Every day in a hospital near you, a patient is critically ill, near death, or unable to make or communicate his or her health care decisions, with no one named by the patient to make these decisions under such circumstances.
In 2017, only slightly more than one-third of adults in the United States had advance health care directives. The highly personal decisions made in moments of crisis involve complex medical, social, and ethical factors and determine the quality of life or death of the patient. In those moments, one hopes the choices being made for the patient most closely reflect the choices the patient would make if he or she were able to do so.
Understanding health care decision-making in the absence of an advance health care directive will help elder and special needs law attorneys in two ways. First, people caring for an incapacitated family member without an advance directive will come to these attorneys for advice. Understanding accepted medical practice and standards of care will help attorneys advocate for families and advise them on what to expect. Second, attorneys can use this knowledge to urge all of their clients to create effective advance directives.
This article is based on a 24-month original research project conducted by the American Bar Association (ABA) Commission on Law and Aging, in collaboration with the Society of Critical Care Medicine (SCCM) and Society of Hospital Medicine (SHM), and funded by the Retirement Research Foundation. Researchers studied health care decision-making in critical care settings when patients lacked capacity and had no advance directive. The research team compared default surrogate consent laws with clinical practice. The team also studied how health care providers respond to conflict among family members and how health care providers make decisions for patients who have no readily identifiable family or friends to make heath care decisions for them.
The research project started with a review of existing statutory provisions and a literature review, followed by a survey of health care providers, 527 of whom responded. After the survey, a roundtable of national experts on health care decision-making was conducted. The researchers found deep differences between accepted medical practice and statutory law. These findings, reflecting the reality of what happens on the front line of health care when a patient is critically ill, are expected to impact the advice that elder and special needs law attorneys offer clients.
II. The Research Project
A. The Question
The research team asked, who makes health care decisions if the patient cannot and how should treatment proceed if a patient who lacks capacity has no advance directive? Every state has laws related to delegating health care decision-making, and all but five states have statutory provisions on who can make health care decisions for a patient who lacks capacity. The goal of this research project was to discover what, if any, impact these laws have on health care decision-making in critical care clinical practice.
B. Project Scope and Limitations
The researchers narrowed the focus of the project to patients who lost capacity and had nothing in writing naming a surrogate. The inquiry focused on three distinct legal issues: (1) default health care surrogates, (2) conflict among potential health care decision-makers, and (3) patients in inpatient critical care settings without readily identifiable family or friends to make health care decisions for them. Studying the effectiveness of advance health care directives was beyond the scope of this project. The researchers instead focused on care in hospital and critical care settings — where people are most likely to receive health care during a crisis. The results of a similar study in noncritical care settings may differ from this study’s results.
The survey sample was limited to members of the two medical societies: SCCM and SHM. The respondents were health care providers with a primary focus on critical care, often end-of-life care, of patients with very complex and life-threatening illnesses. The views of this sample may not reflect those of the broader health care provider community. The researchers received 527 responses, about a 1.7 percent response rate; respondents were self-selecting. The nature of the sample and response rate prevent the researchers from making broad projections, and the results are best characterized as the views of the respondents to this survey. The focus was on global questions about health care decision-making, not on differences among individual health care providers; hence, the researchers chose not to focus on the impact of respondent demographics.
1. Health Care Provider
A health care provider is any person who is involved in providing a patient’s medical care or treatment and who usually is involved in seeking patient consent. More than three-fourths of the respondents were physicians (see Table 1). Nurse practitioners and physician assistants made up slightly more than 10 percent of respondents, who understood that ultimate responsibility for obtaining consent may fall to the physician.
Consent is the authorization of the patient, or someone acting on behalf of the patient, given to the health care provider for the provision, withholding, or withdrawal of medical care. For some treatments, such as surgery, consent is formally obtained in writing. For more routine treatment, consent is provided verbally and recorded in the patient’s medical records. Consent may also be implied when the treatment is described and started and no one objects. In emergency circumstances when the patient is unable to communicate and no one is on hand to speak for the patient, consent is inferred for emergency, lifesaving, and life-prolonging care. The researchers specifically excluded emergency inferred consent in the survey design.
3. Do Not Resuscitate and Do Not Attempt Resuscitation Orders
Do not resuscitate (DNR) and do not attempt resuscitation (DNAR) orders are medical orders directing health care providers not to take steps to restart a patient’s breathing or cardiac function if it stops. In some states, DNR and DNAR orders are being replaced by more detailed physician orders for life-sustaining treatment (POLSTs), also known as medical orders for life-sustaining treatment (MOLSTs). POLSTs and MOLSTs contain more specific directions on medical care that a person wants or does not want at the end of life and are intended to be portable — moving with the patient from health care setting to health care setting.
4. Health Care Surrogate
A health care surrogate or proxy is a person who makes health care decisions for another person. A surrogate may be appointed by the patient, may become a default surrogate by law, or may be assumed to have authority based on common law or generally accepted medical practice.
5. Health Care Wishes/Health Care Values/Health Care Goals
The terms wishes, values, and goals were used somewhat interchangeably in the research process. To be more specific “wishes” refer to expressed instructions for health care treatment. Values are strongly held personal beliefs that shape choices or instructions. Health care goals are the desired outcomes of health care treatment.
III. The Law and Health Care Surrogates
The question of who has control over making decisions about medical treatment has been an evolving area of the law since the 1960s. Although not explicit in the U.S. Constitution, the U.S. Supreme Court in Griswold v. Connecticut inferred a person’s right to control health care decisions as part of a penumbra of constitutional rights. Under the principle of informed consent and accepted medical practice, health care providers have long asked family members for consent for the treatment of children and adults who lack capacity. However, concern about liability for lack of consent and shifting family structures left some providers wary of family consent. As a result, many states codified the common law, expanding it with definitions of who could provide consent.
When this project started, 42 states had laws defining who is authorized to make health care decisions for a patient lacking capacity and with nothing in writing. The survey referred to these persons as “default health care surrogates.” Of these states, 38 had a hierarchy statute. Under a hierarchy statute, the highest-ranking person on the list in the statute who is available and willing to provide consent is authorized to make health care decisions. The laws generally authorize the nearest living relative, or next of kin, as the surrogate. The laws vary from state to state on the order of priority of family members and from how far out into the family tree these surrogates can be chosen. The Colorado and Hawaii laws offer a list of suggested persons and ask the persons on that list to decide who should serve as surrogate. Tennessee and West Virginia laws provide a list of potential surrogates and direct health care providers to select a surrogate based on criteria described in the law.
The researchers wanted to know whether these laws influence health care decisions. Therefore, the researchers explored if these laws result in health care providers seeking consent from the surrogate designated in the laws, how providers navigate multiple surrogates, and whether providers are aware of the laws on health care decision-making.
A. Surrogate Selection/Awareness of Default Surrogate Consent Laws
Survey participants were asked to respond to a series of hypothetical questions involving a medical crisis in which a patient is incapable of making his or her own health care decisions. Following is the first question researchers asked:
• A patient, age 87, has a terminal condition, and a decision needs to be made about a DNR/DNAR order. The spouse and adult child are at the hospital. With whom would you discuss treatment options and from whom would you seek consent?
If close family members are available — specifically, the patient’s spouse and adult child — a small majority of respondents (55.4 percent) said they would discuss treatment options with and seek consent from both the spouse and adult child, whereas 44.6 percent said they would select the spouse alone (see Table 2). No one selected the adult child exclusively. The prevailing law for consent in most states designates the spouse to provide consent.
During the roundtable discussion, the researchers asked about the desire to discuss treatment options with the spouse and adult child. The responses fell into two categories: (1) a concern about obtaining consent from a person who knows the patient’s wishes best and (2) a desire to promote agreement and family harmony. Despite the prevailing legal authority of the spouse, the results suggest the practice of seeking group consensus, at least with immediate family.
The next question follows:
• In a similar situation, the spouse and adult child disagree. You tried to reach consensus but were unsuccessful. From whom would you seek consent?
If the spouse and adult child disagree on a course of treatment and consensus is not possible, 91 percent of respondents said they would frequently seek the spouse’s consent, although more than one-third (35 percent) said they would frequently choose “whoever appears to know the patient best” (see Table 3). The most frequent answer shows a clear preference for asking next of kin for consent. The second most frequent answer reflects a concern about making the best decisions based on the patients’ values.
• Also, in a similar situation, conflict exists between the patient’s only adult child and a close friend familiar with the patient’s values. You tried to reach consensus but were unsuccessful. What would you do?
If an irreconcilable conflict exists between an adult child and a close friend of the patient familiar with the patient’s values, a large majority of respondents (80 percent) said they would frequently accept consent from the adult child; only 0.5 percent said they would accept consent from the close friend (see Table 4). However, more than one-third (36 percent) said they would frequently “seek guidance from an ethics committee or consultant.” This predominant reliance on close family versus nonfamily is consistent with the hierarchical preference dictated in the majority of state statutes. In narrative comments and at the roundtable, health care providers expressed a desire to determine who knows the patient’s values best before accepting consent.
To assess survey participants’ awareness of default surrogate consent laws, researchers asked the following question:
• Are you aware of any laws in your state that designate a surrogate decision-maker when there is no advance directive naming a person to make health care decisions and no guardian?
More than two-thirds (68 percent) of respondents reported awareness of default surrogate consent laws (see Table 5). When isolating data from states with such laws, awareness grew slightly to 71 percent of respondents. Surprisingly, when isolating data from states without these laws, 55 percent of respondents thought that their states did have them. Health care providers are more likely to ask next of kin for consent than they are to be aware of laws directing them to do so.
How does this information impact the advice lawyers give to clients?
Not only did the survey reveal that health care providers are most likely to seek consent from next of kin, it also revealed that respondents in states with default surrogate consent laws are more likely to seek consent from next of kin than they are to be aware of any laws directing them to do so. Nonfamily members can expect to be listened to, but when it comes time to ask for consent, health care providers are most likely to ask the nearest next of kin.
If a client is not next of kin, the client will need to make a strong case that he or she best understands the patient. Even in states without default surrogate consent laws, health care providers are most likely to seek consent from the patient’s next of kin. In fact, more than half of the respondents in states without default surrogate consent laws believe that such laws requiring them to consent next of kin exist.
When more than one family member is present, health care providers are likely to talk to everyone, not just the person authorized by law to make health care decisions. Health care providers try to ensure that everyone understands the treatment plan and seek input on what the patient would want. Clients need to be open to input from everyone; doing this builds trust and understanding with health care providers.
If conflict exists among family members or family and nonfamily members, the discussion should include everyone. Health care providers try to determine who best understands the patient’s values. When there is conflict, health care providers may trust the person who shows the greatest level of understanding. When in doubt, providers call on an ethics committee or consultant for guidance. Lawyers and clients should never assume that health care providers know who is next of kin or that health care providers are aware of the laws of the state.
Health care providers are likely to listen to everyone who shows a genuine interest in the patient. Lawyers and clients should not expect health care providers to limit the discussion to the person responsible for granting consent. Limiting discussion in this manner diminishes trust by health care providers. In conflicts among family members and close friends, health care providers are likely to listen to close friends and try to determine who best understands the patient. This may lead to an extended consultation aimed at determining the patient’s wishes, values or health care goals — even if the law is clear in the jurisdiction that the legal next of kin is authorized to give or withhold consent.
B. Conflict Among Multiple Surrogates
A risk of conflict or disagreement also exists among multiple surrogates. When a conflict or disagreement arises among multiple surrogates of different degrees of consanguinity, prevailing law and medical practice resolve in favor of asking the closest next of kin, or highest-ranking person, for consent. Default surrogate consent laws may result in appointment of joint or multiple surrogates of the same class, such as siblings of the patient.
Four basic models of dispute resolution exist in state default surrogate consent laws. As shown in Table 6, 20 states share the most common model, which is majority rule. Majority rule allows the majority of the surrogates of the same class to make health care decisions. Two states refer conflicts to ethics committees, and two states avoid conflicts by asking the health care provider to select a single surrogate based on criteria in the law.
The researchers explored the usefulness of the majority rule by asking this question:
• The situation is similar to the one presented previously (87-year-old with a terminal condition who is incapable of making his or her own health care decisions), but there is no spouse and the conflict is among the three adult children, with two favoring DNR/DNAR and one against. You tried to reach consensus but were unsuccessful. What would you do?
In this situation, in which there is no spouse and an irreconcilable conflict exists among three adult children over a DNR/DNAR decision, only 14 percent of respondents reported that they would frequently accept consent from a majority of the adult children, whereas a majority of respondents (60 percent) favored seeking guidance from an ethics committee or consultant (see Table 7). Twenty states provide that in the case of such disagreements among surrogates of equal level, the physician should accept the decision of the majority. Looking only at the data from the states with a majority rule mechanism, there were 267 responses. Of these, less than 7 percent said that they would frequently accept consent from the majority of the adult children. By contrast, 49 percent said they would never accept consent from the majority of the adult children, indicating that a majority rule approach to dispute resolution among surrogates is not a generally accepted clinical practice among the respondents.
It is clear that majority rule is largely rejected by health care providers as a solution to family conflict. The narrative comments on this question and input at the roundtable offer insight into this. Providers are concerned about agreement among family members for two reasons: (1) fear of liability or litigation and (2) a desire to promote family harmony. One commentator explained that taking sides breaks the family in ways that may never heal. Others said that they want the family to survive the death of the patient. Even when the law offers a safe harbor to providers accepting consent from the majority, about 90 percent of providers who responded to the survey decline to do so. Feedback in the narrative comments and at the roundtable express the desire to seek consensus and, if not consensus, at least an understanding of the prognosis by all involved and the reasons the choice is being made.
Implications of this information for lawyers counseling clients
If a patient who lacks capacity lives in a state with majority rule and the lawyer’s client would be one or more of the default health care surrogates, the lawyer can offer two options. One is to seek conflict resolution, seeking common ground and understanding with other family members. The other is the adversarial option of seeking to enforce the law. If a lawyer is working with the majority of the class of surrogates, he or she can seek to enforce the majority rule. If the lawyer is representing the minority of the class, he or she can work to convince a court that majority rule will result in an unwanted outcome or an outcome not in the best interests of the patient. Many lawyers tend to default to the adversarial model of proving that the client is right, or that the other party is wrong. Looking at this issue through the clinicians’ model of seeking family harmony, attorneys can offer conflict resolution as a tool for preserving or healing the family. The success of conflict resolution varies based on how far apart the parties are, how deeply entrenched they are, and the level of trust among them. Under the ABA Model Rules of Professional Conduct, the client decides on the objectives of the representation. Options to offer would include family harmony or a purely legal solution.
Similar to health care providers, lawyers can make clients aware of the long-term impact of legal solutions on relationships and family. Although some families are so divided that there is no option short of litigation, other families are divided by a lack of trust and communication and, in the role of counselor-at-law, attorneys can help those families resolve conflict and heal.
Going beyond the scope of this research project, when engaged in advance care planning with a client who still has capacity, it is important to ask about potential or active conflicts among family members. Conflicts that surface after the client is no longer able to be a part of the discussion can be difficult to resolve. If the differences among family members are not great, a facilitated family meeting or family counseling could resolve the issues. If the differences are great or the sides are deeply entrenched, strong advance care planning with instructions to resort to court action if necessary may be the only way to ensure that the client’s wishes are honored.
C. Patients Without Readily Identifiable Family or Friends
The “unbefriended,” patients without readily identifiable family or friends, challenge both the law and health care providers. This research was started using the term “unbefriended,” but in testing the survey instrument with health care providers, the researchers rapidly learned two things. Most providers did not know what “unbefriended” meant, and most of those who did found the term offensive. For purposes of the survey, the researchers used “patients without readily identifiable family or friends.” This topic was the subject of a lengthy discussion during the roundtable portion of this project. Even though a couple of people argued that the term “unbefriended” had passed into common usage, most agreed that the term makes it sound as if the unbefriended are unloved — when in reality many have simply outlived family and friends. Roundtable participants settled on the term “unrepresented.”
Persons without readily identifiable family or friends represent a small percentage of the overall population. In 1998–2010, an estimated 6.6 percent of U.S. adults (8.3 million) age 55 and older had no spouse or children; 1 percent (835,000) had no spouse, children, siblings, or parents. According to the Alzheimer’s Association, “An estimated 70 percent of older adults with Alzheimer’s or other dementias live in the community, compared with 98 percent of older adults without Alzheimer’s or other dementias. Of those with dementia who live in the community, 74 percent live with someone and the remaining 26 percent live alone.”
When the research team asked health care providers how many times in the past 3 months they encountered patients without readily identifiable family or friends, more than half reported seeing one to five such patients (see Table 8).
As presented in Table 9, 13 states have statutes dealing with health care decision-making when a patient has no readily identifiable family or friends. Among those 13 states, six different statutory models exist for decision-making. The most common statutory model is for health care decisions to be made by the attending physician in consultation with an ethics committee or second physician.
The researchers asked the following question to elicit health care providers’ responses to situations involving patients without readily identifiable family or friends:
• Same case, there are no relatives or friends willing to discuss treatment options or give consent. What would you do? (See Table 10.)
This data shows no clear answer. More than half the respondents would frequently seek a second physician’s opinion (55 percent) or seek guidance from an ethics committee or consultant (58 percent). Close behind, 45 percent would frequently seek guidance from the hospital risk management or legal department. The option of going to court for a guardianship was frequently chosen by a substantial minority (39 percent), whereas the option of making a decision by oneself was frequently chosen by only 10.5 percent of respondents. When isolating data from North Carolina and Oregon, the two states that allow the treating physician alone to make health care decisions for patients who have no one to speak for them, only 7 percent said that they would frequently do so.
During the roundtable, health care providers cited the time needed as an obstacle to locating family and friends for patients unable to communicate and with no one to speak for them. The cost and time needed for guardianship was often cited as an obstacle to finding someone to provide consent for these patients.
If health care providers are unable to locate family or close friends, a stranger makes health care decisions for these critically ill patients. Faced with such a situation, most of those responding to the survey said that they would use the laws or turn to other physicians, ethics professionals, or risk management to make decisions.
In narrative comments on the survey and during the roundtable, health care providers spoke of the resources consumed in providing care for patients who have no one to speak for them. While effort is expended trying to find family or friends, medical care is provided without knowing what the person would or would not want. Decision-making for these patients involves multiple consultations. Some medical providers go beyond the expected effort and send staff into the community searching for someone to speak for the patient. In some areas, seeking guardianship takes months and costs thousands of dollars as the patient is being cared for. Several providers reported difficulty transferring patients from hospitals to skilled nursing homes, sometimes waiting months for a guardianship order to approve the transfer.
How does this information impact lawyers?
When an attorney preparing an advance directive or health care power of attorney asks a client whom he or she wants to name as a health care surrogate and the client does not have a clear answer, the client is at risk. Even if the client is able to name someone, lawyers should inquire further by asking about the client’s contact with the named agent. How frequently do the patient and the agent talk? How often do they see each other? How well does the proposed surrogate know the client’s values? If the answers to these questions are “not frequently,” “not often,” and “not well,” respectively, the client is at risk of naming an agent who at best will have little basis on which to make decisions and at worst will be unavailable or decline to help when needed.
What does an attorney do to help a client who has no one to name? A starting point is helping the client document his or her health care values. Clients with the ability to pay can hire professionals to assist with health care decisions. Clients who are unable or unwilling to pay for professional help can be encouraged to share their values with health care providers, neighbors, and others who might be called on for help making health care decisions when needed.
This research shows that (a) health care providers who care for individuals during medical crises generally turn to the same people the laws suggest as default health care surrogates, (b) disputes between surrogates or between family members are not well addressed by statutory attempts, and (c) persons without readily identifiable family or friends to serve as surrogates present a great challenge in law and medicine.
Health care providers are more likely to ask next of kin for consent than they are to be aware of laws directing them to do so. Even in states without such laws, accepted medical practice dictates asking family members for consent. If a client’s wishes are outside the normal assumption of asking next of kin for consent, the client needs to put these wishes in writing and may need the attorney’s help explaining the document and the law to health care providers.
Health care providers dislike family disputes and are much more likely to try to mediate, explain, or seek common ground, consensus, or understanding among family members than they are to be aware of, or to take refuge in, laws offering a clear answer to family disputes about health care decision-making.
Unrepresented clients, or clients without readily identifiable family or friends, are the greatest challenge in the law, with little agreement among the handful of states that have attempted to address this challenge. These patients present a great ethical dilemma for health care providers, with little agreement on how to best make decisions for them. Lawyers can play a role in helping identify clients at risk and suggesting steps — such as leaving instructions for potential health care decisions.