Ethical issues of compliance/adherence in the treatment of hypertension

Definitions and philosophy 

The most common definition of compliance in medical literature is the extent to which behavior follows medical advice.24 Because “compliance” is now the term used to define physician documentation for billing, “adherence” is replacing “compliance” in contemporary language. In this review, both terms are used. There is an ethical conflict within the definition of compliance. Evidence-based medicine provides pathways to clarify decisions for patient care, but these pathways rarely consider the patient’s perspective. Is it ethical to expect compliance and life-style changes without considering the impact on a person’s life? While ethics usually refers to what is good, bad, right, or wrong, one meaning of the word, derived from the Greek thikos, means custom or character. Therefore, if a person is accustomed to a certain practice, is it ethical to insist on a change in that person’s custom or character without consulting him or her? As patients have gained authority, physicians are less likely to lay down the law of treatment and more likely to try to sell the plan to the patient.25 A common belief about compliance/adherence is that the answer to noncompliance lies in patient education. However, what of the patient who knows the facts and decides the risks are not worth the treatment? Also, what of the patient who feels rejected—perhaps someone in the care of a very authoritarian physician—and so agrees with the therapy, despite contradictory personal preferences? Moreover, what seems like an unreasonable demand in treatment based on medical knowledge today might be contradicted by medical information in the future. One need only look to the current controversy over hormone replacement therapy to see that “today’s fact can turn out to be tomorrow’s error.”25

The physician-patient relationship and medical decision-making 

Medical interventions are generally thought to be justified when their benefits outweigh their risks and burdens. In a study of the ethics of interventions for mild hypertension, there is detailed discussion about what is the best treatment, particularly when the immediate threat is low and the intervention may result in long-term morbidity.26 Noted are studies reporting a large reduction in the incidence of serious outcomes, but which may indeed only benefit a very small number of patients. Thus, interventions for common medical problems, such as mild hypertension, may be beneficial for the community at large but are far less likely to benefit each person. Clinicians make decisions for a particular patient based on the medical literature that addresses outcomes for a population, not for an individual patient. Moreover, most studies do not report on an individual’s personal qualities that may affect the positive or negative effects of the intervention. Patient preferences are often not part of this decision-making process. The physician often develops the treatment plan independently and then attempts to make the therapy acceptable to the patient to achieve compliance. Some patients may find that daily medication reinforces their fear of illness; others may find it gives a sense of protection. One patient may have serious concerns about changes in sexual function, whereas another may find it unimportant. Any attempt to prolong life may be highly valuable to some, whereas others place more value on their quality of life. Clearly, an individual’s culture, perspective, wishes, and goals will affect his or her adherence with a medical treatment plan. Patients need to share in these decisions. The best decisions in medicine occur when the physician and patient each contribute to the process.27 Even the marginally competent patient can have a partial role in decision making if the information and possibly the approach are modified.28

A review of the key phrase, “obeying doctor’s orders,” concludes that the ideal patient is a passive and obedient recipient.29 In most studies of noncompliance, it is assumed that patients should comply, implying that the physician is the only one skilled enough to make the right decision and that patient default is irrational. This idea spawns a research approach that is unlikely to consider the social setting and meaning of the behavior of noncompliance. Patients have their own ideas about the use of medications, rationalizing that they do not like to become dependent on a therapy or believing medications are only good when one feels ill.29 The patient’s views of a physician are influenced by their own culture and experience. If the physician does not meet their expectations, regardless of how unreasonable these views are, medical instructions may not be accepted as the absolute truths. In the view of personal and cultural factors, it is suggested that almost everyone chooses not to follow medical advice at some time.

Some believe the doctor-patient relationship explains nonadherence with medical therapies.30 An authoritarian patient with a passive physician, or vice versa, results in tense interactions that foster noncompliance. If we view compliance from a medical perspective, noncompliance is a type of deviation in need of explanation.31 This idea centers on the benevolent authority of the physician and consenting acceptance of the patient, a concept that diminishes patient autonomy. From a medical perspective, it is “a problem in search of a remedy.” Few studies of compliance ask a patient why they did not take their medications. In a study of compliance with epilepsy medication, noncompliant patients were asked why they did not take their prescribed medications.32 They reported their noncompliance as a form of self-regulation and did not consider it to be noncompliance. For those with chronic illnesses, medical regimens are constant reminders of their illness, and many strive to avoid the constant patient role and develop their own medical habits. Some adjust medications to meet social needs, such as reducing the stigma associated with taking a medication during work hours. Some adjust medication times to avoid unpleasant side effects. A study of compliance with arthritis medications found patients tested the efficacy of prescribed drugs by stopping them briefly to evaluate their symptoms.33 Self-regulation of medication may give a person a sense of control over their illness.

What about the issue of trust between a patient and the physician or nurse? The goal of a trusting relationship is to put each other at ease. This is not easily done, because the physician-patient or nurse-patient relationship is inherently asymmetrically weighted toward the health provider. A philosophical discussion asks whether patients have the right to expect physicians to understand their background and experiences, and whether physicians are supposed to learn more about their patients.34 It is suggested that physicians have both the resources to inspire trust and the obligation to put their patients at ease but that it requires more than just good will. In part, this has to do with a patient’s belief that he or she is taken seriously and of the physician or nurse actually listening rather than just waiting for the patient to stop talking to get on with the assessment. Nonverbal cues may convey more than what is actually said. Interviews with rural southern African-American patients about noncompliance found many lacked understanding about their medications and had high levels of noncompliance that may reflect cultural differences between patient and physician.35

Hypertension and compliance/adherence 

Numerous studies have focused on blood pressure control and compliance.1, 2, 25, 36, 37, 38, 39 The next section summarizes the findings of some of these studies. A large multicenter, randomized, double-blind trial of men with 3 different antihypertensive medication regimens examined the effects of medication upon quality of life over 24 weeks.36 Thirteen percent of the patients withdrew from the study because of unacceptable adverse drug reactions. Because monthly pill counts were done, the authors suggest early withdrawal was an index of noncompliance. Withdrawal was 40% to 60% lower in the group reporting the fewest side effects during the first 8 weeks of the study. This study did not include an opportunity to change medications to avoid side effects and as with most studies, random assignment to treatments did not allow for patient preferences, likely increasing the risk of noncompliance. As noted in a study of the problems of uncontrolled blood pressure, physicians commonly believe that patient noncompliance with prescribed antihypertensive therapy is a primary issue in hypertension control.25 Hypertension is more likely to be controlled with the right drug or combination of drugs that the patient tolerates, finds convenient, and finds effective.37 As discussed above, patient participation in treatment plans has rarely been examined but is integral to the success of any hypertension management plan.

The effects of noncompliance in the form of missed or late doses on blood pressure control was examined in a study of 169 mildly hypertensive adults prescribed a single antihypertensive medication.1 Electronic monitoring vials were used for 3 months. Although 23% of participants missed a dose in the 2 weeks before the blood pressure check, neither diastolic nor systolic blood pressure was predicted by adherence to medication. It was concluded that dosing regimens using 1 dose/24 hours that maintained at least 50% peak effect at the end of 24 hours may protect patients from episodic nonadherence. However, because patients may be even more casual about missed doses if they knew this, sharing this information with patients may actually foster nonadherence. Should information shared with patients be restricted by the physician for potential negative effects or is patient autonomy more important? The former position contradicts the assumption that patients should be active and informed partners in their care.

An investigation of patient perspectives of compliance used a questionnaire for 138 hypertensive patients in Finland.38 Self-reported compliance with diet, medication, exercise, and follow-up visits were evaluated. Compliance with medications was better than compliance with lifestyle changes. This suggests that patients may feel medication will manage their hypertension so they do not feel the need to change their lifestyle.

Fifty-two patients prescribed antihypertensive drugs in the past year were interviewed to determine differences in drug treatment plans using decision analysis versus patient preference.39 Patients were asked to evaluate their preference for the state of their health from zero (equal to death) to 1 (equal to perfect health). After this, each patient was shown the decision tree analysis of outcomes for medication or no medication and ranked their preferences. Thus, the patient was asked to make a hypothetical gamble based on their preference for their state of health. Forty-four percent of patients preferred not to take antihypertensive medication. Those who had less than a 10% 5-year risk of a cardiovascular event were slightly more likely to decline medication (62%) than those with a 10% or greater risk (50%).39 Medication compliance was also measured by prescription refills and defined as 80% or more of prescribed medication. Interestingly, there was no relationship between the outcome of decision analysis and a patient’s past adherence to medication. Other barriers to compliance, such as drug side effects, were not evaluated.

A recent review of studies from 1967 to 2001 of interventions to enhance patient adherence to medications found that the studies were too disparate to allow meta-analysis.23 However, 36 interventions were evaluated; most were complex and labor intensive. Only 17% resulted in significant improvement in treatment outcomes, whereas 49% resulted in significant improvement in medication adherence. There was no pattern of types or numbers of interventions that predicted success. The authors conclude that there is little evidence that medication compliance can be consistently improved with current interventions.

Compliance in end-stage renal disease 

Noncompliance with oral medications for 135 hemodialysis patients using a microelectronic monitoring device found 52% were repeatedly noncompliant with antihypertensives and 70% were noncompliant with phosphate binders.3 Interestingly, self-report of noncompliance was significantly lower at 12% and 8%, respectively. Interviews of patients in another study of self-reported compliance showed that patients intended to comply, believed they were compliant, and could not remember specific incidences of noncompliance.40

A study conducted in 1972 with 43 chronic hemodialysis patients revealed that noncompliance with diet was associated with persons with low frustration tolerance, little ability to delay gratification, and high gains from the sick role.4 It is suggested that the abuse of diet by dialysis patients is an attempt to adjust to the stress of dialysis. They are, in other words, too sick to comply, and the steady abuse of their diet ensures their sick role. Most of the patients, regardless of their compliance with diet, complained bitterly about their thirst, which may indicate their frustration with harsh restrictions or perhaps their unhappiness in general.

An interesting strategy was studied in a small group of chronically noncompliant hemodialysis patients.5 The goal was to improve compliance with antihypertensive medication by having the dialysis nurse give the medication on dialysis days. While dosing 3 times a week was less than optimal, it was better than the total doses most of these patients admitted to taking. Alternatively, patients who only attended dialysis once a week rather than their prescribed 3 times a week had a transdermal patch applied at each dialysis session. The result of these interventions was significant improvement in blood pressure control and high patient acceptance of the regimen. Although this is not coerced compliance in the usual sense, it is a variation of the same theme and suggests an alternative for those who are grossly noncompliant and who even admit that they should never be expected to be compliant. Patients did agree to the regimen although they did not share in the decision-making.

Is patient education the answer to noncompliance or part of the problem? Ethical issues of patient education were studied in the context of hypertensive control.41 Educators often are concerned about their right to impose values on a patient. The authoritarian teacher is at one extreme of the spectrum, offering little flexibility for patients and soliciting minimal patient input. At the other extreme is a totally passive teacher who simply presents the information and leaves all decision making to the patient. It is suggested that educators combine 2 major concepts of teaching, contractual approaches and clarification of the patient’s role. This allows the educator to repeatedly evaluate the patient, and then alternatively involve the patient, to help determine how much pressure should be placed on the patient at a given time. Finally, an editorial exploring whether anything works to improve compliance in dialysis patients notes that few interventions have been rigorously tested.6 After describing the findings of various studies since the 1970s, it is suggested that the best method to improve compliance may be to simply pay more attention to noncompliant patients while modifying treatments in a partnership setting with patients.

Conclusions 

Recognizing the individuality and autonomy of patients appears to be a common approach to medical treatment decisions. The attitudes and interpersonal abilities of physicians and nurses towards patients probably play the biggest role in the struggle for improved compliance with medical therapies. Accepting patient behavior in our relationships with patients may also be needed. Would it help to adopt Fred Rogers’ theme, “I like you just the way you are,” to improve relationships with patients who would then be more likely to comply?

We may become more enlightened in future research of compliance if studies are designed to include patient preference in a partnership with physicians. This would represent the ideal method of determining the best treatment for patients and perhaps, at last, improve compliance outcomes.