Avoiding paternalism in medical interpretation for informed consent for deaf patients: Incorporating respect and relational autonomy

Case

A patient visits the Emergency Department (ED) because they experience pain in their lower abdomen. The patient is a 50-year-old man, Edward, who is Deaf and communicates primarily with ASL. When he enters the ED, he has difficulty describing his pain to the in-take nurse because an interpreter is not yet available. Due to his pain, the on-call physician decides to admit the patient to the hospital instead of waiting for an interpreter to become available. After three days in the hospital, the patient has not yet had the opportunity to speak with another person who uses ASL. Finally, on the third day, an interpreter becomes available and begins communicating with the patient. The patient begins communication with the interpreter to explain their symptoms. The interpreter and the patient communicate well together. Later, the patient is screened during intake and must offer a family medical history to rule out some of the possible risks. The patient, however, is unable to answer many of these questions. Next, the on-call physician states that the best way to address the abdominal pain is to have surgery as soon as possible. As the clinician communicates, he primarily faces the interpreter, not the patient. The clinician is also wearing a mask to prevent the spread of the current respiratory illness.

The patient is exacerbated by the pain and lack of communication from not being able to see the clinician's face. However, the next step in the patient's treatment plan is to provide consent for the procedure. While the interpreter has shared technical information with the patient, the patient seems uneasy with the clinician's recommendations. The clinician leaves the room briefly to visit another patient, leaving the interpreter and the patient alone. The patient asks the interpreter for their advice regarding the surgery. The interpreter cannot offer medical advice. However, the interpreter understands the advice the clinician provided, and the interpreter believes that if they support the clinician's recommendation, the patient will likely have the procedure. Should the interpreter support the clinician's recommendation to the patient? Does this support the autonomy of the patient, or does it undermine the autonomy of the patient through paternalistic intervention?

Discussion

To answer these questions, we should first identify the necessary components of informed consent. In this case, the patient does not experience diminished autonomy. He has the capacity to receive and understand relevant health care information. Their case is distinct because there is a communication barrier that may undermine their opportunity to consent.

Communication barriers are addressed legally through the Americans with Disabilities Act (ADA). In scenarios where informed consent is required, an interpreter can be made available. In the United States, the ADA protects health care interactions by stating that “effective communication” ¹⁰ must be provided which may vary in format, length, complexity, and context. ¹⁰ When effective communication occurs it “reduces the occurrence of patient safety events, facilitates timely diagnosis and treatment, and enables high-quality discharge.” ¹¹ Interpreters are not always available, but they can be made available in the case of informed consent. The interpreter's role is to facilitate communication between the patient and the clinician about the medical diagnosis, medical history, procedures, follow-up treatments, and take home directions. ¹⁰ Even though this service is available, this does not always remove barriers to health care access or aid in acquiring informed consent.

Informed consent consists of two components: sharing relevant information and the patient understanding this information. Philosophers and legal scholars may have different ways of classifying the different components of informed consent. Faden and Beauchamp identify informed consent as consisting of five elements: “(1) disclosure, (2) understanding, (3) voluntariness, (4) competence, and (5) consent.” ¹ Debates on what criteria are required to meet each of these elements persist in the literature. However, this is a useful way of categorizing the elements of informed consent as both a philosophical and legal concept.

Millum and Bromwich describe the “standard view” of informed consent as consisting of two parts: information disclosure for the person giving consent, and consent is met when the disclosed information is understood. They identify these as the disclosure and understanding requirements. Without both, consent is not provided. Under the standard view, disclosure is understood as the basis of the understanding requirement. They, however, reject this relationship between the disclosure and understanding requirements in the common view, and instead make the argument that disclosure and understanding require two distinct grounds. To meet the disclosure requirement, the person providing the information must provide information that (1) “she thinks is relevant to the consent decision” and (2) “that the profferer of consent would reasonably be expected to receive.” ¹² The understanding requirement is met when “the person proffering consent must understand three things: (1) that she is giving consent; (2) how to exercise her right to give or refuse consent; and (3) to what she is being asked to consent (Millum and Bromwich, 2018).” ¹³ I will use this definition of informed consent in the case of Deaf, deaf, hard of hearing or deaf-blind patients in this case study.

However, like in most discussions of informed consent, the role is often reduced to two distinct actors: the provider and the patient. The medical care team may represent multiple views, but the goal is the same, to achieve informed consent through the signing of a legal form. While the goal of another type of understanding, autonomous authorization ¹ is desirable, it may not always happen. Informed consent can be understood as autonomous authorization ¹ where informed consent allows a medical intervention to occur due to the authorization of the patient beyond “merely acquiescing in, yielding to, or complying with an arrangement” but instead the patient has “substantial understanding and in absence of control by others, intentionally authorizes a health or research professional to do something.” ¹ For this type of authorization to occur, the patient must meet the understanding requirements from Bromwich and Millum.^12,13 To make these types of decisions both competence and decision-making processes must be possible. While all these conditions must be met for informed consent to occur, the role of the interpreter has not yet been addressed from an ethical analysis perspective.

Faden and Beauchamp define competence and voluntariness as preconditions for the other elements of informed consent. Competence is required for both understanding and decision-making. ¹ If the patient cannot understand the nature of the procedure, or decide using the available facts, then they cannot provide valid consent. Views of the competence requirement for decision-making for individuals who are Deaf, deaf, or hard of hearing in the medical community can be biased, outdated, and inaccurate. Physicians may also feel uncomfortable treating patients, or respect patients’ “intelligence, motivation and desire to understand and participate in their health care.” ¹⁴ Physicians may not always exhibit direct bias but may be unaware of the legal requirements to provide interpreters for patients.¹⁵ Communication barriers arise between patients and clinicians because many do not realize the difficulty of speech reading (also known as lip-reading), where approximately only 30% of words can be effectively lip-read. ³

Although there are legal protections to aid in communication, Deaf persons, and individuals who are deaf, hard of hearing, or deaf-blind experience multiple barriers to health care and risks to their health status. Barriers to health care access are not limited to the Deaf community and individuals who are deaf or hard of hearing. However, barriers to health care impact individuals who are Deaf, deaf, and hard of hearing in different ways. For example, individuals who are Deaf and use ASL are at elevated risks for “obesity, partner violence, and suicide.” ¹⁵ Barriers to health care access include fewer visits to the doctor, ¹¹ lower rates of preventative screenings, ¹⁶ miscommunication of health care information ¹⁷ and higher health care costs for middle-aged to older adults. ¹⁸ Additionally, discrimination in the form of audism limits access to many social determinants of health such as education and fair employment. ¹⁹

In terms of relationships between the provider and patient, Deaf or hard of hearing persons may experience high levels of anxiety due to fear of being misunderstood or not having their health care needs met. One study identified barriers at many of the various stages of the health care visit, beginning with waiting in the office, lack of consensus on modes of communicating between health care staff and patient, lack of understanding of the procedure being performed, and problems with follow-up after the appointment. ¹⁴ One of the important barriers to care stems from lack of or failure to communicate. While not all individuals may want an interpreter, there is an unmet need for those who may. In a 2004 study on Medicare recipients, individuals with a hearing impairment reported being dissatisfied with the quality of health care provided due to inaccessibility of location, difficulty having questions answered about medications over the phone, and level of care provided by their provider. ²⁰ These barriers led to feelings of frustration for both the providers and the patient. ²⁰

The role of the interpreter can fall into four different types: helper, conduit, communication facilitator, and bilingual, bicultural specialist. ²¹ The helper model existed prior to formal organizations to train interpreters where individuals depended on friends and family. The conduit model treats the interpreter as merely passing information back and forth. The latest model “considers situational and cultural factors as relevant to the interpreting task.” ²¹ The implementation of these styles of interpretation varies depending on the interpreter. And while interpreters may not want to impact the “form, content, structure and outcomes of interactive discourse, but the reality is that interpreters, by their very presence do.” ²¹ If the interpreter cannot be truly neutral, then the question is what role should they play, if any, in removing barriers to health care and in the process of achieving informed consent?

The first question is what barriers persist that make informed consent difficult for persons who are Deaf, deaf, hard of hearing, or deaf-blind. Is there a potential violation of the autonomy of the patient in these situations? Again, in this case, there is not a question of capability to consent. A harmful stereotype about people with variations in hearing is that they lack the capacity to understand which I directly dismiss. Instead, I am addressing the more common concern of how to ensure that the information and the conditions for informed consent are possible.

In Edward's case, he experienced delays in care, isolation, frustration, fear, and ultimately distrust of the health care system due to these delays. This distrust may make it difficult to attain his true informed consent. Due to his experiences, Edward invited the interpreter to serve as more than a mere conduit of information (even though this may be at odds with current professional and legal standards). In a code of conduct for interpreters endorsed by the National Association of the Deaf (NAD) and the Registry of Interpreters for the Deaf, Inc. (RID), it clearly states that interpreters should “refrain from providing counsel, advice, or personal opinions.” ²² Their philosophy is that “members of the American Deaf community have the right to informed choice” ²⁰ and one of the central guiding principles is to “do no harm.” ²² In terms of the messages delivered, interpreters are expected to “render the message faithfully by conveying the content and spirit of what is being communicated […].” ²² What may be abstracted from this professional organization is that the primary role of the interpreter is to convey medical information. However, in the case of informed consent, what are the additional requirements from the interpreter that must be met that are not explicitly stated but also do not undermine these professional standards?

In some ways in this case study, the interpreter is asked to play a role that lies between being a concerned family member or a patient advocate. Except in Edward's case, he does not require a third party to make decisions on his behalf. Instead, he relies on a third party to make sure that the decision and understanding criteria are met. If the interpreter makes sure that these two criteria are met, then it does not seem that the interpreter has acted paternalistically. Paternalism is understood as “the intentional overriding of one person’s known preferences or actions by another person, where the person who overrides justifies the action by the goal of benefiting or avoiding harm to the person whose preferences or actions are overridden.” ² If the request (1) comes from the patient and (2) the request is one in which the patient uses the interpreter as another piece of information to ultimately make their decision, then the interpreter is not necessarily acting paternalistically. However, there is a high risk of paternalistic intervention from the interpreter because of the communication role that they have. Acting paternalistically could easily occur in this situation, and the clinician and the patient may not be aware. Due to this risk, the interpreter shares in a set of special obligations to both the patient and the clinician. I argue that the special obligations are to do what is required to meet the disclosure and understanding criteria of informed consent appropriately.

Additionally, the special obligation to do what is required to make the conditions possible for the disclosure and understanding criteria of informed consent also extends to supporting the conditions necessary for shared decision-making. While there are various interpretations regarding the scope of shared decision-making, the general idea of shared decision-making involves the participation of the patient alongside the physician. However, shared decision-making can also include other members of the care team, patient advocates, or other individuals. As a counter to medical paternalism, shared decision-making emphasizes the importance of patient values and preferences. Entwistle and Watt (2016) offer an analysis of shared decision-making which describes it as either broad or narrow. Under narrow interpretations, shared decision-making may involve simply an exchange of information. Under broader interpretations, shared decision-making can involve a collaborative approach to decision-making, which considers the complexity of social relationships. ²³ In my view, neither narrow nor broad accounts of shared decision-making could be possible without first making the conditions for informed decision-making possible. While I will not defend either one of these views as more preferable, it is likely that the broad shared decision-making account is most compatible with the assessment brought forward in this case study. The barriers which prevent these two conditions for informed consent from occurring would also be barriers to shared decision-making.

The second question asks, if the interpreter identifies a barrier to care that they may be uniquely qualified to intervene on, is this an act of paternalistic intervention? If so, does this paternalistic intervention undermine the autonomy of the patient? In Edwards’ case, the interpreter is the only person that Edward directly communicated with and expressed his frustration with his treatment. The interpreter who has worked for many years in this position has witnessed barriers to care for Deaf patients in the past. When Edward asks the interpreter for their advice, the interpreter understands that what he communicates will likely impact Edward's decision. Specifically in this case, the interpreter understands what the clinician communicated, and that the patient's well-being will likely be improved with this intervention. There are two ways that the interpreter could act paternalistically through a violation of the disclosure and understanding requirement, although there are likely other ways to act paternalistically not included in this discussion.

First, if the interpreter withholds information that he thinks is relevant to the consent decision, this may limit the type of information shared. For example, if the clinician phrases the procedure in a certain way, and the translator re-phrases the description of the procedure, they may unknowingly exclude information that the patient may think is relevant. Additionally, if the interpreter excludes information that the patient would reasonably expect to receive, this would be problematic. For example, if the interpreter does not explain all the risks involved that the clinician states, this would be a clear failure.

In the case of the understanding requirement, if the interpreter does not make it clear that Edward can give consent or refuse the procedure with the understanding of what the procedure and risks of the procedure are, this requirement is not met either. The interpreter is in a unique position because they may be able to identify a lack of understanding of these options in a way that the clinician cannot due to language barriers.

There are several examples in the case study where barriers to care occurred due to communication barriers. First, the delay in communication for three days and transfer from facilities is one example. Another example is that the clinician did not direct their gaze towards the patient, and instead looked towards the interpreter resulting in Edward feeling isolated. Third, the limited options provided to the patient for treatment were rooted in biased assumptions about the competency of the patient.

The physician can improve the conditions for informed consent by remembering to face the patient, not the interpreter, during their visit. In one of the first of its kind of studies, Nicodemus et al. (2021) studied the role of trust between physicians, medical interpreters, and patients. In their survey of 811 bilingual English and ASL patients, they identified strategies for physicians to increase trust. One strategy is to employ a “teach-back” method when providing medical information to patients. ²⁴ This way the physician can ensure that the information is understood by the patient and communicated by the interpreter. Unfortunately, a shortage of interpreters is a reality in the American health care system. Additionally, many interpreters who work as medical translators are trained as generalists and not specialists. ²⁴ This may add an additional barrier to communication of complex medical information. However, once present and appropriately trained, interpreters can play a critical role in reducing communication barriers for persons who are Deaf, deaf, hard of hearing or deaf-blind. However, traditional views on autonomy as they relate to informed consent may not justify the role of the interpreter making recommendations to the patient which transitions us into the third question.

The third question asks how the autonomy of the patient best be protected in attaining informed consent. One account of autonomy promoted by Childress and Beauchamp understands autonomy as “self-rule that is free from both controlling interference by others and from limitations, such as inadequate understanding, that prevent meaningful choice.”² However, critics of this definition of autonomy argue that a weakness of these accounts is that they view individuals as solely responsible for setting and achieving their own goals, instead of identifying the social context through which decisions are made. ²⁵ Under a version of the relational autonomy account, it disputes an assumption that patients are on “equal footing” with the clinicians. Only under idealized assumptions, patients and clinicians could be considered on equal footing. ²⁶ Due to barriers in communication about the patient's health, Edward may not be able to achieve more idealized versions of consent because of these social barriers.

Relational autonomy is a distinct view because it identifies how oppression limits autonomous action. Oppression is understood as “function[ing] in complex and often largely indivisible ways, affecting whole social groups rather than simply disrupting individuals” ⁷ and focuses on how autonomy is limited through “social and political structures.” ⁷ In Edward's case, he experienced these barriers to care because of existing social and political structures. The lack of funding for and training of qualified interpreters when requested is a social limitation. The lack of training of the clinician to directly address the patient instead of the interpreter is another. The differences in health care access and health status I ask in my first question is also explained by these types of barriers. As such, if the interpreter is trained to recognize the existence and persistence of these barriers, they may be in a unique position to support the autonomy of the patient. Others have argued for improved training to meet the needs of Deaf persons Queer and Transgender Persons of Color ²⁷ in education which could be extended to the training of medical interpreters.

In addition to these accounts, relational autonomy places an emphasis on the embedded nature of social relationships and social context which can be applied to decision-making in health care, although Beauchamp and Childress do not necessarily view their definition as being at odds with relational autonomy accounts. ² Relational autonomy can include the role of the interpreter as more than a mere conduit of information without opening the interpreter up to the danger of paternalistic action or overstepping their professional role.

The role of the interpreter, as informed by the answer to the second question as well, is to support the autonomous decision-making of the patient by making sure that the disclosure and understanding requirements are met. However, unlike other types of informed consent decisions, where information is directly disclosed from the treating clinician or researcher, the interpreter when invited to participate at the request of the patient can be part of this information sharing process. In this paper, I do not argue that interpreters must always be present, but only insofar as the patient requests them. As such, the role of relational autonomy could support an understanding of their role.