The Patient's Duty to Adhere to Prescribed Treatment: An Ethical Analysis

Abstract

This article examines the ethical basis for the patient's duty to adhere to the physician's treatment prescriptions. The article argues that patients have a moral duty to adhere to the physician's treatment prescriptions, once they have accepted treatment. Since patients still retain the right to refuse medical treatment, their duty to adhere to treatment prescriptions is a prima facie duty, which can be overridden by their other ethical duties. However, patients do not have the right to refuse to adhere to treatment prescriptions if their non-adherence poses a significant threat to other people. This paper also discusses the use of written agreements between physicians and patients as a strategy for promoting patient adherence.

Keywords:

• contracts
• non-adherence
• noncompliance
• patient's duties
• physician-patient relationship

I. Introduction

Consider the following hypothetical (though typical) cases in medicine

A 21-year-old female with type I diabetes has been admitted to the hospital with her Fifth diabetic ketoacidosis (DKA) in the last 12 months. She has not been taking her diabetes medications as directed, checking her blood sugar levels, or adhering to her diet or exercise program.

A 27-year-old male with HIV has been admitted to the hospital for treatment of pneumonia. His disease has progressed rapidly in the last year. He has been prescribed antiretroviral drugs to treat HIV, but he has not taken them consistently and has missed follow-up appointments.

Both of these cases involve what has been called “noncompliance” or “non-adherence.” Some find the word “noncompliance” to be offensive, since it implies that the patient is a like child who must obey the doctor's orders (Lutfey & Wishner, 1999; Jacobson, 2000). To avoid the negative connotations of the term “noncompliance,” this article will use the term “non-adherence” instead. Regardless of how one describes cases like the two above, they all deal with some failure by the patient to follow treatment prescriptions. The patient has come to the doctor seeking help for a medical problem. The doctor has made a diagnosis and has recommended treatment. Although the patient has accepted the doctor's advice, the patient has not taken the steps required to make the treatment effective. The patient has not kept up his or her end of the bargain.

Patient non-adherence is a well-documented medical phenomenon. It has been discussed within the context of HIV/AIDS (Singh et al., 1996), diabetes (Johnson, Bazargan, & Bing, 2000), hypertension (Sharkness & Snow, 1992), organ transplantation (Edwards, 1999), substance abuse treatment (Lash, 1998), chronic pain management (Fishman, Mahajan, Jung, & Wilsey, 2002), rehabilitation (Kirschner et al, 2001), and medical research (Melnikow & Kiefe, 1994).

There are many different psychosocial and economic explanations of why patients often do not follow medical recommendations, including financial difficulties, inadequate education or counseling, poor communication between doctor and patient, unpleasant side effects of medications, bad tasting medications, family turmoil, mistrust of doctors, and mental illnesses and disabilities (Stephenson et al., 1993). There are also a variety of practical solutions to non-adherence, such as better patient education, counseling, and communication; changes in the frequency, dose, or route of intake of medications; prescribing affordable medications and communicating with patients about the costs of medications; providing patients with financial or psychosocial support; clinical monitoring of drug levels in patient's blood streams; and improved management of drug side-effects (Stephenson et al., 1993; Alexander, Casalino, & Meltzer, 2003).

This article will not examine the causes of non-adherence in any detail, since its purpose is to focus on the ethical aspects of the problem. The article will explore the ethical basis for the patient's duty to adhere to treatment prescriptions. The article will argue that patients have a moral duty to adhere to treatment prescriptions, once they have accepted the treatment. Since patients still retain the right to refuse medical treatment, the duty to adhere to treatment prescriptions is a prima facie duty, which can be overridden by the patient's other ethical duties to one's self or others. However, patients do not have the right to refuse to adhere to treatment prescriptions if their non-adherence poses a significant threat to other people. This article will also assess the use of written agreements between doctors and patients as a strategy for dealing with non-adherence.

II. What is Non-Adherence?

At the outset, it will be useful to define “non-adherence” and distinguish it from other concepts in medical ethics. The first point to observe is that there is a difference between non-adherence and unhealthy lifestyle choices. Consider a man who has been smoking cigarettes for 40 years but has never seen a doctor. His smoking habit is unhealthy, but it does not constitute non-adherence per se. Suppose that the man makes an appointment with a doctor because he is having a persistent cough and shortness of breath. After conducting various tests, the doctor informs the man that he has cardiovascular disease and should stop smoking. The man accepts this treatment recommendation. The doctor refers the man to a stop-smoking clinic and prescribes a nicotine patch. If the man continues to smoke after accepting this specific treatment prescription, we would call this non-adherence.

One should also distinguish between non-adherence and treatment refusal. It is a clearly established point of medical ethics and law that a competent patient has a right to refuse medical treatment, including life-saving treatment (Gostin, 1997, Beauchamp & Childress, 2001). Suppose the doctor advises the cigarette smoking man (from the example above) that he should have major surgery to remove a tumor from one of his lungs. After careful consideration, the man decides that he does not want the surgery. Even though the man has not heeded his doctor's specific treatment recommendation, we would call this refusal of treatment rather than non-adherence. The person who refuses treatment does not accept the treatment at all; the person who does not adhere to treatment accepts the treatment but fails to adhere to the regimens needed to implement it.

Some cases that look like non-adherence from the physician's perspective appear to be refusal of treatment from the patient's perspective. For example, suppose that a patient is not taking his or her medication because of its unpleasant side effects, but the patient also does not tell the doctor why he or she is not taking the medicine. This situation would look like non-adherence from the doctor's perspective, but it would look like refusal of treatment from the patient's perspective. There is probably a significant number of cases, especially those involving problems with the side effects of drugs, which could be described as refusal or treatment or non-adherence, depending on one's perspective. Better communication between doctor and patient would seem to be the best way to approach these types of cases.

Having made these two important distinctions, we can define non-adherence as follows: non-adherence occurs when

1. a patient (or the patient's legal representative, such as a guardian or health care agent) seeks help from a physician for a medical problem;

2. the physician evaluates the patient and makes a treatment prescription;

3. the patient accepts the treatment prescription, that is the patient promises to implement the treatment prescription and the patient intends to implement the treatment prescription; but

4. the patient fails to implement the treatment prescription.

This definition helps us to understand the difference between non-adherence and treatment refusal. Someone who refuses treatment does not accept the treatment. A patient may manifest his or her non-acceptance of the treatment by not promising to implement the treatment or not intending to implement it. A patient may manifest his or her non-acceptance after initially accepting the treatment. For example, a patient could try a medication, have unpleasant side effects, and then decide that he or she no longer intends to take the medication. Or a patient could seek a second opinion about his or her situation. Neither of these situations would constitute non-adherence. The patient has the right to change his or her mind and even break his or her promise to the doctor, except in specific situations, which we will discuss below. The patient could even change his or her mind again and tell the doctor that he or she wants to try the medication again. In doing so, the patient would be reaffirming his or her intention to take the medication. As long as the patient honestly communicates with the doctor about the decision to not take the medication, then we should view his or her behavior as treatment refusal, not as non-adherence.

This definition also sheds some light on why doctors find non-adherence to be so frustrating. When a patient fails to adhere to a prescription, a doctor may feel betrayed or exploited because the patient is not upholding his or her end of the bargain. The doctor may view the patient as defiant because the patient is refusing to follow medical advice; lazy, because the patient does not seem to have the willpower to implement medical advice; or even irrational, because the patient apparently accepts the goals of the treatment plan but is not following effective means to implement it. ¹ Unless the doctor can find a satisfactory explanation of or justification for the patient's behavior, for example, that the patient cannot afford the treatment or does not know how to take it properly, the doctor may view non-adherence as a moral problem. This is the main reason why doctors often describe non-adherent patients in morally loaded terms, such as “lazy,” “difficult,” “defiant,” “non-compliant,” or “crazy.”

Since doctors have moral duties to benefit patients and promote public health, they have an obligation to try to treat non-adherent patients, which often requires considerable patience (Scofield, 1995). Doctors should continue to work with patients to address the various problems that may be causing non-adherence, such as lack of access to medications, side effects of medications, difficulties with taking medications, forgetfulness, transportation problems, and so on. However, if the pattern of non-adherence continues despite the doctor's concerted efforts help the patient implement the treatment, the doctor may feel that he or she is wasting time and society's resources. Non-adherence eventually takes its toll. At some point, many doctors decide that they would rather not treat non-adherent patients.

III. Non-Adherence and Patient Duties

To get a better understanding of the ethics of non-adherence, it will be useful to think about how non-adherence relates to the patient's duties, since non-adherence appears to involve the patient's failure to uphold his/her ethical duties in the physician-patient relationship (Jacobson, 2000; Brown, Dickson, & Van der Wal, 2003). Before the demise of medical paternalism, doctors expected patients to follow all of their treatment prescriptions, even when patients did not agree with them. The Code of Medical Ethics adopted by the American Medical Association (AMA) in 1847 discussed the patient's duties to doctors at length and asserted that the patient has a duty to promptly obey the physician's orders (Meyer, 1992).

Medical ethics opinions, codes, and articles written in the last three decades have emphasized patient's rights, such as the right to refuse treatment, the right to make medical decisions, the right to be informed, and so on. Even so, two powerful organizations in the United States (U.S.), the American Hospital Association (AHA) and the AMA, have both published documents that address the patient's duties to cooperate with medical recommendations. The AHA's Patient's Bill of Rights, which has been adopted by virtually every hospital in the United States, includes a section on the patient's responsibilities that makes a vague reference to non-adherence:

The effectiveness of care and patient satisfaction with the course of treatment depend, in part, on the patient fulfilling certain responsibilities…Patients are responsible for informing their physicians and other caregivers if they anticipate problems in following prescribed treatment. (American Hospital Association, 1992)

While the AHA's reference to “problems” with meeting prescribed treatment avoids the use of offensive language, it does not clearly state the patient's duties. There are many types of “problems” with following a prescribed treatment, such as refusal of treatment or financial difficulties, which may not constitute non-adherence.

The AMA's Council on Ethical and Judicial Affairs has an opinion on patient responsibilities that specifically addresses the problem of non-adherence but uses the word “compliance”:

While physicians have the responsibility to provide health care services to patients to the best of their ability, patients have the responsibility to communicate openly, to participate in decisions about the diagnostic and treatment recommendations, and to comply with the agreed-upon treatment program…Once patients and physicians agree upon the goals of therapy and a treatment plan, patients have a responsibility to cooperate with that treatment plan…Compliance with physician instructions is often essential to public and individual safety. (AMA, 2001a)

Unlike the AHA, the AMA makes it clear that the patient has a duty to “cooperate with the treatment plan” and “comply” with physician instructions. The AMA opinion also implies a distinction between refusal of treatment and non-adherence, because it holds that the duty of compliance arises only when the physician and patient have agreed upon the goals of therapy and a treatment plan.

While the AHA and AMA documents provide some useful guidance, they do not provide an ethical analysis of the patient's duties that could help us better understand non-adherence. Surprisingly, the bioethics literature also has very little to say about the ethical basis of the patient's duties. Many articles and books focus on the rights of the patient but only a few, for example, Meyer (1992), Sider and Clements (1984), address the duties or obligations of patients in any depth. Indeed, a MEDLINE search of abstracts, conducted on August 15, 2003, found that 432 articles (96%) addressed patients’ rights, while only 18 (4%) addressed patients’ duties, obligations, or responsibilities. Clearly, bioethics scholars need to do more thinking and writing about patient's duties. ²

IV. Patient's Duties: A Kantian Analysis

To better understand patients’ duties, it will be useful to draw some insights from moral theory. Since there is not sufficient space in this article to review even a short list of moral theories, this article will apply Immanuel Kant's (1724–1804) theory concerning the questions at hand. Kant's theory has inspired such contemporary philosophers and political theorists as John Rawls, Alan Donagan, Christine Korsgaard, Onora O'Neill, Thomas Nagel, Alan Gewirth, and Bernard Williams, and it provides the theoretical grounding for the principles of autonomy and respect for persons in biomedical ethics (Beauchamp & Childress, 2001; Marshall, 2001).

Kant's fundamental insight is that moral conduct consists in following a rule (or maxim) that is valid for all rational beings. For Kant, the rightness (or wrongness) of an action derives from its motivation. Rational beings (or moral agents) can act from the motive of inclination, self-interest, or duty. Rational beings act from the motive of duty when they do their duty for duty's sake. Moral duties are imperatives that are valid for all rational beings—categorical imperatives. According to Kant, all categorical imperatives can be derived from a single moral principle, the categorical imperative (CI). Kant states three different versions of the CI, which are materially equivalent because they imply the same types of duties. ³ Freedom of the will (or autonomy) consists not in doing whatever one chooses to do; it consists in choosing actions that do not violate the moral rules (Kant, 1753).

To decide whether an action would be right or wrong, one should formulate a maxim describing the action and determine whether it is consistent with the CI. If the maxim is consistent with the CI, the action is morally permitted; if the maxim is not consistent with the CI, the action is morally forbidden (O'Neill, 1975). For example, consider the action, “I will make a promise I do not intend to keep in order to get what I want.” The maxim of this action would not be consistent with the CI; promises would be useless if everyone always made promises they did not intend to keep. The maxim would be invalid for all rational beings, and the action it represents would be wrong.

Kant made two distinctions between moral duties. First, he distinguished between duties to one's self and duties to others (Kant, 1753). Kant argued that one may have duties to one's self, such as a duty not to commit suicide or a duty not to harm one's self. Rational agents often promote their own good when they act from inclination or self-interest, but this is not the same as promoting one's own good from the motive of duty. Duties to other people include the familiar list of duties one finds in widely accepted moral codes, such as the duties not to lie, cheat, steal, break promises, harm, or murder. A rational agent treats every human being, including himself or herself and other people, as having intrinsic value and worth (Kant, 1753).

Second, Kant distinguished between perfect and imperfect duties (Kant, 1753). A perfect duty is a duty that one should never violate. For example, Kant believed that people have a perfect duty not to commit murder or suicide. An imperfect duty is an obligation that one may sometimes violate in order to meet other obligations. For example, Kant believed that people have a duty to help others. This obligation is an imperfect duty because we must sometimes refrain from helping others in order to help ourselves, keep promises, and so on. When perfect and imperfect duties conflict, one should always adhere to perfect duties. Thus, imperfect duties are like W.D. Ross’ prima facie duties in that they have some moral justification, even though they can be overridden (Ross, 1930).

How might one apply this framework to the moral duties of patients? ⁴ Let's consider duties to one's self first. Kant held that one has a duty to develop one's talents and abilities:

[A rational being] finds himself in comfortable circumstances and prefers to indulge in pleasure rather than to bother himself about broadening and improving his natural aptitudes…But he cannot possibly will that this should become a universal law of nature or be implanted in us as such a law by natural instinct. For as a rational being, he necessarily wills that all his faculties should be developed, inasmuch as they are given him for all sorts of possible purposes. (Kant, 1753, section 423, p. 31)

This passage supports an imperfect moral duty to maintain one's own health, since health plays an essential role in developing one's natural talents and abilities. A rational being would perform the actions necessary to promote his/her own health, such as eating a balanced diet, getting enough exercise, practicing good hygiene, avoiding risky activities, and seeking the help and advice of health care professionals when needed. But how should one balance the duty to maintain health, which is an imperfect duty, against other duties? Does Kant's moral philosophy require us to devote our lives to maintaining our health? Moreover, what is health?

This article will not explore questions about the definition of health in detail. ⁵ However, it will sketch a Kantian answer to questions about balancing the duty to maintain one's health against other duties. Consider the maxim, “I will not engage in activities that place my health at risk.” Rational agents would not choose this as a law for all people because they would recognize that to develop their talents and abilities, they may need to engage in some activities, such as work or recreation, which place their health at risk. Rational agents may also sometimes need to place their health at risk in order to help other people. Thus, a rational agent would place his or her health at risk in order to honor moral obligations to himself or herself or others.

Next, let's examine duties to other people. The CI implies that we have a duty not to harm people or place other people at a significant risk of harm. Consider a very risky activity with very little benefit for other people, such as drag racing. A rational being would not choose as a rule for all people a maxim that permits drag racing, because adoption of such a rule by society would result in many injuries and fatalities, due to recklessness.

Patients who ignore medical advice may place other people at risk. For example, a person who takes antiretroviral medications inconsistently may develop resistant strains of HIV, which could pose a public health risk (Senak, 1997). Patients sometimes have a duty to follow medical prescriptions in order to protect other people from harm (Jacobson, 2000). Of course, recognition of this general duty does not settle all of the important questions relevant to harm to others, such as questions about the probability, severity, and preventability of the harm, or how to balance benefits and harms. There is not sufficient space to answer these important questions in this article. ⁶

The CI also implies that moral agents have a duty not to lie to other people. Indeed, Kant held that people have a perfect duty not to lie to one another because it is impossible for a rational being to conceive of lying becoming a general law. Lying contains a contradiction in thought: if everyone lied, all lies would be useless, because no one would believe them (Kant, 1753). Although I agree with Kant that people have a moral duty not to lie, I do not regard truth telling as perfect duty. I think lying could be justified for an important purpose, such as to save a human life. I will not explore this issue further in this article. ⁷

Regardless of whether one views the obligation not to lie as an imperfect or perfect duty, it is clear that Kant's theory condemns lying. If we apply this insight to health care, it implies that doctors and patients both have a moral duty to tell the truth to one another. Although many articles in the bioethics literature have focused on the physician's duty to tell the truth to patients, it is important to recognize that patients have a parallel duty not to lie to their doctors. Patients should not lie to their doctors about their medical history, social history, symptoms, and financial arrangements (Meyer, 1992).

The last type of moral duty that this article will address concerns the duty to keep one's promises. We have already noted Kant holds that one has a perfect duty not to make a promise that one does not intend to keep, since this action would be based on a self-contradictory maxim (Kant, 1753). But would it be wrong to break a promise that one intended to keep if the obligation to keep the promise conflicts with another moral obligation? For instance, suppose that I promise to meet a friend for lunch, but on the way I stop to rescue someone from a car accident. One might argue that a rational agent would not legislate the maxim, “I will not break a promise to help someone in desperate need” as a universal law for all people, because a rational agent would recognize that he/she might sometimes be in desperate need of aid from someone who made a promise to be somewhere else. Suppose that I have promised to meet someone for lunch but I become very dizzy and feverish and need to lie down at home for a while. A rational agent would not endorse a maxim that prohibited a person from taking care of him/her self in order to avoid breaking a promise.

Finally, one of the key tenets of ethical reasoning is that “ought implies can”: one cannot have a moral duty to do something that one cannot do (Prior, 1967). If a promise is logically or physically impossible to keep, one does not have moral duty to keep it. One might also argue that one does not have a duty to keep promises that are practically impossible to keep, since one might need to violate other duties to one's own self or others to keep highly impractical promises. For example, suppose that I promise to paint my neighbor's fence on Saturday but it rains all day. One might argue that I do not have a duty to paint his fence on that day because it is highly impractical to paint a fence while it is raining. If I paint the fence when it is raining, I will do a poor job of painting it, and I may need to paint it again. I would therefore break my implicit promises to do a good job of painting the fence by painting it in the rain. Even though one does not have an obligation to keep promises that one cannot keep, one can still have a moral duty to avoid making promises that one knows that one cannot keep. Keeping a promise that one knows that one cannot keep is like making a false or invalid promise, since one cannot honestly intend to do something that one knows is impossible to do.

How might these Kantian insights apply to the patient's duties? Many bioethicists view the physician-patient relationship as a type of agreement between two parties. Different commentators use different analogies to portray this agreement. Some regard the agreement as like a contract (Veatch, 1981; Brody, 1987); others regard it as a like partnership (Quill, 1983; Emanuel & Emanuel, 1992; Meyer, 1992). The U.S. legal system holds that a physician-patient relationship is based on an explicit or implicit contract between the physician and patient (Hall, Ellman, & Strouse, 1999). All agreements, whether legally binding contracts or informal arrangements, are based on reciprocal promises between parties who are capable of giving their consent. One can view the physician-patient relationship as based on explicit or implicit promises made by the physician and the patient, which promote cooperation and trust.

The most basic promise that the physician makes to the patient is a promise to help the patient with his or her medical problem; the most basic promise that the patient makes to the physician is the promise to accept the physician's help, which implies a promise to cooperate with the physician and communicate honestly with the physician. All of the other promises made in the relationship depend on these two promises: if the doctor has not offered to help, then he or she has no relationship with the patient; if the patient has not accepted the physician's help, then he or she has no relationship with the physician. Moreover, the agreement is not set in stone. Both the physician and the patient can terminate or renegotiate the relationship, provided that they discharge the duties they have undertaken in the relationship (Meyer, 1992).

Some of the implicit or explicit promises that the physician makes to the patient include ⁸ : the promise to treat the patient with dignity and respect; the promise to inform the patient; the promise to honor the patient's right to make medical decisions; the promise to act in the patient's best interests; the promise to diagnose and treat the patient's medical problems; the promise to promote the patient's health; the promise not to abandon the patient; the promise to maintain patient confidentiality; and the promise to exhibit skill, competence, and professionalism. The patient also makes a variety of implicit or explicit promises to the doctor, including: ⁹ the promise to provide the physician with the information required to make a diagnosis, including symptoms, medical history, and social history; the promise to tell the truth to the physician and communicate honestly and openly; the promise to receive and ask for information about his or her medical condition; the promise to participate in medical decisions, if able; the promise to adhere to treatment prescriptions, if the patient accepts the treatment; the promise to come to scheduled appointments, or to give reasonable notice if one cannot make an appointment; the promise to meet financial obligations to the physician; the promise to respect the health care staff and other patients; the promise to inform the physician or medical staff of any legal documents, such as advance directives, that may be important in making medical decisions.

Even though the patient has a duty to keep his or her promises to the doctor and help to implement a treatment regimen, the patient would be morally justified in breaking his or her promises to the doctor

1. in order to fulfill a more important moral duty, or

2. if keeping the promise is impossible or highly impractical.

For example, suppose that a doctor prescribes a terminally ill patient an opioid medication for pain, which makes the patient drowsy and confused. The patient may decide to not take the medication in order to avoid these side effects and be awake and alert for conversations with family members. It would be acceptable for the patient to break this promise to the physician in order fulfill his or her duties to himself or herself and others. If a patient does not have access to a medication as a result of financial hardship, then he or she may find it impossible or highly impractical to keep his or her promise to take the medication. Even when a patient has a good reason to not follow treatment prescriptions, he or she should still tell the doctor why he or she did not take the medication as prescribed, since the patient has a duty to communicate honestly and openly with the doctor.

V. The Patient's Duties and Non-Adherence

With the possible exception of the duty not to lie, each of the patient's moral duties addressed in the previous section implies a duty to adhere to medical recommendations and advice. First, the patient's duties to himself or herself imply that the patient should seek medical advice and help, when needed, and adhere to the doctor's prescriptions. A patient who does not follow treatment prescriptions places his or her own health at risk and does not fulfill his or her obligation to take care of himself or herself. Even though patients have a duty to follow medical prescriptions, this duty is not a perfect one. They can still place their health at risk to honor other moral obligations. It is rarely the case that a patient must not follow treatment prescriptions in order to honor competing moral duties.

However, sometimes failing to heed medical advice may honor a competing moral obligation. For instance, suppose that a man becomes impotent when he takes his blood pressure medication, and that his sexual dysfunction is having a negative impact on his sense of self-worth and his relationship with his spouse. Suppose that the man has decided to address this problem by taking his medication regularly, except on those days that he plans to have sex. One might argue that the man has a good reason to not adhere the doctor's prescription, because of the medication's negative impact on his self-image and his relationship with his wife. Even so, he should still discuss the side effects with his doctor, since it may be possible to prescribe a medication that does not have these undesirable side effects. Since patients have a duty to communicate honestly and openly with their doctors, they have an obligation to talk to their doctors about problems they may be having with implementing a therapy.

Second, the patient's duties to not place others at serious risk of harm imply that the patient should adhere to treatment prescriptions that are required to prevent harm to others. For example, a patient with tuberculosis who does not take his or medications as directed places himself or herself as well as other people at risk because he or she could spread the disease. A patient with HIV who does not takes his antiretroviral medications consistently places himself or herself as well as other people at risk because he/she could develop a drug-resistant form of HIV. In situations like these, the patient does not have a good reason, such as a duty to himself or herself or others, for failing to adhere to treatment prescriptions. Patients with diseases that pose a threat to public health have an obligation to follow prescribed treatment. Society may take appropriate measures, such as forced treatment or quarantine (in the case of tuberculosis), or denial of medications (in the case of HIV), in order to protect people from the risks caused by non-adherence (Senak, 1997; Jacobson, 2000). ¹⁰

Although forced treatment conflicts with the patient's right to refuse treatment, and denying medications to a patient undermines access to care, there is a sound ethical basis for both of these policies. According to the harm principle, the government may interfere with the rights of the individual in order to prevent that person from harming other people (Feinberg, 1973). Although the harm principle is usually associated with John Stuart Mill, Kantian ethics also provides a moral foundation for the principle. Consider the maxim, “If I pose a significant risk of harm to other people, society may not take appropriate steps to prevent me from harming others.” A rational agent would not choose this as a rule for all people, because he or she would not want to be placed at a significant risk by someone else, and he or she would want society to be able to take appropriate steps to prevent that risk. The opposite maxim is a rule that a rational agent could and would adopt as a rule for all people.

Third, let's consider the patient's (or his or her representative's) promises to the doctor as a source of moral duties to adhere to treatment prescriptions. As noted earlier, the physician-patient relationship is a type of agreement where the parties (or their representatives) make implicit or explicit promises to each another. One of the promises that the patient makes to the doctor is that he or she will adhere to the doctor's prescriptions, if he or she accepts the medical care. When a patient accepts medical therapy from a doctor, he or she makes a commitment to help implement that therapy. If a doctor prescribes blood pressure medication, and the patient agrees to take the medication, understands how to take the medication, and can afford the medication, then the doctor may legitimately expect that the patient will take the medication as prescribed. This expectation is not paternalistic, domineering, or arrogant; it is simply what partners may expect of each other.

Although the patient has a moral duty to keep a promise to adhere to treatment prescriptions, provided that he or she has accepted the treatment, there are some exceptions to this general rule. First, a patient may break a promise to the doctor in order to honor a more important moral obligation to himself or herself or other people. For example, if a patient is taking a medication that makes him or her drowsy, and he or she needs to drive a vehicle, then he or she may refrain from taking the medication in order to avoid harming himself or herself or other people. Second, a patient may break a promise to the doctor if it is impossible (or impractical) for him or her to keep the promise. For example, if a patient cannot afford a medication, he or she is not required to take it. In both of these exceptions, the patient's duty to communicate honestly and openly implies that he or she should tell the doctor why he or she is having some difficulties with implementing the treatment recommendations. Telling the truth to the doctor will also allow him or her to understand and better address the reasons for non-adherence.

VI. Using Written Agreements to Improve Patient Adherence

This article will now explore briefly a promising strategy for promoting patient adherence in some situations: using written agreement to promote patient adherence. Written agreements are based on the notion that the relationship between doctor and patient is like a partnership. Partners cooperate together to achieve a common goal and trust each other. In the case of physicians and patients, that common goal is promoting the patient's health, dignity, and welfare. To achieve this common goal, partners make promises to each other. These promises also establish rights, responsibilities and expectations within the partnership. It is often useful to set forth these promises, rights, responsibilities, and expectations in a written agreement. A written agreement can also establish procedures for modifying or rescinding the agreement, as well as the consequences that may follow from failing to uphold one's end of the agreement.

Many doctors have used written agreements with patients to deal with a variety of issues in the doctor-patient relationship, including non-adherence. Pain management specialists have had some success in using written agreements to promote adherence to medication regimens in treating patients with chronic pain (Fishman, Mahajan, Jung, & Wilsey, 2002). The appendix at the end of this article includes a sample agreement for using opioid analgesics to treat chronic pain (see Appendix A).

A written agreement can address the problem of non-adherence by encouraging the patient to take his or her responsibilities seriously. The agreement treats both parties as autonomous, moral agents who are free to make promises and agreements. Both parties can freely enter or exit the agreement, and both parties assume the duties and responsibilities stated in the agreement. Neither party can, in good faith, deny these responsibilities or evade them. Since both parties can be held responsible for their failure to adhere to the terms of the agreement, they may be more inclined to honor their duties and responsibilities than they would in the absence of a written agreement. Signing a written agreement with a physician may encourage a patient to think twice before making a conscious choice to go against treatment prescriptions that he or she has already accepted.

A written agreement can encourage patient adherence by spelling out possible consequences of non-adherence, which might include termination of the treatment regimen, termination of the physician-patient relationship, or forced treatment, depending on the circumstances. Although these consequences may sound harsh and even uncaring, they can be justified on the grounds that they are fair. First, non-adherence can affect many people other than the patient. Non-adherence with HIV or tuberculosis medications can pose a public health threat. Non-adherence with instructions for taking opioid analgesics, such as asking for early prescription refills or selling medications to third parties, can pose significant legal liability issues for physicians. Non-adherence with immune-suppressing drugs following an organ transplant can waste valuable resources. In general, non-adherence can also cause considerable frustration for the physician and can harm the patient.

Second, the consequences that a patient might face for non-adherence are no different, in principle, from the consequences that any person might face who fails to abide by an agreement. If you fail to pay the mortgage payments on your house, then a bank can foreclose on your mortgage and evict you from your house. If you fail to adhere to an honor code that you accept when you enroll in a university, you can be expelled. If someone is capable of making an agreement, then he or she should also be prepared to faces the consequences of failing to adhere to the agreement.

In addition to helping deal with non-adherence, a written agreement can have other beneficial impacts on the physician-patient relationship. First, since a written agreement can spell out the patient's responsibilities, as well as any specific warnings or instructions, the agreement may itself serve as a basis for patient education and additional communication and negotiation between doctor and patient. Second, since the agreement may include provisions for non-adherence, such as terminating the physician-patient relationship, a written agreement may make physicians more inclined to help patients with documented problems with adherence, such as chronic pain patients, since the agreement would give them a legitimate way to stop prescribing opioid analgesics to a patient who abuses them.

A written agreement between a physician and a patient may or may not be recognized as a legally binding contact. Not all written agreements are legally binding contracts. A written agreement will not be recognized as a legally binding if one of the parties is not capable of consenting to the contract; if the agreement involves fraud, deception, duress, or undue influence; if the agreement does not require one party to change his or her legal rights, or if agreement is deemed unconscionable (Calamari & Perillo, 1998). Conversely, a valid contract need not be spelled out in a written agreement. Many types of informal agreements are recognized as legally binding (Calarmari & Perillo, 1998). Indeed, a legally valid contract between a physician and a patient can be established when the patient accepts the physician's offer of help, or the physician accepts the patient's request for services (Hall, Ellman, & Strouse, 1999).

Whether a court will hold that a particular written (or unwritten) agreement is a legally binding contract depends on the legal principles at issue as well as the facts and circumstances of the case. Since a court of law may, under the right conditions, recognize a written agreement between a doctor and a patient as a legally binding contract, physicians and patients should understand that their agreements could have legal consequences. Physicians should seek legal advice before signing written agreements with patients, and they should make such advice available to patients.

VII. Objections and Replies

Before concluding, it will be useful to respond to some objections to my views.

Objection: The use of written agreements is too legalistic. It will encourage suspicion, distrust, and litigation. Many patients will not want to sign these agreements or will find them uncaring and offensive.

Reply: I agree that written agreements are more legalistic than unwritten ones. However, written agreements may not encourage suspicion, distrust, or litigation provided that

1. they are written in language that is clear and easy to understand; and

2. the doctor uses tact, care, and compassion in explaining the benefits of the written agreement to the patient.

A doctor could explain to the patient that he or she would like to help the patient with his or her medical problem(s), but that he or she has some concerns related to adherence to prescribed treatment that can be best addressed by a written agreement. If used appropriately, a written agreement could encourage trust and understanding between doctor and patient, and minimize litigation. Additionally, many patients are already very familiar with all of the paperwork they already must sign to receive health care, such as insurance releases, privacy notifications, informed consent documents, and so on. Many patients will not be offended by having to sign one more piece of paper.

Objection: Mentally incapacitated patients cannot enter written agreements. Many patients lack the ability to reach an agreement due to mental illness, mental disability, emotional disturbances, the effects of medications or physical ailments, and lack of sufficient cognitive development. Incompetent (or mentally incapacitated) cannot sign these agreements.

Reply: This is a common issue in healthcare, which is not unique to written agreements. Indeed, mentally incapacitated patients will have problems with all types of decision-making in medicine. Physicians should carefully assess the mental capacity of each patient to determine whether he/she is capable of making medical decisions or signing a written agreement. If the patient is mentally incapacitated, then the physician may be able to reach an agreement with the patient's legal representative. In some cases, the patient and his or her representative could sign the agreement. For example, a physician might develop a written agreement that is signed by a 14-year-old diabetic patient and his or her parents.

Objection: The Kantian view holds that patient's duties are based on his/her autonomy, but many patients lack autonomy, due to mental incapacity, and so on. Patients who are not autonomous have no moral duties.

Reply: This is indeed a weakness of the Kantian view: it does not apply to people who are not moral agents. However, almost all moral theories that attempt to formulate moral duties suffer from the same problem. Someone who cannot comprehend moral duties does not have moral duties. Even though mentally incapacitated patients do not have moral duties, their representatives may have moral duties. If a patient's representative may exercise the patient's rights, then the guardian or representative is also responsible for fulfilling the patient's duties that the patient would have, if the patient could fulfill them. For example, even though a child does not have a moral duty to take his or her medication as directed, the child's parents have a moral duty to ensure that the medication is administered to the child as directed.

Objection: Any agreement between physician and patient, written or unwritten, is limited by the ethical standards of medicine. For example, a physician and a patient cannot enter into an agreement in which the doctor promises to help a health patient commit suicide or prescribe opioids inappropriately. Therefore, the physician-patient relationship is based on the standards of medicine, not on any agreement between doctor and patient.

Reply: The fact that the agreement between physician and patient is limited by the ethical standards of medicine does not prove that there can be no agreement. The objection only proves that the agreement itself cannot serve as the sole ethical basis for the relationship for the physician and patient. Any agreement between a professional and a client can be limited by the standards of the profession. For example, a banker cannot enter into an agreement in which he promises to launder money for a client. Even though the banker's duties are limited by the nature of his/her profession, we would still say that the banker can make agreements or contracts with his/her clients.

Objection: The Kantian approach assumes that patients (or their legal representatives) can make free choices, but freedom of the will is an illusion. People do not actually make free choices. Thus, they do not freely choose to obey moral rules or enter into agreements.

Reply: This is an objection to the possibility of moral duties for anyone in society. This article will not attempt to defend a solution to the age-old philosophical problem of free will. ¹¹ It will assume that adult human beings who are not mentally disabled can make free choices and act freely. Free choices are decisions that are caused by a person's internal, cognitive processes, such as judgment, reflection and deliberation. Un-free choices are those that are caused by external influences, such as coercion, manipulation, fraud, or undue influence or overwhelming internal compulsions, such as addiction or emotional distress.

Many patients are vulnerable as a result of their physical or mental ailments, or their economic or social status. However, vulnerable people can still make free choices if they have adequate mental capacity. The mere fact that a patient (or his or her representative) is vulnerable does not prove that he or she cannot make a free choice. Since many patients are vulnerable, doctors have a special duty to ensure that they do not take advantage of patients. Doctors should not use their power and influence to induce patients to make particular choices, nor should they manipulate the information they present to the patient in order to induce patients to make particular choices. The patient's vulnerabilities place ethical constraints on doctors, but they do not vitiate the possibility of an agreement between the doctor and the patient (Meyer, 1992).

VIII. Conclusion

Patients have a variety of moral duties, including the duty to develop their talents and abilities, to tell the truth, to keep promises, and to avoid placing other people at significant risk of harm. These duties imply that patients should maintain their health and seek medical help, when necessary. When they seek medical help, patients should tell the truth to the physician, keep the promises they make to the physician, and avoid transmitting dangerous diseases to other people. Once a patient accepts medical treatment, he or she has a prima facie moral duty to adhere to the doctor's treatment prescriptions. Since the patient still retains the right to refuse treatment, he or she may deviate from treatment prescriptions in order to honor conflicting duties to himself or herself or other people, or if it is impossible (or impractical) to adhere to the doctor's prescriptions.

However, patients should still inform their doctors when they have decided to not implement therapy as directed. The patient does not have the right to refuse medical treatment that is required to prevent him or her from posing a significant threat to other people. If the patient's failure to adhere to treatment recommendations makes him or her a significant threat to the health and safety of other people, then the government is morally justified in taking appropriate steps to reduce the threat that the patient poses to the public, including forced medical treatment, quarantine, or denial of medications. Physicians should consider using written agreements with patients as a way of promoting adherence as well as of enhancing the doctor-patient relationship. Physicians who deal with diseases where non-adherence is a common problem, such as chronic pain or diabetes, have pioneered the use of written agreements. However, all physicians should consider using these agreements to deal with adherence problems related to other diseases.

Notes

I follow the means/ends conception of rationality: a “rational” person is someone who chooses effective means to his or her goals; an irrational person someone who does not. See Audi (2001).

This search was conducted using MEDLINE's advanced function to search for phrases in abstracts. MEDLINE was accessed at: www.biomednet.com on August 15, 2003.

This is a controversial issue. Kant believed that his three versions of the CI were equivalent, but some writers have argued that they are not. See O'Neill (1975).

Meyer (1992) also developed a Kantian approach to patient's duties. He argued that the patient's moral duties are based on the patient's moral autonomy. The AMA's opinion on patient responsibilities has a Kantian theme even though it does not mention Kant: “Like patients’ rights, patients’ responsibilities are derived from the principle of autonomy…Autonomous, competent patients assert some control over the decisions which direct their health care. With that exercise of self-governance and free choice comes a number of responsibilities” (AMA, 2001b).

For further discussion, see Humber and Almeder (1997), Callahan (2000).

For further discussion, see Feinberg (1973) and Beauchamp and Childress (2001).

For further discussion, see Bok (1978).

The list is based on the AMA's Principles of Medical Ethics (2001a).

The list is based on the AMA's Opinion E 10.0 2 (2001b) and the AHA's Patient's Bill of Rights (1992).

If a patient is not adhering to antiretroviral drug regimens and is not likely to do so in the future, the safest course of action is to stop taking antiretroviral medications completely, since continuing to take the medications inconsistently will only increase the chance of developing a drug-resistant strain of HIV. If a doctor decides to stop prescribing antiretroviral drugs to a patient, he would still recommend other forms of therapy appropriate for treatment of the other diseases and symptoms associated with HIV, such as yeast infections, pneumonia, and tuberculosis. See Singh (1996). Although the policy of denying patients antiretroviral medications would appear to violate the physician's duty to treat the patient, it does not, since continued treatment with these medications is likely to harm, not benefit, the patient. See also Orentlicher (1991).

For further discussion, see Dennett (1984).

Appendix A

Sample Written Agreement Between Physician and Patient for the Use of Opioid Analgesics to Treat Chronic Pain

Agreement

I. Purpose

The purpose of the agreement between __________(physician) and ________(patient) is to define the rights, responsibilities, and expectations of the physician and the patient concerning the treatment of the patient's chronic pain condition, using opioid analgesics. The physician has already assessed the patient and determined that the use of opioid analgesics is a medically appropriate method of treating the patient's pain, provided that the patient adheres to the terms and conditions of this agreement.

II. Physician's Duties

1. The physician will prescribe opioid medications to the patient for the treatment of chronic pain according to the standard of care for chronic pain management.

2. The physician will monitor the patient's medical care and periodically reassess the patient's condition as well as the need for opioid medications or other forms of treatment.

3. The physician will use his or her experience, skill and care for the good of the patient.

4. The physician will provide the patient with the information that he or she requires to make decisions relating to his or her chronic pain condition and to understand the nature of his or her condition and its treatment. This information will include the benefits and risks of treatment with opioids as well as alternative forms of treatment. Physical or psychological dependence on opioids, as well as addiction, are three significant risks of using opioids to treat chronic pain. The risk of addiction is much higher if the patient has a previous history of alcohol or substance abuse. Suddenly stopping the use of opioid medications may lead to opioid withdrawal, which is an uncomfortable but not a physically life-threatening process.

5. The physician will respect the patient's rights to autonomy, dignity, and privacy.

III. Patient's Duties

1. The patient will take his or her medication(s) as directed by the physician. The patient will not seek to have his or her prescription refilled before the physician determines that it is time for refill. The patient will not use any drugs (legal or illegal) to treat his or pain condition, other than those that have been prescribed by the physician. The patient will not consume alcohol or other mood altering drugs, except nicotine or caffeine, unless the physician gives his or her permission. The patient will not give or sell his or her medication(s) to other people or allow them to be stolen or misplaced. The patient will store his or her medication(s) in a safe place. Medications that are lost or stolen will not be replaced until it is time for a refill.

2. The patient will not seek a treatment for this condition from any other physician or seek opioid analgesic medications or mood altering medications from any other physician, unless permitted to do so by the physician. If the patient requires emergency medical care for pain, he or she will inform the physician.

3. The patient will obtain opioid medications only from:_______________________________(list up to three pharmacies).

4. The patient will meet all of his or her scheduled appointments with the physician.

5. The patient will honestly and openly communicate all information to the physician that is required to assess and treat the patient's condition, including past medical history, social history, any changes in symptoms or overall health, and any difficulties that arise in taking his or her medication as prescribed or meeting the terms of this agreement.

IV. Changes in this Agreement

1. The physician and patient may renegotiate or change this agreement at any time, provided that the party seeking to change the agreement gives the other at least 5 days notice and both parties consent to the change(s).

V. Consequences for Failing to Adhere to this Agreement

1. If the physician does not adhere to the terms of this agreement, the patient may terminate his or her relationship with the physician.

2. If the patient does not adhere to the terms of this agreement, the physician may stop prescribing analgesic medications to the patient. He or she may also terminate his or her relationship with the patient, provided that he or she gives the patient reasonable notice and an opportunity to receive medical care from another physician.

Signed this at_____________(location) on __________(date)

_________________(physician)

_________________(patient)

_________________(witness)