When persons living at home or in care facilities (referred to here as “patients”) choose to engage in activities that put themselves or others at risk of harm, health care providers must find approaches to support both patient autonomy and the safety of patients and others. This also applies when substitute decision-makers make choices on behalf of patients.

Determining when and how to intervene can be a complex task. A literature review carried out by the authors found that processes have been developed to address risky behaviors in specific locations (e.g., use of negotiated risk agreements in assisted living),[1] specific circumstances (e.g., choices of competent persons living at home),[2] and specific populations (e.g., those with mental health diagnoses).[3] However, this literature review did not identify an ethics-based approach that is practical, easy to use, and applicable to all patients and settings, as noted in a 2010 report calling for a decision-making tool that incorporates a framework and checklist of what clinicians should consider.[4] The ethical approach outlined below for managing patients choosing to live at risk aims to make a contribution to this important literature.

Living at risk

Living at risk is defined as acting in a way that impacts the person (risk to self) or others (risk to others) in physical, emotional, or psychological ways. This may involve a wide variety of activities such as eating when at risk of aspiration, living at home without adequate support, going on unsupervised outings, smoking around oxygen supplies, or refusing to use a walker needed to prevent falls. Cases such as the fictionalized one that follows arise on a regular basis.

Jahal is a 64-year-old with a history of homelessness, cognitive impairment affecting short-term memory, and depression. She was recently placed in residential care and has adapted to her new environment. However, Jahal likes to go on long walks and this is causing her care providers concern. When they attempt to keep her in the facility, she becomes agitated and lashes out verbally and physically. Once she knocked down another resident while trying to leave the facility. Jahal generally manages on her walks, although she has fallen a few times and on one occasion was returned by police when a store owner reported her sitting on a bench outside for several hours.

These kinds of cases are complex for a number of reasons:

• In Western society, respect for autonomy is considered a primary bio-ethical principle. Any intervention contrary to the patient’s wishes must be justified.[5,6]
• Not all patients who have cognitive deficits or mental illness are incapable of decision-making. Assessment of the capacity of a patient to understand and assume risks can take time, involve judgment, and require a large interdisciplinary team.[7]
• Determining whether a given risk is “too risky” is a challenge in terms of defining parameters, setting aside personal biases, and striving for objectivity.[7]
• Health care providers are trained to place a premium on patient safety and may struggle with allowing someone to live at risk.[6] They may also find it challenging to view risk taking as making a positive contribution to quality of life.[4]
• Harming patients is contrary to the codes of ethics of health professions and contrary to the bioethical principle of non-maleficence. Health care providers may believe that they contribute to harm by allowing a patient to take risks or by participating themselves in a risk activity (e.g., by feeding a patient who has swallowing difficulties).[2]
• There may be legal implications if patients are permitted to engage in risk activities and they or others are harmed. Tools that are developed to address risk may be construed as intended primarily to protect facilities.[1]

So how can complex cases involving risks be evaluated using an ethical approach?

Decision-making process

Health care providers can benefit from using a practical decision-making process that is informed by BC legislation, a literature review, and the bioethical principles of respect for autonomy, non-maleficence, beneficence, and justice. The main features of the decision-making process are described below and in the Figure [10]. The timing of steps may be adjusted (e.g., determining patient capacity may occur earlier than suggested), but whatever the timing, each step must be completed before coming to a decision.

The team making the decision will consist of those who work with and know the patient, such as a physician, nurse practitioner, nurse, occupational therapist, social worker, and care aide, and will include the patient or substitute decision-maker and family members as appropriate.

The BC Mental Health Act[8] and Adult Guardianship Act[9] may apply to persons who are at risk to themselves or others. If either Act is applicable to the situation at hand, it should be taken into account. Even when this legislation is relevant, however, the proposed ethical analysis and process may prove useful in any approach taken to address risk.

Evaluating and assessing risk

In the interests of respecting patient autonomy, the process begins by considering the patient’s wishes and how these can be met in relation to the patient’s life context, goals, and values. If a decision is made to override the patient’s wishes, the onus is on health care providers to justify this decision.

The team starts by assessing the risk activity in question and establishing the nature of the possible harm (physical, emotional, or psychological) and the probability and severity of the harm. Harm can be serious, permanent, and likely, or it can be not serious, not permanent, and not likely and can range through various degrees of seriousness, duration, and likelihood.[10] Those assessing risk should be aware of their own personal biases and tolerance for risk, and should use objective and reliable evidence, eschewing speculation and emotion, and ensuring that the activity is actually harmful rather than merely offensive.

After evaluating the risk activity, the team must decide if intervention is required. To trigger intervention, risk should be significant: that is, not a risk that is highly likely but with minor effect or a risk with major effect but so unlikely as to be merely theoretical. The goal is not to remove all risk but to achieve a tolerable level of risk. Risk can never be totally eliminated and all persons choose to live with some degree of risk.[3]

Addressing tolerable and intolerable risk

If risk is deemed to be tolerable, no further action is needed beyond monitoring. If the risk is deemed intolerable and intervention is required, all options should be explored by the team and patient or substitute decision-maker as appropriate, even when some options may seem extraordinary, outside standard budgets, or controversial.

Options should be considered based on whether they satisfy the principles at the heart of an ethical approach.[11] The intervention must:

1. Be effective (i.e., satisfy the principle of respect for autonomy). Ask: Is the goal of the intervention clear, and if it does infringe on the patient’s autonomy will it achieve the goal? For example, will use of bed rails prevent falls or might they be ineffective or even increase risk?
2. Be least intrusive (i.e., satisfy the principle of respect for autonomy). Ask: Are there less intrusive ways to meet the goal? For example, is a wheelchair essential or would a walker be adequate to reduce risk of falls?
3. Not cause greater harm than it prevents (i.e., satisfy the principles of non-maleficence and beneficence). Ask: Are the potential harms of the intervention greater than the potential harms of the risk activity itself? For example, will restricting a resident to a wheelchair to reduce falls lead to depression and lower quality of life?
4. Be nondiscriminatory (i.e., satisfy the principle of justice). Ask: Is the patient being held to a higher standard than similarly situated members of the public? For example, should a patient at high risk of falls who can make capable decisions for herself be allowed to reject the use of assistive devices as would an individual living at home?
5. Be fair (i.e., satisfy the principle of justice). Ask: Should the patient be involved in the discussions about the care plan and if not, why not? For example, should a patient incapable of understanding the risk posed by smoking around oxygen supplies be involved in the discussion?

Establishing patient capacity

If the patient or substitute decision-maker agrees to the proposed intervention, it can be implemented. If the patient rejects the intervention, the team must establish whether the risk activity poses a risk only to the patient or to others as well. If the risk poses harm to others, the risks must be reduced to a tolerable level regardless of patient context or capacity as intolerable risk to others is never acceptable.

If the activity poses a risk solely to the patient, patient capacity needs to be established regarding the activity in question. This involves assessing the patient’s ability to:

• Understand the nature, degree, and consequences of the risk.
• Demonstrate preferences.
• Act free of undue influence.

If the patient is clearly capable or clearly incapable (e.g., has significant cognitive impairment), a formal capacity assessment may not be necessary. If there is uncertainty about capacity, however, the patient’s interest is best served by determining capacity for the particular activity or decision.

Capable patients have the right to make decisions for themselves. Teams are obliged to advise the patient how to reduce risks and may coax, persuade, or possibly offer incentives, but a capable patient must never be forced to accept interventions. If a patient is incapable, the team and substitute decision-makers should consider the patient’s current choices and previous capable wishes and values when making a decision. Whenever possible, the patient should think the intervention is reasonable.

Implementing the decision

Finally, in deciding whether to intervene in a patient’s risk activities, health care providers often worry about legal liability. Completing a rigorous ethical process that comes to a thoughtful, collective decision that can be justified does not eliminate the risk to the patient or the possibility of legal action, but such an analysis is often central to risk management. It is always advisable to check with a physician’s insurer or a facility risk manager with details of a specific case.

The team should agree to follow the care plan so that the patient receives consistent care, and the care plan should be documented and revisited when the patient’s condition changes.

Application of ethical considerations

Despite the complexity involved when choosing the most appropriate intervention, a well-considered, ethically justifiable course of action must be taken. Failing to intervene is unjust because it leaves patients responsible for choices they may not be capable of making or allows health care providers to act on their own biases or fears.

In the fictional case described above, what should be done about Jahal’s wish to go on outings from the facility? Confining an ambulatory patient to a locked unit will almost certainly cause emotional harm, and others may be harmed by a resulting increase in physical aggression. With the aim of honoring Jahal’s wishes, the team should develop a plan for walks that reduces risk of harms to a tolerable level. Options include:

• Taking Jahal’s photograph before she goes out so that police will have this information if needed.
• Providing her with ID, a cellphone, and a GPS bracelet.
• Setting times when the team will call police if she has not returned.
• Giving her supplies to support her well-being, such as a sweater and snacks.

Although Jahal is unlikely to understand the risks posed by her walks and is incapable in this regard, if she accepts the team’s interventions for making the risk tolerable a formal assessment of capacity is not necessarily needed. A trial of these interventions can be undertaken and, if incidents occur, the plan can be revised to further reduce risks.

Summary

When persons living at home or in care facilities choose to engage in activities that put themselves or others at risk of harm, health care providers must find approaches to support both patient autonomy and the safety of patients and others. Living at risk is best addressed by analyzing the risks involved, considering all options available to reduce risks to a tolerable level, and implementing interventions based on the ethical principles of respect for autonomy, non-maleficence, beneficence, and justice. Risk can never be totally eliminated and all persons choose to live with some degree of risk. When health care providers support patients who choose to live at risk, a terrible outcome, including death, may occur. While this must be acknowledged as a risk inherent in patient-centred care, energy should be directed to ensuring that risks of harm are reduced to a tolerable level.

Acknowledgments

We thank Drs Alister Browne, David Rabin, and David Unger for their assistance with this article.

Competing interests

None declared.

This article has been peer reviewed.

Ms Young is an ethicist with Providence Health Care and an adjunct professor in the Department of Occupational Science and Occupational Therapy at the University of British Columbia. Dr Everett is a clinical ethicist with Vancouver Coastal Health and a clinical professor in the Faculty of Medicine at the University of British Columbia.