Special Article
When Physicians Treat Members of Their Own Families
N Engl J Med 1991; 325:1290-1294October 31, 1991
- Abstract
- Abstract
Background.
Little is known about the circumstances under which physicians care for family members. We sought to examine current practice and, in particular, to learn how often family members request medical care or treatment, whether physicians accede to such requests, and what concerns, if any, physicians have about caring for their family members.
Methods.
In late 1990 we distributed a pretested, structured questionnaire to all members of the active medical staff (physicians with M.D. or D.O. degrees) of a large suburban community teaching hospital. Of 691 eligible members of the medical staff, 465 physicians responded.
Results.
Of the 465 respondents, 461 (99 percent) reported requests from family members for medical advice, diagnosis, or treatment. A total of 386 (83 percent of the respondents) had prescribed medication for a family member, 372 (80 percent) had diagnosed medical illnesses, 334 (72 percent) had performed physical examinations, 68 (15 percent) had acted as a family member's primary attending physician in the hospital, and 44 (9 percent) had operated on a family member. In addition, 152 (33 percent) reported that they had observed another physician "inappropriately involved" in a family member's care, and 103 (22 percent) had acceded to a specific request about which they felt uncomfortable.
Conclusions.
Practicing physicians often attend and treat their family members and diagnose their illnesses. They may often experience some disquiet in doing so. For physicians, complete medical data, proper training, and sound judgment are essential when family members request treatment. (N Engl J Med 1991;325:1290–4.)
Media in This Article
Table 1
Services Provided to Family Members by the 465 Respondents.*Table 2
"Most Important" Operation or Procedure Performed on a Family Member.*Article Activity
- Article
STORIES have been told about physicians who treat their own family members, but to the best of our knowledge, this practice has not been studied. Family members may benefit; they may avoid the inconvenience and expense of an office visit and gain an especially caring, available expert who is able to interpret medical language and help them maneuver through medical systems. Physicians may also benefit; accustomed to caring for patients and surrounded by books, tools, and pharmaceuticals, they may consider attending to ill family members a natural and rewarding opportunity.
Ethical questions have been raised, however, about physicians who treat members of their own families. McSherry found incomplete physical examinations, medical records, and immunizations to be undesirable consequences of physicians' treating their own children.1 The 1901 code of ethics of the American Medical Association (AMA) noted that a family member's illness "tends to obscure [the physician's] judgement and produce timidity and irresolution in his practice."2 In 1977, a revised admonition was dropped by the AMA with other "outmoded matters of medical etiquette."3 In 1989, fearing financial abuse by unscrupulous providers, Medicare barred payment to physicians who care for "immediate family members."4
In this study, in an attempt to understand current practice, we asked these questions: Which family members request advice, diagnosis, or treatment? How do physicians respond to such requests? Which, if any, requests make physicians feel uncomfortable? And under what circumstances do physicians refuse requests from family members for medical assistance?
Methods
In the fall of 1990, the Lutheran General Hospital Center for Clinical Ethics developed a questionnaire designed to determine attitudes and activities of practicing physicians in response to family members' requests for medical assistance. No hospital or medical staff policy governed physicians' medical activities in caring for family members.
The self-administered questionnaire, designed to elicit anonymous responses, was pretested and then distributed to 693 active physicians on the medical staff of a large, university-affiliated teaching hospital in a suburban community. Two of the questionnaires were undeliverable. After three weeks, a second questionnaire was mailed to the physicians who had not responded, and after three more weeks a third questionnaire was mailed.
In responding to the questionnaire, physicians were asked to indicate which of their family members had asked for advice, consultation, or a second opinion; the frequency of such requests; and their responses to them. Physicians were asked if they had ever diagnosed illness in family members, examined them, prescribed medication, provided prescription samples, acted as a primary or consulting attending physician in the hospital, performed elective or emergency surgery, administered cardiopulmonary resuscitation, performed a Heimlich maneuver, or given immunizations. Physicians also listed what they considered the "most important" diagnosis they had made for a family member and the "most important" operation or procedure they had performed. Physicians were asked if they had ever refused a specific request, if they had acceded to a request that made them feel uncomfortable, and if they had observed another physician "inappropriately involved" in a family member's care. They were then asked to indicate why they had refused or acceded to the request or believed the observed involvement to be inappropriate. Responses were grouped according to empirically derived categories. The questionnaire also included four background items: age, sex, primary professional role, and number of years in practice.
Statistical analysis was performed with ABC software. Statistical comparisons of group means were made with use of Student's t-test. The chi-square statistic was used to test differences in categorical variables. Physicians' participation was voluntary, and the confidentiality of their responses was ensured. The institutional review board of the Lutheran General Hospital approved the research protocol.
Results
Of the 691 physicians surveyed, 465 (67 percent) returned completed questionnaires. The age (mean ±SD, 45.7±11.9 years) and sex distribution of the respondents (81 percent were male) did not differ from those of the hospital medical staff as a whole (P>0.43 and P>0.10, respectively). The number of respondents in various specialties ranged from 6 (for pathology) to 141 (for internal medicine). Physicians had been in practice for 1 to 60 years (mean, 14.9±11.5).
Requests by Family Members
Four respondents (1 percent) reported that they had never been asked for medical advice by a family member. Those who were reported as making requests included spouses (reported by 65 percent of the respondents), children (59 percent), mothers (50 percent), siblings (44 percent), fathers (43 percent), nieces and nephews (41 percent), and mothers-in-law (40 percent). Respondents who listed "other" family members (24 percent) were usually referring to brothers- or sisters-in-law, aunts, uncles, and cousins.
Physicians' Responses to Requests
Of the 461 respondents asked for advice, 265 (57 percent) reported "almost always" providing it; 174 (38 percent) reported "sometimes" providing it. Physicians' responses about which services, if any, they provided to family members are shown in Table 1Table 1
Services Provided to Family Members by the 465 Respondents.*. The respondents had performed a mean (±SD) of 4.4±2.1 (median, 4.2) of the 12 services listed; 386 of the 465 respondents (83 percent) had prescribed medication, 372 (80 percent) had diagnosed medical illnesses requiring treatment, 334 (72 percent) had performed physical examinations, 68 (15 percent) had attended a family member as primary physician in the hospital, 32 (7 percent) had consulted in the hospital, 44 (9 percent) had performed elective surgery on family members, and 17 (4 percent) had performed emergency surgery. Nineteen respondents reported that they had attended family members on different occasions as both primary and consulting physician; 11 had performed both elective and emergency surgery. The distribution of family members receiving care paralleled that of the family members requesting it.In general, the proportion of respondents who provided services to family members rose with the respondents' age and the number of years in practice. Since beginning practice, respondents 45 years of age and older had provided more services (mean, 5.0±2.2) to their relatives than respondents younger than 45 (mean, 4.0±1.8; P<0.001). Relatives' requests were nearly evenly distributed between older and younger physicians, although fewer older physicians than younger physicians (53 percent vs. 64 percent, P<0.05) reported refusing a request.
Physicians' responses were categorized according to whether they were in primary care specialties (family medicine, internal medicine, obstetrics and gynecology, or pediatrics) or other specialties (anesthesiology, emergency medicine, neurology, pathology, psychiatry, radiology, or surgery). Respondents in primary care (n = 312) had performed a mean of 4.9 services listed in Table 1; respondents in specialties other than primary care (n = 153) had performed a mean of 3.5 services (P<0.001). Although significantly higher proportions of primary care respondents had diagnosed illnesses (P<0.001), examined family members (P<0.001), or prescribed medication (P<0.005), more than half of the non-primary care respondents had provided each of these services. No significant difference was observed between the numbers of respondents in primary care and other specialties who had consulted on a family member's care in the hospital (P>0.25).
Operations and Diagnoses
Forty-three of the physicians who had operated on family members specified what they considered their "most important" operations, listed in Table 2Table 2
"Most Important" Operation or Procedure Performed on a Family Member.*. Procedures were most commonly performed on parents (reported by seven physicians), spouses (seven), children (seven), and parents-in-law (four). Most of the procedures listed do not usually require general anesthesia; many were dermatologic. Ten of the reported procedures were obstetrical, including six cesarean sections.The respondents recorded what they considered to be their "most important" diagnosis requiring medical treatment. In Table 3Table 3
"Most Important" Diagnosis Made for a Family Member.*, the 332 reported diagnoses are shown. The conditions diagnosed ranged in severity from actinic keratosis (n = 1) to cancer (n = 27). The family members for whom these diagnoses were made were most often children and spouses.Of the 391 respondents who answered the question, 188 (48 percent) referred the patient for whom they had made their most important diagnosis; 82 (21 percent) diagnosed the illness and treated and then referred the patient; and 121 (31 percent) diagnosed the illness and treated the patient themselves. Of these 121 respondents, 10 treated family members with trivial medical conditions (e.g., ingrown toenail); 55 treated family members with minor medical conditions (e.g., cystitis, pharyngitis, or conjunctivitis); and 38 treated patients with major medical illnesses (e.g., pneumonia, colon cancer, or asthma). In the 38 cases of major illness, 8 of the patients were children, 6 were nieces or nephews, 5 were parents, and 19 were other relatives. One respondent made a surgical diagnosis (abdominal ptosis) and operated on the patient; 17 respondents did not identify their most important diagnosis.
Physicians' Refusal and Discomfort
Most respondents (262) reported refusing a family member's request for diagnosis or treatment; of these, 214 gave a reason. Some respondents (103) reported acceding to a request about which they felt uncomfortable; 87 of these gave a reason.
Table 4Table 4
Reasons for Refusal of and Discomfort with Requests by Family Members. shows the reasons given by the respondents for refusing a request and for acceding to a request even when uncomfortable. The most common reasons for refusing were that the clinical problem was not within the physician's area of expertise (34 percent) and that examination and follow-up might be inadequate (18 percent). Respondents who had acceded to requests that made them feel uncomfortable also cited the inadequacy of examination and follow-up (23 percent), the absence of a medical indication for the request (20 percent), and their own lack of objectivity (20 percent).Of the 152 respondents who had observed a physician whom they considered "inappropriately involved" in the care of a family member, 117 reported the circumstances, which fell into four broad categories. Respondents described individual physicians who had lost their objectivity (not performing procedures needed to investigate serious illness or pursuing care that the respondents considered futile [44 percent]); physicians who had interfered directly or indirectly with appropriate diagnostic or therapeutic measures (29 percent); physicians who had performed medically contraindicated procedures (15 percent); and physicians who had examined or followed family members inadequately (11 percent).
Discussion
To understand current norms, we gathered empirical data on which family members ask physicians for advice, diagnosis, or treatment and how physicians respond to these requests. We also attempted to understand the dilemmas inherent in this practice. In general, physicians reported providing services to relatives in proportion to how often and by whom they were asked.
Family Members' Requests
The services family members request and the services their physician relatives offer are probably different. Family members may request care that requires a complete history and physical examination, new knowledge, or facilities that are unavailable, thus sometimes embarrassing and frustrating their physician relatives. Conversely, many young children are simply given medical care by their parents. Although most respondents reported requests from their children, we did not distinguish among children's own requests, another parent's requests for them, and the physician parent's own wish to provide care.
Physicians' Responses
Caring for family members has advantages for patients with minor illnesses, especially when the physician is in primary care, although advantages for patients near the end of life have also been described.5 Recurrent problems such as conjunctivitis and pediatric pharyngitis are usually minor, have predictable courses, and may seem too trivial to trouble an unrelated doctor about. For all patients, our medical—financial complex is best negotiated by a strong patient advocate, conveniently located and readily accessible, who is both altruistic and self-interested.
The most important diagnoses physicians gave to family members ranged from trivial to grave. The breadth of these diagnoses and the relatively narrow range of operations performed suggest that personal, psychological, and familial factors contributed to the assessment of "most important." It is uncertain whether respondents made these diagnoses informally, as knowledgeable spouses (for example, a wife asks her physician husband about a breast lump she has found, which he then palpates), or formally, as attending physicians (a physician daughter takes her father's blood pressure regularly, reviewing his age, diet, weight, medications, electrocardiogram, and family history).
Sources of Discomfort
The central reason for physicians' refusal of and discomfort with requests appears to be missing medical information. About a third of the respondents mentioned colleagues who appeared to be inappropriately involved in the care of family members, describing the care provided as inadequate or contraindicated or reporting the obstruction of other providers' care. Respondents inferred that this poor quality of care resulted from their colleagues' closeness to the patient, but this inference may be scientifically unjustified. Comparative process and outcome evaluations of care provided to patients who are family members and to those who are not would permit a more complete assessment of the quality of care.
Our data suggest that along with limiting their active participation, physicians attempt to limit their emotional involvement in family members' care. Setting limits may reflect physicians' recognition of the emotional complexity of having dual roles,6 physicians' difficulty in providing reassurance when a serious illness is suspected,2 or the problems anticipated when there is a family relationship instead of a potentially therapeutic doctor–patient relationship.
Who counts as a family member? Respondents listed only relatives, but not all relatives are emotionally close, and some people who are emotionally close are not relatives. Many physicians are devoted to others: friends, lovers, or favorite patients. Medicare's list of "immediate relatives" includes spouses, parents, children, siblings, stepparents, stepchildren, stepbrothers, stepsisters, children-in-law, siblings-in-law, grandparents, grandchildren, and spouses of grandparents or grandchildren. Since 1989, Medicare has not paid for patient care charges by immediately related physicians, their associates, or their professional corporations.4 Blue Cross—Blue Shield, which has a stricter definition of "family member," has not paid for these charges since 1976.
Is this ethics, etiquette, or just sound judgment? The issues that arise run the gamut of modern medical ethics: when to breach confidentiality or refuse treatment; how to obtain informed consent, assess decision-making capacity, or provide end-of-life care; how to appeal unfair economic constraints or ensure access to care; and whom to consider as a patient and whom as family. The care of family members was formerly part of medical etiquette; professional courtesy7 was intended to permit physicians to avoid the difficulties of making medical judgments for members of their own families.
Sound judgment is required on the part of a physician who treats a relative. Considerations include the adequacy of one's medical training to meet a relative's needs; one's willingness to examine a relative's intimate history, perform a complete physical examination, and, if necessary, convey bad news; the ability to assess signs and symptoms objectively8 , 9; and one's ability to anticipate and negotiate family conflicts. Whether family members will receive high-quality medical care from a related doctor or whether they would be better off seeing someone else probably depends on the judgment of the physician, the medical urgency of the case, and the availability of medical colleagues.
Our descriptive study of family members' requests and physicians' responses has limitations. Only 465 practicing physicians returned usable questionnaires (for a response rate of 67 percent); nonrespondents may have had different experiences. Respondents were members of the medical staff of a community teaching hospital; a national sample of physicians in different practice settings should be studied to assess the generalizability of our data. Respondents reported the frequency of requests from family members and their own actions; to verify these responses and to gather data on incidence, interviews with physicians or reviews of medical records might be useful.
These data suggest that practicing physicians examine, treat, and prescribe for their family members. The data do not reveal how physicians regard family members' signs and symptoms as compared with those of other patients. The 1901 AMA code suggests that physicians may act "with timidity and irresolution," but our respondents diagnosed serious illnesses, prescribed powerful medications, and performed major operations. The clinical contexts of physicians' decisions about whether to treat family members, including the acuteness, urgency, and severity of illness and the availability of the necessary expertise are unknown but should be carefully researched. At the least, physicians need a forum for the discussion of the medical care of family members, since some requests for care are troublesome to practitioners.
Other research efforts should address what seems to be the most important clinical question present: How does the doctor–patient relationship affect the quality of care when doctor and patient are related? The same question should be asked of doctor–patient relationships that are close but not familial. Taking care of family members begins at home; whether such care should include medical care delivered by a physician relative10 provides many opportunities for further research.
Supported by Lutheran General Hospital and the Lutheran General Medical Group. The views expressed are those of the authors and do not necessarily reflect the views of the supporting institutions.
Presented at the 14th annual meeting of the Society for General Internal Medicine, Seattle, May 1, 1991, and at the second annual International Conference on Physician Health, Toronto, June 7, 1991.
We are indebted to Ruth Priest, M.D., for clinical insight, to Elena Butkus and Anita Schweickart, J.D., for legal research, and to Patricia Wiederer and Mary Fortman for assistance in the preparation of the manuscript.
Source Information
From the Center for Clinical Ethics (J.L., C.A.D.) and the Department of Family Practice (D.L.), Lutheran General Hospital, Park Ridge, Ill., and the Center for Clinical Medical Ethics and the Section of General Internal Medicine, University of Chicago, Chicago (J.L., C.B.S.). Address reprint requests to Dr. La Puma at the Center for Clinical Ethics, Lutheran General Hospital, 1775 Dempster, Park Ridge, IL 60068.
- References
References
1
McSherry
Web of Science2
Code of ethics of the American Medical Association. Chicago: American Medical Association, 1901:15.
3
Caveat. In: Opinions and reports of the Judicial Council. Chicago: American Medical Association, 1977.
4
Blue Cross and Blue Shield of Illinois. Immediate relative services excluded. Medicare B Bulletin. Marion: Blue Cross and Blue Shield of Illinois, December 1990:10–1.
5
Cranford RE. The role of the ethics consultant in personal ethical dilemmas. In: Culver CM, ed. Ethics at the bedside. Hanover, N.H.: University Press of New England, 1990:194–206.
6
Freeman
Web of Science | Medline7
Professional courtesy survey . JAMA 1966;195:299–301.
CrossRef | Web of Science8
La Puma
Web of Science | MedlineErratum, Ann Intern Med 1988;108:315.
Web of Science9
Tumulty
Full Text | Web of Science | Medline10
Spock
Web of Science | Medline
- Citing Articles (12)
Citing Articles
1
Jonathan R. Scarff, Steven Lippmann. (2012) When Physicians Intervene in Their Relatives’ Health Care. HEC Forum
CrossRef2
Susan Regan, Timothy G. Ferris, Eric G. Campbell. (2010) Physician Attitudes Toward Personal Relationships With Patients. Medical Care 48:6, 547-552
CrossRef3
Sara A. Slavin, Sumner A. Slavin, Robert M. Goldwyn. (2010) A Family Operation: Plastic Surgeons Who Perform Aesthetic Surgery on Spouses or Other Family Members. Plastic and Reconstructive Surgery 125:3, 1018-1023
CrossRef4
Gregory L. Eastwood. (2009) When Relatives and Friends Ask Physicians for Medical Advice: Ethical, Legal, and Practical Considerations. Journal of General Internal Medicine 24:12, 1333-1335
CrossRef5
Kenneth Oberheu, James W. Jones, Robert M. Sade. (2007) A Surgeon Operates on His Son: Wisdom or Hubris?. The Annals of Thoracic Surgery 84:3, 723-728
CrossRef6
James W. Jones, Laurence B. McCullough, Bruce W. Richman. (2005) The ethics of operating on a family member. Journal of Vascular Surgery 42:5, 1033-1035
CrossRef7
Elizabeth Campbell, Lainie Friedman Ross. (2003) Professional and Personal Attitudes about Access and Confidentiality in the Genetic Testing of Children: A Pilot Study. Genetic Testing 7:2, 123-130
CrossRef8
Amalia M. Issa. (2002) Taking Off the White Coat: Can Family Members Who Are Physicians Be Good Surrogate Decision-Makers?. Journal of the American Geriatrics Society 50:5, 946-948
CrossRef9
P P Jonge, C M Thijssen, P M Boendermaker. (2002) Zieke dokters anno 1998huisarts onderzoek. Huisarts en Wetenschap 45:1, 833-836
CrossRef10
Isabel Ruiz, Ildefonso Hernandez-Aguado, Pablo Garrido. (1998) Variation in Surgical Rates. Medical Care 36:9, 1315-1323
CrossRef11
Tom J. Wachtel, Victoria L. Wilcox, Anne W. Moulton, Dominick Tammaro, Michael D. Stein. (1995) Physicians’ utilization of health care. Journal of General Internal Medicine 10:5, 261-265
CrossRef12
(1992) When Physicians Treat Their Own Families. New England Journal of Medicine 326:13, 895-896
Full Text
- Letters
