Sorry, you need to enable JavaScript to visit this website.
Skip to main content

Doctor, Don't Treat Thyself

Elin Olaug Rosvold, MD, PhD | September 1, 2004
View more articles from the same authors.

The Case

A 50-year-old radiologist presented to the emergency department of the community hospital where he worked and reported increasing shortness of breath over the past several days. His medical history was notable for hyperlipidemia, which had been discovered 10 years earlier after he presented with unstable angina and required coronary bypass surgery. He stated that he had no chest pain whatsoever and repeatedly emphasized that he had been completely free of cardiac symptoms since his surgery 10 years ago.

Physical examination revealed tachypnea (25 respirations per minute), tachycardia (110 beats per minute), and bilateral lower lung field crackles. A chest x-ray was obtained and read by the patient himself, who declared it to show bilateral infiltrates clearly indicative of pneumonia. Because it was after midnight and no other radiologist was in-house, the admitting internist did not question the reading of the radiologist–patient. The admitting physician ordered blood cultures and prescribed intravenous ceftriaxone as well as oral levofloxacin as empiric coverage for community-acquired pneumonia.

Several hours later, the nurse paged the admitting physician to say that the patient's blood pressure was plummeting. The physician ordered a 500-cc intravenous bolus of normal saline for presumed hypovolemia or early sepsis. He came to see the patient after the bolus had been administered (30 minutes later) and found the patient profoundly dyspneic. A blood gas revealed a pH of 7.2, pCO2 of 50, and pO2 of 50 on high-flow oxygen.

The physician told the nurse to call a code, in order to alert the emergency physician on duty, and to bring the "crash cart." The emergency physician arrived promptly and began preparations to intubate, but the patient went into cardiac arrest. The initial rhythm was ventricular fibrillation, but degenerated into asystole, from which the patient never recovered. He was pronounced dead after 30 minutes of resuscitative efforts.

The emergency physician had asked that a troponin be added to the blood work drawn at the time of the code. The next morning, on reviewing these results, the internist saw that the troponin had been markedly elevated. The incident prompted an internal review by the hospital, during which several radiologists reviewed the initial chest x-ray and reported it as clearly consistent with pulmonary edema. Autopsy confirmed a large anterior myocardial infarction and prominent pulmonary edema.

The Commentary

This case dramatically and tragically illustrates what may befall physicians who treat their own illnesses. Although the radiologist prudently went to the emergency department to seek help for his illness, problems arose when he misinterpreted his own x-rays, and the admitting internist followed his advice. This situation likely was difficult for both parties, as the radiologist probably had no experience being a patient at his own hospital, and the internist was placed in the tricky situation of either accepting the judgment of his patient-expert or questioning an experienced radiologist's reading of a simple chest x-ray.

Physicians seem to find it difficult to assume the patient role.(1,2) This difficulty may stem from denial; perhaps dealing with patients' illnesses and suffering makes them feel partly immune from diseases. Although such denial may help them carry out their work, it can also be hazardous to their health, by leading doctors to minimize and intellectualize their symptoms, thereby delaying treatment and help-seeking.(3,4) In addition, physicians may be afraid to show weakness to colleagues and patients. Since physicians have the skills and opportunities to initiate testing and treatment, apprehension about seeking help can lead to self-treatment.

Physicians often self-treat, usually by self-prescribing medicines.(4-8) Many physicians find it inconvenient and unnecessary to consult another physician for a disease that they are competent to treat. However, "self-treatment" removes the objectivity and distance necessary in a physician–patient relationship. Anxiety and denial may blur accurate symptom self-evaluation. In this case, we are given the impression that the x-ray was fairly typical for pulmonary edema; the radiologist's failure to entertain this possibility may well have reflected his own (probably subconscious) denial of the possibility of recurrent coronary heart disease.

For some physicians, a separate but special concern is self-treatment with addictive drugs. Physicians use hypnotic and anxiolytic agents more often than comparable groups in the general population.(4,9) Such self-prescription risks both drug dependency and failure by the physicians to seek professional help for mental problems.

In general, physicians take less sick time than other employees (7,10), and many report working while ill.(4,10,11) Reasons for not taking sick leave have been related to both psychological (sense of responsibility) and organizational factors (difficulty in finding suitable replacements).(10) Some physicians work while having possibly contagious diseases, thus placing their patients and colleagues at risk of becoming infected.(4)

Physicians have a responsibility to take care of their health properly, both for their own good and that of their patients. To do so, doctors should be aware of warning signs for when to seek help. Inspired by La Puma and Priest (12), who created a checklist for physicians who consider treating a family member, I suggest the following checklist for physicians who are ill and consider treating themselves (Table 1).

When physicians do seek help, they often engage in informal consultations with colleagues or friends.(2,4,11) Such "curbside consultations" can be hazardous, as neither the patient nor his physician follows the rules for a normal consultation. The ill physician might give limited information, either because of time pressure or because he does not want to expose himself to a colleague. The examination will usually be insufficient, or nonexistent, because the setting is inappropriate. In addition, the consultant might not respond adequately or take responsibility for the patient since the encounter is not viewed as a formal consultation. In the present case, the radiologist, by presenting to the hospital and being evaluated by an internist, did undergo a more formal consultation. However, because it was his own hospital, he probably had personal ties with the treating physicians, which may well have influenced his care.

In the United States, one in three physicians report not having a personal physician (13). In the United Kingdom, where the entire population is required to have a personal physician, up to 99% do (5,6). However, even those physicians who do have their own doctor often register with a personal friend or a practice partner, which might influence the physician patient relationship. Moreover, even when they are formally registered, physicians still may be reluctant to consult their general practitioner. Thus, there is a need to find new strategies to help physicians obtain optimal treatment. One example is from Norway, where a group of general practitioners are specially trained by the Norwegian Medical Association to be physicians for physicians (14). [Editors' Note]

Being the physician for a physician is a difficult role.(1,2,14) A physician patient can be reluctant to let another physician be in charge of treatment. His use of medical terms when describing the illness, and his interpretation of his own symptoms can mislead the treating physician. In the present case, the physician–patient himself read the x-rays and thus diagnosed himself. In other cases, the physician–patient might reveal only the symptoms that are in line with the diagnosis he has already given himself, and the treating physician might forget to ask all necessary questions.

When caring for a patient trained in medicine, the treating physician may feel professionally insecure, which may precipitate extensive examinations. On the other hand, the treating physician might offer less information than explained to 'ordinary' patients, due to a belief that the physician–patient knows the medical facts of the disease and the treatment procedure. There is also a risk that treating physicians over-identify with physician–patients and thus are influenced by their own anxiety about becoming ill. This can lead to distancing oneself from the patient, possibly resulting in poor treatment.

Professional courtesy, ie, the provision of care to colleagues without charging them, may engender a feeling of being a 'VIP' patient. However, it might also have a negative effect on the physician–patient relationship. The physician–patient might feel indebted to the treating physician or be afraid of using too much of the physician's time.(1,14)

Physicians who treat other doctors need knowledge about physicians' reaction to illness and the problems that can occur in consultations with such patients (Table 2). The following advice applies (1,2,14,15): When treating physicians, you must reassure them about confidentiality and clarify the physician–patient relationship as early as possible. Thorough examinations should be done in optimal circumstances. It is important to ask about self-treatment and self-diagnosis and to discuss the diagnostic and treatment plan in detail with the patient. You also need to be aware of your own feelings. If you are anxious about treating other physicians, you should not accept such patients. You should avoid engaging in corridor encounters, but not refuse to help a colleague who is ill. Instead, you can use the informal consultation to encourage the colleague to seek appropriate help.

Take-Home Points

  • Physicians are reluctant to enter the patient role and may avoid seeking help.
  • Both self-treatment and informal consultations bypass the objectivity necessary for optimal care, and can be risky.
  • Each physician should have a primary care physician to call on. Physicians who take on such patients must be educated about important issues in caring for other physicians.
  • Even when physicians do seek help, they may describe their symptoms in medical terms that apply only to the diagnosis already self-ascribed. Physicians caring for physician–patients must remember to ask all routine questions anyway.

Elin Olaug Rosvold, MD, PhD
Assistant Professor
Department of General Practice and Community Medicine
University of Oslo, Norway


1. Stoudemire A, Rhoads JM. When the doctor needs a doctor: special considerations for the physician-patient. Ann Intern Med. 1983;98:654-9.[ go to PubMed ]

2. Schneck SA. "Doctoring" doctors and their families. JAMA. 1998;280:2039-42.[ go to PubMed ]

3. Hahn RA. Between two worlds: physicians as patients. In: Sickness and healing. An anthropological perspective. New Haven and London: Yale University Press; 1995:234-261.

4. Rosvold EO. Physicians in illness and health. An epidemiological study on health- and illness behaviour among 19th and 20th century Norwegian physicians [dissertation]. Oslo, Norway: Institute of General Practice and Community Medicine, University of Oslo; 2002.  Accessed August 9, 2004.

5. Fiebach JB, Schellinger PD, Gass A, et al. Stroke magnetic resonance imaging is accurate in hyperacute intracerebral hemorrhage: a multicenter study on the validity of stroke imaging. Stroke. 2004;35:502-6.[ go to PubMed ]

6. Forsythe M, Calnan M, Wall B. Doctors as patients: postal survey examining consultants and general practitioners adherence to guidelines. BMJ. 1999;319:605-8.[ go to PubMed ]

7. Töyry S, Rasanen K, Kujala S, et al. Self-reported health, illness, and self-care among finnish physicians: a national survey. Arch Fam Med. 2000;9:1079-85.[ go to PubMed ]

8. Christie JD, Rosen IM, Bellini LM, et al. Prescription drug use and self-prescription among resident physicians. JAMA. 1998;280:1253-5.[ go to PubMed ]. Accessed August 9, 2004.

9. Domenighetti G, Tomamichel M, Gutzwiller F, Berthoud S, Casabianca A. Psychoactive drug use among medical doctors is higher than in the general population. Soc Sci Med. 1991;33:269-74.[ go to PubMed ]

10. McKevitt C, Morgan M, Dundas R, Holland WW. Sickness absence and 'working through' illness: a comparison of two professional groups. J Public Health Med. 1997;19:295-300.[ go to PubMed ]

11. Baldwin PJ, Dodd M, Wrate RM. Young doctors' health--II. Health and health behaviour. Soc Sci Med. 1997;45:41-4.[ go to PubMed ]

12. La Puma J, Priest ER. Is there a doctor in the house? An analysis of the practice of physicians' treating their own families. JAMA. 1992;267:1810-2.[ go to PubMed ]. Accessed August 9, 2004.

13. Gross CP, Mead LA, Ford DE, Klag MJ. Physician, heal thyself? Regular source of care and use of preventive health services among physicians. Arch Intern Med. 2000;160:3209-14.[ go to PubMed ]

14. Tyssen R. [The physician-patient relationship when the patient is a physician]. [Article in Norwegian]. Tidsskr Nor Lægeforen. 2001;121:3533-5.[ go to PubMed ]

15. Anderson W. Doctoring doctors. British Medical Journal. 1999;319:2.[ Free full text ]


Table 1. Checklist for Physicians Feeling Ill

Am I uncertain about what my symptoms might represent?

Am I deep inside afraid that I am seriously ill?

Do I explain away my symptoms?

Do I need medical tests or examinations?

Do I need prescription drugs?

Do I need to discuss my case with another physician?

Is my spouse or are my friends worried about my health?

If you answer "Yes" to at least one of these questions, you should make a formal appointment with your personal physician.

Table 2. Checklist for Physicians Caring for Fellow Physicians

Discuss confidentiality and clarify the physician–patient relationship as early as possible.

Perform thorough examinations in formal circumstances.

Ask about self-treatment and self-diagnosis and discourage these practices.

Discuss diagnostic and treatment plans in detail: do not assume that a physician's professional knowledge makes such discussion unnecessary.

Avoid engaging in corridor consultations, but do not refuse to help a colleague who is ill. Instead, encourage the colleague to seek appropriate help.

This project was funded under contract number 75Q80119C00004 from the Agency for Healthcare Research and Quality (AHRQ), U.S. Department of Health and Human Services. The authors are solely responsible for this report’s contents, findings, and conclusions, which do not necessarily represent the views of AHRQ. Readers should not interpret any statement in this report as an official position of AHRQ or of the U.S. Department of Health and Human Services. None of the authors has any affiliation or financial involvement that conflicts with the material presented in this report. View AHRQ Disclaimers
Related Resources From the Same Author(s)
Related Resources