermatologic specialty care. Journal of the American Academy of Dermatology. 1997:36:250-256.

(28.) Stroke Unit Trialists' Collaboration. Collaborative systematic review of the randomized trials of organized inpatient (stroke unit) care after stroke. British Medical Journal. 1997, 314:1151-1159.

(29.) Rosenthal GE, Harper DL, Quinn LM, Cooper GS. Severity-adjusted mortality and length of stay in teaching and non-teaching hospitals. Results of a regional study. JAMA. 1997:278:485-490.

(30.) Benninger MS, King F, Nichols RD. Management Guidelines for Improvement of Otolaryngology Referrals from Primary Care Physicians. Otolaryngology- Head & Neck Surgery. 1995:113:446-452.

(31.) Griffith SP, Freeman WL, Shaw CJ, Mitchell WH, Olden CR, Figgs LD, Kinyoun JL, Underwood DL, Will JC. Screening for Diabetic Retinopathy in A Clinical Setting: A Comparison of Direct Ophthalmoscopy by Primary Care Physicians with Fundus Photography. The Journal of Family Practice. 1993:37:49-56.

(32.) Levetan CS, Salas JR, Whets IF, Zumoff B. Impact of endocrine and diabetes team consultation on hospital length of stay for patients with diabetes. American Journal of Medicine. 1995:99:22-28.

(33.) Peters RM, After-hours telephone calls to general and subspecialty internists: an observational study. Journal of General Internal Medicine. 1994:9:554-557.

Peter Cram, MD, is a House Officer in The Department of Medicine at the University of Michigan in Ann Arbor. He can be reached at 313/665-7943.

Walter H. Ettinger, Jr., MD, MBA, is Professor of Internal Medicine and Public Health Sciences and Director of The J. Paul Sticht Center on Aging at Wake Forest University Baptist Medical Center in Winston-Salem, North Carolina. He can be reached at 910/713-8583 and via email at [e-mail].