Treatment Refusal

Abstract

Treatment refusal, which MeSH classifies under treatment adherence and compliance, is a patient's declining of a recommended test, procedure, or therapy. It is at once a legal right grounded in autonomy, a clinical event to be understood, and, when a patient's decision-making capacity is in doubt, a judgment a clinician must make about whether the refusal may stand. This article separates a competent refusal, which must be honored, from one compromised by a disorder of understanding or appreciation, setting out the four-abilities standard and the instruments that operationalize it. It then examines why capable patients refuse: the psychological reactance provoked by pressure, the beliefs and behavioral intentions that precede the choice, and the shared-decision process in which a declined option is still a reached one. A worked example and three interactive demonstrations follow.

Keywords: treatment refusal, decision-making capacity, psychological reactance, shared decision making

A competent adult may decline any medical treatment, however beneficial, and a clinician may not override that decision merely because it is unwise. This is settled law and settled ethics, and it makes refusal a problem of a particular kind: not a failure to be corrected but a choice to be understood, and sometimes a decision whose validity must be assessed. The clinician faced with a refusal must ask two different questions at once. Is this patient able to make the decision? And, taking the decision as made, why was it made, so that a recommendation can be revisited rather than merely repeated?

Treatment refusal names the patient's declining of recommended care. The sections below separate the two questions the clinic confronts, the capacity question of whether a refusal is competent and so binding, and the psychological question of why patients, competent or not, decline what is offered, before turning to the collaborative process that reframes refusal as the far end of choice rather than its breakdown.

What Treatment Refusal Is

Treatment refusal is the act of declining a recommended medical intervention. The right to refuse follows directly from the principle that governs consent: if valid medical treatment requires the patient's informed agreement, then the patient who withholds that agreement has, by the same logic, declined the treatment. The autonomy that makes consent meaningful is the same autonomy that makes refusal binding, so a refusal by a capable patient is not a lesser or provisional decision but a full exercise of the right to determine what is done to one's own body (Scholten et al., 2021).

What complicates refusal is that the right belongs only to a patient who can exercise it, and the capacity to decide is not guaranteed by the capacity to speak. A patient may decline treatment for reasons a clinician respects, for reasons a clinician disagrees with but must accept, or because a disorder has compromised the very abilities the decision requires. Distinguishing these is the central clinical task, because the first two must be honored and only the third may, after formal assessment, be set aside (Appelbaum, 2007). The right to refuse is therefore inseparable from the question of competence, and the two have been studied together since the modern law of consent took shape.

Types of Treatment Refusal

In the Medical Subject Headings vocabulary, treatment refusal sits among the headings that index how patients act on and exercise rights over their care, and it has one narrower descriptor directly beneath it. MeSH is an indexing classification, not a theory of the construct, so the placement below reflects how the literature is catalogued rather than a claim that refusal decomposes cleanly into this single child; the reasons patients refuse, set out in the sections that follow, cut across it.

The single child understates how much structure refusal has. The more consequential distinction, and the one that organizes this article, is not catalographic but functional: between a refusal whose validity is in question because the patient's capacity is in doubt, and a competent refusal whose reasons are in question because the clinician hopes to address them. The first is a matter of assessment, the second a matter of understanding, and they call for entirely different responses.

Competence and the Right to Refuse

Whether a refusal is binding turns on the patient's decision-making capacity, and the modern standard decomposes that capacity into four abilities. The patient must be able to understand the relevant information, to appreciate how it applies to their own situation, to reason with it in weighing options, and to express a choice. Loren Roth, Alan Meisel, and Charles Lidz first laid out the competing legal tests of competency and showed that they varied from a bare ability to communicate to a demanding scrutiny of the reasoning behind a choice (Roth et al., 1977). Paul Appelbaum and Thomas Grisso consolidated this into the four-abilities model that became the operational standard, deliberately separating the quality of the reasoning from the content of the conclusion so that an unpopular choice is not mistaken for an incapable one (Appelbaum & Grisso, 1988).

The decisive feature of the standard is that it assesses the process, not the outcome. A capable patient may refuse treatment for reasons the clinician finds mistaken, and the refusal still stands; incapacity is established only by a failure in one of the four abilities, not by the imprudence of the decision. To make this reproducible at the bedside, Grisso, Appelbaum, and Carolyn Hill-Fotouhi built the MacArthur Competence Assessment Tool for Treatment (MacCAT-T), a semi-structured interview that scores each ability against the specific decision at hand (Grisso et al., 1997). Later reviews catalogued the proliferation of such instruments and the difficulty of validating any single one, since capacity is decision-relative and no cut score settles a borderline case on its own (Sturman, 2005; Palmer & Harmell, 2016).

Capacity is judged on four abilities, and it is all-or-nothing in a precise way: it is present only when every ability clears the threshold (the navy line). A single deficit removes capacity even when the other three are strong — a high average cannot buy back a failed ability.

Toggling “unwise” changes nothing: the verdict never moves, because capacity assesses the process of deciding, not the wisdom of the conclusion. A competent patient may refuse for reasons the clinician finds mistaken, and the refusal still stands.

Why Capable Patients Refuse: Reactance

Once a refusal is established as competent, the clinical question shifts from whether it may stand to why it was made, and here the psychology of refusal begins. The most studied mechanism is one that turns the clinician's own urgency against the recommendation. Jack Brehm's theory of psychological reactance holds that when a person perceives a threat to a behavioral freedom, a motivational state arises that is directed at restoring that freedom, often by doing the very thing that was discouraged (Brehm, 1966). A strong directive to accept treatment is, to the patient, a threat to the freedom to decline, and reactance predicts that the harder the push, the harder the refusal can become.

Reactance is not a fringe effect. Decades of research have established it as a robust and general motivational response, with the magnitude of the reactance rising with the importance of the threatened freedom and the force of the threat (Miron & Brehm, 2006). A half-century review found the theory's predictions borne out across domains from health messaging to interpersonal persuasion, and drew the applied lesson directly: communications that are experienced as controlling tend to backfire, producing less of the intended behavior than gentler approaches (Rosenberg & Siegel, 2018). The clinical paradox follows immediately. The instinctive response to a refusal, to press harder, is precisely the one most likely to entrench it, which is why the management of refusal has moved away from persuasion and toward collaboration.

Why Capable Patients Refuse: Beliefs and Intentions

A refusal also has antecedents that precede the encounter entirely, in the patient's beliefs about illness and treatment. Irwin Rosenstock's health belief model held that whether a person takes a recommended health action depends on their perceived susceptibility to the condition, the perceived severity of it, and the perceived benefits of the action weighed against its perceived barriers (Rosenstock, 1974). A patient who judges a condition mild, or a treatment's burdens to outweigh its benefits, arrives already disposed to refuse, and no amount of bedside insistence addresses a belief formed long before.

Icek Ajzen's theory of planned behavior gives the more general account. A behavior is predicted by the intention to perform it, and that intention is the joint product of three things: the person's attitude toward the behavior, the subjective norm they perceive around it, and their perceived behavioral control over carrying it out (Ajzen, 1991). On this account a treatment decision is not formed at the moment of the offer but assembled from standing attitudes, the perceived expectations of family and community, and the patient's sense of whether they can actually follow through. A refusal is then the visible end of an intention that was already in place, which is why understanding the components of that intention is more useful to a clinician than contesting the conclusion.

Shared Decision Making

If refusal cannot be overcome by pressure and is rooted in beliefs and intentions that predate the encounter, the constructive response is to make the decision jointly. Shared decision making is the model that Cathy Charles, Amiram Gafni, and Tim Whelan defined as requiring that both the clinician and the patient participate, that information flows in both directions, and that both parties work toward an agreed decision they each endorse (Charles et al., 1997). On this model a refusal is not a breakdown of the process but a possible and legitimate output of it: a declined option that the patient reached with the clinician's help is a shared decision, not a failed persuasion.

Glyn Elwyn and colleagues made the model operational with a three-talk structure, moving from team talk, in which the clinician establishes that a choice exists and offers support, through option talk, comparing the alternatives, to decision talk, eliciting and acting on the patient's informed preferences (Elwyn et al., 2012). Anne Stiggelbout and colleagues assembled the evidence that this process improves the match between care and patient values without, on balance, harming outcomes (Stiggelbout et al., 2015), and a wave of health-policy argument held that shared decision making had become not merely good practice but an ethical and systemic imperative (Spatz et al., 2017). Within this frame a refusal elicited through shared decision making carries information the clinician needs, and is the point at which care can be realigned rather than abandoned.

The patient’s leaning is fixed and points to declining the treatment, so the outcome is a refusal regardless. What the three stages change is what kindof refusal it is. Collaboration is the combined quality of the stages, and one neglected stage drags the whole process down.

Raising all three stages flips the identical refusal from a breakdown of the encounter into a decision made together. The goal is not to avert the refusal but to reach it jointly, where it carries the information a clinician needs to realign care rather than merely repeat the recommendation.

Worked Example: Identical Recommendation, Opposite Decision

Consider two competent patients given the same strong recommendation to begin a preventive medication. Both have the capacity to decide; the question is why one accepts and the other refuses. Following the theory of planned behavior, each patient's intention to adhere is modeled on a 0-to-100 scale as a weighted combination of three antecedents: their attitude toward the treatment, the subjective norm they perceive, and their perceived behavioral control. The weights are fixed at 0.40 for attitude, 0.35 for norm, and 0.25 for control, and a clinic takes an intention below 50 as a predicted refusal.

- Patient A holds a skeptical attitude (30), perceives weak normative support (40), and modest control (50). - Patient B holds a favorable attitude (70), perceives supportive norms (60), and high control (80).

Patient A's intention is `0.40×30 + 0.35×40 + 0.25×50 = 12.0 + 14.0 + 12.5 = 38.5`, below the threshold, so a refusal is predicted. Patient B's is `0.40×70 + 0.35×60 + 0.25×80 = 28.0 + 21.0 + 20.0 = 69.0`, well above it, so acceptance is predicted. The recommendation was identical and both patients were fully competent; the opposite decisions are driven entirely by the attitudes, norms, and control beliefs each brought to it (Ajzen, 1991). A clinician who reads Patient A's refusal as a gap in information will make no progress, because the gap is in attitude and perceived norm, not in understanding, and reactance warns that pressing the recommendation harder may lower the intention further rather than raise it (Rosenberg & Siegel, 2018).

Discussion

The study of treatment refusal has had to hold two commitments together. The first is that a competent patient's refusal is sovereign: it is the exercise of a right, not a symptom to be treated, and the clinician's disagreement with it carries no authority to override it (Scholten et al., 2021). The second is that competence cannot simply be assumed, because some refusals arise from a disorder that has compromised the very abilities the decision requires, and honoring such a refusal uncritically abandons the patient as surely as overriding a competent one violates them (Appelbaum, 2007). The four-abilities standard and the instruments built on it exist precisely to keep these two commitments from collapsing into each other, by making the capacity determination explicit, decision-relative, and reviewable rather than a matter of whether the clinician agrees with the choice (Appelbaum & Grisso, 1988; Grisso et al., 1997).

That division is why MeSH files treatment refusal where it does. Refusal is the negative pole of treatment adherence and compliance: both describe what a patient does with a recommendation, and both are governed by the same beliefs, intentions, and relationship with the clinician. The reframing the field has converged on, from persuasion toward shared decision making, treats refusal not as adherence's failure but as a decision to be understood and, where the patient wishes, revisited together (Charles et al., 1997; Elwyn et al., 2012).

Cognitive Implications

Treatment refusal brings together two currents of cognitive psychology that are usually studied apart. The capacity question is, at bottom, a question about specific cognitive abilities: understanding is comprehension and retention of information, appreciation is the application of that information to the self, and reasoning is the manipulation of it in service of a choice. The four-abilities standard is in effect a domain-specific battery, and the difficulty of validating any capacity instrument reflects the general difficulty of measuring a bounded cognitive competence against a decision whose demands vary (Sturman, 2005; Palmer & Harmell, 2016). This is why capacity is decision-relative rather than global: a patient may have the cognitive resources to decline a blood draw and lack those to weigh a complex surgical trade-off.

The reasons question draws on the psychology of motivation and decision making. Reactance is a motivational state, not a belief, which is why it responds to the manner of a recommendation rather than its content and why it can push a decision against the person's own considered interest (Brehm, 1966; Miron & Brehm, 2006). The health belief model and the theory of planned behavior, by contrast, model the decision as the output of standing attitudes, perceived norms, and control beliefs, so that an intention is assembled well before the moment of choice (Rosenstock, 1974; Ajzen, 1991). A refusal is therefore overdetermined in a precise sense: a competent patient's no can be, simultaneously, the discharge of a pre-formed intention and a reactive defense of a threatened freedom, and the two call for different responses even though they produce the same word.

Current Directions

The active frontier is the hardest category of refusal, where capacity and psychopathology are entangled. Irene Elzakkers and colleagues studied the capacity of patients with anorexia nervosa to refuse treatment and found that standard capacity assessment often judged them competent on the four abilities even as the disorder shaped the values driving the refusal, exposing a gap the four-abilities model does not fully close (Elzakkers et al., 2018). The open question is whether appreciation, the ability to apply information to one's own case, can capture a disorder that distorts valuation without impairing comprehension, and the case has become the testing ground for the standard's limits.

A second current is explicitly normative. Jakov Scholten, Jochen Gather, and Jochen Vollmann argued that the informed-consent process must treat patients with mental disorders as equal decision-makers, and that capacity thresholds and substitute decision-making must be scrutinized for the discrimination they can smuggle in when a refusal is too readily attributed to illness (Scholten et al., 2021). Running alongside this is the maturing consensus that shared decision making is the default frame for any consequential choice, including a refusal, with the open work being how to realize it under real clinical constraints rather than whether to adopt it (Spatz et al., 2017; Stiggelbout et al., 2015). The questions are practical and sharp: how to assess capacity when a disorder acts on values rather than cognition, how to keep the capacity determination free of bias, and how to make collaborative decision-making routine.

Reactance rises with how important the threatened freedom is and how controlling the directive feels. As the push grows, the chance of refusal climbs toward certainty — so pressing harder can entrench the very refusal it was meant to overcome.

The same recommendation, delivered gently or forcefully, produces very different refusal rates when the freedom at stake matters to the patient. This is why managing refusal has moved away from persuasion and toward a collaboration that does not threaten the freedom to decide.

Common Misconceptions

An unwise refusal proves the patient lacks capacity.

No. Capacity is judged on the process of deciding, not the wisdom of the conclusion. A competent patient may refuse for reasons the clinician finds mistaken, and the refusal still stands; incapacity is established only by a failure of understanding, appreciation, reasoning, or the ability to express a choice (Appelbaum & Grisso, 1988; Appelbaum, 2007).

Pressing harder is the way to overcome a refusal.

Often the opposite. Psychological reactance means that a recommendation experienced as controlling can intensify the refusal it was meant to overcome, so a more forceful push can lower the chance of acceptance rather than raise it (Brehm, 1966; Rosenberg & Siegel, 2018).

Capacity is a fixed, global trait a patient either has or lacks.

No. Capacity is decision-relative and can fluctuate: a patient may have the capacity to make a simple choice and lack it for a complex one, and capacity lost to a reversible condition can return. The assessment is always tied to the specific decision at hand (Grisso et al., 1997; Palmer & Harmell, 2016).

A refusal is the failure of shared decision making.

Not so. In the shared-decision model a declined option that the patient reached with the clinician's help is a legitimate output of the process, not a breakdown of it. The goal is a decision the patient endorses, which may be a refusal (Charles et al., 1997; Elwyn et al., 2012).

Glossary

Appreciation.

In the four-abilities standard, the patient's ability to apply the relevant medical information to their own situation, as distinct from understanding it in the abstract.

Attitude.

In the theory of planned behavior, a person's favorable or unfavorable evaluation of a behavior, one of the three antecedents of the intention to perform it.

Decision-making capacity.

The clinical judgment that a patient possesses the abilities a particular decision requires; the condition under which a refusal is binding. Compare mental competency, its legal counterpart.

Four-abilities standard.

The operational model of capacity comprising understanding, appreciation, reasoning, and the ability to express a choice, consolidated by Appelbaum and Grisso.

Health belief model.

Rosenstock's account that a health action depends on perceived susceptibility, perceived severity, and the perceived benefits of the action weighed against its barriers.

Informed consent.

The patient's voluntary, informed agreement to a treatment; refusal is the withholding of that agreement, grounded in the same right to self-determination.

Intention.

In the theory of planned behavior, the immediate antecedent of a behavior, formed from attitude, subjective norm, and perceived behavioral control.

MacCAT-T.

The MacArthur Competence Assessment Tool for Treatment, a semi-structured interview that scores each of the four abilities against the specific decision a patient faces.

Mental competency.

The legal determination that a person is able to make a given decision; the juridical counterpart to the clinical judgment of decision-making capacity.

Perceived behavioral control.

In the theory of planned behavior, a person's sense of how easily they can perform a behavior, shaping both the intention and the behavior itself.

Psychological reactance.

The motivational state, described by Brehm, that arises when a freedom is threatened and is directed at restoring it, often by doing the discouraged thing.

Reasoning.

In the four-abilities standard, the patient's ability to weigh and manipulate information in the service of comparing options and reaching a choice.

Shared decision making.

A model in which clinician and patient both participate, exchange information in both directions, and work toward a decision they each endorse, of which a refusal can be a legitimate output.

Subjective norm.

In the theory of planned behavior, a person's perception of the social expectations surrounding a behavior, one of the three antecedents of intention.

Treatment refusal.

A patient's declining of a recommended test, procedure, or therapy; binding when made by a patient with decision-making capacity.

Understanding.

In the four-abilities standard, the patient's comprehension and retention of the information relevant to the decision.

Vaccination refusal.

The declining of a recommended immunization; the narrower MeSH descriptor beneath treatment refusal, where individual choice carries population-level consequences.

Key Researchers

Icek Ajzen

(University of Massachusetts Amherst). Social psychologist whose theory of planned behavior links attitudes, subjective norms, and perceived control to behavioral intention, the framework that explains why a patient who understands a recommendation and could comply may still refuse when attitude and perceived norms point the other way. Wikipedia · Wikidata · Google Scholar

Paul S. Appelbaum

(Columbia University). Psychiatrist and legal scholar who, with Thomas Grisso, operationalized decision-making capacity into four assessable abilities, understanding, appreciation, reasoning, and expressing a choice, the standard by which a refusal is judged competent rather than treated as a symptom to be overridden. ORCID · Wikipedia

Jack W. Brehm

(University of Kansas, 1928–2009). Social psychologist who formulated the theory of psychological reactance, the motivational rebound against a perceived threat to freedom, which is the most cited mechanism for a refusal that intensifies precisely as the pressure to comply increases. Wikidata

Glyn Elwyn

(Dartmouth College). Physician-researcher who developed the three-talk model of shared decision making and the OPTION instruments, reframing refusal as the far end of a preference-elicitation process rather than a breakdown of it. ORCID · Google Scholar

Thomas Grisso

(University of Massachusetts Chan Medical School). Forensic psychologist who, with Appelbaum, built the MacArthur Competence Assessment Tool that made capacity assessment reproducible at the bedside, letting a clinician distinguish a competent refusal from one compromised by a disorder of understanding or appreciation. ORCID · Wikipedia

Frequently Asked Questions

What is treatment refusal?

It is a patient's declining of a recommended test, procedure, or therapy. When the patient has decision-making capacity, the refusal is a binding exercise of the right to determine what is done to one's own body, and MeSH classifies it beneath treatment adherence and compliance (Scholten et al., 2021).

Can a competent patient refuse life-saving treatment?

Yes. An adult with decision-making capacity may decline any treatment, including one the clinician judges life-saving, and the refusal must be honored. Disagreement with the choice is not grounds to override it; only a failure of capacity can set a refusal aside (Appelbaum, 2007).

How is decision-making capacity assessed?

Against four abilities: understanding the relevant information, appreciating how it applies to one's own situation, reasoning with it to compare options, and expressing a choice. The MacCAT-T interview scores each ability against the specific decision rather than the patient's global state (Appelbaum & Grisso, 1988; Grisso et al., 1997).

Does an unwise decision mean a patient lacks capacity?

No. Capacity turns on the process of deciding, not the wisdom of the outcome. A capable patient may choose for reasons the clinician finds mistaken; incapacity is shown only by a deficit in one of the four abilities (Roth et al., 1977).

Why might pressing a patient harden their refusal?

Because of psychological reactance. When a recommendation is experienced as a threat to the freedom to decline, a motivational state arises that defends that freedom, so a more controlling push can entrench the refusal rather than overcome it (Brehm, 1966; Rosenberg & Siegel, 2018).

What makes a capable patient refuse treatment?

Beliefs and intentions formed before the encounter. The health belief model points to perceived susceptibility, severity, benefits, and barriers; the theory of planned behavior points to attitude, subjective norm, and perceived control as the antecedents of the intention a refusal expresses (Rosenstock, 1974; Ajzen, 1991).

How does shared decision making change how refusal is handled?

It reframes refusal as a possible, legitimate output of a joint process rather than a failure of persuasion. Through team, option, and decision talk, the clinician elicits the patient's informed preferences, so a declined option is a decision reached together (Charles et al., 1997; Elwyn et al., 2012).

When is capacity assessment hardest?

When a disorder shapes the values behind a refusal without impairing comprehension. Studies of anorexia nervosa find patients can be judged competent on the four abilities while the illness drives the refusal, exposing a limit of the standard that remains an active research problem (Elzakkers et al., 2018).

References