Practice recommendations
- The process for evaluating treatment refusal should be consistent regardless of setting, problem (ie, medical or psychiatric), type of patient, or practitioner.
- Assess decisional capacity, psychiatric dangerousness, and medical risk in all cases of treatment refusal while addressing potential causes of treatment refusal.
- Based on these assessments, choose between: a) respecting the treatment refusal, b) obtaining a surrogate, or c) mandating hospitalization and possibly treatment.
An 80-year-old woman with diabetes who has been your patient for many years is in failing health and may need dialysis for deteriorating kidney function. She refuses even to consider further evaluation.
A 37-year-old man has, according to his family, become increasingly depressed and makes comments suggesting suicidal ideation. They are afraid for his safety. He says he’s just going through a rough period and doesn’t need help.
Would you be prepared to handle instances of treatment refusal such as these? Treatment refusal can be challenging, creating conflicts among patients, families, and health care providers, and raising important ethical considerations. A patient’s autonomy may be undermined if her wishes are overridden. Inappropriately confining, restraining, or treating patients may cause harm. Failing to obtain a surrogate when indicated may result in a missed opportunity to benefit a patient. Refusal of treatment, when unaddressed or mishandled, may lead to patient dissatisfaction, substandard care, increased litigation, or disparities in care.
Clearer guidelines are needed to help clinicians evaluate and manage patients who refuse treatment. Building on previous work in treatment refusal, informed consent, and competency theory, we propose an approach to treatment refusal that provides 3 unique contributions:
- First, our model integrates medical and psychiatric treatment refusal practices—usually found in separate literature bases—into one model.
- Second, it emphasizes that evaluations of both decisional capacity and dangerousness (both defined later) are crucial to determining the appropriate response to treatment refusal.
- Third, it provides concrete guidance for how to decide between 3 actions: 1) respect the treatment refusal, 2) obtain a surrogate, or 3) mandate hospitalization and possibly treatment.
How to assess and manage treatment refusal
Step 1: Evaluate both decisional capacity and dangerousness
Grisso and Appelbaum have described an empirically tested model of how to evaluate decisional capacity.1Decisional capacity refers to the ability to make a choice about treatment, and it is determined by the presence and extent of functional abilities, which are hierarchical, from the simplest to the most complex:
- Making a choice (“I’d like to have the cardiac catheterization”)
- Understanding relevant details, such as diagnosis, prognosis, the benefits and burdens of different treatment alternatives, and what will happen without treatment
- Appreciating that the relevant details apply to oneself and will mean something for one’s own future
- Rationally describing why a choice was made (such as explaining why one would prefer a particular set of risks, benefits, or burdens from one treatment alternative over another).
Many factors may interfere with these abilities and they should be assessed,1 but additional discussion is beyond the scope of this article.
Not all patients with mental illness have impaired decisional capacity. On the contrary, there is significant variability in the decisional capacity of people with serious mental illness.2 One instrument to evaluate decisional capacity for treatment decisions has been studied, but time constraints may limit its widespread application in clinical practice.3
Dangerousness in the clinical setting is generally defined as the intent to harm oneself or others, creating imminent risk. For the purposes of our model, this will be called “psychiatric dangerousness.” Other sources describe the assessment of psychiatric dangerousness in more detail,4 but the assessment typically includes a thorough psychosocial history, an evaluation for mental illness, a determination of risk factors for suicide, and often corroborating history from family or friends. “Medical dangerousness” is defined as the risk of morbidity or mortality that accompanies medical intervention or non-intervention.
In cases of treatment refusal, explicitly assess both decisional capacity and dangerousness. There are 3 benefits in doing so. First, it helps avoid the tendency to assess dangerousness only in the psychiatric context and decisional capacity in the general medical context. Second, it provides a useful way to approach treatment refusal when the cause of symptoms or complaints is ambiguous. Third, it helps when a patient exhibits both medical and psychiatric symptoms of illness.
Step 2: Determine the need for involuntary hospitalization and treatment
The ideal threshold for involuntary confinement is the point at which those who will harm themselves or others are confined and protected, while those who will not harm themselves or others are not. Determining the ideal threshold for involuntary treatment is difficult. Besides the challenges in predicting dangerousness,5 this determination involves a balance between potentially competing goals: respecting individual liberties, enhancing quality of life, and protecting patients and others from harm.