When Adolescents "Mismanage" Their Chronic Medical Conditions: An Ethical Exploration

Insoo Hyun *

Abstract. Many adolescent patients with chronic medical conditions do not manage their illnesses very closely and often put themselves at risk for serious health complications. Setting aside cases of nonadherence that are due to practical difficulties involving the implementation of a management plan, a deeply problematic question remains. How should health care providers respond to adolescent patients who express a conscious and value-driven decision to pursue other goals and interests that are incompatible with their doctors' recommended directives? Using two guiding ethical principles, the "relevant difference principle" and the "principle of noninterference," as well as available empirical data on adolescent decision making and risk perception, the paper concludes that most adolescents ages 14 and older should be allowed to make self-determining decisions regarding the management of their chronic medical conditions.

How should health care providers respond when adolescent patients with chronic medical conditions fail to comply with their recommended management plans? What ethical considerations ought to guide the reactions of professionals entrusted with the medical care of this population? Before I address these questions specifically, I think it is helpful to consider briefly the defensible policies for similar cases involving adults.

Many adult patients with chronic medical conditions, such as type I (insulin-dependent) diabetes and hypertension, willingly accept the recommendations of their health care providers and carefully adhere to life-long plans of treatment. Others, however, do not manage their conditions so closely and often, to the dismay of their physicians, put themselves at risk for serious health complications. Such patients are of particular concern [End Page 147] to health care professionals, since some complications resulting from poorly-managed chronic conditions exhibit a delayed time of actualization and can be very difficult or even impossible to reverse. As a case in point, some insulin-dependent diabetics who do not keep their condition under tight control may later suffer renal failure and/or blindness. Due to the serious potential consequences of this and other inadequately-controlled chronic conditions, there exist good prima facie reasons for helping such patients to understand the dangers of their present behavior and for assisting them in adhering to an effective treatment plan. ¹

Of course, how one should respond to any particular case of nonadherence is contingent on the specific reasons behind that patient's failure to control his or her chronic medical condition effectively. As health care providers well know, there can be a multitude of possible explanations in this regard, ranging anywhere from unintended patient error, to diagnosable depression, to just plain weakness of will. Yet, as long as patients and their providers share the same health-related values and goals for treatment, different strategies can be explored to target and overcome these obstacles. In short, the task in these cases is usually a practical one. ²

In other cases, however, patients' nonadherence can be both free and willful. Many adult patients make conscious and value-driven decisions to pursue goals and interests that are incompatible with their doctors' expressed directives. For example, a patient might offer unique personal reasons (including cultural or religious motives) for not adhering to a recommended plan of treatment. Such a decision would fall under the aegis of personal autonomy provided it is informed and carefully considered, rooted in that patient's system of values, and not likely to cause significant harm to others. The morally appropriate response for health care professionals in these cases, I submit, is to respect the patient's wishes and retreat to a policy of nonintervention, assuming the foregoing conditions have been satisfied and all legitimate attempts to persuade the patient rationally have failed. ³ Although some observers may feel uneasy with this particular approach, it is consistent with the moral view that mentally competent persons should be allowed to formulate and pursue freely their own conceptions of the good, even if doing so will place their physical well-being at risk (Brock 1991). Both in the clinic and in everyday life, adults have a recognized fundamental right to lead their lives as they see fit, regardless of whether others agree with their particular choices.

But what if the noncompliant patient in question is an adolescent? Are there ethically compelling reasons for handling an adolescent's "mismanagement" [End Page 148] of his or her chronic medical condition any differently than an adult's? The answer is not immediately obvious. Although some situations arise in which health care professionals clearly ought to employ the same approach for both adults and adolescents, there may be definite limits to how similarly these two groups of patients should be managed. Surely in cases where an adolescent's nonadherence is attributable to practical difficulties that impede his or her ability to carry forth the treatment plan (such as a chaotic home environment or difficulties adjusting to a new treatment schedule), the correct response would be to help the patient overcome these obstacles. As mentioned earlier, patients and physicians who share the same basic goals and purposes of treatment should focus their attention on the former's specific barriers to compliance. ⁴ Leaving aside all practical cases of patient nonadherence, however, a separate issue remains of how providers should respond when adolescents, like their adult counterparts, appeal to personal and value-driven reasons for not adhering to their recommended plans of treatment. Should health care professionals apply the same standards and policies to both groups or might they be justified in exercising a measure of paternalism in the case of adolescents that would be deemed unacceptable in comparable situations involving competent adults? ⁵ In other words, how alike should the approaches be when dealing with instances of adolescent and adult nonadherence of the latter sort?

Initially, one may be tempted to view this subject as an open and shut case. That is, one might maintain that adolescents simply are not adults and therefore should not be treated as such. But what can be said in support of this speedy conclusion? A careful examination of the issue should begin with the guiding ethical principle that any unequal treatment between two given groups or parties must be justified on the basis of some clear and morally significant difference between them. For convenience, I will call this the relevant difference principle.

In the following sections, I consider exactly what these differences may be with regard to adults and adolescents and examine whether they are sufficient to justify the adoption of a paternalistic approach toward the latter. It bears explicit mention here that, as a philosopher, I am concerned mainly with the ethical (and not legal) concerns surrounding adolescents' noncompliance. Also, I am examining the problem as it relates to health care professionals only and not as it pertains to family members and others who may stand in a more intimate relationship with the adolescents in question. Whether there are any separate and unique moral [End Page 149] considerations facing this second group is an issue worth investigating on another occasion. Finally, I am assuming for the purposes of this paper that the adolescent's parents or guardians have authorized the physician to provide medical treatment for their dependent patient and to intervene paternalistically if the adolescent is incapable of self-managing his chronic condition. Having made these preliminary remarks, I begin by discussing briefly some popular attitudes toward adolescence.

The Ambiguities of Adolescence

People's general view of adolescence in American society seems to have remained relatively unchanged since G. Stanley Hall first published his comprehensive and ambitious treatise on the subject in 1904. In contemplating the gradual transition from childhood to adulthood, Hall eloquently described adolescence as a period of Sturm und Drang (storm and stress). Although some later theorists have disputed Hall's thesis that adolescence is necessarily a tumultuous journey (Mead 1953), popular contemporary attitudes toward this phase of life nevertheless tend to conform to the position that adolescents must negotiate several disparate developmental and social pressures during their metamorphosis into full-fledged adults. Often these pressures pull in opposite directions. For example, young people initiate their exit from childhood by gaining greater independence of thought and by entering new social relationships; however, like children, they typically remain under the care, supervision, and influence of their parents or guardians. Also, despite their extensive physical changes and increasing psychological maturation, they are not yet socially recognized as adults. Due to these and other tensions, one might surmise that American adolescents today face a centaurian existence, one that consists of a combination of two distinct ways of being, neither of which dominates the other.

Although such metaphors might aid speculations about the nature of adolescence, we should remain cautious of the ease with which such figurative language can slip into exaggeration. In attempting to describe vividly the experiences of modern adolescents, one may inadvertently exaggerate the difficulties corresponding to this stage of life. To complicate matters further, there is also the danger that popular attitudes involve a tendency to project what we wish to believe rather than what is actually the case. Whether adolescents typically encounter the level of stress and uncertainty commonly attributed to them is ultimately an empirical issue that I cannot settle here. [End Page 150]

Adolescents' Cognitive Capacities

Assume for the sake of argument that adolescence is a period especially replete with multiple pressures and ambiguities. Given this supposition, are there good reasons for limiting the extent to which adolescents should be allowed to make independent decisions about their own health and medical treatment? Since we are considering the question of whether adolescents with chronic medical conditions should be allowed to follow their own ideas of what, on balance, is in their overall best interests, it is appropriate to consider first the normal developmental status of adolescents' cognitive capacities. If adolescents generally exhibit a level of cognitive maturity that is significantly lower than the standard met by most adults, then perhaps a paternalistic approach to handling the former's nonadherence to treatment can be justified, at least on the basis of the relevant difference principle. ⁶ Interestingly, however, much of the well-known research in the area of adolescents' cognitive development seems to deny the antecedent of the claim above. I will present two prominent examples from the psychological literature.

First, there is the famous thesis advanced by Piaget and Inhelder (1958), who argued that adolescence marks the period where formal operational thought develops in most individuals. Unlike very young children, who are capable of thinking only in terms of the concrete and the particular, individuals who possess formal operational thought are able to reason beyond specific cases to consider the possible, the probable, and the abstract. I find this view of adolescent cognitive development of great importance to the present inquiry, since a person's formal operational thought (as conceived by Piaget and Inhelder) enables her to choose among various alternatives after imagining where each course of action might lead. Since competent medical decision making largely involves patients' abilities to understand and deliberate about the consequences of each proposed treatment alternative, adolescents who have developed formal operational structures would seem to be capable of making their own medical choices. (By contrast, young children who are not yet able to deliberate abstractly are rightly excluded from having a significant role in their own medical decision making due to their level of cognitive immaturity.)

Although not everyone wholly accepts Inhelder's and Piaget's developmental theory, the plausibility of their overall thesis was supported more recently in a study conducted by Weithorn and Campbell (1982), who discovered that 14-year-old subjects did not differ substantially from adults [End Page 151] with respect to their ability to reason about and understand their treatment alternatives when faced with different (hypothetical) medical situations. A comparison of the thought processes of 14-year-olds and legally recognized adults revealed that adolescents in this age range possess a level of competence equal to that of adults when measured along four standards of competency (evidence of choice, reasonable outcome, rational reasons, and understanding) and for four possible conditions (diabetes, epilepsy, depression, and enuresis). Weithorn and Campbell concluded their study with the observation that a denial of the right to autonomous medical decision making could not be justified simply on the basis of a presumption of adolescents' cognitive incapacity.

Empirical research of the sort just mentioned has provided crucial support for the view that adolescents 14 years and older usually have attained a level of hypothetical reasoning and understanding that is on par with that of most adults. In short, I propose that in the absence of any significant evidence to the contrary, it is reasonable to maintain that individuals in middle adolescence (14 years) typically possess the mental abilities necessary for making self- determining medical choices.

Adolescent Risk and Risk Perception

Critics might object, however, that the preceding discussion omits at least one crucial factor that may justify a paternalistic approach to adolescent patients. Specifically, adolescents may exhibit a much greater tendency than adults to under-appreciate their own vulnerability to harm. Citing adolescents' propensity to perceive themselves as invincible and impervious to danger, such critics may argue that the range of adolescent medical decision making should be limited when these young patients place themselves at risk for bodily harm. In other words, since adolescents are unable to see realistically the possibility of negative personal outcomes and consequences, they should not be allowed to make potentially hazardous decisions for themselves. Notice that the thrust of the critics claim lies in their acceptance of the "invulnerability hypothesis," which asserts that adolescents are unable to perceive risk accurately, despite the otherwise relatively mature status of their cognitive capacities.

The invulnerability hypothesis is not a novel ideal in the literature on adolescent behavior and development, and it seems to be supported, at least indirectly, by some researchers who study risk perception in adults. As Kahneman and Tversky (1973) showed nearly three decades ago, people's perception of risk depends on a variety of factors, including their [End Page 152] ability to recall and imagine negative outcomes based on their own personal experience or on how vividly the outcomes are presented to them. Proponents of the invulnerability hypothesis might utilize this finding to argue that adolescents who have had fewer life experiences with negative outcomes than adults therefore may perceive them as less likely to occur. Adolescent risk perception thus seems to be a specific relevant difference between adult and adolescent patients that would justify differences in the way these two groups are treated.

At face value, this line of argument appears quite plausible, especially when considered against the popular and persistent societal belief, which I consider in the next section, that adolescents, more than adults, tend to pursue short-term goals and interests. Despite its initial credibility, however, the overall contention of such critics is weakened significantly by the following considerations.

First, there is no compelling empirical evidence to support the claim that adolescents are so prone to see themselves as invulnerable to harm. ⁷ Although many commentators on adolescent risk taking simply accept the idea that adolescents, by nature of their developmental status, perceive themselves as invincible, there currently is no substantial proof that they truly are any more disposed to having this perception than adults (Dolcini et al. 1989; Enright, Lapsley, and Shukla 1979; Lapsley et al. 1986). In fact, according to some recent studies, adolescents aged 12 to 18 were found to be more likely than college students and other adults aged 18 to 65 to believe they could experience drug and alcohol problems, relative to others of their age and gender (Millstein 1993). Furthermore, there are numerous studies--like the research conducted by Kahneman and Tversky (1973)--that reveal that adults operate with several perceptual biases and greatly tend to underestimate their own vulnerability to negative events (Weinstein 1980; 1987). Most people believe, for instance, that they are better than average drivers, more likely than others to live past the age of 80, and less likely to be harmed by the products they use (Slovic, Fischhoff, and Lichtenstein 1980). Given the information contained in these studies, many adults may be shocked to discover the degree to which they routinely and unwittingly expose themselves to risks of various sorts. Upon closer examination of the available data on adult and adolescent risk perception, there appears to be good reason to question both whether adults should serve as the ideal standard against which to measure adolescents and whether the invulnerability hypothesis truly deserves the authoritative power often accorded it. [End Page 153]

Second, by framing the issue solely in terms of adolescents' failure to perceive risk accurately, critics gloss over a very important distinction. It is important to differentiate two salient concepts: (1) risk in the technical, probabilistic sense, which relates to the frequency of a particular negative outcome; and (2) risk tolerance, which is a subjective notion based on the priorities, values, and goals that one has set out for one's life. To illustrate, an instance of the former is the statistical probability of an individual's being the victim of an airline accident, while the latter might be specified by an individual's willingness to take up skydiving as an exciting weekend hobby. The key difference between these two concepts lies in the fact that, unlike technical risk, which can be objectively calculated, the values of potential outcomes and consequences are inherently subjective. Given this difference, a person's awareness of the frequency of some negative outcome--i.e., her knowledge of the technical risk-will not necessarily diminish her tolerance for the risk in question. If one understands the concept of risk as involving both the possibility of personal loss and the possibility of personal gain, one can better appreciate how persons can tolerate a particular risk when they judge its potential benefits to be sufficiently attractive so as to motivate the relevant behavior, even when they understand that the statistical likelihood of loss may be rather high. In short, it is clear that risk tolerance (like risk aversion) need not be affected by a greater awareness of the probabilities.

These points can greatly aid one's thinking about adolescent risk. The distinction between technical risk and risk tolerance is an important consideration, since an adolescent patient's alleged inability to perceive the dangers of her nonadherence to a treatment plan may be due less to a mistaken belief about the probabilities than to her personal level of risk tolerance. In other words, she may be motivated by her value structure to accept the risk in question even when she understands the probabilities. If this is the case, then she would be more like the adult patient who consciously pursues aims and interests that are incompatible with a recommended treatment plan than the young child who does not understand, in any meaningful way, the probabilities of certain negative outcomes. Although others may disapprove of the adolescent's values, such disapproval is a very different kind of criticism than the accusation that she does not understand the technical risk associated with her behavior. If she pursues her goals in an organized and deliberate fashion, one cannot so quickly dismiss her choices as irrational. ⁸ [End Page 154]

In summary, I conclude that the critics have failed to establish the contention that adolescents who take risks with their chronic medical conditions do not realistically comprehend the possibility of harm and therefore need paternalistic intervention. Essentially, this position rests on an unconfirmed (and likely false) hypothesis regarding adolescent risk perception and a reluctance to see alternative explanations of adolescent risk behavior. I have argued that some of these patients, like their adult counterparts, may in fact be operating with clear values and personal convictions, not magical thinking. Thus, the critics' assertion regarding adolescent risk perception does not seem to pass the test of the relevant difference principle.

Adolescents and Their Future Interests

Another popular belief contends that adolescents are worse prudential planners than adults because they generally choose more immediate gains and pleasures over their interests in the distant future. One might refer to this tendency as adolescents' bias toward the nearer future. ⁹ Would this claim, if true, provide a sufficiently relevant difference to justify a paternalistic approach to noncompliant adolescent patients?

We can begin to answer the question by categorizing chronic medical conditions into two (general) classes. Some conditions, such as asthma, are immediately symptomatic in that they provide patients with vivid and unpleasant consequences every time they stray from their management plans. (Many in this group might experience chronic discomfort in spite of their careful adherence.) Other conditions, such as type I diabetes, are not immediately symptomatic and patients may not always experience discomfort when they fail to follow their recommended treatment plans. Indeed, patients in the second group may not notice any complications from their nonadherence until the damage is manifested later, usually as severe organ failure. Having drawn this distinction, one could suggest that chronic medical conditions with hidden or delayed symptoms pose an insidious threat to adolescent patients who tend to favor immediate benefits over long-term interests. In short, one might argue that adolescents' bias toward the nearer future can have devastating long-term health effects for patients with this latter class of chronic condition. Therefore, paternalistic intervention is necessary to protect them from disaster.

I find this argument invalid the following reasons. First, the empirical claim that adolescents as a group are more biased than adults toward the nearer future seems to be supported only anecdotally. Those who maintain [End Page 155] this view tend selectively to ignore the fact that many teenagers, like many adults, forgo immediate pleasures in order to secure their future well-being as well as they can. For example, many high school students are concerned with gaining admission to a quality college and apply themselves very hard to their studies, shunning distracting amusements. Others worry deeply about the environment and work to promote conservation efforts, even when this undertaking makes their present lives more arduous. In addition, proponents of the view outlined above also seem to ignore the fact that many adults are unwilling to take on discomforts now for later gains, as evidenced by the high rates of smoking, obesity, and sedentary habits in this population. In short, I believe speculation can go either way on this issue. Whether adolescents in fact are more prone than adults to be biased toward the nearer future is a question best addressed by the kind of psychological studies mentioned in the previous section. (It is worth noting, however, that the research on adolescent decision making cited above did not identify any compelling data that would support the popular assumption that there is a significant and relevant difference between adolescents and adults in this regard.)

Second, even if it were proven true that adolescents are more likely than adults to have this bias or that their bias is more severe, it does not follow from either of these propositions that paternalistic intervention would be justified. Suppose it is the case that the bias in question is caused by one's failure to imagine pains in the more distant future or by one's false belief that the negative outcomes are somehow less real. It is clear that such individuals do not understand or fully appreciate the consequences of their present course of behavior, and as a result of this fail to deliberate adequately. Paternalism would of course be justified in these cases only if it were true that the bias could not possibly be corrected. But this last claim is highly unlikely given what is known about adolescent cognitive development. As I argued earlier, adolescents, like adults, have the mental power to imagine potential negative outcomes and to correct their false beliefs with the aid of a health care expert. The appropriate response in these cases is detailed and frank discussion, not paternalism.

Consider next the possibility that adolescents' bias toward the nearer future amounts simply to their inability to refrain from actions that they know will be worse for them later. This would constitute a classic case of weakness of will. The challenge here is to help such patients overcome their internal hindrances and achieve what they themselves acknowledge would be best for them in the long run. To this end, providers must work [End Page 156] with their patients to identify strategies that might help them fulfill their more carefully considered desires. The actions taken by providers in such cases would not be paternalistic, but would be better classified as autonomy enhancing due to the consensual nature of the help being provided. So, on either interpretation of adolescents' bias toward the nearer future, the arguments for paternalistic intervention do not succeed.

Protecting Adolescents from Themselves: Some Words of Caution

Finally, critics might assert that providers should adopt a firm protective attitude toward adolescent patients simply in order to prevent them from causing harm to themselves. To put the point bluntly, adolescents need protection. The justification for this position, one might argue, lies in the fact that adolescents generally lack life experience and often end up doing things they regret later. And since adolescents do not yet have the wisdom of experience to guide them, providers should intervene when necessary and restrict their liberty for the purposes of enhancing their welfare and minimizing harm. It might be said that paternalistic intervention in these cases not only prevents future evils, but also helps usher these patients safely through a phase of life replete with multiple social pressures.

I respond first by calling attention to a troubling consequence of this argument. Notice that on this line of reasoning particular adults who also happen to lack a "sufficient" degree of life experience would become candidates for having others make important decisions for them and otherwise protect them from their own voluntary wishes, at least insofar as their choices endanger their physical well-being. I believe the demands of consistency alone produce this unhappy result.

Second, there is a general, practical limitation to providers' ability to take a paternalistic approach to patients' noncompliance. The vast majority of health-related behavior is simply outside the control of physicians, whether it is taking medication daily, following a diet, or following an exercise program. Providers are engaging in self-delusion if they think that a more paternalistic position translates into better compliance. Overall, an approach that encourages patients to be more self-motivated may achieve better results.

However, the critics' position is not completely without merit; their concerns seem to be motivated by an admirable social ideal, namely, that all persons ought to possess a sincere concern for the well-being of others, [End Page 157] especially those who may be experiencing a period of stress and uncertainty. As valuable as this ideal may be, one must take care not to violate any equally important moral rights in its pursuit. One way to guard against this danger is to utilize the principle of noninterference. ¹⁰ According to this moral principle, no one may legitimately prevent a person's acting as she wishes without an open and compelling reason. Importantly, this principle places the burden of justification on the one who intervenes, not the one who expresses an intention to do such and such. Given the principle of noninterference, the key question is whether providers are adequately justified in forcefully preventing an adolescent patient from self-managing her chronic medical condition. This last question should not be passed over lightly. Although concern for the physical well-being of a willful adolescent might count as a reason for intervention, is it sufficient to outweigh her legitimate interest in making self-determining choices about her own medical care?

Many people seem to have a dark view of adolescence, seeing it as a period of enormous pressure and anxiety. However, adolescence is also a time for experimentation and new experiences. Teenagers are engaged in a process of self-discovery, and they need a firm sense of independence in order to flourish later in life. To this end, risk taking and testing limits are important components of the developmental transitions involved in identity formation. As Erikson (1968) has pointed out, a constant lack of exploring different options and possibilities may lead the adolescent to premature identity foreclosure. In other words, preventing an adolescent from taking risks may bode less well for her in the long run than allowing her to learn how to balance managing her chronic medical condition with her other goals and values and her relationships with her peers.

One should not suppose, however, that providers are therefore faced with an either/or choice between giving adolescents freedom for self-discovery and ensuring that they will be around later in life (or around without serious impairments). I believe providers can minimize this potential conflict by helping their patients to elucidate their own competing interests and by helping them to recognize the value of their long-term physical well-being (Emanuel and Emanuel 1992).

Most of us would agree that the ethically appropriate response to a situation in which one disagrees with another's values is not to run roughshod over the other to prevent her from following her own ideas of what is best for her; instead, one should attempt to persuade her rationally to change her mind. I suggest that, when confronted with a willful [End Page 158] adolescent patient, health care providers should try hard to persuade the patient by disclosing fully the technical risks involved in her choice, making the potential harms vivid, and emphasizing the value of improving one's health. But if the adolescent stands firmly by her ideals in spite of these sincere efforts, I believe providers ultimately should respect that choice and not resort to paternalistic measures. ¹¹ Since the relevant psychological data confirm middle adolescence as the period when most persons come to possess the capacities necessary for meaningful self-determination, my position applies to adolescents ages 14 and older, while leaving open the possibility that some exceptional younger patients may also be found capable of making their own decisions about how to manage their chronic conditions. Certainly not every individual patient (adult or adolescent) will be competent to make self-determining medical choices, but there should be little difference overall in the way health care professionals manage adult and older adolescent patients who do not maintain optimal medical control over their chronic conditions. ¹²

Conclusion

Although it is clear that chronic conditions can cause adolescents to remain more dependent on adults at a time when their peers are becoming increasingly independent, health care providers should embrace the fact that they can play a very important role in aiding their teen patients to become even more self-determining. Successfully managing chronic illness in adolescents should not include forcing or tricking them to be compliant with treatment; rather, patients should be given options that will maximize their sense of control over their own lives. By allowing adolescents from, say, the age of 14 the freedom to make their own decisions about what is best for them and by providing steady guidance and advice along the way, providers can set a pattern for self-management whereby adolescents gradually learn to treat their conditions satisfactorily without having to abandon their values or their subjective sense of identity. In this way, the treatment of a chronic medical condition actually can enhance further the development of personal autonomy.

Although these final suggestions may seem applicable only to the adolescent patient population, there is no reason why providers should not engage adult patients in a comparable manner. As many of us know, social pressures and personal anxieties do not cease automatically upon legal adulthood but persist (albeit in other forms) throughout the life span. All patients with chronic medical conditions must learn how to determine [End Page 159] for themselves the best way to balance the efforts required to ameliorate their illness with their other goals and interests. This important task is preempted by treatment strategies that do not elicit the proper use of a patient's rational capacities. Successful self-management depends crucially on persons' capacities for self-determination, and, as such, patients need to be motivated by reasons they themselves can accept and follow voluntarily. I conclude that health care professionals can be much more effective overall when they approach their adolescent patients less as "half-formed" adults and more as new participants in the moral community.

* I thank Howard Brody, Martin Draznin, David Newman, Michael Pritchard, John R. Stone, and the anonymous reviewers of this journal for their comments on an earlier draft of this paper.

Insoo Hyun, Ph.D., is Assistant Professor of Philosophy and Associate Director of the Center for the Study of Ethics in Society, Western Michigan University, Kalamazoo, MI.

Notes

1. In this paper, I use the terms "compliance," "management," and "adherence" in their most obvious senses. Although persons legitimately may question the degree to which these purportedly neutral terms are themselves value-laden and subject to controversy, my subsequent arguments do not depend on a resolution of these theoretical debates.

2. I do not mean to suggest that physicians are thereby permitted to use any means to remove practical barriers to an adult patient's treatment adherence. Coercive and deceptive methods for achieving compliance, for example, may violate the autonomy and dignity of competent patients. Rather, I am assuming that the patient and physician are both involved in openly formulating and implementing a plan to correct the practical difficulties in question.

3. I agree with the recommendation advanced by Emanuel and Emanuel (1992) that providers should follow a deliberative model of the physician-patient relationship and try hard to engage their patients in an open and honest dialogue of what course of action would best promote their health interests. However, if disagreement persists despite these efforts, I maintain that the patient should have the final say. As I elaborate below, I believe this position is both morally and practically justified. Emanuel and Emanuel, on the other hand, do not consider at much length such matters of fundamental conflict between provider and patient except to say that patients should seek the care of another physician if they so happen to reach an impasse.

4. Of course, many practical barriers to patient compliance are not easy to surmount and can raise complex moral problems in their own right for health care professionals. For instance, an adolescent's family may face difficulties paying for the required treatment, or a chaotic home environment may indicate that the parents are neglecting the medical needs of their dependent patient. Regrettably, I do not have room to examine the moral issues raised by these and other scenarios. I acknowledge them here only to separate them from the type of nonadherence cases I wish to explore in depth in this paper. I thank an anonymous reviewer of this journal for encouraging me to make this point more explicit.

5. The sort of paternalistic measures I have in mind involve the use of deceptive methods such as exaggerating the likelihood, immediacy, and severity of the possible health risks to frighten adolescents into adhering more closely to their recommended management plans. Aside from the considerations I will detail later, one possible weakness with deception is that adolescents are likely to find out about it, leading them to a general distrust of their health care providers. Of course, other measures can avoid this particular disadvantage by being more forthright, such as the implementation of an ad hoc system of rewards and punishments in order to alter the desirability matrix of the adolescent's current course of behavior.

6. According to one prominent analysis, all instances of competent medical decision making require (1) a capacity to understand and communicate, (2) a capacity to reason and deliberate, and (3) the possession of a set of values and goals (Buchanan and Brock 1989). Although this analysis is intended to aid determinations of whether a patient is competent to make a particular medical decision, these three requirements also may be reasonably taken to constitute some of the core attributes necessary for a patient's self-managing his chronic medical condition. Thus, if adolescents generally have serious deficiencies in understanding, reasoning, deliberation, and the like, when compared with adults, then we will have found a justification for adopting a paternalistic approach toward adolescents with chronic conditions.

7. It is worth emphasizing here that Weithorn and Campbell (1982) also did not discern any such difference between the medical decision-making processes of 14-year-olds and legally recognized adults.

8. One might question whether adolescents possess a sufficiently mature or stable set of values to allow them to make competent self-management decisions. In response, I think it is important not to set a higher standard arbitrarily for adolescents than for adults with regard to this particular facet of competence in medical decision making. Although all patients (both adult and adolescent) need personal values as a basis for evaluating and comparing their different choices, their values need not comprise a fully detailed and developed conception of the good; rather, their values should possess some minimum of coherence and be stable over the course of the decision-making process (Buchanan and Brock 1989). Regardless of where one ultimately sets the minimum level of stability and coherence for decision-making competence, it would be unfair to demand more from adolescents than from adults who are judged to be competent.

9. I borrow this term from Derek Parfit (1984), who argues that it may not be irrational in all cases for one to care less about one's more distant future.

10. This principle is wonderfully detailed by Stanley Benn (1975).

11. I should emphasize, however, that this is under the condition that the adolescent's choice would not cause significant harm to others, including her family. Although this raises the controversy over what should count as a significant harm (should financial burdens count?), my point here is merely to acknowledge that all patients, both adult and adolescent, cannot legitimately follow their own ideas of what is best for them regardless of their effect on other people. Considerable and identifiable harm to others provides a clear limitation on one's moral right to personal autonomy.

12. I acknowledge that people's capacities for self-determination are often a matter of degree and can raise difficult borderline questions. With regard to adolescent and adult patients with chronic conditions, I think that a variable standard of competence may be utilized to handle especially tough cases. According to this standard, the level of decision-making competence required should vary directly in proportion to the degree of danger involved in the decision (Buchanan and Brock 1989). For example, patients who take on serious, life-threatening complications for the sake of what seem to be merely cosmetic reasons would have to satisfy a very high standard of decision-making competence.

References

Benn, Stanley I. 1975. Freedom, Autonomy, and the Concept of a Person. Proceedings of the Aristotelian Society n.s. 76: 109-112.

Brock, Dan W. 1991. The Ideal of Shared Decision Making Between Physicians and Patients. Kennedy Institute of Ethics Journal 1: 28-47.

Buchanan, Allen E., and Brock, Dan W. 1989. Deciding for Others: The Ethics of Surrogate Decision Making. Cambridge: Cambridge University Press.

Dolcini, M. Margaret; Cohn, Lawrence D.; Adler, Nancy E.; et al. 1989. Adolescent Egocentrism and Feelings of Invulnerability: Are They Related? Journal of Early Adolescence 9: 409-18.

Emanuel, Ezekiel J., and Emanuel, Linda L. 1992. Four Models of the Physician-Patient Relationship. Journal of the American Medical Association 267: 2221-26.

Enright, Robert D.; Lapsley, Daniel K.; and Shukla, Diane G. 1979. Adolescent Egocentrism in Early and Late Adolescence. Adolescence 14: 687-95.

Erikson, Erik. 1968. Identity: Youth and Crisis. New York: Norton.

Hall, G. Stanley. 1904. Adolescence: Its Psychology and Its Relations to Physiology, Anthropology, Sociology, Sex, Crime, Religion, and Education. Vols. 1-2. New York: Appleton.

Inhelder, Bärbel, and Piaget, Jean. 1958. The Growth of Logical Thinking From Childhood to Adolescence. New York: Basic Books, Inc.

Lapsley, Daniel K.; Milstead, Matt; Quintana, Stephen M.; et al. 1986. Adolescent Egocentrism and Formal Operations: Tests of a Theoretical Assumption. Developmental Psychology 22: 800- 807.

Mead, Margaret. 1953. Coming of Age in Samoa: A Psychological Study of Primitive Youth for Western Civilization. New York: Modern Library.

Millstein, Susan G. 1993. Perceptual, Attributional, and Affective Processes in Perceptions of Vulnerability Through the Life Span. In Adolescent Risk Taking, ed. Nancy J. Bell and Robert W. Bell, pp. 55-65. Newbury Park: Sage Publications.

Parfit, Derek. 1984. Reasons and Persons. New York: Oxford University Press.

Slovic, Paul; Fischhoff, Baruch; and Lichtenstein, Sarah. 1980. Facts Versus Fears: Understanding Perceived Risk. In Societal Risk Assessment: How Safe is Safe Enough?, ed. Richard Schwing and Walter A. Albers, pp. 181-216. New York: Plenum Press.

Tversky, Amos, and Kahneman, Daniel. 1973. Availability: A Heuristic for Judging Frequency and Probability. Cognitive Psychology 5: 207-32.

Weinstein, Neil D. 1980. Unrealistic Optimism About Future Life Events. Journal of Personality and Social Psychology 39: 806-20.

------, ed. 1987. Taking Care: Understanding and Encouraging Self-Protective Behavior. New York: Cambridge University Press.

Weithorn, Lois A., and Campbell, Susan B. 1982. The Competency of Children and Adolescents to Make Informed Treatment Decisions. Child Development 53: 1589-98.