Priority-setting ethics in public health

Priority-setting ethics in public health

Michaelis, Annie P


THE most fundamental goal of public health is to increase the well-being of populations, regardless of distinctions in social and economic status among population groups. This goal is difficult to achieve because socioeconomic characteristics play an inescapable role in most aspects of daily life-individuals are either limited or empowered by the circumstances of their demographic and social groups. In the United States, popular prejudices allow the health problems of those who are viewed as politically or socially less desirable to be given less attention than the problems of those who command power or sympathy within mainstream politics and society. Marginalized population groups are at risk of having their public health needs inadequately addressed.

While it is important that the field of public health be accountable to all citizens and sensitive to their values, public health ought not be subordinate even to majority opinions regarding the “social worth” of certain segments of society. Instead, public health professionals should develop and utilize tools with which to deliver services fairly, and to counter prejudiced arguments that declare certain groups undeserving of public assistance. A powerful argument can be made that it is unethical and dangerous to allow distinctions of “social worth” to influence decisions regarding the allocation of public health resources. Public health professionals must collaborate to establish a standardized ethical system of priority-setting with which to consistently and clearly refute social-worth arguments.

Ethical principles have long been established for the delivery of clinical care to individuals. An informative analogy can be made between the difficult ethical choices in allocating public health funding and the ethical quandaries faced by doctors and hospitals regarding organ transplantation. As in public health, resources for organ transplants are limited, and the demand exceeds the resources. This paper utilizes the principles underlying allocation decisions in the field of organ transplantation to explore the possibility of a similar set of standards designed to guide and to explain public health funding allocation decisions.


The AIDS Example

Deciding whether to intervene in a public health crisis based upon the perceived social value of those most immediately at risk is disrespectful of the high-risk group and dangerous to the general population. When cases of AIDS were first recognized in 1981, the disease was thought to be limited to homosexuals and intravenous drug users. Reflecting strong, popular prejudices, officials in the fields of medical care, politics, and justice were slow to address the problem (1). An excerpt from a 1985 article in the Washington Post (2) gives a concise but revealing impression of the social climate in 1981:

At first the deadly virus, discovered in 1981, had been thought to be confined to discrete groups, primarily male homosexuals and drug addicts, for whom there was limited sympathy in society at large. Some uncharitable persons even suggested that there was no public responsibility to search for a cure, since the afflicted had voluntarily chosen to engage in the conduct that leads to the disease.

Such a suggestion is typical of worth-based arguments; the groups affected by the public health emergency were seen to be at fault and suffered social stigma because of their lifestyles. Looking at the beginning of the AIDS epidemic in retrospect, it is easy to see how a public health problem that is perceived to be isolated can become devastating for the country and the world. AIDS is one among many examples of public health crises that have been addressed slowly-to the detriment of society as a whole-because the primary victims were socially marginalized or economically disadvantaged. The nineteenth-century history of tuberculosis’ massive toll on the poorest segments of society is illustrative of this point (3), as are contemporary impediments to gun-violence prevention in the United States’ inner cities (4,5).

Although there were bureaucratic miscommunications and infrastructure inadequacies that contributed to the slow beginning for AIDS research (1,6), such obstacles would likely have been more quickly hurdled if “innocent” victims had been affected from the start. Consider the acceleration of media coverage and government concern for the growing AIDS epidemic when it was first discovered that blood transfusion victims, such as child hemophiliacs, were also contracting the disease (1).

Perhaps there also would have been less delay if public health professionals had functioned under an acknowledged set of ethical guidelines that gave precedence to such criteria as the number of victims and the severity of the disease’s effects on its victims, turning a blind eye to the social and political power of the affected groups. In striking contrast with the response to AIDS, the history of Legionnaire’s Disease demonstrates that the federal government could have mobilized itself more quickly and efficiently if it had believed AIDS to be a public health problem of significance. In July 1976, members of the American Legion attending a conference at the Bellevue-Stratford Hotel in Philadelphia began to fall sick from an acute pneumonia illness. Ultimately, 221 people were stricken and 34 died. Believing the mysterious deaths to be the beginning of a Swine Flu epidemic that was spreading through Asia, Congress immediately confronted the issue and President Ford signed the National Swine Flu Immunization Program of 1976, to ensure mass immunization for all American citizens (7). However, it soon became apparent that the mystery illness was not Swine Flu. Even after this relieving news, the federal government continued to pursue the mystery and the initial Center for Disease Control (CDC) investigation was expanded so that all the disease survivors, their available relatives, and more than 4,400 Legionnaires and their families were questioned, and cadavers were examined at the microscopic level (7). In January 1977, it was discovered that the infection is spread by bacteria thriving within certain types of air conditioning systems. Since that time, federal agencies have required more stringent cleaning and hygiene provisions for large-scale air conditioning systems.

This example of federal mobilization in the face of a potential epidemic is a stark contrast to the government’s reaction to the AIDS epidemic. By 1983, the year during which the CDC finally determined that AIDS should become a reportable condition (8), at least 1,000 people, most of whom were homosexuals, had died from the disease and 1,500 more were known to have the disease (9). Although health officials from the government realized that growing numbers of gay men were dying from a mysterious disease and many individuals pursued the issue, the response of the federal government as a whole was not characterized by the alacrity that it showed when the victims of a mysterious infection were United States veterans. Even by 1986, as the epidemic was continuing to escalate, a report from the Institute of Medicine and the National Academy of Sciences concluded that U.S. research efforts remained inadequate (10).

If worth-based decision-making had been as taboo in the field of public health as it is in organ transplantation, perhaps early advocates of AIDS research and prevention would not have had such a difficult task in convincing politicians and the public that AIDS work needed to be funded immediately. It would be overly idealistic to believe that a system could be established in which politics and social status did not matter, but we can and should attempt to create a public health system in which a comprehensive ethical framework helps counterbalance the occasionally unjust leanings of majority or vocal minority opinions.

Social Worth and Ethics

Today’s public health system is susceptible to worth-based allocation arguments. Some argue that considering the social worth of a program’s beneficiaries is not necessarily an unethical method of deciding between competing program proposals (11). However, “worth” is essentially subjective, and worth-based decisions are ethically tenuous. Given the multitude of variables involved in most public health problems, there are usually inadequate resources and time to discover all of the facts that would, if known, influence our value judgments. Even if we were able to identify all relevant factors, it would be dangerous to assume that we, as subjective and fallible human beings, could consistently make unbiased and just worth-based decisions. Despite these limitations, special-interest groups or political actors with specialized agendas too often have success using worth-based arguments to undermine the efforts of public health professionals on more controversial program initiatives. To improve this situation, it is critical that public health institutions collaborate to establish an explicit and public consensus about theoretical priority-setting standards for the delivery of public health services. Beginning from the extensive work that has already been done by bioethicists and physicians in the field of organ transplantation, it is possible to envision a more comprehensive and widely accepted system of priority setting for public health interventions.


When dealing with organ transplantation, prioritization is a difficult and emotionally charged issue. There are never enough organs to fully supply the need, and patients who do not receive an organ in time will die. Physicians and bioethicists from many nations have examined various methods of making priority-setting decisions (12-16). Although there is not agreement on all details, there seems to be a consensus regarding which types of criteria are acceptable considerations and which are not. Various acceptable and unacceptable criteria are thoroughly examined in a 1995 article written by the Council on Ethical and Judicial Affairs of the American Medical Association (16). According to this article, the acceptable criteria for resource allocation are:

* Likelihood of benefit to the patient

* Impact of treatment on the patient’s quality of life

* Duration of the benefit

* Urgency of the patient’s need

* All else being equal: the amount of resources required for successful treatment

Although each of these factors refers specifically to organ transplantation, each may also be relevant to priority setting on the broader scale of public health. From a public health standpoint, the “patient” refers to the population that stands to gain from a specific program, and “treatment” includes all measures taken by that program. While considering these criteria, it is important to keep in mind that each criterion should be used in the context of the others and that they are not intended to be ranked hierarchically-decision-makers must use their judgment, case-by-case, to decide which considerations outweigh others. In this judgment, decision-makers should be guided by a principle of significant differences: the greater the differences between potential beneficiaries with regard to a certain criterion, the more weight that criterion may be given in the final decision. It should be noted that, in real-life situations, the individual criteria often overlap and must be considered jointly.

Likelihood of Benefit

According to the AMA article, likelihood of benefit is an acceptable criterion because it helps “maximize the number of lives saved as well as the length and quality of life” (p. 29). In organ transplantation, this means that, all else being equal, a person who has a ninety percent chance of benefiting from a transplant should have priority over a person with a twenty percent chance of benefiting from the same organ. Physicians are encouraged to determine the likelihood of benefit by examining physiological realities but not situational factors (such as a patient’s difficulties with communication or a non-supportive home environment), which could affect transplant success. Situational factors are not to be considered because reliance on such criteria may systematically disadvantage those members of society who most need assistance (16).

In public health, this criterion is a difficult one because most relevant factors are situational. However, it is not impossible to use the “likelihood of benefit” criterion in an ethically sound manner. First, predictions about likelihood of benefit should be based primarily on the structure of the proposed program, rather than on the attributes of the beneficiaries themselves. For example, if a program has been designed to spread information about AIDS prevention in a low-income community, it is important to consider whether the structure of the program lends itself to the situation (will the information be disseminated by a source respected by community members?) and to consider whether similar initiatives have had success in other communities (do AIDS education programs truly lower the incidence of risky behavior?).

On the other hand, it may be necessary to consider possible situational barriers that might gravely affect a program’s success. If it is found that, due to social or cultural traits of the community, a program is unlikely to be effective, it is the duty of public health professionals to attempt modifications of the program in order to overcome those obstacles. If they can be remedied, the program’s likelihood of benefit will be substantially increased. Regarding this issue, the AMA Council on Ethical and Judicial affairs states:

In general, if patient traits or behaviors that may adversely affect the patient’s likelihood of benefit are taken into account, at least two conditions should be met. First, reasonable extra efforts on the part of the physicians or others must fail to overcome the obstacles posed by the patients’ traits or behaviors; and second, the traits or behaviors must directly and substantially detract from the patient’s likelihood of benefit, (p. 30)

Thus, it is the duty of public health professionals to recognize which situational factors would significantly detract from a program’s goals and to construct programs designed to overcome the most detrimental obstacles. Because this is a difficult task and because it is not always possible to accurately predict a program’s likelihood of benefit, this criterion should not be used without reference to other considerations. Nonetheless, this is one element of an ethical priority-setting system and is useful on a theoretical level to remind public health decision-makers that it is imperative they analyze any available statistics about the potential outcomes and typical setbacks of program designs.

Quality of Life

Quality of life can be a problematic criterion in both medicine and public health. Although physicians can make educated assessments of the ways in which a certain procedure will help a patient, the idea of “quality” is necessarily subjective. For example, one person may value eyesight above all else, while another person may believe that the ability to walk is the most important determinant of health and quality of life. On a level more relevant to public health, one person may believe that accessible medical care is the best determinant of a population’s quality of life while another person may argue that universal access to quality education is most important.

The authors of the AMA article address this dilemma by stating, “improvements in quality of life would be measured for each patient by comparing functional status with treatment and functional status without treatment” (p. 30). In other words, the comparison should involve the before-and-after status of each case, rather than a judgment regarding which types of benefits (among separate cases) constitute the greatest improvement in quality of life.

A hypothetical example will demonstrate the relevance of this method in the discipline of public health: one community suffers from a high rate of teenage delinquency and resulting social problems such as vandalism and alcohol-related injury. A second community suffers from an affordable housing crisis, forcing extended families to live together in crowded and insufficient space. A charitable foundation must decide whether to fund a youth center for the first community or to build an affordable apartment complex for the second community. Based on relevant statistics and data from similar programs, it is predicted that a youth center could lower teenage delinquency rates by 20 percent. The affordable housing complex would provide new living opportunities for 5 percent of the families in the second community. Regardless of personal feelings about the importance of youth safety and development as compared to the importance of healthy living situations for families, the funding-all else being equal-should go to the youth center program with its 20 percent improvement rate. However, if the situations of the two communities were not otherwise equal, the quality of life improvement might not be as important. For example, if the community in need of a youth center was a wealthy, suburban neighborhood and the community in need of affordable housing was an inner-city ghetto, the funding allocation decision-makers would have a more complex choice.

This raises a new criterion that should be considered in public health priority setting: available alternatives. If a community or group has surplus resources that are not being utilized to their fullest extent towards solving a problem, or are being used unproductively, it should be the duty of public health professionals to work with the community in such a way that the community’s own resources (rather than, or in addition to, external resources) may be used to alleviate existing public health problems. In such cases, public funding would be most wisely used on a limited scale, providing the community with guidance and instruction from experts, so that the community might establish its own sustainable, self-motivated program. For example, before using public resources to establish a new youth center in the wealthy community, public health professionals might first start a focused consciousness-raising program, designed to culminate in a community-led fundraising campaign that would build the center. This strategy promotes efficient use of community resources and leaves public health resources available for other important programs to be initiated in areas with fewer internal resources. Although this scenario neglects some of the complexities of a real-life funding priority dilemma, it isolates the quality of life criterion and clearly demonstrates the need to consider quality of life improvements and available alternatives.

Duration of Benefit

The third factor indicated by the AMA article is duration of benefit. It is often possible to predict accurately the longevity of a program’s projected benefits. In organ transplantation, the ethical issue of age discrimination is an important consideration. The AMA council tackles this dilemma by stating, “Patients should be assessed according to their own medical histories and prognoses, not aggregate statistics based on membership in a group” (p. 31). Thus, if an organ transplant candidate is 47 years old and he or she stands to gain many more years of life than a particular younger candidate, the ethical priority-according to this category alone-would be to give the older patient the organ.

Duration of benefit considerations are also relevant to public health. For example, if a community program is launched to give free flu vaccinations at the beginning of a particularly harsh winter season, it may prevent hundreds or thousands of people from contracting the flu. The program’s direct benefit lasts approximately one year. In comparison, a program to provide flu shots to public elementary school students each winter for a projected period of at least ten years would benefit smaller numbers of people for a longer period of time. With this example, it becomes obvious that, for public health purposes, we must create an additional criterion: number of beneficiaries. The duration of benefit must be linked to the number of beneficiaries to ensure that the true scope of each program is known. In addition, duration of benefit can be closely tied with quality of life: if the choice were between a flu vaccine for children and a polio vaccine for children (all else being equal), priority would go to the polio immunization program, because polio vaccines provide a lifetime of immunity and also because, in children, the negative health consequences of polio are usually far more severe than those of the flu.

Urgency of Need

In organ transplantation, urgency can be measured most basically by the degree to which a person will suffer physical harm if he or she is not given an organ at a certain time (16). In public health, urgency can be considerably more complex to assess because public health programs work at a population level rather than an individual level. Urgent public health problems can remain unaided if they lack a champion who has the resources and initiative to organize the compilation of statistics and data that are necessary for the scope of a problem to be understood.

The most objective manner of determining the urgency of a public health issue is to consider the number of people falling victim to a certain social problem and the severity of the medical and/or social consequences suffered by those victims. While considering numbers, however, it should be kept in mind that prevention is often the most consequential product of public health efforts. As the AMA authors write, “If patients with a less urgent need are set aside until their condition deteriorates to the point of dire emergency, treatment may not be as beneficial as it would have been had it been begun much earlier” (p. 31). Public health professionals should devote energy to considering the future implications of current threats, even when the present urgency of a problem does not seem great.

Amount of Resources

The final criterion indicated by the AMA article is the amount of resources required. As the authors carefully note, this criterion is not useful in all cases. However, on occasion, assigning higher priority to patients who need less of a scarce resource can maximize the number of lives saved (16). For example, if two patients are in equal circumstances except that one needs a heart transplant and the other needs both a heart and a liver transplant to survive, priority should go to the patient who needs only a heart; instead of using two organs to save one life, the liver can then be used for another patient in need. Unlike the other four AMA criteria, this criterion should not be applied in public health. Because public health programs work on a population level rather than an individual level, there is not necessarily a correlation between resources used and lives saved; in some cases, more resources will save more lives, while in other cases, a relatively simple and inexpensive program might be the key to solving a grave public health problem. In public health, considering the amount of resources required is ethically troubling because, oftentimes, the problems that would require the most resources to fix are the problems affecting the poorest and most underserved social groups.

Inappropriate Criteria for Resource Allocation

In addition to the five acceptable criteria, the AMA authors list several inappropriate criteria that they have found to be prevalent in medical literature and in allocation decisions: ability to pay, social worth, perceived obstacles to treatment, contribution of the patient to his or her own medical condition, and past use of resources. While ability to pay is not relevant to public health priority setting, and social worth has been discussed previously, the other three criteria have significant implications that warrant mention.

Perceived obstacles to treatment is an unacceptable decision-making criterion because it would systematically give lower priority to people whose circumstances pose special challenges to successful treatment. For example, because it is more difficult to locate and communicate with the indigent, public health programs designed to serve them must go to extra lengths to overcome the obstacles inherent in indigent lifestyles.

A patient’s contribution to his or her own condition is a factor that can be associated with the social worth argument. Some assert that, when patients (or program beneficiaries) have failed to take action to protect themselves, or have actively contributed to their own condition, they are to blame for their predicament and have forfeited the right to be given the same priority for treatment as others. The AMA Council points out two ways in which this reasoning is flawed: first, the degree to which particular factors are responsible for a condition is often unclear. Second, it tends to be the case that, of all the possible contributing factors, only some actions are singled out as blameworthy, while other harmful behaviors are viewed as either innocent or unavoidable.

Past use of resources can be applied in such a way that blame for the failure of a program falls on the program’s intended beneficiaries. For example, numerous federal and state programs have focused on alleviating the problems associated with alcohol abuse on Native American reservations (17). Many of these programs have been largely unsuccessful and, as a result, Native Americans have been criticized as unwilling or unable to change their own destructive behaviors and have become stereotyped as “drunken Indians” (18). Despite obvious bureaucratic mismanagement of resources by the government programs-including the “loss” of millions of dollars that were meant to fund programs on reservations (18)-individuals inside and outside the government have blamed program failures on Native American biological and cultural “predisposition” to alcohol abuse, arguing that Native Americans are not deserving of future government resources because they have failed to take advantage of past programs (17,18).

Individuals making arguments about past use of resources may point out that certain groups have been given numerous chances to benefit from past programs but have not changed their lifestyles; therefore, other groups, who have not already received as many public health resources, should be given priority. This kind of argument fails to consider the fact that groups do not start out on equal ground (16). If the programs in one community have not paid off as quickly as programs in another community, it is possible that the first community suffers from deeper problems, or that the programs directed at that group were flawed in design or execution and therefore were inadequate for successfully addressing the issues.


In public health’s determined pursuit of social justice, funding prioritization is a central but often overlooked quandary. Although it is a daunting task, United States public health professionals have the potential to fortify the ethical standards by which resource allocation decisions are made and explained to the public. Using variations on the criteria that bioethicists and physicians have agreed upon for priority setting in organ transplantation, an ethical framework for public health funding would include evaluations of the likelihood that a program would benefit its recipients, the degree of improvement to the beneficiaries’ quality of life, the available alternatives within the community, the duration of the benefit, the number of people assisted by the program, and the urgency of the problem addressed by the program. Social worth decision-making is ethically tenuous and can be detrimental to society as a whole. It is vital that public health professionals work to change social norms about priority setting by formulating and publicizing a standardized set of prioritization criteria following the model of the now widely accepted measures used in organ transplantation.

Acknowledgments: I extend my deepest thanks to Stephen Teret for his invaluable contributions to the brainstorming process and for his assistance in editing the many early drafts of this article. I would also like to thank Martin McKneally, Julie Mair, James Hodge, and Shannon Frattaroli for their thorough readings of the article and extremely helpful suggestions.


1. Shilts, R. And the Band Played On: Politics, People, and the AIDS Epidemic. New York, NY: St. Martin’s Press, 1987.

2. “AIDS: Risks and Responsibilities,” Washington Post. Wed. Aug. 14, 1985: A22.

3. Rosen, G. A History of Public Health. New York, NY: MD Publications, Inc., 1958.

4. Cooper, J. “Cooper’s Corner,” Guns and Ammo. April 1991: 104.

5. Murphy, S. L. Deaths: Final Data for 1998. National Vital Statistics Reports 48 (11) (2000).

6. Lee, P. R., and Arno, P. S. “The Federal Response to the AIDS Epidemic,” Health Policy 6 (1986): 259-67.

7. Garrett, L. The Coming Plague: Newly Emerging Diseases in a World Out of Balance. New York, NY: Farrar, Straus, Giroux, 1994.

8. Report of the Presidential Commission on the Human Immunodeficiency Virus Epidemic. “Chapter 5: The Public Health System,” June 1988. James D. Watkins, Chairman.

9. Sargent, E. D. “1,500 March to Hill Seeking Funds to Fight AIDS,” Washington Post. Sun. Oct. 9, 1983: B1.

10. Walters, L. “Ethical Issues in the Prevention and Treatment of HIV Infection and AIDS,” Science 239 (1988): 597-603.

11. Suter, E. A., Waters, W. C., Murray G. B., et al. “Violence in America: Effective Solutions.” J. Med. Assoc. of Georgia 84 (1995): 253-63.

12. Mullen, M. A., Kohut, N., Sam, M., Blendis, L., and Singer, P. A. “Access to Adult Liver Transplantation in Canada: A Survey and Ethical Analysis.” Can. Med. Assoc. J. 154 (3) (1996): 337-42.

13. Harris, J. “Cardiac Surgery in the Elderly,” Heart 82 (1999): 119-20.

14. Holm, S. “Goodbye to the Simple Solutions: the Second Phase of Priority Setting in Health Care,” BMJ 317 (1998): 1000-1007.

15. Klein, R. “Puzzling Out Priorities,” BMJ 317 (1998): 959-60.

16. Council on Ethical and Judicial Affairs, American Medical Association. “Ethical Considerations in the Allocation of Organs and Other Scarce Resources Among Patients,” Arch. Intern. Med. 155 (1995): 29-40.

17. Brady, M. “Alcohol Policy Issues for Indigenous People in the United States, Canada, Australia and New Zealand,” Contemporary Drug Problems 27 (3) (2000): 435-509.

18. Miller, J. M. and Hazlett, M. “The ‘Drunken Indian’: Myth Distilled into Reality Through Federal Indian Alcohol Policy,” Arizona State Law J. 28: 223-98.

Copyright Journal of Public Health Policy 2002

Provided by ProQuest Information and Learning Company. All rights Reserved