AMA Journal of Ethics®

Illuminating the art of medicine

Journal of Ethics Header

AMA Journal of Ethics®

Illuminating the art of medicine

Virtual Mentor. January 2009, Volume 11, Number 1: 38-42.

Journal Discussion

  • Print
  • |
  • View PDF

Preferences for End-of-Life Care: A Physician and Homeless Patient Comparison

Physicians who base end-of-life care decisions for patients on their own preferences may offer less treatment than the patients themselves would have wanted.

Carolyn Bramante and John Song, MD, MPH, MAT

Norris WM, Nielsen EL, Engelberg RA, Curtis JR. Treatment preferences for resuscitation and critical care among homeless persons. Chest. 2005;127(6):2180-2187.

The 2005 Chest article, “Treatment Preferences for Resuscitation and Critical Care Among Homeless Persons,” identified differences between physician and patient desires regarding end-of-life care, specifically mechanical ventilation and cardiopulmonary resuscitation (CPR) [1]. Norris and co-authors compared the preferences of different groups of homeless men and women to each other, to physicians, and to a group of patients with chronic obstructive pulmonary disease (COPD) who were not homeless. This last group was included to control for the difference in education level between the homeless and physician groups. The study found that each group of homeless individuals preferred more care than either the COPD patients or physicians would have chosen for themselves. This is significant because, when physicians treat patients who have no advance directives or surrogate decision makers, they tend to choose for those patients the level of care they would want for themselves [1]. Thus, Norris's article suggests that when physicians make end-of-life decisions for a homeless patient, they are likely to choose less care than the patient desires.

Learning ObjectiveUnderstand the argument that physicians who base end-of-life care decisions for patients on their own preferences may offer less treatment than the patients themselves would have wanted.

Homeless individuals face a greater burden of morbidity and mortality and have a life expectancy much lower than members of the general population [2-6]. For example, the average life expectancy in the U.S. is 77.8 years, but the average life expectancy of homeless people can be as low as 41 years, with mortality rates (deaths per 1,000 individuals per year) as high as 10 times greater than the general population [2, 4-7]. Physicians are likely to encounter a disproportionately high number of situations involving end-of-life decisions for homeless patients, and thus there is an urgent need to address the question of how to properly meet the desires of homeless patients [4-7].

The need is also urgent because there is consistent evidence of poor attitudes toward homeless persons [2, 8-12]. This author has heard that “doctors don't treat homeless people well,” in informal conversations with homeless people. Poor awareness of these patients' needs and wants is significant because, as Norris and others have reported, homeless patients are less likely to have a surrogate decision maker or a family member who can be reached [2, 3]. Thus, difficult end-of-life decisions may fall upon physicians, and it is possible they would make choices that they themselves would prefer in such a situation [2, 3, 8].

The authors offer several solutions to resolve the difference in patient and physician desires, such as the use of ethics committees. “We believe these data suggest that institutions should consider developing explicit policy for involving additional persons, such as hospital ethics committees and/or advocates for homeless persons” [13]. They also suggest “asking every patient at the time of hospital admission who their surrogate decision maker would be and how to contact that person” [13]. Both suggestions would help honor the wishes of homeless patients—an important step in fighting for the rights of those who have had little access to health care.

Norris et al. noted several limitations to their study. The participants represent a convenience sample from a limited geographical area with a variety of relationships with family, and the control group of COPD patients was not matched to the other groups. This last limitation is significant because the COPD patients in the control group suffered from significant disease. Their experiences with poor health and the health care system might have influenced their desire for less end-of-life care. A more valuable comparison might be made between a random sample of homeless individuals and a random sample of domiciled individuals in similar states of health.

Norris's mention of advance care planning and advance directives as the most effective ways to honor the end-of-life wishes of homeless individuals is important. Power over what happens to a homeless person's body at the end of life is significant for the individual who experienced so little power over much of his or her life [2, 3]. Rather than doing this upon admission to the hospital, however, when levels of stress might be overwhelming or the patient might not have decisional capacity—a distinct possibility given the fact that homeless persons tend to seek medical care later in the course of their illnesses, or if there are overlying mental health or substance abuse issues—health care professionals should strive to have such directives on-record for homeless people in their county [14]. Record keeping would also greatly reduce the burden on physicians and other health care professionals, as well as the strain on bioethics committee consultations.

Is there a danger that this approach might include advance directives that would require medically futile intervention? Might clinicians be forced by these directives to provide care they might not ordinarily provide? This potential is interesting and possible inasmuch as today's medical technology allows for unnatural longevity at the end of life. This argument cuts both ways, however; physicians might feel more comfortable allowing a natural death if there were a living will indicating that heroic measures were not desired. It is well-established that physicians and medical staff are often reluctant to remove life-sustaining treatments in the absence of an advance directive [14].

Are advance directives the best way to meet the desires of homeless individuals? Would the process of engaging in advance-care planning prompt an individual to make healthier decisions and engage in less risky behavior? Homeless persons often have extensive experience with death at all ages [2, 3]. Researchers have found that homeless people are interested in having advance directives on-record, and appreciate the chance to talk about these issues with willing listeners [15]. There are conceptual reasons to believe advance directives might better serve a population that is often alienated and estranged from potential surrogate decision makers [16, 17]. These data lend credit to Norris's conclusion that physicians should be made aware that “as a group, homeless persons prefer more aggressive life-sustaining treatments than physicians and some other patient populations” [4].

Because of these limitations and the inherent limitations of generalizing care to a group, it seems that a valuable answer for possible differences in end-of-life wishes is to have a record of advance directives. In fact, shouldn't all patients have advance directives on file, as well as access to the medical resources to honor them? This would help respect wishes of patients, relieve physicians from the pressure of such decisions, and potentially reduce unnecessary waste in medical care.

The article does not address the perhaps unanswerable underlying reasons behind this discrepancy in end-of-life wishes. One possible contributing factor is that African Americans are less likely to forgo life-sustaining treatment. The sample of homeless patients that Norris et al. reported was 47.7 percent African American [18], and white homeless persons often chose answers that correlated closer to the physicians than to their nonwhite counterparts (e.g., CPR in dementia). Another explanation might be that, as groups, physicians and patients judge quality of life differently [19-21]. But that's another research question.


  1. Norris WM, Nielsen EL, Engelberg RA, Curtis JR. Treatment preferences for resuscitation and critical care among homeless persons. Chest. 2005;127(6):2180-2187.
  2. Song J, Ratner ER, Bartels DM, Alderton L, Hudson B, Ahluwalia JS. Experiences with and attitudes toward death and dying among homeless persons. J Gen Intern Med. 2007;22(4):427-434.
  3. Kushel MB, Miaskowski C. End-of-life care for homeless patients: “she says she is there to help me in any situation.” JAMA. 2006;296(24):2959-2966.
  4. Kung HC, Hoyert DL, Xu J, Murphy SL. Deaths: final data for 2005. Natl Vital Stat Rep. 2008;56(10)1-120.
  5. Dembling BP, Chen DT, Vachon L. Life expectancy and causes of death in a population treated for serious mental illness. Psychiatr Serv. 1999;50(8):1036-1042.
  6. Saitz R, Gaeta J, Chen DM, Richardson JM, Larson MJ, Samet JH. Risk of mortality during four years after substance detoxification in urban adults. J Urban Health. 2007;84(2):272-282.
  7. Barrow SM, Herman DB, Cordova P, Struening EL. Mortality among shelter residents in New York City. Am J Public Health. 1999;89(4):529-534.
  8. Buck DS, Monteiro FM, Kneuper S, et al. Design and validation of the health professionals' attitudes toward the homeless inventory (HPATHI). BMC Med Educ. 2005;5(1):2.
  9. Zrinyi M, Balogh Z. Student nurse attitudes towards homeless clients: a challenge for education and practice. Nurs Ethics. 2004;11(4):334-348.
  10. O'Toole TP, Gibbon J, Harvey J, Switzer G. Students' attitudes toward indigent patients. Acad Med. 2002;77(6):586.
  11. Buchanan D, Rohr L, Stevak L, Sai T. Documenting attitude changes towards homeless people: comparing two st and ardised surveys. Med Educ. 2007;41(4):346-348.
  12. Wen CK, Hudak PL, Hwang SW. Homeless people's perceptions of welcomeness and unwelcomeness in healthcare encounters. J Gen Intern Med. 2007;22(7):1011-1017.
  13. Norris et al., 2186.
  14. Tillyard AR. Ethics review: ‘Living wills' and intensive care—and overview of the American experience. Crit Care. 2007;11(4):219.
  15. Song J, Bartels DM, Ratner ER, Alderton L, Hudson B, Ahluwalia JS. Dying on the streets: homeless persons' concerns and desires about end of life care. J Gen Intern Med. 2007;22(4):435-441.
  16. Shalowitz DI, Garrett-Mayer E, Wendler D. The accuracy of surrogate decision makers: a systemic review. Arch Intern Med. 2006;166(5):493-497.
  17. Miura Y, Asai A, Matsushima M, et al. Families' and physicians' predictions of dialysis patients' preferences regarding life-sustaining treatments in Japan. Am J Kidney Dis. 2006;47(1):122-130.
  18. Norris et al., 2183.
  19. Sneeuw KC, Sprangers MA, Aaronson NK. The role of health care providers and significant others in evaluating the quality of life of patients with chronic diseases. J Clin Epidemiol. 2002;55(11):1130-1143.
  20. Petersen MA, Larsen H, Pedersen L, Sonne N, M. Assessing health-related quality of life in palliative care: comparing patient and physician assessments. Eur J Cancer. 2006;42(8):1159-1166.
  21. Bruera E. Poor correlation between physician and patient assessment of quality of life in palliative care. Nat Clin Pract Oncol. 2006; 3(11): 592-593.

Further Reading

  • American Thoracic Society, CDC, and Infectious Diseases Society of America. Treatment of tuberculosis. MMWR Recomm Rep. 2003;52(RR-11):1-77.

Carolyn Bramante is a second-year medical student at the University of Minnesota School of Medicine in Minneapolis.

John Song, MD, MPH, MAT, is an assistant professor in the Center for Bioethics and the Department of Medicine at the University of Minnesota in Minneapolis. He completed a fellowship in general internal medicine at the Johns Hopkins University School of Medicine and the Greenwall Fellowship in ethics and public policy at Johns Hopkins and Georgetown Universities. Dr. Song’s research interests include bioethics education, homeless people and other underserved populations, end-of-life care, and medical professionalism.

The viewpoints expressed on this site are those of the authors and do not necessarily reflect the views and policies of the AMA.