Palliative Care for Latino Patients and Their Families: Title and subTitle BreakWhenever We Prayed, She Wept

Ms C, a young woman from Central America who spoke only Spanish, initially presented in December 2006, during her 34th week of pregnancy, with marked leukocytosis and was diagnosed with leukemia. After delivering a healthy full-term infant daughter, she achieved remission with chemotherapy. However, 6 months later, a repeat bone marrow biopsy revealed a relapse. Ms C was transferred to a university hospital in 2007 where she was treated with a second round of chemotherapy. A month later, a bone marrow biopsy showed persistent disease. In discussions with Ms C and her partner, Mr M (the infant's father), the attending oncologist stated that there were no further curative treatment options.

Ms C was the oldest of many siblings, all of whom lived in a rural part of Central America. Prior to her hospitalization, the patient lived with her infant daughter, Mr M, and his family. Ms C and Mr M were evangelical Christians and were undocumented immigrants. Mr M worked at a blue-collar job early in the day and took over child care duties while Ms C worked as a cook. Ms C was often deferential to her partner, allowing him to make medical decisions for her, although she had not officially designated him (or anyone else) as her health care proxy. Despite intensive efforts by the hospital social workers, it was not possible to obtain emergency visas for family in Central America to visit Ms C.

Shortly after relapse, after private conversations with the oncology team, the patient agreed to a do-not-resuscitate–do-not-intubate (DNR/DNI) order. At the time of the palliative care service (PCS) consultation on the following day, the patient had been in respiratory distress for several days but was still able to discuss goals of care and make medical decisions. Discussion with the PCS in Spanish clarified that Ms C's primary goal was to spend as much time as possible with her daughter in her remaining weeks, which would not be feasible were she transferred to the intensive care unit. Mr M, however, requested that “everything be done,” saying, “We continue to hope and pray for a miracle.” In subsequent meetings, the PCS consultants reinforced the oncologist's assessment that curative options had been exhausted and encouraged her partner to begin his good-byes. The PCS team was less and less able to engage Ms C in discussions due to somnolence, and Mr M wished only to discuss further curative treatment options and declined to consider discussions about end-of-life care plans. Mr M adamantly maintained that there was still hope for her recovery and continued to inquire about therapeutic options such as “surgery for her lungs” and “washing her blood.”

Despite supplemental oxygen and escalating doses of morphine and lorazepam, the patient's dyspnea worsened. Early in the morning a week after the initial PCS consult, a nurse noted that the patient had stopped breathing. When the oncology team arrived to pronounce her dead, they found Mr M trying to revive her by performing mouth-to-mouth resuscitation.

The palliative care attending, the medical student, the social worker, and the chaplain were interviewed by a Perspectives editor 5 months after the patient's death.

PALLIATIVE CARE ATTENDING: This was an incredibly difficult case. When we first went to see this patient . . . it was clear that she was profoundly dyspneic and uncomfortable. She only spoke Spanish, and I don't speak Spanish. We were lucky enough to have a resident who was Latino, so he and the medical student, who also spoke Spanish, translated. The first thing that we tried to address was her dyspnea. . . . Her husband was very concerned that she would not be able to talk to him [if she were treated with opioids]. He really believed that she would get better and there was very little opening to even consider that she wouldn't get better. There were no holes that I could find to even bring it up.

MEDICAL STUDENT: He was adamantly maintaining that a miracle was possible and that anything could happen. This contributed to his inability to accept what the DNR/DNI meant.

SOCIAL WORKER: He was very suspicious of any questions about her [immigration] status. He felt that we were asking questions that were inappropriate because he was really only interested in treatment options. If we asked anything that wasn't directly addressing those, he would question why we were asking.

CHAPLAIN: In my own personal notes about this case, I had written that this was a tragic case. . . . I remember her infant daughter lying on the bed with her. It was a very stark image. My notes read, “Whenever we prayed, she wept.”

A young pregnant woman develops a fatal cancer and dies before her daughter reaches her first birthday. Her partner cannot bear the thought of her dying. The suffering experienced by this patient, her partner, and the health care team suffuses all of the care providers' descriptions of this case. The deep sadness here is universal.

What is also clear is the profound influence of culture and immigration on this young woman's experience at the end of her life and on that of her partner. Below we discuss ways in which Latino culture can influence the experience of terminal illness. Unfortunately, research directed at understanding and improving end-of-life care in the Latino community is relatively limited. Most studies of racial and ethnic difference in end-of-life care have focused on differences between African Americans and whites, and studies that include Latinos are often underpowered to find differences.^{1 - 3} Despite these limitations, we cite the best available evidence, sharing our own experiences and suggestions where data do not exist.

Available evidence about rates of advance care planning in the Latino community is mixed. Rates of hospice use among Latinos appear to be similar to whites. In a recent multicenter study of patients with advanced cancer, only 29 of 62 of Latino patients (47%) vs 249 of 312 of white patients (80%) had an advance care plan.⁴ However, in a study of 239 elderly white and 224 elderly Latino patients in New York City, only 35% had completed a health care proxy and rates did not differ by ethnicity.⁵ A third study of 40 930 elderly patients with advanced cancer found that Latinos had slightly lower rates of hospice use compared with whites (38% vs 42%).⁶

We first describe the diversity within the Latino community and discuss the influence of the immigrant experience and then focus on communication issues including language discordance, use of interpreters, and health literacy. Next, we discuss Latino cultural norms and beliefs that may influence end-of-life decision making, appreciating the importance of religion and spirituality, and exploring experiences of discrimination. Finally, we provide concrete suggestions to assist clinicians in using interpreters, fostering clear health communication, promoting trustworthiness, and understanding cultural interpretations of illness. Our goal is to educate clinicians of every specialty to improve culturally competent care of seriously ill Latinos and their families.

CHAPLAIN: I don't think that there's any question that we could have done a better job [from a cultural perspective]. . . .We're talking about foundational beliefs and about comfort—not the medical comfort, but comfort with the space, the presence, the people. Think about if we were receiving our end-of-life care in [Latin America], surrounded by Spanish-speaking people. . . .

Latinos are the largest ethnic minority group in the United States, currently accounting for 15% of the US population, a proportion that is projected to grow to 30% by the year 2050.⁷ Latin American culture is diverse and influenced by national origin and social class. Latin America today is a product of 500 years of population and cultural admixture of the indigenous peoples, the European colonialists and immigrants, and the forced migration and enslavement of Africans.⁸ Within the Latino community in the United States, there is considerable variation in health care access and utilization patterns by country of origin, English-language proficiency, insurance status, and length of time in the United States, emphasizing the heterogeneity of the population, much like any other racial or ethnic group.⁹ However, the predominant use of the Spanish language and shared cultural themes create definite similarities among Latinos in the United States.¹⁰ Clinically, making assumptions about individuals based on ethnic and cultural identity risks stereotyping. However, ignoring the profound influence of culture and social situation on patients' experiences of illness, expectations of medical interventions, communication styles, and ways of coping can lead to misunderstanding, conflict, anger, resentment, and lower quality of care.^{1 ,11}

CHAPLAIN: They seemed fairly isolated. I don't recall anybody other than the partner being present when I was there. . . . I know that there was a lot of distress about the future for this baby. The patient and her partner weren't married. I'm not sure about their immigration status. . . . I would assume that there were a lot of fears in play around that. . . . 

For many Latinos in the United States, the experience of illness is influenced by their experience as immigrants. More than half of immigrants to the United States are from Latin America, and 40% of Latinos residing in the United States are immigrants.¹² In 2000, there were 16 million immigrants to the United States from Latin America: 11 million from Central America and Mexico, 3 million from the Caribbean, and 2 million from South America.¹² Quiz Ref IDAccording to the Pew Hispanic Center, 57% of undocumented immigrants are from Mexico, and 24% from elsewhere in Latin America.¹³ Research on the experiences of undocumented Latinos is limited and is challenged by difficulties identifying and sampling participants due to fear of deportation.¹⁴

Fear of deportation likely contributed to conflict between the PCS and Mr M. Deportation is a constant reality for undocumented Latinos that often precludes access to health care and may lead to negative perceptions of the health care environment.^{14 - 16} Recent, well-publicized cases of “hospital deportation” may further exacerbate fears among undocumented immigrants.¹⁷ Rates of uninsurance are high among undocumented immigrants (59% in 2004 according to a recent report by the Urban Institute).¹⁸ The Federal Personal Responsibility and Work Opportunity Reconciliation (Welfare Reform) Act of 1996 limits federal funding for health services for undocumented immigrants to emergency services (under Medicaid).¹⁹ The degree of support for nonemergency services is limited and varies by state. Quiz Ref IDAlthough government agencies are required to verify immigration status in determining Medicaid eligibility, health care professionals, including social workers and interpreters, are not legally mandated to report undocumented immigrants to US Citizenship and Immigration Services (USCIS, formerly the INS).²⁰ Because all immigrants are eligible for nonfederally funded health benefits, hospitals and health care workers should not attempt to verify patients' citizenship status as a routine practice for providing care. Proposition 187, a voter-approved California ballot initiative later overturned in the courts, would have mandated that health care providers report undocumented immigrants to immigration authorities.²¹

From the patient's perspective, inquiring about immigration status should only be done when the benefits outweigh the risks. In this case, it could have caused a fear of deportation, a sense of discrimination, diminished trust, and worsening of communication barriers. However, awareness of immigration status could have been used in planning for future care of the infant, might have helped bring into the open Mr M's distrust and the fears, and would have directly addressed Mr M's perception that curative care was being withheld due to their undocumented status. In cases such as this, clinicians might assume that the patient is undocumented unless the patient volunteers information otherwise. Social workers may be best positioned to ask about immigration status so as to separate clinical care from the immigration issue. In this case, a social worker discovered that both the patient and her partner were undocumented immigrants, but this information was not communicated to the PCS teams. If clinicians do choose to ask, they might begin by stating, “I want to ask you about your immigration status. I am not asking for legal reasons and anything you say will not be reported by me or other care providers to the immigration authorities. I ask because I care about you and worry that you might have fears or concerns about how your immigration status is affecting the care you (or your loved one) are receiving.” Involving local legal and social support groups and seeking out more information resources may have helped illuminate issues concerning custody of the child and educating the medical team about relevant cultural issues (see Web Resources at the end of this article).²²

MEDICAL STUDENT: [The partner’s] English was better [than the patient’s]. We could kind of talk in “Spanglish.” . . . I think we could have done better [in terms of cultural sensitivity]. For instance, I think we could have used a professional translator all the time.

Thirty-two million of the estimated 44 million Latinos residing in the United States speak Spanish at home.²³ English proficiency varies by generational status; in a recent Pew Hispanic Center study of 14 057 Latino adults, 23% of immigrant first-generation Latinos reported being able to speak English very well compared with 88% among the second generation, and 94% among later generations.²⁴ Errors in understanding are common when clinicians attempt to communicate in English with Spanish-speaking patients with limited English proficiency.²⁵ Even when Latinos do not understand, they may signal understanding to appear cooperative and not defy the authority figure.^{26 - 28}

The number of Spanish-speaking health professionals is growing, and some evidence suggests that Spanish-speaking patients give higher ratings of care to Spanish-speaking clinicians.^{29 - 30} However, health professionals' level of fluency in Spanish varies greatly. Language fluency is not a yes-or-no skill but a spectrum that includes facility of language, familiarity with medical terminology, and understanding of cultural context. An imbalance between speaking ability and understanding may lead to situations in which clinicians can ask simple questions in Spanish, but not understand complex responses.^{31 - 32} Using ad hoc interpreters, such as bilingual staff or adult family members, may be expedient, but is also potentially problematic. In a recent qualitative study, professional interpreters made fewer significant errors than were made by ad hoc interpreters (53% vs 77%, P < .0001).³³ In addition to important errors, clinicians may be concerned that ad hoc interpreters, particularly family members, may alter the interpreted content to suit their own beliefs or otherwise affect the patient's responses. Title VI of the 1964 Civil Rights Act provides the basis for a legal right to an interpreter, a principle reinforced by the courts and President Clinton's Executive Order 13 166.³⁴ Quiz Ref IDUnfortunately, many patients with limited English proficiency are unaware of their legal right to a trained medical interpreter.³⁴ Therefore, Latino patients should be informed about their legal right to an interpreter at no cost and be offered one if desired.³¹ Health care professionals do not need to obtain informed consent from patients prior to disclosing protected health information to interpreters.³⁵

Although not realistic that all encounters with patients who have limited English proficiency be facilitated by a professional interpreter, there are particularly sensitive, complicated, or important conversations during which professional interpretation should be mandatory. These include conversations about goals and preferences for care at the end of life or, as in this case, situations with significant conflict between the patient, caregiver, and health care team. Physician training in the use of interpreters is advised,³⁶ although a 2004 survey of US resident physicians found that 35% reported receiving no or very little instruction in how to deliver services effectively through a medical interpreter.³⁷ When used, clinicians should follow established recommendations (Table 1).³⁸ In a 2006 national survey, 68% of US hospitals reported trained interpreters on staff, and 92% reported that telephone interpretation services were available.⁴⁰ In hospitals without professional interpreter services, telephone interpretation services can be critical for communication (see Web Resources at http://www.jama.com). Videoconferencing technology may also be available. In pressing clinical scenarios, use of dual telephones with a professional interpreter intermediary can facilitate rapid communication. With Ms C and Mr M, the clinicians should have used professional interpreters for all key interactions and telephone interpreters when staff interpreters were unavailable.

Even professional interpretation has important limitations in the context of complex end-of-life decision making. Two studies demonstrated that interpreter-led family meetings in the intensive care unit, when compared with meetings that did not require an interpreter, were 6 minutes shorter, contained fewer supportive statements, and commonly included interpretation errors.^{41 - 42} Furthermore, language fluency and cultural competence are not the same. Although an interpreter may be fluent in Spanish, she or he may not understand the cultural context of a patient's country of origin. Consequently, clinicians need to be vigilant in checking for understanding and exploring patient and family members' cultural understanding of illness (Box 1 and Box 2). Patient navigators, who are trained advocates from within the Latino community, may facilitate culturally competent communication, helping seriously ill patients navigate the health care system and identifying barriers to optimum palliative care.⁵⁰ However, the effectiveness of patient navigators has not been evaluated in end-of-life settings.

Assess Prior Knowledge

Use Simple Language

Encourage Questioning

Use the Teach-Back Method

Acculturation is the modification of a group's behaviors, values, and beliefs as a result of contact with a dominant culture. Assessment of acculturation status can help clinicians understand how Latino patients have adapted their perspectives in response to the dominant local culture (Anglo European-American in much of the United States).

In general, be open, curious, and respectful.

Preface with such comments as

Questions to ask about acculturation include

Questions to ask about patient's culture of origin (can replace “your culture” with the name of the country the patient came from)

Based on Carrillo et al⁴⁸ and Kleinman.⁴⁹

MEDICAL STUDENT: To be frank, I'm not quite sure [about reasons for the partner's misunderstanding of the DNR/DNI status]. It was probably partly his health literacy.

Health literacy has been defined as “the degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions.”⁵¹ Close to half of the US population has limited health literacy⁵² ; however, the prevalence of limited health literacy is higher among Spanish-speakers (62%) compared with English-speakers (46%).⁵³ In addition, health literacy is tightly linked to educational attainment and reading skill, all 3 of which are the lowest among Latinos compared with all other US racial and ethnic groups.^{52 ,54} Notably, health literacy is a dynamic construct that is influenced by the demands and expectations placed on a patient by the health care system, the nature of the patient's disease process, emotional stresses, and prior experiences with medical decision making.^{55 - 56} Patients with limited health literacy are more likely to have difficulty processing written and oral health information^{45 ,57 - 58} and to have difficulty navigating the health care environment^{51 ,59 - 61} and are less likely to engage in shared decision making or to report that interactions with clinicians are helpful or empowering.^{60 ,62 - 63}

Optimal communication techniques include assessing prior knowledge, using simple language without jargon, encouraging questions, and using the teach-back method (Box 1).⁶⁴ Using proper communication techniques may be one of the most important interventions to reduce health disparities related to limited health literacy.^{46 - 47 ,65} Due to issues of shame and potential stigma, it is recommended that clinicians adopt clear health communication techniques for all patients rather than screening for or measuring patients' health literacy.⁶⁶

PALLIATIVE CARE ATTENDING: I only had a few minutes alone with her when her husband wasn't there. . . . I got very different concerns from her than I got from him, and that definitely impeded her care. She really wanted her symptoms controlled and wanted to talk about her baby. She was desperate to see her parents. He would always focus back on the medical issues, so we weren't able to address her concerns. It was very frustrating and very hard.

MEDICAL STUDENT: I think [there was a sense of fatalism on the part of the patient], in the sense of her resignation, particularly in regard to her daughter.

Appreciating the influence of cultural expectations of care and acculturation is critically important.⁶⁷ In Latin America, physicians often do not disclose bad news or poor prognosis with their patients and are expected to keep up the patients' hope.^{68 - 70} In a recent study of 91 patients in Cuba with advanced cancer, for example, only 41% knew that they had cancer, and patients who knew their diagnosis reported less anxiety compared with patients who did not know.⁷⁰ However, in a US survey, not informing the patient of a poor-outcome fatal disease was preferred by about a third of older Mexican Americans compared with less than 15% of whites or African Americans.⁷¹ Evidence suggests that Latinos expect physicians to take a more directive role with regard to decisions regarding life-sustaining treatment.^{72 - 73} Furthermore, although growing, access to palliative care programs and to liberal use of opioids are limited in Latin America, particularly in resource-poor nations such as those in Central America. Efforts aimed at increasing opioid availability face significant challenges, including high costs of commercial opioid preparations, bureaucratic requirements, restrictive legislation, and lack of physician education.^{74 - 75} Clinicians should attempt to bridge such gaps stemming from cultural perspectives and expectations (Box 2).

Regardless of country of origin or degree of acculturation in the United States, a number of culturally based beliefs and attitudes that permeate the Latino community may influence illness experience, decision making, and care at the end of life. These cultural scripts (culturally based patterns of social interaction) include familismo, machismo, and fatalismo.^{10 ,26} Although these cultural scripts present themselves in many clinical situations, they may be particularly prominent at the end of life (Table 2). Clinicians should note that these cultural scripts may exist and operate to a greater or lesser extent in many Latinos, but there is marked individual variation, particularly according to social class and acculturation.^{67 ,76} Although clinicians should be attuned to these cultural scripts (cultural “sensitivity”), they should refrain from stereotyping and assuming that these scripts are applicable to all Latino patients. A broad approach may be a useful starting point (Box 2).

Latinos place great value on the family, which may include not only the nuclear family but also godparents, family friends, neighbors, distant relatives, and even a former spouse.^{27 ,77} The extended family forms a collective support network with strong ties of solidarity, interdependence, and loyalty. Clinicians should be prepared for large families to be present at the bedside and to be involved in care. Clinical decision making needs to be considered in the collective context of the family unit; this runs counter to the individual-based paradigm in US mainstream culture. In this case, the family unit was Ms C and Mr M (Ms C's partner and father of their child) and the clinicians might have managed the patient's care with this in mind. In addition, Ms C desperately wanted support from her parents in Central America in her last month of life. For some Latinos, particularly recent immigrants such as Ms C, family involvement in medical decision making may be hindered by immigration status, vast geographical distance, and serious political and economic constraints. Quiz Ref IDDespite the efforts of the medical team, the rapidity of Ms C's decline and the political reality that made it nearly impossible to obtain visas meant that her family could not be with her at the time of death. In retrospect, clinicians in this case may have missed opportunities to transport her home to be with her family before she became unstable and could perhaps have contacted Ms C's family there by telephone to assist with decision making. If a patient's loved ones have access to the Internet and a webcam, patients may be able to see, as well as talk to, loved ones. In addition, Latino community organizations or the patient's church may be able to provide volunteers to act as support in the family's stead.

Traditional gender roles are maintained in much of Latin America, particularly among those of less privileged social classes.^{26 ,76 ,78} Despite a strong male predominance in society, women are more responsible for health issues in a family. Although machismo is a term with negative connotations of male oppression, it also refers to the need men feel to serve as the family provider and protector. Machismo can have an overwhelming influence on decision making for Latina women, when male family members assert their caring by making most of the decisions.^{78 - 79} To truly assess Latina patients' goals and desires for care, clinicians may need to elicit their concerns and preferences in private, if this is acceptable to the patient. In this case, Ms C was seriously ill, profoundly dyspneic, and did not speak English. The extent to which Ms C's deference to Mr M resulted from machismo or powerlessness due to her medical condition or other social or religious factors is unknown. Latino male patients, on the other hand, may feel shame and disappointment when symptoms or functional decline limit their ability to fulfill patriarchal expectations.¹⁵ Clinicians should respectfully inquire about these concerns, and consider creative ways of helping Latinos contribute to their family at the end of life. For example, physicians may describe their recommended treatments as a way to provide for the family by saying, “These medications will help you be as pain free as possible, so you can continue to care for your family.”

Quiz Ref IDFatalismo, or fatalism, is a belief in fate—that one's future is preordained or not under one's control.Fatalismo has been primarily studied as beliefs and attitudes about cancer that may deter Latinos from participating in cancer screening and other preventive services.^{80 - 81} Fatalismo probably played a limited role in this particular case, especially for Mr M, but when patients are seriously or terminally ill, fatalismo may manifest as pessimistic attitudes about the future, leading patients to delay seeking treatment and potentially to suffer alone.^{82 - 83} Whether fatalismo contributes as an independent factor to health-related behavior is currently unknown. Fatalismo may also be confused with the spirituality of devoutly religious persons.^{84 - 85}

CHAPLAIN: On the intake form, where they ask for religion, somebody had put in “Catholic.” . . . But it became clear from the family that they were not Catholic. . . . They were evangelical Christians.

Religion and spirituality (la espiritualidad) have a powerful influence on many Latinos. According to a recent Pew study of 4016 Latinos, 92% consider themselves religious—68% are Roman Catholic and 15%, Evangelical or “born-again” Protestants (Evangelistas).⁸⁶ Evangelicals pray more, attend religious services more frequently, and evangelize more frequently than Catholics, and they are more likely to believe the Bible to be literally true.⁸⁶ The incorrect categorization of Ms C's religious affiliation highlights the need to specifically ask patients about their religion and to refrain from stereotyping.

Mr M made clear to multiple members of the medical team that he did not want to engage in discussions not focused on curative treatment, stating that he was hoping for a miracle. Interviews reveal the team's frustration because they felt unable to discuss issues identified as important by Ms C, particularly planning for care for the baby after her death. The extent to which Mr M's belief in miracles represented a religious belief as opposed to an expression of denial or frustration at his loss of control is unclear.⁸⁷

Many religious and spiritual considerations apply similarly across cultures.⁸⁷ In a recent study of a diverse group of patients with advanced cancer, religiousness was associated with wanting all measures to extend life, even after adjustment for race and ethnicity (odds ratio, 1.96; 95% CI, 1.08 to 3.57).⁸⁸ In focus groups, Latino surrogates indicated that suffering was to be borne as part of a test of faith.⁸⁹ This concept has also been noted frequently in literature on attitudes of African Americans toward end-of-life care and may represent a common expression of perseverance and struggle in the face of discrimination and subjugation.^{90 - 92} Also striking is the degree to which death is acknowledged and even celebrated in some Latino cultures, such as El Dia de los Muertos (Day of the Dead) in Mexico. This ancestral holiday is focused on honoring and celebrating the dead and the coming together of family when there is a death.¹⁰ Clinician exploration of these beliefs may start with the simple question, “Can you tell me about religious or spiritual beliefs or practices that are important to you?” When patients agree, involving the patients' community faith ministers, who have been briefed about the medical situation, can be helpful in understanding religious perspectives that differ from the clinician’s, fortifying and bolstering patients' religious and spiritual support systems, and establishing the trustworthiness of the medical team.

Distressingly high proportions of Latinos state that they have experienced health care discrimination. In a recent, diverse, population-based survey of 42 044 adults residing in California, 13% of all Latinos and 16% of immigrant Latinos reported experiencing racial or ethnic discrimination in the health care setting.⁹³ Latino immigrants have qualitatively described negative patient-clinician interactions, including feeling that physicians did not take enough time to educate or examine them and did not care about them personally.⁹⁴ Perceived discrimination can have detrimental consequences for health-related behavior; for example, patients who perceive medical discrimination are less likely to undergo recommended screening for early detection of cancer.⁹⁵

Physicians and health care institutions have not always demonstrated trustworthiness. Studies have consistently found that Latinos receive lower quality care than do white patients.⁹⁶ One study of patients presenting with pain in US emergency departments found that only 24% of Latinos received an opioid compared with 31% of whites (P < .001).⁹⁷ Crawley et al^{92 ,98} advocate reframing discussions about “distrust” (which focuses the burden of change on the patients' attitudes of distrust) into a discussion of establishing the “trustworthiness” of physicians and health care systems in end-of-life care situations. Clinicians in this case might have gently explored Mr M's experiences of discrimination and concerns about discriminatory withholding of medical resources and used language that promoted trustworthiness (Table 3). It may be helpful for clinicians to state outright that they do not tolerate discrimination in their practices and that they strive to treat all patients equally.¹

As demonstrated by the case of Ms C and Mr M, Latino culture exerts a profound influence on the end-of-life experience of patients and their loved ones. Cultural factors can be adequately addressed when clinicians use universal strategies of clear health communication and understand Latino cultural themes prevalent at the end of life. In this case, clinicians might have improved her care by using professional translators for important conversations; learning more about her experience prior to hospitalization (immigration experience and relationship with her partner); contacting the patient's family in Central America early in the course of decision making; and investigating options for transporting her home to her family before she became too ill for the journey. By taking steps to enhance communication and understanding and displaying an open, interested, and respectful attitude, clinicians can learn and grow professionally while providing high-quality, culturally congruent care for dying Latino patients and their families.