Advertisement

Cultural Factors Influencing Advance Care Planning in Progressive, Incurable Disease: A Systematic Review With Narrative Synthesis

Open ArchivePublished:July 16, 2018DOI:https://doi.org/10.1016/j.jpainsymman.2018.07.006

      Abstract

      Context

      Advance care planning (ACP) can improve end-of-life outcomes, but low uptake indicates it is less acceptable to patients of some cultural backgrounds.

      Objectives

      The objectives of this study were to explore how cultural factors influence ACP for patients with progressive, incurable disease and how ACP might be made cross-culturally appropriate.

      Method

      We conducted a systematic literature review using narrative synthesis. Protocol was registered prospectively (PROSPERO CRD42017060441). Key words and subject headings of six databases (AMED, PsycINFO, Embase, Ovid MEDLINE, CINAHL, and Cochrane) were searched without time restrictions. Eligible studies reported original research published in full that included adult participants with progressive, incurable disease or their formal or informal caregivers. Study quality was assessed using the Mixed Methods Appraisal Tool.

      Results

      Eight hundred and eighteen studies were screened. Twenty-seven were included: 20 quantitative, four qualitative, and three mixed methods. Most (20/30) studies were conducted in the U.S., where nonwhite ethnicity was associated with lower acceptability of formal, documented ACP processes. Cultural factors affecting ACP acceptability included religiosity, trust in the health care system, patient and clinician comfort discussing death, and patient attitudes regarding decision-making. Informal, communication-focused approaches to ACP appear more cross-culturally acceptable than formal processes. Clinician education in cultural competence is recommended. Study limitations included use of unvalidated tools and convenience samples and lack of reflexivity.

      Conclusion

      Many interconnected cultural factors influence the acceptability of ACP in progressive, incurable disease, although specific mechanisms remain unclear. A communication-focused approach to ACP may better meet the needs of culturally diverse populations.

      Key Words

      Introduction

      Significant advancements in medical technologies and modern biomedical medicine have led to widened availability of life-sustaining treatments.
      However, the use and withdrawal of these treatments pose ethical challenges for all involved in end-of-life (EOL) care
      • Asch D.A.
      • Hansen-Flaschen J.
      • Lanken P.N.
      Decisions to limit or continue life-sustaining treatment by critical care physicians in the United States: conflicts between physicians' practices and patients' wishes.
      • Breen C.M.A.
      • Abbott A.P.
      • Katherine H.
      • et al.
      Conflict associated with decisions to limit life-sustaining treatment in intensive care Units.
      • Berlinger N.
      • Jennings B.
      • Wolf S.M.
      • Truog R.D.
      • Campbell M.L.
      • Curtis J.R.
      • et al.
      Recommendations for end-of-life care in the intensive care unit: a consensus statement by the American College [corrected] of Critical Care Medicine.
      and can have adverse impacts on patients and their families.
      • Covinsky K.E.G.
      • Lee Cook
      • Francis E.
      • et al.
      The impact of serious illness on patients' families.
      In this context, advance care planning (ACP) offers potential to improve patient and family outcomes, through identifying, documenting, and enacting patients' EOL preferences.
      A recent consensus statement defined ACP as “a process that supports adults at any age or stage of health in understanding and sharing their personal values, life goals, and preferences regarding future medical care,” with the goal of helping to “ensure that people receive medical care that is consistent with their values, goals, and preferences during serious and chronic illness.”
      • Sudore R.L.
      • Lum H.D.
      • You J.J.
      • et al.
      Defining advance care planning for adults: a consensus Definition from a Multidisciplinary Delphi Panel.
      Documented benefits of ACP include improvements in patient and family satisfaction and well-being,
      • Detering K.M.
      • Hancock A.D.
      • Reade M.C.
      • Silvester W.
      The impact of advance care planning on end of life care in elderly patients: randomised controlled trial.
      improved concordance between preferences for care and delivered care,
      • Silveira M.J.
      • Kim S.Y.
      • Langa K.M.
      Advance directives and outcomes of surrogate decision making before death.
      ,
      • Houben C.H.M.
      • Spruit M.A.
      • Groenen M.T.J.
      • Wouters E.F.M.
      • Janssen D.J.A.
      Efficacy of advance care planning: a systematic review and meta-analysis.
      and health care savings.
      • Dixon J.
      • Matosevic T.
      • Knapp M.
      The economic evidence for advance care planning: systematic review of evidence.
      These benefits are reflected in endorsement of ACP by the Australian,
      British,
      and American
      AMA Code of Medical Ethics
      Opinions on care at the end of life.
      medical associations.
      Given globalization and increasing migration, requirements for cross-culturally appropriate EOL care are set to increase,
      • Gysels M.
      • Evans N.
      • Menaca A.
      • et al.
      Culture and end of life care: a scoping exercise in seven European countries.
      as reflected in palliative care policy guidance.
      ,
      However, there is evidence of significant disparities in uptake (i.e., acceptance of or engagement with) of EOL decision-making and care planning support services, including ACP, among minority ethnic groups in multicultural settings in the UK,
      • Calanzani Natalia K.J.
      • Higginson Irene J.
      Palliative and end of life care for Black, Asian and Minority Ethnic groups in the UK, Demographic profile and the current state of palliative and end of life care provision.
      New Zealand,
      • Frey R.
      • Raphael D.
      • Bellamy G.
      • Gott M.
      Advance care planning for Maori, Pacific and Asian people: the views of New Zealand healthcare professionals.
      the U.S., and Australia.
      • Johnstone M.J.
      Ethics and advance care planning in a culturally diverse Society.
      Despite this, little attention has been given to cross-cultural issues in ACP to date.
      • Johnstone M.J.
      Ethics and advance care planning in a culturally diverse Society.
      Understanding ACP's impact in different populations, settings, and contexts is needed to maximize the full potential of ACP.
      • Geronimo Jimenez W.S.T.
      • Amrit K Virk
      • Chan Kee
      • Low J.C.
      • Andy Hau
      • Yan Ho
      Overview of systematic reviews of advance care planning: Summary of evidence and global Lessons.
      In particular, if all patients and families are to benefit from ACP, there is a need to better understand how cultural factors influence its acceptability. “Culture” can be understood as a system of ideas, rules, meanings, and ways of living and thinking that are built up, shared, and expressed by a particular group of people, often of the same ethnic background.
      • Taylor E.J.
      Spirituality, culture, and cancer care.
      • Martsolf D.S.
      Cultural aspects of spirituality in cancer care.
      • Keesing R.M.
      AJSE. Cultural anthropology: A contemporary perspective.
      In 2014, Lovell and Yates
      • Lovell A.
      • Yates P.
      Advance Care Planning in palliative care: a systematic literature review of the contextual factors influencing its uptake 2008-2012.
      published a systematic review examining the influence of contextual factors on the uptake of ACP among adults and identified ethnicity as one of these, but research on the influence of ethnicity and culture included in their review was limited to five U.S.-conducted studies. Another review in this area examined the attitudes of culturally diverse groups toward EOL decision-making
      • Kwak J.
      • Haley W.E.
      Current research findings on end-of-life decision making among racially or ethnically diverse groups.
      but focused on community-dwelling older adults, who may have varied levels of health or illness. There is evidence that patients' EOL wishes, such as treatment preferences or desire for truth-telling, are influenced and changed by the following: time and health state,
      • Fried T.R.
      • O'Leary J.
      • Van Ness P.
      • Fraenkel L.
      Inconsistency over time in the preferences of older persons with advanced illness for life-sustaining treatment.
      greater experience of illness,
      • Ko E.
      • Lee J.
      End-of-life communication: ethnic differences between Korean American and non-Hispanic White older adults.
      and extent of available information about advance directives (ADs).
      • Froman R.
      • Owen S.
      Randomized study of stability and change in patients' advance directives.
      There is therefore a need to synthesize evidence related to the influence of cultural factors in ACP for patients with progressive, incurable conditions, who are likely to have different needs and views of ACP than healthy adults.
      This systematic review aimed to answer the following questions:
      • 1.
        How, if at all, does patients' and families' cultural background influence the acceptability of ACP for seriously ill patients?
      • 2.
        How might ACP need to be adapted to make it more cross-culturally appropriate?
      • 3.
        What cultural factors do clinicians need to be aware of in approaching communication and planning with patients and families about EOL issues?

      Methods

      Design, Protocol, and Registration

      We conducted a systematic literature review with narrative synthesis, a primarily textural approach to synthesizing and presenting the findings of multiple mixed-methods studies as a narrative, as outlined by Popay et al.
      • Popay J.
      • Roberts H.
      • Sowden A.
      • et al.
      Guidance on the conduct of narrative synthesis in systematic reviews.
      Given that the research conducted to date into cultural factors in ACP features both qualitative and quantitative studies, narrative synthesis was an appropriate approach for the first systematic review in this area. The review protocol was registered prospectively with PROSPERO.
      • Selman L.
      • McDermott E.
      How do cultural factors influence advance care planning in progressive, incurable disease? A systematic review with narrative synthesis - protocol: PROSPERO.

      Inclusion Criteria

      Participants

      We included studies of adults with progressive, incurable disease and their formal (staff) or informal (family) caregivers. Progressive, incurable conditions included but were not limited to: advanced or late stage (3+) cancer, HIV/AIDS, dementia, Parkinson's, organ failure (including renal and heart failure), and chronic obstructive pulmonary disease. Mixed study samples in which at least 50% of participants had incurable, progressive illness were included.

      Studies

      To reflect the complexity of the concept of culture, we included research that measured, adjusted for, recorded, and/or explicitly considered the influence of “cultural” factors on ACP, including: ethnicity/“race,” religion/spirituality, nationality, or country of permanent residence. We operationalized “culture” in these broad terms, to help ensure important factors related to the acceptability of ACP were not missed. We define ethnicity as “the fact or state of belonging to a social group that has a common national or cultural tradition.”

      Ethnicity (noun) definition Oxford Dictionaries, accessed online. Available from https://en.oxforddictionaries.com/definition/ethnicity.

      “Race” is a more difficult concept to define, often tied to crude categorization based on skin color, and is considered scientifically discredited
      • Bhopal R.
      Glossary of terms relating to ethnicity and race: for reflection and debate.
      ; our use of the term in our narrative reflects its use by study authors. We explicitly included research that considered the influence of religion/spirituality on ACP uptake because religion and spirituality are intimately related to and expressed via culture and can influence how and whether a patient identifies with a particular culture.
      • Speck P.
      Culture and spirituality: essential components of palliative care.
      All study designs were included if inclusion criteria were met and the research was published in full, to achieve a comprehensive understanding of cultural factors in ACP. Only original research was included, as the review aimed to synthesize empirical evidence.

      Exclusion Criteria

      Studies were excluded if they met any of the following criteria.

      Participants

      We excluded children and adolescents (under 18 years of age); cancer survivors (people whose cancer is in remission and who are no longer being treated); elderly populations without progressive, incurable disease; people with chronic but not incurable conditions, for example, diabetes, arthritis; mixed study samples in which <50% had incurable, progressive disease or where the proportion with such disease could not be determined; and studies of trainee (unqualified) health care providers.

      Studies

      Studies not reported in English were excluded because of resource limitations. Research studies that did not include consideration of ACP or culture, as operationalized above, were excluded. Reports of studies only available as abstracts or letters to the editor were excluded, as were review articles. Studies were not excluded based on institutional setting or country.

      Information Sources and Searches

      On March 26, 2017, we searched six electronic databases (Table 1). Initial keywords of sentinel papers
      • Lovell A.
      • Yates P.
      Advance Care Planning in palliative care: a systematic literature review of the contextual factors influencing its uptake 2008-2012.
      ,
      • Kwak J.
      • Haley W.E.
      Current research findings on end-of-life decision making among racially or ethnically diverse groups.
      ,
      • Johnson K.S.
      • Kuchibhatla M.
      • Tulsky J.A.
      What explains racial differences in the use of advance directives and attitudes toward hospice care? [References].
      • Mebane E.W.
      • Oman R.F.
      • Kroonen L.T.
      • Goldstein M.K.
      The influence of physician race, age, and gender on physician attitudes toward advance care directives and preferences for end-of-life decision-making.
      • Perkins H.S.
      • Geppert C.M.A.
      • Gonzales A.
      • Cortez J.D.
      • Hazuda H.P.
      Cross-cultural similarities and differences in attitudes about advance care planning.
      • Smith A.K.
      • McCarthy E.P.
      • Paulk E.
      • et al.
      Racial and ethnic differences in advance care planning among patients with cancer: impact of terminal illness acknowledgment, religiousness, and treatment preferences.
      • Smith A.K.
      • Davis R.B.
      • Krakauer E.L.
      Differences in the quality of the patient-physician relationship among terminally ill African-American and White patients: impact on advance care planning and treatment preferences.
      and corresponding subject headings were used to identify synonyms for three elements of our topic area: [ACP] AND [Cultural factors] AND [Acceptability]. The search strategy (Table 2) was developed in OvidSP's and adjusted for the CINAHL and Cochrane databases, to reflect the different subject headings.
      Table 1Databases Searched
      Via OvidSPAMED (Allied and Complementary Medicine) 1985 to March 2017
      PsycINFO 1806 to March, Week 4, 2017
      Embase 1974 to 2017 March 26
      Ovid MEDLINE(R) Epub Ahead of Print, In-Process & Other Non-Indexed Citations, Ovid MEDLINE(R) Daily, Ovid MEDLINE and Versions (R)
      IndividuallyCINAHL
      Cochrane Library
      Table 2Search Strategy
      Database(s)Advance Care PlanningCultural FactorsAcceptability
      OvidSP: AMED, PsycINFO, Embase, Ovid MEDLINE, CINAHL, Cochrane(exp Advance Care Planning/) or (advance care plan*.tw.) or (care plan.tw.)(exp Ethnic Groups/ or exp cross-cultural comparison/ or exp cultural characteristics/ or exp cultural diversity/)(exp Patient Preference/eh, px [Ethnology, Psychology]) or (exp "Patient Acceptance of Health Care"/eh, px, sn [Ethnology, Psychology, Statistics & Numerical Data]) or (exp Attitude to Death/) or ((acceptability or acceptable or uptake or engage or engagement or preference or preferences).tw.)
      CINAHL(MM "Advance Care Planning") OR TI ("care plan" or "advance care plan") AND AB ("care plan" or "advance care plan")(MH "Ethnic Groups+") OR (MH "Ethnological Research") OR (MH "Culture+") OR (MH "Cultural Diversity") OR (MH "Cultural Values")(MH "Patient Satisfaction") OR "Patient Acceptance of Health Care" OR (MH "Attitude to Death") OR (MH "Attitude to Medical Treatment") OR TI (acceptability or acceptable or uptake or engage or engagement or preference or preferences) AND AB (acceptability or acceptable or uptake or engage or engagement or preference or preferences) OR TI patient preference AND AB patient preference
      Cochrane LibraryMeSH descriptor: [Advance Care Planning] explode all trees OR advance care plan*:ti,ab,kw (Word variations have been searched) OR care plan:ti,ab,kw (Word variations have been searched)MeSH descriptor: [Ethnic Groups] explode all trees OR MeSH descriptor: [Cross-Cultural Comparison] explode all trees OR MeSH descriptor: [Cultural Characteristics] explode all trees OR MeSH descriptor: [Cultural Diversity] explode all treesMeSH descriptor: [Patient Preference] explode all trees OR MeSH descriptor: [Patient Acceptance of Health Care] explode all trees OR MeSH descriptor: [Attitude to Death] explode all trees OR acceptability or acceptable or uptake or engage or engagement or preference or preferences:ti,ab,kw (Word variations have been searched)

      Study Selection

      Titles and abstracts of papers retrieved were screened independently to identify studies that potentially met the inclusion criteria (E. M.). The full texts were retrieved and assessed for eligibility (E. M.), with a random sample of 10% independently screened to check consistency in application of the criteria (L. E. S.). Disagreements between E. M. and L. E. S. over the eligibility of studies were resolved through discussion and reference to and refinement of the inclusion and exclusion criteria.

      Data Collection Process and Data Items

      A standardized, prepiloted form was used to extract data from the included studies for assessment of study quality and evidence synthesis. Extracted information included study setting, country, study population, ethnicity/race of participants, details of the ACP intervention, study methodology, findings regarding acceptability of ACP, recommendations regarding ACP and clinical implications, and information for assessment of the risk of bias/limitations (see Table 3). E. M. extracted data, with L. E. S. extracting data independently for a random 10% of studies. Discrepancies at this stage were minor and resolved through discussion.
      Table 3Description of Included Papers
      No (#)AuthorTitleCountry/SettingPopulationEthnicity/Race of PatientsMethodology
      Methodology = QUAL (qualitative), QUAN (quantitative), or MIXED METHODS (combining qualitative and quantitative methods).
      ACP Process StudiedMain Finding Regarding Acceptability of ACPQuality Appraisal Score; Missing Items
      1Balboni et al.
      • Balboni T.A.
      • Vanderwerker L.C.
      • Block S.D.
      • et al.
      Religiousness and spiritual support among advanced cancer patients and associations with end-of-life treatment preferences and quality of life.
      Religiousness and spiritual support among advanced cancer patients and associations with end-of-life treatment preferences and quality of lifeU.S.—eight cancer centers/hospitals/PC centers/oncology departments/veteran health care centers in CT, NY, TX, MA, and NHn = 230, advanced cancer patients with failure of first-line chemotherapy, aged ≥20 yrs, with adequate stamina to complete interview, without dementia or delirium, spoke English or Spanish, and unpaid caregiverWhite 140 (61%), black 44 (19%), Hispanic 39 (17%), Asian 5 (2%), and other 2 (1%). Of 100 participants from southern institutions, 61% were nonwhite; of 130 from northeastern institutions, 22% were nonwhite (self-identified)QUAN—survey, hospital chart review, statistical analysis (McNemar's test, linear regression, univariate models, multivariate analysis, logistic regression)LW, DPOA, and DNRGreater religiousness/spirituality is associated with lower levels of formal ACP completion50; (4.2) samples not representative of the population under study and (4.4) less than 60% response rate (42% [230/538])
      2Braun et al.
      • Braun U.K.
      • Ford M.E.
      • Beyth R.J.
      • et al.
      The physician's professional role in end-of-life decision-making: voices of racially and ethnically diverse physicians.
      The physician's professional role in end-of-life decision-making: voices of racially and ethnically diverse physiciansU.S.—Texas Medical Center, Houstonn = 26, internists and specialists (doctors) who commonly treated patients in end stages of disease, purposively sampled to include ethnic minority doctors, seven racially/ethnically homogenous focus groups and moderatorsThree focus groups with Caucasian patients, n = 11. Two focus groups with AA patients, n = 8. Two focus groups with Hispanic patients, n = 7 (self-identified)QUAL—focus groups, semistructured interviews, thematic analysis using grounded theoryDoctor's EOL discussion and decision-making style and valuing of addressing patient's religious/spiritual needsDoctors felt that patient ethnicity/race is clinically relevant to some aspects of informal ACP. Concordance/discordance of clinician' and patient' ethnicity/race was thought more important by AA than Caucasian doctors.75; (1.4) no consideration of how findings relate to researchers' influence
      3Braun et al.
      • Braun U.K.
      • Beyth R.J.
      • Ford M.E.
      • Espadas D.
      • McCullough L.B.
      Decision-making styles of seriously ill male veterans for end-of-life care: Autonomists, Altruists, Authorizers, Absolute Trusters, and Avoiders.
      Decision-making styles of seriously ill male veterans for end-of-life care: autonomists, altruists, authorizers, absolute trusters, and avoidersU.S.—Michael E. DeBakey VA Medical Center (MEDVAMC) in Houston, Texasn = 44, “seriously ill” patients (lists included diagnoses in Table 2), arranged in eight racially/ethnically homogenous focus groups, purposively sampledTwo focus groups with AA patients n = 14, three focus groups with Hispanic patients n = 17, and focus groups with Caucasian patients n = 13 (self-identified)QUAL—focus groups, inductive thematic analysisEOL decision-making styleFive different types of EOL decision-making style (autonomists, altruists, authorizers, absolute trusters, and avoiders) influence acceptability of formal ACP. EOL decision-making style was suggested to be associated with different racial/ethnic groups (i.e., no white patients were “Avoiders”), although this relationship was not confirmed.75; (1.4) no consideration of how findings relate to researchers' influence
      4Collins et al.
      • Collins M.
      • Lehane E.
      Perspectives on death, dying and advance care planning from patients receiving haemodialysis.
      Perspectives on death, dying and advance care planning from patients receiving haemodialysisIreland—acute renal unit in Irish hospitaln = 50 (67% of eligible population) individuals receiving hemodialysis for more than three months“Irish patients”—those who were not “Irish” (5% of the unit) and those who could not speak/understand English were excludedQUAN—questionnaire and descriptive statisticsImportance of 20 wide-ranging EOL care issues (Table 5)Most Irish dialysis patients wished to know their prognosis if they had less than six months to live (60%) and wanted honest answers from their doctors.50; (4.2) samples representative of population under study and (4.3) validity of instrument used for measurements not known
      5DeSanto-Madeya et al.
      • DeSanto-Madeya S.
      • Nilsson M.
      • Loggers E.T.
      • et al.
      Associations between United States acculturation and the end-of-life experience of caregivers of patients with advanced cancer.
      Associations between United States acculturation and the end-of-life experience of caregivers of patients with advanced cancerU.S.—four comprehensive cancer centers in the Northeast and two in the Southwestn = 167 identified primary, unpaid, informal caregivers of patients with diagnosed advanced cancer, where both patient and caregiver could speak English or Spanish and neither met criteria for dementia or deliriumWhite n = 79, black n = 8, Asian n = 8, Hispanic n = 71, and other n = 1 (self-reported by caregiver)QUAN—chart review and questionnaire, descriptive statistics (t-test, Pearson correlation, logistical regression)EOL preferences and medical decision-making, physician-caregiver communication including doctor's listening and information providedCaregivers' level of acculturation influenced their EOL preferences and EOL medical decision-making, for example, being less acculturated was associated with being more likely to perceive that doctors had provided too much information.50; (4.2) reasons for eligible participants not participating are not addressed and (4.4) less than 60% response rate (32% [167/448])
      6Frey et al.
      • Frey R.
      • Raphael D.
      • Bellamy G.
      • Gott M.
      Advance care planning for Maori, Pacific and Asian people: the views of New Zealand healthcare professionals.
      Advance care planning for Maori, Pacific and Asian people: the views of New Zealand healthcare professionalsNZ—Hospitals, general practice, Maori community support services, and Pacific and Asian hospital support servicesn = 40, health and social care professionals with a wide range of knowledge and experience in PCNZ European (19), Maori ethnicity (3), or membership in a Pacific (3), Asian (5), Southeast Asian (7), or other European (3) (self-reported)QUAL—interviews, focus groups, thematic analysisACP defined as “a voluntary process of discussion about future care between an individual and their [sic] care providers, irrespective of discipline”Maori, Pacific, and Asian families/whanau in NZ are reluctant to engage in ACP conversations50; (1.3) no consideration of how findings relate to context/setting and (1.4) no consideration of how findings relate to researchers' influence
      7Gabbay et al.
      • Gabbay B.B.
      • Matsumura S.
      • Etzioni S.
      • et al.
      Negotiating end-of-life decision making: a comparison of Japanese and U.S. residents' approaches.
      Negotiating end-of-life decision making: a comparison of Japanese and U.S. residents' approachesU.S. + Japan—two U.S. sites in LA, California (one with predominantly poor, veteran population), and five Japanese sites in Central Honshu, Kyushu, Okinawan = 347 (103 internal medicine residents in U.S. and 244 general medical practice residents in Japan)103 U.S. residents and 244 Japanese residents (ethnicity not recorded)QUAN—nonrandomized cross-sectional questionnaire and descriptive statistics (chi-square tests and t-tests)EOL decision-making, diagnostic and prognostic disclosures, and involvement of familyDifferences were found in American and Japanese clinicians' approaches to EOL decision-making. Japanese clinicians were more likely to inform a patient's family members about diagnosis/prognosis, before the patient themselves (44%:2%, P < 0.001).75; (3.3)—they could not fully assess for differences between sites in U.S./Japan and therefore do not control for this
      8Garrido et al.
      • Garrido M.M.
      • Harrington S.T.
      • Prigerson H.G.
      End-of-life treatment preferences: a key to reducing ethnic/racial disparities in advance care planning? [References].
      End-of-life treatment preferences: a key to reducing ethnic/racial disparities in advance care planning?U.S.—outpatient clinics in five states (Connecticut, Massachusetts, New Hampshire, New York, and Texas)n = 606 patients from the five centers who had advanced cancer and spoke English or Spanish and were cognitively able to participateNon-Latino whites n = 437, blacks n = 93, and Latinos n = 76QUAN—nonrandomized cross-sectional baseline interviews, descriptive statistics (bivariate tests)DNR order completionWhite non-Latino patients were more likely to have a DNR order (45%) than black (25%) and Latino (20%) patients (P < 0.001). Individuals with DNR orders had lower levels of positive religious coping than those without DNR orders (30% vs. 40%; P = 0.048).75; (3.2) as methods for measuring each preference not validated
      9Hopp et al.
      • Hopp F.P.
      • Duffy S.A.
      Racial variations in end-of-life care.
      Racial variations in end-of-life careU.S.—older persons living in continental U.S. outside of institutional settings at onset of AHEAD studyn = 540 relatives (proxy respondents) for persons who died between 1993 and 1995; including 169 spouses, 247 children, and 124 other relativesWhite n = 454 and black n = 86 (others excluded because of small sample size of respondents)QUAN—secondary analysis of phone or in-person survey data, descriptive statistics (logistical regression modeling)LW, DPOA, and EOL discussion including treatment preferencesDeceased AAs were 63% less likely than WAs to have had an EOL discussion before death (P < 0.001). Ethnicity/race continued to be a significant predictor of ACP uptake even after logistic regression modeling to control for sociodemographic factors.75; (4.2) samples not representative of population under study (decedents included in this study were more likely to be older, male, and to have lower educational attainment)
      10Janssen et al.
      • Janssen D.J.
      • Spruit M.A.
      • Schols J.M.
      • Wouters E.F.
      A call for high-quality advance care planning in outpatients with severe COPD or chronic heart failure.
      A call for high-quality advance care planning in outpatients with severe COPD or chronic heart failureThe Netherlands—one academic and five general hospitalsn = 185 outpatients with severe COPD (n = 105) or chronic heart failure (n = 80)DutchQUAN—nonrandomized prospective cross-sectional survey, logistic regression analysisDiscussion of preferences for CPR or MV, prognosis of survival, dying process, and PC, according to patients and clinicians. Quality of communication.Doctors reported greater rate of having had ACP communication than patients. Most Dutch COPD or heart failure patients were able to indicate their preferences regarding life-sustaining treatments. Doctors rarely discussed prognosis, dying, or PC.50; (3.2) validity of survey tools used not stated and (3.4) less than 60% response rate (for patients with heart failure)
      11Kataoka-Yahiro et al.
      • Kataoka-Yahiro M.R.
      • Conde F.A.
      • Wong R.S.
      • Page V.
      • Peller B.
      Advance care planning among Asian Americans and Native Hawaiians receiving haemodialysis.
      Advance care planning among Asian Americans and Native Hawaiians receiving haemodialysisHawaii—four outpatient dialysis centers in the island of Oahu, Honolulun = 50 CKD hemodialysis patients of Asian American or Native Hawaiian ethnicity with ability to read/understand English and capacity for informed consent62% Asian Americans, 24% Native Hawaiians, and 14% mixed Asian American + Native HawaiianQUAN—survey and descriptive statisticsAD, LW, and ACP (process of communication among patients, families, and health care providers addressing EOL care)Most patients' attitudes about ACP were positive; however, less than half (40%) had completed formal ACP such as AD or LW, despite most having heard of these processes.50; (4.2) samples not representative of understudy population and (4.3) as questionnaire tool not validated
      12Kurella Tamura et al.
      • Kurella Tamura M.
      • Goldstein M.K.
      • Perez-Stable E.J.
      Preferences for dialysis withdrawal and engagement in advance care planning within a diverse sample of dialysis patients.
      Preferences for dialysis withdrawal and engagement in advance care planning within a diverse sample of dialysis patientsU.S.—University of California San Francisco Mt. Zion Hemodialysis Center and the San Francisco Veterans Affairs Medical Center Hemodialysis Centern = 62 end-stage renal disease (ESRD) patients who were fluent in English and without an active psychiatric disorder including schizophrenia and dementia or a learning disability (about 23% of the clinic did not meet these criteria)Black n = 23, Latino n = 7, white n = 21, Asian n = 10, (1 participant failed to complete questionnaire and was excluded) (self-reported)QUAN—questionnaire, medical chart review, Fisher's exact testEOL discussions, AD, DNR/DNIRates of engagement in informal ACP (EOL discussion) were generally higher than for formal ACP (documentation). Racial/ethnic minorities were not less likely to engage in EOL discussions: AAs were more likely to engage in EOL discussions with health care professionals than whites (70% vs. 38%) (P < 0.01).75; (4.2) samples not representative of understudy population (only recruited English-speaking participants)
      13Laguna

      Laguna J. Racial/ethnic variation in care preferences and care outcomes among United States hospice enrollees. Dissertation Abstracts International: Section B: The Sciences and Engineering Vol. 2015;76(3-B(E).

      Racial/ethnic variation in care preferences and care outcomes among United States hospice enrolleesU.S.—657 hospice-providing agencies in the U.S.n = 3661 Medicare hospice patients (representing 788,872 patients) (Medicare is 82% of hospice care in U.S.)White, black, and Hispanic (as reported on charts)QUAN—NHHCS hospice care survey statistical analysis (F-statistics, chi-squared tests, ANOVA)AD, DNR, HC proxyBlack hospice patients were less likely to complete an AD than whites (80%:93%) (P = 0.004). Hispanic hospice patients were 3.68 times more likely to have signed a DNR order than whites (P < 0.01) and 3.67 times as likely as blacks (P < 0.001).75; (4.3) “unique” unvalidated variable of emergent care use
      14Loggers et al.
      • Loggers E.T.
      • Maciejewski P.K.
      • Jimenez R.
      • et al.
      Predictors of intensive end-of-life and hospice care in Latino and white advanced cancer patients.
      Predictors of intensive end-of-life and hospice care in Latino and white advanced cancer patientsU.S.—eight cancer centers/hospitals/PC centers/oncology departments/veteran health care centers in five statesn = 292 stage IV cancer patients with metastatic disease refractory to first-line chemotherapyLatino (n = 58) and white (n = 234) (self-reported)QUAN—nonrandomized cohort study[Coping with Cancer (CWC)], statistical analysis (stratified logistical regression)EOL discussion, DNRLatino patients were less likely to acknowledge their condition was terminal than whites (19%:47.4%, AOR 0.18) (P = 0.0001) and less likely to have DNR orders than whites (22.4%:50.4%, AOR 0.23) (P = 0.0003). No differences in the rate of positive religious coping, EOL discussions, or preferences for life-extending care among Latino and white patients.100
      15Mack et al.
      • Mack J.W.
      • Paulk M.E.
      • Viswanath K.
      • Prigerson H.G.
      Racial disparities in the outcomes of communication on medical care received near death.
      Racial disparities in the outcomes of communication on medical care received near deathU.S.—eight cancer centers/hospitals/PC centers/oncology departments/veteran health care centers in five statesn = 332 advanced cancer patients who participated in and died during the CWC studyWhite (n = 261) and black (n = 71) (self-identified)QUAN—nonrandomized cohort study (CWC), statistical analysis (stratified logistical regression, bivariable and multivariable)EOL discussion regarding “wishes about care would want to receive if dying,” achievement of communication goals, DNR ordersSimilar rates of EOL discussions in black and white patients (35%:38%). EOL discussions associated with achieving some communication goals among black patients, including placement of DNR orders (AOR 4.25, P = 0.04), but this did not translate into a reduction of life-prolonging care, as occurred with white patients.100
      16Nakazawa et al.
      • Nakazawa K.
      • Kizawa Y.
      • Maeno T.
      • et al.
      Palliative care physicians' practices and attitudes regarding advance care planning in palliative care units in Japan: a nationwide survey.
      Palliative care physicians' practices and attitudes regarding advance care planning in palliative care units in Japan: a nationwide surveyJapan—99 palliative care units (PCUs)n = 99 doctors working in PCUs in Japan who responded to survey requestJapanese doctors/residents, ethnicity not recordedQUAN—survey, descriptive statistics (chi-square test, Fisher's exact test)AD: “directions recorded by competent individuals to allow them to influence treatment decisions in the event of serious illness and subsequent loss of competence.”

      ACP: “making decisions about patient's future health care by a patient in consultation with health care providers, family members, and important others, should they become incapable of medical treatment decisions.”
      86.9% of doctors agreed that “ACP is an effective way for patients to influence their medical treatment, should they lose competence”; however, many failed to implement aspects of ACP such as recommending completion of ADs and designation of health care proxies or implementing existing ADs.50; (4.2) samples not representative of understudy population as self-selecting participants and (4.4) less than 60% response rate (49%)
      17Periyakoil et al.
      • Periyakoil V.S.
      • Neri E.
      • Kraemer H.
      No Easy Talk: a mixed methods study of doctor reported barriers to conducting effective end-of-life conversations with diverse patients.
      No easy talk: a mixed methods study of doctor reported barriers to conducting effective end-of-life conversations with diverse patientsU.S.—two large training hospitals in California (Stanford Hospital and Clinics and the VA Palo Alto)n = 1040 multispecialty doctors who care for seriously ill patients (n = 8 did not identify any barriers, so did not participate, n = 1032 remaining), development cohort n = 29, validation cohort n = 966Caucasian n = 485, Latino-American n = 58, AA n = 34, Asian n = 335, and mixed ethnicity/race n = 84 (presumed self-reported in survey, although this not clearly stated)MIXED METHODS—sequential exploratory design: survey then grounded theory analysesEOL conversations99.9% doctors reported barriers to conducting EOL conversations with patients, especially those of different ethnicity/race, including language, religious/spiritual beliefs, health literacy, and cultural differences in truth handling and decision-making. Doctor's ethnicity did influence their level of reported barriers (Chi-square = 12.77, DF = 4, P = 0.0125).75; (5.3) no triangulation or consideration of divergence of qualitative/quantitative results and (1.4) no consideration of how findings relate to researchers' influence
      18Perry et al.
      • Perry E.
      • Swartz J.
      • Brown S.
      • Smith D.
      • Kelly G.
      • Swartz R.
      Peer mentoring: a culturally sensitive approach to end-of-life planning for long-term dialysis patients.
      Peer mentoring: a culturally sensitive approach to end-of-life planning for long-term dialysis patientsU.S.—21 dialysis centers across Michigann = 203 dialysis patients38% were AAQUAN—RCT and generalized least squares estimation for multilevel binomial modelsIntervention: peer mentoring on EOL decision-making—dialysis patients trained to help other patients through eight contacts over two- to four-month period, semistructured, detailed on p113 of paperPeer mentoring intervention increased completion of ADs (P < 0.001), comfort discussing ADs (P < 0.01), and improved subjective well-being (P < 0.05) among AA patients compared with printed materials and control group. (These effects did not occur for white patients, although printed materials did decrease reported suicidal ideation in whites [P < 0.05]).75; (2.2) no description of allocation concealment from researchers or participants
      19Phipps et al.
      • Phipps E.
      • True G.
      • Harris D.
      • et al.
      Approaching the end of life: attitudes, preferences, and behaviors of African-American and white patients and their family caregivers.
      Approaching the end of life: attitudes, preferences, and behaviors of African-American and white patients and their family caregiversU.S.—patients from cancer centers, tumor registries, or recruited through oncologists/doctorsn = 68 patients with stage III-B or IV lung or stage IV colon cancer and n = 68 patient-designated family caregivers (of 159 eligible) (of 198 identified but 38 died before time of contact)AA n = 38 and white n = 30MIXED METHODS—interviews and medical records review, statistical analysis and coding of qualitative dataLW, completion of an HC proxy directive, DPOA for health care and EOL discussionsWhite patients were more likely than AA patients to have completed formal ACP including DPOA (34%:8%, P = 0.01) and LW (41%:11%, P = 0.004).75; (5.3) no triangulation or consideration of divergence of qualitative/quantitative results and (4.4) less than 60% response rate (42% 68/159)
      20Sharma et al.
      • Sharma R.K.
      • Dy S.M.
      Documentation of information and care planning for patients with advanced cancer: associations with patient characteristics and utilization of hospital care.
      Documentation of information and care planning for patients with advanced cancer: associations with patient characteristics and utilization of hospital careU.S.—Johns Hopkins Hospital and Sidney Kimmel Comprehensive Cancer Center, Baltimoren = 238 deceased patients with advanced cancer, primarily Medicaid, who died between two and 15 months after diagnosis and had at least three hospital visitsWhite n = 151, black n = 76, and other n = 11QUAN—retrospective cross-sectional hospital chart review, descriptive statistics (multivariate logistic regression, chi-squared tests, ANOVA)AD, DPOA, DNR order, hospice discussion, and hospice referralRates of ACP documentation were generally low. Black patients had a higher odds of hospice discussion than white patients in the multivariate analysis (AOR 2.11; 95% CI 1.18 to 3.76) (P = 0.01).75; (3.1) participants not recruited in a way that minimizes selection bias (all from one specialist cancer hospital, so not representative of other settings)
      21Smith et al.
      • Smith A.K.
      • Davis R.B.
      • Krakauer E.L.
      Differences in the quality of the patient-physician relationship among terminally ill African-American and White patients: impact on advance care planning and treatment preferences.
      Differences in the quality of the patient-physician relationship among terminally ill African-American and white patients: impact on advance care planning and treatment preferencesU.S.—patients recruited through local physicians in six sitesn = 803 terminally ill patientsAA (n = 115) and white (n = 688) (no distinction between Hispanic or non-Hispanic) (self-reported)QUAN—nonrandomized cross-sectional analysis of Commonwealth–Cummings Project data (nationwide survey), bivariable sequential statistical analysisACP defined as “the patient reporting on one or more of the following: a living will, a health care proxy, or having talked with family or physician about plans for end-of-life care” (i.e., LW, HC proxy, or EOL discussion)AAs were less likely than whites to have an ACP (ARR = 0.66, 95% CI = 0.52–0.84) (P < 0.001). AAs' ratings of the quality of their patient-doctor relationships were found to be poorer than those of WAs for several measures, but the difference was not statistically significant for level of trust in doctors.50; (3.1) participants not recruited in a way that minimizes selection bias (recruited through doctors who may be more likely to recruit patients with whom they had good relationship) and (3.2) poor validity of tool used (questions about doctor-patient relationship were skewed toward “agree”)
      22Smith et al.
      • Smith A.K.
      • McCarthy E.P.
      • Paulk E.
      • et al.
      Racial and ethnic differences in advance care planning among patients with cancer: impact of terminal illness acknowledgment, religiousness, and treatment preferences.
      Racial and ethnic differences in advance care planning among patients with cancer: impact of terminal illness acknowledgment, religiousness, and treatment preferencesU.S.—eight cancer centers/hospitals/PC centers/oncology departments/veteran health care centers in five statesn = 449 patients with advanced cancer (clinician's estimate that the patient would live less than six months and failure of first-line chemotherapy) and had an unpaid caregiver312 non-Hispanic white, 75 non-Hispanic black, and 62 Hispanic patients (self-identified)QUAN—nonrandomized cross-sectional CWC study, statistical analysis (chi-square test, Fisher's exact test, modified Poisson approach)ACP defined as “patient's report of LW or DPOA/HC proxy), completed DNR order (not measured in above study), or having discussed wishes for EOL care with their physician” (i.e., LW, DPOA/HC proxy, DNR order, EOL discussion)Compared with whites, blacks and Hispanics were less likely to have an ACP (W 80%; B 47%; and H 47%). Compared with whites, blacks and Hispanics were more likely to consider religion very important (W 44%; B 88%; and H 73%) (P < 0.001). Although black and Hispanic patients were less likely to consider themselves terminally ill, this factor did not explain observed disparities in ACP.75; (3.2)—unknown validity of measurements used (e.g., religiousness scale)
      23Song et al.
      • Song M.K.
      • Ward S.E.
      • Happ M.B.
      • et al.
      Randomized controlled trial of SPIRIT: an effective approach to preparing African-American dialysis patients and families for end of life.
      Randomized controlled trial of SPIRIT: an effective approach to preparing African-American dialysis patients and families for end of lifeU.S.—six outpatient dialysis clinics in Western Pennsylvanian = 58 dyads: AAs with end-stage renal disease and their chosen surrogate decision-makersAll AA, n = 58MIXED METHODS—sequential explanatory design: RCT, surveys, interviews. Statistical analysis (Fisher's exact test, Mann-Whitney U) and thematic analysis qualitative data.Intervention: Sharing Patients' Illness Representations to Increase Trust (SPIRIT) involving identifying and exploring concerns, creating conditions for discussion and conceptual change about death/dying, introducing replacement information, encouraging future discussions, assessing need for additional supportSPIRIT intervention was well received by both patients and surrogates. SPIRIT was found to increase quality of communication scores reported by patients: intervention vs. control at T2 (one week) (U = 283.50, P = 0.03) and T3 (three months) (U = 165.00, p < 0.01) and by surrogates: intervention vs. control at T2 (U = 95.00, P < 0.01) and T3 (U = 139.00, P = 0.03).50; (5.3) no triangulation or consideration of divergence of qualitative/quantitative results and (1.3) no consideration of how context/settings could influence findings and (1.4) no consideration of how findings relate to researchers' influence and (2.3) outcome data completed below 80% (54% participants agreed to be included in follow-up data)
      24True et al.
      • True G.
      • Phipps E.J.
      • Braitman L.E.
      • Harralson T.
      • Harris D.
      • Tester W.
      Treatment preferences and advance care planning at end of life: the role of ethnicity and spiritual coping in cancer patients.
      Treatment preferences and advance care planning at end of life: the role of ethnicity and spiritual coping in cancer patientsU.S.—patients from cancer centers, tumor registries, or recruited through oncologists/doctorsn = 68 patients with an advanced stage of lung or colon cancerAA n = 38 and white n = 30 (self-defined)QUAN—nonrandomized cross-sectional secondary analysis of medical records and survey data using coding and statistics (Fisher's exact test)LWAAs more likely than whites to report belief in divine intervention (rs = 0.27, P = 0.02) or that their fate was in the hands of higher power (rs = 0.25, P = 0.04). Patients who reported divine intervention belief less likely to have a LW (P = 0.007).50; (3.2) unknown validity of measurements used and (3.4) less than 60% response rate (42% [68/159])
      25Walczak et al.
      • Walczak A.
      • Mazer B.
      • Butow P.N.
      • et al.
      A question prompt list for patients with advanced cancer in the final year of life: development and cross-cultural evaluation.
      A question prompt list for patients with advanced cancer in the final year of life: Development and cross-cultural evaluationAustralia + U.S.—one U.S. and two Australian treatment centersn = 34 oncology patients with advanced cancer (LE < 12 months as assessed by doctor) and n = 13 health professionals treating patientsPatients: Australia n = 15, U.S. n = 19. Health care professionals: Australia n = 7, U.S. n = 6QUAL—interviews, focus groups, and questionnaires, coding and thematic analysisIntervention: question prompt list (QPL)—booklet containing categorized lists of useful questions patients can ask their clinicianBoth the U.S. and Australian versions of the QPL intervention were well accepted and endorsed by patients and health care professionals. Differences in Australian + U.S. approaches to EOL and prognosis discussions emerged in the interviews.50; (1.3) no consideration of how findings relate to context and (1.4) no consideration of how findings relate to researchers' influence
      26Wallace et al.
      • Wallace M.P.
      • Weiner J.S.
      • Pekmezaris R.
      • et al.
      Physician cultural sensitivity in African American advance care planning: a pilot study.
      Physician cultural sensitivity in African American advance care planning: a pilot studyU.S.—three major teaching hospitals in NYn = 183 convenience sample of all physicians in attendance at medicine grand rounds on one specified dayWhite (n = 92), Asian (n = 48), and black Americans (n = 8) and non-U.S.–born black (n = 21)QUAN—nonrandomized cross-sectional questionnaire and statistics (Student's t-test, chi-squared test)Clinician knowledge and perceptions of AA patient preferences, cultural competence (measured by awareness of differences between ethnic groups in AD completion rates, knowledge of ACP, trust in clinicians)72% of doctors agreed that different racial/ethnic groups have distinct attitudes toward ADs. 58% acknowledged lack of familiarity with end-of-life preferences of AA patients, although this was more commonly reported among white doctors than black doctors. 88% of U.S.-born black doctors (7/8), vs. 35% of white doctors (32/91), and 26% (5/19) of non-U.S.–born black doctors, perceived that the Tuskegee experiment has impacted AA medical decision-making (P = 0.014).50; (3.1) participants not recruited in a way that minimizes selection bias (recruited from grand round participants, to ensure high response rate, however, limited generalizability) and (3.3) differences between groups being compared not accounted for (small groups, so multivariable analysis for potential confounders, e.g., level of training, not possible)
      27Zaide et al.
      • Zaide G.B.
      • Pekmezaris R.
      • Nouryan C.N.
      • et al.
      Ethnicity, race, and advance directives in an inpatient palliative care consultation service.
      Ethnicity, race, and advance directives in an inpatient palliative care consultation serviceU.S.—Long Island Jewish (LIJ) Medical Center, an academic tertiary hospital in New Hyde Park, New Yorkn = 400 patients who took part in a PCC at LIJ during the nine-month study period where chart had complete documentations of AD variables (533 total, 133 did not have full data recorded)White n = 187, AA n = 142, and Asian n = 44QUAN—nonrandomized cross-sectional hospital chart review, descriptive statistics (t-test, Mann-Whitney test, chi-squared test, Fisher's exact test, Bonferroni-adjusted pairwise comparisons)Intervention: PCC with multidisciplinary team to develop a plan of care including pain/symptom relief, establishing goals of care, integration of psychological and spiritual aspects of the patient and family experience and offering support to the family during the patient's illness and the bereavement period. Outcomes: completion of AD defined as DNR or DNIWhite patients completed more ADs than did nonwhite (AA, Asian, and Hispanic) patients before the intervention (25.67% vs. 12.68%) (P = 0.021). Further analysis showed AAs differed from white patients in the likelihood of AD execution rates pre-PCC, but not post-PCC. PCC intervention significantly reduced differences between white patients and AAs in completing ADs.75; (3.4) outcome data below 80% (75% [400/533])
      ACP = advance care planning; PC = palliative care; LW = living will; DPOA = durable power of attorney; DNR = do not resuscitate; AA = African American; EOL = end of life; NZ = New Zealand; WA = white American; COPD = chronic obstructive pulmonary disease; CPR = cardiopulmonary resuscitation; MV = mechanical ventilation; CKD = chronic kidney disease; AD = advance directive; DNI = do not intubate; NHHCS = National Home Health and Hospice Care Survey; HC proxy = health care proxy; AOR = adjusted odds ratio; RCT = randomized control trial; ARR = adjusted relative risk; PCC = palliative care consultation.
      a Methodology = QUAL (qualitative), QUAN (quantitative), or MIXED METHODS (combining qualitative and quantitative methods).

      Risk of Bias

      The quality of all included studies was assessed independently by E. M. and L. E. S. using the Mixed Methods Appraisal Tool (MMAT), version 2011.
      • Pluye P.
      • Robert E.
      • Cargo M.
      • et al.
      Proposal: A mixed methods appraisal tool for systematic mixed studies reviews.
      This validated tool was appropriate for this review as it can be applied to qualitative, quantitative (randomized, nonrandomized, and descriptive), and mixed-methods study designs. The tool uses a set of questions specific to study design, converted into four possible binary scores (worst to best: 25/50/75/100). Disagreements between the review authors over the quality appraisal of three particular studies were resolved through discussion, without requiring a third reviewer's consultation. No studies were excluded based on quality, but we reflect on the quality of the identified studies in the narrative synthesis.

      Analysis

      We conducted a narrative synthesis of the studies' findings, following the framework stages proposed by Popay et al.,
      • Popay J.
      • Roberts H.
      • Sowden A.
      • et al.
      Guidance on the conduct of narrative synthesis in systematic reviews.
      namely developing a preliminary synthesis, exploring relationships in the data, assessing the robustness of the synthesis product, and developing a theoretical model of how cultural factors influence the acceptability of ACP. The narrative synthesis was guided by our three review questions and was structured around study population characteristics, study settings, quality of studies, different cultural factors assessed/arising in the studies, different types of ACP examined in the studies, acceptability of ACP, and clinical implications and recommendations. To develop and present the narrative, we used several of the tools suggested by Popay et al.,
      • Popay J.
      • Roberts H.
      • Sowden A.
      • et al.
      Guidance on the conduct of narrative synthesis in systematic reviews.
      including textural description (noting findings that were unusual or relevant to emerging themes); conceptual mapping/tabulation (this included quantifying which studies featured different factors/settings/participant types, to ensure appropriate weighting was given to each in the discussion, as presented in Fig. 1); and thematic analysis (a common technique from qualitative data analysis that involves identifying and exploring the most important and/or recurrent themes or concepts appearing in the literature being reviewed
      • Popay J.
      • Roberts H.
      • Sowden A.
      • et al.
      Guidance on the conduct of narrative synthesis in systematic reviews.
      ). In our narrative, we gave greater weight to the findings and recommendations occurring in the greatest number of individual studies of higher quality.
      Figure thumbnail gr1
      Fig. 1Prisma (2009) flow diagram showing stages of study selection. ACP = advance care planning.
      E. M. led the synthesis, with regular meetings with L. E. S. to review emerging findings and patterns. Researchers' backgrounds inevitably influence interpretative aspects of analysis, particularly in research on cultural factors, and awareness of this contributed to the analysis. E. M. is a white British Medical and Global Health iBSc student who has previously worked nonclinically with a palliative care team; L. E. S. is a white British academic specializing in qualitative and mixed-methods social science, with a PhD in palliative care, experience of cross-cultural research, and an interest in cultural factors in care provision.

      Results

      Study Selection and Characteristics

      Eight hundred eighteen studies were identified and screened. A PRISMA flow chart detailing each stage of exclusion is shown in Fig. 2. The main reasons for exclusion were as follows: not about ACP as defined in our review question (n = 478), not original, published research (n = 125), and participants who did not have progressive, incurable disease (n = 95).
      Figure thumbnail gr2
      Fig. 2Concept map. ACP = advance care planning.
      Twenty-seven studies were included. Of these, 20 were quantitative (eight descriptive, nine nonrandomized cross-sectional, two nonrandomized cohort, and one randomized control trial [RCT]), four were qualitative, and three used mixed methods. The concept map in Fig. 1 details the distribution of types of participant, types of ACP, cultural factors measured, and setting by country.
      We characterized ACP processes as either “formal,” that is, involving production of written documentation or completion of legal processes, or “informal,” that is, involving the broader, holistic process of discussion and decision-making involving patients and/or their caregivers/families and health care professionals; we use this terminology in the synthesis. The distinction between formal and informal ACP processes has been detailed before.
      • Sinclair C.
      • Auret K.A.
      • Evans S.F.
      • et al.
      Advance care planning uptake among patients with severe lung disease: a randomised patient preference trial of a nurse-led, facilitated advance care planning intervention.
      • Brunnhuber K.N.
      • Stephen N.
      • Meier D.E.
      • Weissman D.E.
      • Woodcock J.
      Putting evidence into practice: Palliative care.
      Fifteen studies measured formal ACP processes: do not resuscitate (DNR) orders, ADs including living wills (LWs) and durable power of attorneys (DPOAs), and formal designation of a surrogate decision-maker/health care proxy. Fourteen of these were conducted in the U.S.; one was conducted in Japan.
      • Nakazawa K.
      • Kizawa Y.
      • Maeno T.
      • et al.
      Palliative care physicians' practices and attitudes regarding advance care planning in palliative care units in Japan: a nationwide survey.
      The most common features of studies were as follows: patients only as participants (n = 17), study of informal ACP only (n = 12), and consideration of the cultural factor of ethnicity/race (n = 18). Seventy four percent of studies (n = 20) were conducted in the U.S., and all studies were conducted in high-income countries. A full description of the 27 included studies and their main findings regarding the acceptability of ACP is given in Table 3.

      Risk of Bias

      On the MMAT, 10 studies scored 50, 15 studies scored 75, and two studies scored 100. Common limitations across studies were use of unvalidated tools and convenience samples. For qualitative studies, a common limitation was a lack of reflexivity, that is, “appropriate consideration … to how findings relate to researchers' influence” (question 1.4
      • Pluye P.
      • Robert E.
      • Cargo M.
      • et al.
      Proposal: A mixed methods appraisal tool for systematic mixed studies reviews.
      ). Of the four qualitative and three mixed-methods studies, only one study
      • Phipps E.
      • True G.
      • Harris D.
      • et al.
      Approaching the end of life: attitudes, preferences, and behaviors of African-American and white patients and their family caregivers.
      scored a “yes” for this question, as it gave a detailed explanation for using race-concordant interviewers. However, even this study did not address other factors, such as the researchers' disciplinary background(s).

      Narrative Synthesis

      The narrative synthesis is structured to reflect our three research questions.

      How, If at All, Does Patients' and Families' Cultural Background Influence the Acceptability of ACP for Seriously Ill Patients?

      Formal ACP

      Ethnicity

      Of the 14 U.S. studies measuring formal ACP processes, all but one

      Laguna J. Racial/ethnic variation in care preferences and care outcomes among United States hospice enrollees. Dissertation Abstracts International: Section B: The Sciences and Engineering Vol. 2015;76(3-B(E).

      found that nonwhite patients had lower levels of completion than white patients. Self-reported ethnicities included in these studies were white American (WA), African American (AA), Hispanic American (HA), and Asian American. For example, in a survey of bereaved relatives (n = 540) by Hopp et al.,
      • Hopp F.P.
      • Duffy S.A.
      Racial variations in end-of-life care.
      white patients were more likely to complete a LW (P = 0.001) and to designate a DPOA (P = 0.032) than black patients. One AA patient offered this reason for not feeling a need to complete an AD:“Uh, I don't have anything in writing, because when I ask my sisters' that's just like printing it in gold, stacking it in gold. They're going to do it.”
      • Braun U.K.
      • Beyth R.J.
      • Ford M.E.
      • Espadas D.
      • McCullough L.B.
      Decision-making styles of seriously ill male veterans for end-of-life care: Autonomists, Altruists, Authorizers, Absolute Trusters, and Avoiders.
      In contrast, Laguna's multivariate analysis

      Laguna J. Racial/ethnic variation in care preferences and care outcomes among United States hospice enrollees. Dissertation Abstracts International: Section B: The Sciences and Engineering Vol. 2015;76(3-B(E).

      found that HA hospice patients were 3.68 times as likely as WA hospice patients to have signed a DNR order (P < 0.01). However, this unexpected result is likely explained by the hospice setting, in which patients are likely to be more accepting of their poor prognosis. The extent to which formal ACP documentation is binding varies between states (U.S.)
      • Walczak A.
      • Mazer B.
      • Butow P.N.
      • et al.
      A question prompt list for patients with advanced cancer in the final year of life: development and cross-cultural evaluation.
      and countries,
      • Collins M.
      • Lehane E.
      Perspectives on death, dying and advance care planning from patients receiving haemodialysis.
      and this may also influence completion rates.

      Religiousness

      Several studies found evidence that religiousness may be either a mediating or a confounding factor in the association between nonwhite ethnicity/race and lower levels of formal ACP completion.
      • Smith A.K.
      • McCarthy E.P.
      • Paulk E.
      • et al.
      Racial and ethnic differences in advance care planning among patients with cancer: impact of terminal illness acknowledgment, religiousness, and treatment preferences.
      • Balboni T.A.
      • Vanderwerker L.C.
      • Block S.D.
      • et al.
      Religiousness and spiritual support among advanced cancer patients and associations with end-of-life treatment preferences and quality of life.
      • Garrido M.M.
      • Harrington S.T.
      • Prigerson H.G.
      End-of-life treatment preferences: a key to reducing ethnic/racial disparities in advance care planning? [References].
      • True G.
      • Phipps E.J.
      • Braitman L.E.
      • Harralson T.
      • Harris D.
      • Tester W.
      Treatment preferences and advance care planning at end of life: the role of ethnicity and spiritual coping in cancer patients.
      A quote from an AA patient demonstrates how these factors could be linked:“You don't have no say. The doctors have no say. Only the master has a say. So, you just wait on it.”
      • Braun U.K.
      • Beyth R.J.
      • Ford M.E.
      • Espadas D.
      • McCullough L.B.
      Decision-making styles of seriously ill male veterans for end-of-life care: Autonomists, Altruists, Authorizers, Absolute Trusters, and Avoiders.
      However, two studies that performed adjusted analysis found that religiousness did not explain the association.
      • Smith A.K.
      • McCarthy E.P.
      • Paulk E.
      • et al.
      Racial and ethnic differences in advance care planning among patients with cancer: impact of terminal illness acknowledgment, religiousness, and treatment preferences.
      • Balboni T.A.
      • Vanderwerker L.C.
      • Block S.D.
      • et al.
      Religiousness and spiritual support among advanced cancer patients and associations with end-of-life treatment preferences and quality of life.
      Furthermore, some tools used to assess “religiousness” were not validated
      • Smith A.K.
      • McCarthy E.P.
      • Paulk E.
      • et al.
      Racial and ethnic differences in advance care planning among patients with cancer: impact of terminal illness acknowledgment, religiousness, and treatment preferences.
      • True G.
      • Phipps E.J.
      • Braitman L.E.
      • Harralson T.
      • Harris D.
      • Tester W.
      Treatment preferences and advance care planning at end of life: the role of ethnicity and spiritual coping in cancer patients.
      and some response rates were below 60%.
      • Balboni T.A.
      • Vanderwerker L.C.
      • Block S.D.
      • et al.
      Religiousness and spiritual support among advanced cancer patients and associations with end-of-life treatment preferences and quality of life.
      • True G.
      • Phipps E.J.
      • Braitman L.E.
      • Harralson T.
      • Harris D.
      • Tester W.
      Treatment preferences and advance care planning at end of life: the role of ethnicity and spiritual coping in cancer patients.
      Further research is needed to determine how ethnicity and religion interact in relation to ACP uptake. Overall, our included studies confirm that formal ACP processes are less acceptable to nonwhite U.S. patients.

      Informal ACP

      Trust

      Findings regarding the acceptability of informal ACP processes were more mixed. The most commonly researched aspect of informal ACP was patients' engagement in EOL discussions. In the U.S., Hopp et al.
      • Hopp F.P.
      • Duffy S.A.
      Racial variations in end-of-life care.
      found that deceased AAs (n = 86) were 63% less likely than WAs (n = 454) to have had an EOL discussion before death (P < 0.001). However, this finding may not be generalizable to the general U.S. population as participants were more likely to be older, male, and have lower educational attainment. In addition, Hopp et al. do not specify if discussions were with clinicians or caregivers/family, which may plausibly differ in acceptability to AAs. A commonly hypothesized reason why AAs may be less likely to engage in ACP surrounds the community's widespread mistrust of the health care system. This mistrust has been attributed to historical events such as the Tuskegee Syphilis Study
      • Gamble V.N.
      Under the shadow of Tuskegee: African Americans and health care.
      and AAs' persistently limited access to health care due to socioeconomic/insurance barriers.
      • Crawley L.
      • Payne R.
      • Bolden J.
      • et al.
      Palliative and end-of-life care in the African American community.
      For example, one of the doctors interviewed by Periyakoil et al. stated, “some groups feel more marginalized in the community at large and this makes them more distrustful of the medical system as a whole.”
      • Periyakoil V.S.
      • Neri E.
      • Kraemer H.
      No Easy Talk: a mixed methods study of doctor reported barriers to conducting effective end-of-life conversations with diverse patients.
      Wallace et al. found that 88% of U.S.-born black physicians versus 35% of white physicians perceived that the Tuskegee experiment had impacted AA medical decision-making (P = 0.014), although the sample of black physicians was small (n = 8 compared with n = 91 white physicians).
      • Wallace M.P.
      • Weiner J.S.
      • Pekmezaris R.
      • et al.
      Physician cultural sensitivity in African American advance care planning: a pilot study.
      Contrastingly, Smith et al. found that, although AAs rated the quality of their patient-doctor relationships lower than WAs for several measures, difference in level of trust in doctors was not statistically significant.
      • Smith A.K.
      • Davis R.B.
      • Krakauer E.L.
      Differences in the quality of the patient-physician relationship among terminally ill African-American and White patients: impact on advance care planning and treatment preferences.
      However, the authors questioned the validity of the questionnaire used, suggesting it may have been skewed toward encouraging positive/“agree” responses.

      Comfort Discussing Death

      In contrast to Hopp et al.'s
      • Gamble V.N.
      Under the shadow of Tuskegee: African Americans and health care.
      finding that AAs were less likely to have had an EOL discussion before death, a small comparative study by Kurella et al. found that AA patients were more likely to engage in EOL discussions with health care professionals than WAs (70% vs. 38%; n = 23 and n = 21, respectively; P < 0.01).
      • Kurella Tamura M.
      • Goldstein M.K.
      • Perez-Stable E.J.
      Preferences for dialysis withdrawal and engagement in advance care planning within a diverse sample of dialysis patients.
      However, Kurella et al.'s use of patient self-reporting creates the possibility of recall bias. The authors suggest that, although unlikely, recall of an EOL discussion could plausibly vary between ethnic groups. One reason for this finding could be that patients' level of comfort discussing death influences how memorable EOL discussions are for them and therefore likelihood of recall.
      There is evidence that patients' and families' attitudes toward frank EOL discussions vary between cultural and ethnic groups. For example, in Ireland, Collins et al. found that 78% (n = 50) of hemodialysis patients reported being comfortable discussing death.
      • Collins M.
      • Lehane E.
      Perspectives on death, dying and advance care planning from patients receiving haemodialysis.
      Research in other settings/groups assessed clinicians' views of patients' comfort and identified cultural differences. In the U.S., doctors commented that“certain groups feel that honesty and frankness is not good for the emotional outlook”“some cultures approach (conversations about) death as something to be avoided at all costs.”
      • Periyakoil V.S.
      • Neri E.
      • Kraemer H.
      No Easy Talk: a mixed methods study of doctor reported barriers to conducting effective end-of-life conversations with diverse patients.
      In New Zealand, health care professionals reported that patients and families from Maori and Polynesian/Pacific cultures were reluctant to discuss death
      • Frey R.
      • Raphael D.
      • Bellamy G.
      • Gott M.
      Advance care planning for Maori, Pacific and Asian people: the views of New Zealand healthcare professionals.
      :“(to them) it's like hastening that possibility of their death” (Focus group with Maori patients)“shows that we don't have faith that this person's going to live.” (Focus group with Pacific patients)
      However, the latter findings arose from exploratory research in which there were two clinicians for each focus group (Maori and Pacific); it is unlikely that data saturation was achieved and therefore transferability is limited.
      As communication is a two-way process, comfort of staff discussing death is equally essential for informal ACP. In The Netherlands, patients reported that most doctors (>80%) did not engage them in EOL discussions about prognosis, the dying process, or religious/spiritual beliefs (n = 185).
      • Janssen D.J.
      • Spruit M.A.
      • Schols J.M.
      • Wouters E.F.
      A call for high-quality advance care planning in outpatients with severe COPD or chronic heart failure.
      These findings suggest that the need, in both formal and informal ACP, to discuss death may be a barrier to acceptability for patients and clinicians from some cultural groups, both within the U.S. and globally.

      Patient Attitudes and Additional Factors

      Additional cultural factors influencing the acceptability of ACP that were investigated on a lesser scale/by fewer studies than those discussed previously were caregivers' acculturation level
      • DeSanto-Madeya S.
      • Nilsson M.
      • Loggers E.T.
      • et al.
      Associations between United States acculturation and the end-of-life experience of caregivers of patients with advanced cancer.
      ; patients' attitude to acknowledging a terminal diagnosis
      • Loggers E.T.
      • Maciejewski P.K.
      • Jimenez R.
      • et al.
      Predictors of intensive end-of-life and hospice care in Latino and white advanced cancer patients.
      ; collective (family-centered) versus autonomous (patient-centered) approaches to decision-making in American veterans
      • Braun U.K.
      • Beyth R.J.
      • Ford M.E.
      • Espadas D.
      • McCullough L.B.
      Decision-making styles of seriously ill male veterans for end-of-life care: Autonomists, Altruists, Authorizers, Absolute Trusters, and Avoiders.
      and New Zealanders
      • Frey R.
      • Raphael D.
      • Bellamy G.
      • Gott M.
      Advance care planning for Maori, Pacific and Asian people: the views of New Zealand healthcare professionals.
      ; and patient preference for who to involve in EOL conversations and decision-making in Japan
      • Gabbay B.B.
      • Matsumura S.
      • Etzioni S.
      • et al.
      Negotiating end-of-life decision making: a comparison of Japanese and U.S. residents' approaches.
      and Hawaii.
      • Kataoka-Yahiro M.R.
      • Conde F.A.
      • Wong R.S.
      • Page V.
      • Peller B.
      Advance care planning among Asian Americans and Native Hawaiians receiving haemodialysis.
      Two studies suggested how clinicians discuss EOL care might differ depending on patient ethnicity: Sharma et al.
      • Sharma R.K.
      • Dy S.M.
      Documentation of information and care planning for patients with advanced cancer: associations with patient characteristics and utilization of hospital care.
      found black race was associated with higher odds of hospice discussion, whereas Mack et al.
      • Mack J.W.
      • Paulk M.E.
      • Viswanath K.
      • Prigerson H.G.
      Racial disparities in the outcomes of communication on medical care received near death.
      found EOL discussions resulted in increased awareness of illness being terminal among white patients, but not black patients. Further research into these factors is required before solid conclusions can be drawn about their respective influence on the uptake of ACP.

      How Might ACP Need to Be Adapted to Make It More Cross-Culturally Appropriate?

      Recommendations for a Focus on Communication in ACP

      The wide variety of findings discussed in answer to the first research question, particularly regarding the acceptability of informal ACP, reflect the complex and intricate nature of EOL communications. Many authors of included studies made recommendations for how ACP could be adapted to be more cross-culturally appropriate that emphasized a broader, communication-focused approach to ACP, as opposed to a process focused on formal documentation. For example, Hopp et al. concluded that “ACP needs to be part of a broader strategy of communication with patients concerning EOL care.”
      • Hopp F.P.
      • Duffy S.A.
      Racial variations in end-of-life care.
      Communication-focused interventions were investigated by four of our included papers:
      Song et al. conducted an RCT with semistructured interviews to evaluate the SPIRIT (Sharing Patients' Illness Representations to Increase Trust) intervention.
      • Song M.K.
      • Ward S.E.
      • Happ M.B.
      • et al.
      Randomized controlled trial of SPIRIT: an effective approach to preparing African-American dialysis patients and families for end of life.
      SPIRIT aims to improve EOL communication between AA end-stage renal disease patients and their chosen surrogate decision-makers and involves a single-session interview with a patient-surrogate dyad. SPIRIT was found to significantly improve communication, reported by both patients and surrogates (n = 58 dyads), at one week (OR 4.40, P = 0.02) and three months (OR 3.17, P = 0.03), compared with baseline, although these findings are limited by only 54% of patient-surrogate dyads agreeing to be included in follow-up.
      • Song M.K.
      • Ward S.E.
      • Happ M.B.
      • et al.
      Randomized controlled trial of SPIRIT: an effective approach to preparing African-American dialysis patients and families for end of life.
      Walczak et al. tested the acceptability of two versions of a question prompt list (QPL) intervention for patients with advanced cancer, one developed in Australia and one in America.
      • Walczak A.
      • Mazer B.
      • Butow P.N.
      • et al.
      A question prompt list for patients with advanced cancer in the final year of life: development and cross-cultural evaluation.
      The QPL aims to improve prognosis and EOL discussions between patients and clinicians outside of palliative care settings, in turn leading to better understanding of prognosis and more informed decision-making. Interviews and focus groups with patients (n = 34) and health care professionals (n = 13) in both countries found that patient and staff participants from both settings generally endorsed the QPL tool. However, there were contrasts between Australian and U.S. responses, with U.S. participants generally more hesitant to discuss sensitive content (e.g., prognosis) and using more euphemisms. These contrasts reinforce the need for population-tailored and culturally appropriate versions of communication aides such as QPL.
      Perry et al. conducted an RCT of peer mentoring on EOL decision-making for dialysis patients (n = 203).
      • Perry E.
      • Swartz J.
      • Brown S.
      • Smith D.
      • Kelly G.
      • Swartz R.
      Peer mentoring: a culturally sensitive approach to end-of-life planning for long-term dialysis patients.
      They found that AD completion rate was 35% in the group who received peer mentoring, compared with 12% in the group that received standard printed materials and 10% in the group receiving no specified intervention (P < 0.01). The greatest increase was among AAs (P < 0.001), for whom peer mentoring also appeared to improve subjective well-being (P < 0.05). They suggest that oral, rather than written, EOL education may be more acceptable for AAs. However, the study report is limited in that randomization and allocation concealment are not described.
      Zaide et al. conducted a cross-sectional medical chart review over a nine-month period for patients seen by the palliative care consultation (PCC) service in a tertiary hospital in New York, exploring whether the PCC intervention influenced rates of AD completion for different racial/ethnic groups.
      • Zaide G.B.
      • Pekmezaris R.
      • Nouryan C.N.
      • et al.
      Ethnicity, race, and advance directives in an inpatient palliative care consultation service.
      They found that AD completion rates were higher for both AA (n = 142) and WA (n = 187) patients following the intervention (pre-PCC: AA 11.3% and WA 25.7%; post-PCC: AA 28.9% and WA 33.7%; P < 0.001). The difference between the two groups' completion rates appears to be significantly smaller following the intervention (14.4 before vs. 4.8 after), suggesting the PCC may “level the playing field” regarding acceptability of ADs for WAs and AAs.
      • Zaide G.B.
      • Pekmezaris R.
      • Nouryan C.N.
      • et al.
      Ethnicity, race, and advance directives in an inpatient palliative care consultation service.
      In sum, these findings suggest that communication-focused EOL interventions have potential to increase acceptability and uptake of both informal
      • Song M.K.
      • Ward S.E.
      • Happ M.B.
      • et al.
      Randomized controlled trial of SPIRIT: an effective approach to preparing African-American dialysis patients and families for end of life.
      • Walczak A.
      • Mazer B.
      • Butow P.N.
      • et al.
      A question prompt list for patients with advanced cancer in the final year of life: development and cross-cultural evaluation.
      and formal
      • Perry E.
      • Swartz J.
      • Brown S.
      • Smith D.
      • Kelly G.
      • Swartz R.
      Peer mentoring: a culturally sensitive approach to end-of-life planning for long-term dialysis patients.
      • Zaide G.B.
      • Pekmezaris R.
      • Nouryan C.N.
      • et al.
      Ethnicity, race, and advance directives in an inpatient palliative care consultation service.
      ACP processes, particularly for AA patients.

      What Cultural Factors Do Clinicians Need to Be Aware of in Approaching Communication and Planning with Patients and Families About EOL Issues?

      Clinician Training and Education in Cultural Competence

      Several authors of included studies were called for staff training and education on topics including communication skills,
      • Nakazawa K.
      • Kizawa Y.
      • Maeno T.
      • et al.
      Palliative care physicians' practices and attitudes regarding advance care planning in palliative care units in Japan: a nationwide survey.
      initiating discussions with diverse and potentially hesitant patients,
      • Phipps E.
      • True G.
      • Harris D.
      • et al.
      Approaching the end of life: attitudes, preferences, and behaviors of African-American and white patients and their family caregivers.
      ,
      • Zaide G.B.
      • Pekmezaris R.
      • Nouryan C.N.
      • et al.
      Ethnicity, race, and advance directives in an inpatient palliative care consultation service.
      careful use of language and avoidance of medical jargon,
      • Periyakoil V.S.
      • Neri E.
      • Kraemer H.
      No Easy Talk: a mixed methods study of doctor reported barriers to conducting effective end-of-life conversations with diverse patients.
      the potential influence of religiousness/spirituality at EOL,
      • Balboni T.A.
      • Vanderwerker L.C.
      • Block S.D.
      • et al.
      Religiousness and spiritual support among advanced cancer patients and associations with end-of-life treatment preferences and quality of life.
      • True G.
      • Phipps E.J.
      • Braitman L.E.
      • Harralson T.
      • Harris D.
      • Tester W.
      Treatment preferences and advance care planning at end of life: the role of ethnicity and spiritual coping in cancer patients.
      and more general promotion of staff's “cultural competence” and sensitivity.
      • Gabbay B.B.
      • Matsumura S.
      • Etzioni S.
      • et al.
      Negotiating end-of-life decision making: a comparison of Japanese and U.S. residents' approaches.
      ,
      • Braun U.K.
      • Ford M.E.
      • Beyth R.J.
      • et al.
      The physician's professional role in end-of-life decision-making: voices of racially and ethnically diverse physicians.
      ,
      • Hopp F.P.
      • Duffy S.A.
      Racial variations in end-of-life care.
      ,
      • Phipps E.
      • True G.
      • Harris D.
      • et al.
      Approaching the end of life: attitudes, preferences, and behaviors of African-American and white patients and their family caregivers.
      Wallace et al.
      • Wallace M.P.
      • Weiner J.S.
      • Pekmezaris R.
      • et al.
      Physician cultural sensitivity in African American advance care planning: a pilot study.
      suggested that clinicians should incorporate existing culturally sensitive guidelines, such as the U.S. Federal Cultural and Linguistic Appropriate Services (CLAS),
      National CLAS Standards - The Office of Minority Health: U.S
      Department of Health and Human Services Office of Minority Health.
      into their EOL care.

      Avoiding Stereotyping

      Another common recommendation was that clinicians working cross-culturally should aim to avoid stereotyping patients and caregivers based on racial/ethnic or cultural groups and not lose sight of the individual. Some studies mentioned this explicitly; for example, Braun et al. suggested that clinicians “should not stereotype a patient into a specific decision-making variant based on their ethnicity/race”
      • Braun U.K.
      • Beyth R.J.
      • Ford M.E.
      • Espadas D.
      • McCullough L.B.
      Decision-making styles of seriously ill male veterans for end-of-life care: Autonomists, Altruists, Authorizers, Absolute Trusters, and Avoiders.
      and should only engage with ethnicity/race when “clinically relevant”
      • Braun U.K.
      • Ford M.E.
      • Beyth R.J.
      • et al.
      The physician's professional role in end-of-life decision-making: voices of racially and ethnically diverse physicians.
      (e.g., to EOL decision-making). Similarly, Frey et al. caution that health care professionals should not make “blanket judgments” about an individual's level of interest in ACP based on cultural background.
      • Frey R.
      • Raphael D.
      • Bellamy G.
      • Gott M.
      Advance care planning for Maori, Pacific and Asian people: the views of New Zealand healthcare professionals.
      Hopp et al. highlight the importance of considering the variations within, as well as between, racial/ethnic groups with regard to “beliefs, behaviors, and decisions related to health care at the end of life.”
      • Hopp F.P.
      • Duffy S.A.
      Racial variations in end-of-life care.
      The importance of not losing sight of the individual patient when working cross-culturally links to the recommendation of “cultural competence” training. For example, Wallace et al. suggest that clinicians should strive to know their patients as “individuals within their larger social context” and that clinician training should go beyond “simply learning about epidemiologic differences among racial and ethnic groups because this will not capture the subtle variations from person to person.”
      • Wallace M.P.
      • Weiner J.S.
      • Pekmezaris R.
      • et al.
      Physician cultural sensitivity in African American advance care planning: a pilot study.

      Discussion

      This is the first review to comprehensively synthesize evidence regarding the cultural acceptability of ACP. We found that most research to date in this area has been conducted in the U.S. The main cultural factor examined by researchers was ethnicity/race, and the main types of ACP studied were formal, documented processes. The included studies indicate the influence of a wide range of interconnected cultural factors on the acceptability of ACP for patients and caregivers/families. A common finding was that, for seriously ill patients in the U.S., nonwhite ethnicity was associated with lower acceptability of formal ACP processes. Greater levels of religiosity appear to be a factor in this association, although further investigation is required to confirm and define this relationship.
      Additional cultural factors thought to influence the acceptability of ACP were patients' degree of trust in clinicians and the wider health care system, and their comfort discussing death and EOL issues. Less frequently studied factors included caregivers' acculturation level, collective versus autonomous approaches to decision-making, patients' attitude to acknowledging a terminal diagnosis and preference for who to involve in EOL conversations and decision-making, and differences in how clinicians discuss EOL care. Further research into these factors, particularly patient attitudes, is required to better understand the nature and mechanisms of their influence on the acceptability of ACP.
      Although the concept of culture exceeds ethnic, national, and linguistic boundaries, ethnicity is commonly used as a proxy for culture in research
      • Moscou S.
      The conceptualization and operationalization of race and ethnicity by health services researchers.
      and is recommended as an appropriate proxy for culture.
      • Chin M.H.
      • Humikowski C.A.
      When is risk stratification by race or ethnicity justified in medical care?.
      A key strength of this review is that we expanded its focus beyond just ethnicity, by including religion/spirituality, nationality, or country of permanent residence. Our broad operationalization of “culture” will have influenced our findings; for example, including religion/spirituality as a search term will have likely increased the degree to which religion features in the findings, themes, and conclusions. However, none of the included studies defined study population solely by religious groups—it was always discussed in relation to ethnicity. Our use of narrative synthesis methodology allowed for the inclusion and comparison of quantitative, qualitative, and mixed-method studies. Other strengths include our comprehensive search strategy developed using MeSH terms and our use of dual independent quality appraisal and data extraction.
      A researcher's own cultural background inevitably shapes the research they conduct, as our understanding of the world is inherently shaped by our cultural assumptions.
      • Realism Madill A.
      In the context of this review, our cultural backgrounds influenced our interpretations of findings, which underpinned the narrative synthesis we produced. While accepting this influence, particularly considering the topic of the review, we took steps to minimize unconscious bias by using concept mapping to give greatest weight to common findings reported by most studies. In addition, we were mindful to iteratively ground all our conclusions and interpretations in the included studies' empirical data. However, as with any synthesis, our interpretations are limited by being “a step removed” from the context and participants of the original studies, therefore relying on how studies were reported. Although participants' characteristics were generally well described, several studies lacked reflexivity, omitting to consider the potential influence of setting or of the researchers themselves. In some cases, this limited the interpretation of authors' conclusions, which is reflected in our synthesis.
      Finally, although a systematic approach was taken in our search strategy, the possibility remains that some relevant papers may have been missed. Our exclusion of papers not written in English was a limitation, given that we were searching for global research surrounding culture. However, only two papers were excluded for this reason, one written in French and one in Hungarian. Nevertheless, research conducted in European countries is underrepresented in our review, so these papers could have been potentially valuable.
      Findings from this review suggest that first, formal ACP processes (ADs, LWs, and DPOAs) need better evaluation for cultural sensitivity, as some forms of formal ACP are not acceptable to some cultural groups; and second, more informal, discussion-based ACP may be more acceptable in some cultural groups. Research with patients and caregivers has resulted in similar recommendations that ACP should move “beyond ADs” toward a broader decision-making process involving patients, surrogates, and clinicians.
      • McMahan R.D.
      • Knight S.J.
      • Fried T.R.
      • Sudore R.L.
      Advance care planning beyond advance directives: perspectives from patients and surrogates.
      The importance of communication is already at the heart of palliative care ethos.
      • Hutton N.
      Palliative care, time, and core values.
      However, given that EOL care and ACP are often delivered by nonspecialist clinicians, staff outside the specialty should be encouraged and supported to follow suit in their approach. Common recommendations for how ACP could be adapted to become more culturally sensitive related to shifting toward a more communication-focused approach to EOL care and ACP. Research into communication-focused interventions found promising results in increased acceptability and uptake of both informal and formal ACP processes.
      • Perry E.
      • Swartz J.
      • Brown S.
      • Smith D.
      • Kelly G.
      • Swartz R.
      Peer mentoring: a culturally sensitive approach to end-of-life planning for long-term dialysis patients.
      • Song M.K.
      • Ward S.E.
      • Happ M.B.
      • et al.
      Randomized controlled trial of SPIRIT: an effective approach to preparing African-American dialysis patients and families for end of life.
      • Walczak A.
      • Mazer B.
      • Butow P.N.
      • et al.
      A question prompt list for patients with advanced cancer in the final year of life: development and cross-cultural evaluation.
      ,
      • Zaide G.B.
      • Pekmezaris R.
      • Nouryan C.N.
      • et al.
      Ethnicity, race, and advance directives in an inpatient palliative care consultation service.
      Many of the included studies called for clinician training and education in “cultural competence,” including how cultural background might influence patients' views on EOL issues and ACP. Where absent, this education should be integrated into existing training and curricula. Established initiatives such as the End-of-Life Nursing Education Consortium (ELNEC) (www.aacnnursing.org/ELNEC), which provides training adapted for many different cultural groups, also play an important role. However, criticisms have been leveled at some current approaches to clinical cultural competence, including in EOL care. For example, Tervalon and Murray-Garcia suggest that the traditional idea of medical competence as “detached mastery” is not appropriate to the inevitable complexity of cross-cultural care and should instead be replaced with “cultural humility.”
      • Tervalon M.
      • Murray-García J.
      Cultural humility versus cultural competence: a critical distinction in defining physician training outcomes in multicultural education.
      The complex reality of cross-cultural care is also reflected in Gunaratnam's suggestion that current approaches to cultural competence marginalize the “role of the nonrational and visceral.”
      • Gunaratnam Y.
      Intercultural palliative care: do we need cultural competence?.
      Gunaratnam argues that these marginalized aspects should be recognized in policy and education that aims to promote cultural competence. There is clearly a need for clinicians, educators, and researchers to reflect on what “cultural competence” really means in EOL care.
      • Paal P.
      • Buekki J.
      If I had stayed back home, I would not be alive any more - exploring end-of-life preferences in patients with migration background.
      Many authors highlighted a further, related complexity of cross-cultural care: avoiding stereotyping or losing sight of the individual patient. Recognizing the importance of culture must not lead to cultural essentialism. These complexities and criticisms should be engaged with as the concept of cultural competence becomes adopted into EOL care education for nonspecialists.
      Further research is needed into the cultural acceptability of ACP in countries other than the U.S.: the UK, other European countries, and developing country settings are underrepresented. A recent study from Singapore, published after our literature search, illustrates the benefits of examining ACP in different cultural settings.
      • Menon S.
      • Kars M.C.
      • Malhotra C.
      • Campbell A.V.
      • van Delden J.J.M.
      Advance care planning in a multicultural family centric community: a qualitative study of health care professionals', patients', and caregivers' perspectives.
      The study highlights the need for a nuanced approach to ACP that considers the family network in multicultural, family-centric communities. Further research into cultural adaptations of ACP is needed. To build on research conducted in the U.S., studies are needed to determine causal and directional relationships between religiousness, nonwhite ethnicity and lower levels of formal ACP completion, and the mechanisms involved. The role of discomfort or fear of discussing death as a barrier to ACP also warrants further research. Although beyond the scope of the current review, some included studies discussed the influence of a clinician's own cultural background on ACP and EOL care delivery.
      • Braun U.K.
      • Ford M.E.
      • Beyth R.J.
      • et al.
      The physician's professional role in end-of-life decision-making: voices of racially and ethnically diverse physicians.
      ,
      • Gabbay B.B.
      • Matsumura S.
      • Etzioni S.
      • et al.
      Negotiating end-of-life decision making: a comparison of Japanese and U.S. residents' approaches.
      ,
      • Nakazawa K.
      • Kizawa Y.
      • Maeno T.
      • et al.
      Palliative care physicians' practices and attitudes regarding advance care planning in palliative care units in Japan: a nationwide survey.
      ,
      • Periyakoil V.S.
      • Neri E.
      • Kraemer H.
      No Easy Talk: a mixed methods study of doctor reported barriers to conducting effective end-of-life conversations with diverse patients.
      ,
      • Wallace M.P.
      • Weiner J.S.
      • Pekmezaris R.
      • et al.
      Physician cultural sensitivity in African American advance care planning: a pilot study.
      Given that EOL communication is a two-way process, the influence of clinicians' cultural background is another area that requires evidence synthesis. Future research into communication-focused ACP interventions and solutions to overcome resource barriers to their implementation would also be valuable. The volunteer-led peer mentoring intervention trialed by Perry et al.
      • Perry E.
      • Swartz J.
      • Brown S.
      • Smith D.
      • Kelly G.
      • Swartz R.
      Peer mentoring: a culturally sensitive approach to end-of-life planning for long-term dialysis patients.
      may be a useful model here. More generally, research on the cultural acceptability of ACP would benefit from improved study designs and reporting. Qualitative and mixed-methods researchers should reflect on their cultural background and discipline, as well as the characteristics of those collecting data, in line with the MMAT guidance
      • Pluye P.
      • Robert E.
      • Cargo M.
      • et al.
      Proposal: A mixed methods appraisal tool for systematic mixed studies reviews.
      and COREQ checklist.
      • Tong A.
      • Sainsbury P.
      • Craig J.
      Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups.

      Conclusions

      Many interconnected cultural factors influence the acceptability of ACP. Although causal relationships and mechanisms of action remain unclear, nonwhite ethnicity is associated with lower acceptability of formal ACP, at least in the U.S. ACP does need to be adapted to meet the needs of culturally diverse populations. A less formal, communication-focused approach to ACP may be valuable in this regard.

      Disclosures and Acknowledgments

      The authors thank Sarah Herring at the University of Bristol Medical Sciences library for her support developing the search strategy. This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. The authors have no competing interests.

      References

      1. End of Life Care: An Ethical Overview: Center for Bioethics - University of Minnesota. 2005 (Available from)
        • Asch D.A.
        • Hansen-Flaschen J.
        • Lanken P.N.
        Decisions to limit or continue life-sustaining treatment by critical care physicians in the United States: conflicts between physicians' practices and patients' wishes.
        2012 (2012)
        • Breen C.M.A.
        • Abbott A.P.
        • Katherine H.
        • et al.
        Conflict associated with decisions to limit life-sustaining treatment in intensive care Units.
        J Gen Intern Med. 2017; 16: 283-289
        • Berlinger N.
        • Jennings B.
        • Wolf S.M.
        The Hastings Center guidelines for decisions on life-sustaining treatment and care near the end of life: revised and expanded second edition. Oxford University Press, 2013
        • Truog R.D.
        • Campbell M.L.
        • Curtis J.R.
        • et al.
        Recommendations for end-of-life care in the intensive care unit: a consensus statement by the American College [corrected] of Critical Care Medicine.
        Crit Care Med. 2008; 36: 953-963
        • Covinsky K.E.G.
        • Lee Cook
        • Francis E.
        • et al.
        The impact of serious illness on patients' families.
        JAMA. 2017; 272: 1839-1844
        • Sudore R.L.
        • Lum H.D.
        • You J.J.
        • et al.
        Defining advance care planning for adults: a consensus Definition from a Multidisciplinary Delphi Panel.
        J Pain Symptom Manage. 2017; 53: 821-832.e1
        • Detering K.M.
        • Hancock A.D.
        • Reade M.C.
        • Silvester W.
        The impact of advance care planning on end of life care in elderly patients: randomised controlled trial.
        BMJ. 2010; 340
        • Silveira M.J.
        • Kim S.Y.
        • Langa K.M.
        Advance directives and outcomes of surrogate decision making before death.
        N Engl J Med. 2010; 362: 1211-1218
        • Houben C.H.M.
        • Spruit M.A.
        • Groenen M.T.J.
        • Wouters E.F.M.
        • Janssen D.J.A.
        Efficacy of advance care planning: a systematic review and meta-analysis.
        J Am Med Dir Assoc. 2014; 15: 477-489
        • Dixon J.
        • Matosevic T.
        • Knapp M.
        The economic evidence for advance care planning: systematic review of evidence.
        Palliat Med. 2015; 29: 869-884
      2. AMA Backs Advance Care Planning by Patients. 2006 (Available from)
      3. Advance decisions and proxy decision-making in medical treatment and research Guidance from the BMA's Medical Ethics Department. 2007 (Available from)
        • AMA Code of Medical Ethics
        Opinions on care at the end of life.
        Virtual Mentor. 2013; (Available from)
        • Gysels M.
        • Evans N.
        • Menaca A.
        • et al.
        Culture and end of life care: a scoping exercise in seven European countries.
        PLoS One. 2012; 7: e34188
      4. Improving Supportive and Palliative Care for Adults with Cancer: National Institute for Clinical Excellence - NHS. 2004 (updated 16.5.17. Available from)
      5. Ethics and Advance Planning for End-of-Life Care: American Academy of Family Physicians. 2017 (Available from)
        • Calanzani Natalia K.J.
        • Higginson Irene J.
        Palliative and end of life care for Black, Asian and Minority Ethnic groups in the UK, Demographic profile and the current state of palliative and end of life care provision.
        2013 (19.5.2017. Available from)
        • Frey R.
        • Raphael D.
        • Bellamy G.
        • Gott M.
        Advance care planning for Maori, Pacific and Asian people: the views of New Zealand healthcare professionals.
        Health Soc Care Community. 2014; 22: 290-299
        • Johnstone M.J.
        Ethics and advance care planning in a culturally diverse Society.
        J Transcultural Nurs. 2009; 20: 405-416
        • Geronimo Jimenez W.S.T.
        • Amrit K Virk
        • Chan Kee
        • Low J.C.
        • Andy Hau
        • Yan Ho
        Overview of systematic reviews of advance care planning: Summary of evidence and global Lessons.
        J Pain Symptom Manage. 2018;
        • Taylor E.J.
        Spirituality, culture, and cancer care.
        Semin Oncol Nurs. 2001; 17: 197-205
        • Martsolf D.S.
        Cultural aspects of spirituality in cancer care.
        Semin Oncol Nurs. 1997; 13: 231-236
        • Keesing R.M.
        AJSE. Cultural anthropology: A contemporary perspective.
        3rd ed. Harcourt Brace College Publishers, Fort Worth, TX1998
        • Lovell A.
        • Yates P.
        Advance Care Planning in palliative care: a systematic literature review of the contextual factors influencing its uptake 2008-2012.
        Palliat Med. 2014; 28: 1026-1035
        • Kwak J.
        • Haley W.E.
        Current research findings on end-of-life decision making among racially or ethnically diverse groups.
        Gerontologist. 2005; 45: 634-641
        • Fried T.R.
        • O'Leary J.
        • Van Ness P.
        • Fraenkel L.
        Inconsistency over time in the preferences of older persons with advanced illness for life-sustaining treatment.
        J Am Geriatr Soc. 2007; 55: 1007-1014
        • Ko E.
        • Lee J.
        End-of-life communication: ethnic differences between Korean American and non-Hispanic White older adults.
        J Aging Health Vol. 2009; 21: 967-984
        • Froman R.
        • Owen S.
        Randomized study of stability and change in patients' advance directives.
        Res Nurs Health [Internet]. 2005; 28 (Available from): 398-407
        • Popay J.
        • Roberts H.
        • Sowden A.
        • et al.
        Guidance on the conduct of narrative synthesis in systematic reviews.
        A Product ESRC Methods Programme Version. 2006; 1: b92
        • Selman L.
        • McDermott E.
        How do cultural factors influence advance care planning in progressive, incurable disease? A systematic review with narrative synthesis - protocol: PROSPERO.
        2017 (Available from)
      6. Ethnicity (noun) definition Oxford Dictionaries, accessed online. Available from https://en.oxforddictionaries.com/definition/ethnicity.

        • Bhopal R.
        Glossary of terms relating to ethnicity and race: for reflection and debate.
        J Epidemiol Community Health. 2004; 58: 441-445
        • Speck P.
        Culture and spirituality: essential components of palliative care.
        Postgrad Med J. 2016; 92: 341-345
        • Johnson K.S.
        • Kuchibhatla M.
        • Tulsky J.A.
        What explains racial differences in the use of advance directives and attitudes toward hospice care? [References].
        J Am Geriatr Soc Vol. 2008; 56: 1953-1958
        • Mebane E.W.
        • Oman R.F.
        • Kroonen L.T.
        • Goldstein M.K.
        The influence of physician race, age, and gender on physician attitudes toward advance care directives and preferences for end-of-life decision-making.
        J Am Geriatr Soc. 1999; 47: 579-591
        • Perkins H.S.
        • Geppert C.M.A.
        • Gonzales A.
        • Cortez J.D.
        • Hazuda H.P.
        Cross-cultural similarities and differences in attitudes about advance care planning.
        J Gen Intern Med. 2002; 17: 48-57
        • Smith A.K.
        • McCarthy E.P.
        • Paulk E.
        • et al.
        Racial and ethnic differences in advance care planning among patients with cancer: impact of terminal illness acknowledgment, religiousness, and treatment preferences.
        J Clin Oncol. 2008; 26: 4131-4137
        • Smith A.K.
        • Davis R.B.
        • Krakauer E.L.
        Differences in the quality of the patient-physician relationship among terminally ill African-American and White patients: impact on advance care planning and treatment preferences.
        J Gen Intern Med Vol. 2007; 22: 1579-1582
        • Balboni T.A.
        • Vanderwerker L.C.
        • Block S.D.
        • et al.
        Religiousness and spiritual support among advanced cancer patients and associations with end-of-life treatment preferences and quality of life.
        J Clin Oncol. 2007; 25: 555-560
        • Braun U.K.
        • Ford M.E.
        • Beyth R.J.
        • et al.
        The physician's professional role in end-of-life decision-making: voices of racially and ethnically diverse physicians.
        Patient Education Couns. 2010; 80: 3-9
        • Braun U.K.
        • Beyth R.J.
        • Ford M.E.
        • Espadas D.
        • McCullough L.B.
        Decision-making styles of seriously ill male veterans for end-of-life care: Autonomists, Altruists, Authorizers, Absolute Trusters, and Avoiders.
        Patient Educ Couns. 2014; 94: 334-341
        • Collins M.
        • Lehane E.
        Perspectives on death, dying and advance care planning from patients receiving haemodialysis.
        J Ren Care. 2013; 39: 5-11
        • DeSanto-Madeya S.
        • Nilsson M.
        • Loggers E.T.
        • et al.
        Associations between United States acculturation and the end-of-life experience of caregivers of patients with advanced cancer.
        J Palliat Med. 2009; 12: 1143-1149
        • Gabbay B.B.
        • Matsumura S.
        • Etzioni S.
        • et al.
        Negotiating end-of-life decision making: a comparison of Japanese and U.S. residents' approaches.
        Acad Med. 2005; 80: 617-621
        • Garrido M.M.
        • Harrington S.T.
        • Prigerson H.G.
        End-of-life treatment preferences: a key to reducing ethnic/racial disparities in advance care planning? [References].
        Cancer Vol. 2014; 120: 3981-3986
        • Hopp F.P.
        • Duffy S.A.
        Racial variations in end-of-life care.
        J Am Geriatr Soc. 2000; 48: 658-663
        • Janssen D.J.
        • Spruit M.A.
        • Schols J.M.
        • Wouters E.F.
        A call for high-quality advance care planning in outpatients with severe COPD or chronic heart failure.
        Chest. 2011; 139: 1081-1088
        • Kataoka-Yahiro M.R.
        • Conde F.A.
        • Wong R.S.
        • Page V.
        • Peller B.
        Advance care planning among Asian Americans and Native Hawaiians receiving haemodialysis.
        Int J Palliat Nurs. 2010; 16: 32-40
        • Kurella Tamura M.
        • Goldstein M.K.
        • Perez-Stable E.J.
        Preferences for dialysis withdrawal and engagement in advance care planning within a diverse sample of dialysis patients.
        Nephrol Dial Transplant. 2010; 25: 237-242
      7. Laguna J. Racial/ethnic variation in care preferences and care outcomes among United States hospice enrollees. Dissertation Abstracts International: Section B: The Sciences and Engineering Vol. 2015;76(3-B(E).

        • Loggers E.T.
        • Maciejewski P.K.
        • Jimenez R.
        • et al.
        Predictors of intensive end-of-life and hospice care in Latino and white advanced cancer patients.
        J Palliat Med. 2013; 16: 1249-1254
        • Mack J.W.
        • Paulk M.E.
        • Viswanath K.
        • Prigerson H.G.
        Racial disparities in the outcomes of communication on medical care received near death.
        Arch Intern Med. 2010; 170: 1533-1540
        • Nakazawa K.
        • Kizawa Y.
        • Maeno T.
        • et al.
        Palliative care physicians' practices and attitudes regarding advance care planning in palliative care units in Japan: a nationwide survey.
        Am J Hosp Palliat Med Vol. 2014; 31: 699-709
        • Periyakoil V.S.
        • Neri E.
        • Kraemer H.
        No Easy Talk: a mixed methods study of doctor reported barriers to conducting effective end-of-life conversations with diverse patients.
        PLoS One. 2015; 10: e0122321
        • Perry E.
        • Swartz J.
        • Brown S.
        • Smith D.
        • Kelly G.
        • Swartz R.
        Peer mentoring: a culturally sensitive approach to end-of-life planning for long-term dialysis patients.
        Am J Kidney Dis. 2005; 46: 111-119
        • Phipps E.
        • True G.
        • Harris D.
        • et al.
        Approaching the end of life: attitudes, preferences, and behaviors of African-American and white patients and their family caregivers.
        J Clin Oncol. 2003; 21: 549-554
        • Sharma R.K.
        • Dy S.M.
        Documentation of information and care planning for patients with advanced cancer: associations with patient characteristics and utilization of hospital care.
        Am J Hosp Palliat Med. 2011; 28: 543-549
        • Song M.K.
        • Ward S.E.
        • Happ M.B.
        • et al.
        Randomized controlled trial of SPIRIT: an effective approach to preparing African-American dialysis patients and families for end of life.
        Res Nurs Health. 2009; 32: 260-273
        • True G.
        • Phipps E.J.
        • Braitman L.E.
        • Harralson T.
        • Harris D.
        • Tester W.
        Treatment preferences and advance care planning at end of life: the role of ethnicity and spiritual coping in cancer patients.
        Ann Behav Med Vol. 2005; 30: 174-179
        • Walczak A.
        • Mazer B.
        • Butow P.N.
        • et al.
        A question prompt list for patients with advanced cancer in the final year of life: development and cross-cultural evaluation.
        Palliat Med. 2013; 27: 779-788
        • Wallace M.P.
        • Weiner J.S.
        • Pekmezaris R.
        • et al.
        Physician cultural sensitivity in African American advance care planning: a pilot study.
        J Palliat Med. 2007; 10: 721-727
        • Zaide G.B.
        • Pekmezaris R.
        • Nouryan C.N.
        • et al.
        Ethnicity, race, and advance directives in an inpatient palliative care consultation service.
        Palliat Support Care. 2013; 11: 5-11
        • Pluye P.
        • Robert E.
        • Cargo M.
        • et al.
        Proposal: A mixed methods appraisal tool for systematic mixed studies reviews.
        2011 21.03.2017 (Available from)
        • PRISMA
        PRISMA 2009 flow diagram template.
        2009 (Available from)
        • Sinclair C.
        • Auret K.A.
        • Evans S.F.
        • et al.
        Advance care planning uptake among patients with severe lung disease: a randomised patient preference trial of a nurse-led, facilitated advance care planning intervention.
        2016
        • Brunnhuber K.N.
        • Stephen N.
        • Meier D.E.
        • Weissman D.E.
        • Woodcock J.
        Putting evidence into practice: Palliative care.
        2008 (Available from)
        • Gamble V.N.
        Under the shadow of Tuskegee: African Americans and health care.
        Am J Public Health. 1997; 87: 1773-1778
        • Crawley L.
        • Payne R.
        • Bolden J.
        • et al.
        Palliative and end-of-life care in the African American community.
        JAMA. 2000; 284: 2518-2521
        • National CLAS Standards - The Office of Minority Health: U.S
        Department of Health and Human Services Office of Minority Health.
        2016 (Available from)
        • Moscou S.
        The conceptualization and operationalization of race and ethnicity by health services researchers.
        Nurs Inq. 2008; 15: 94-105
        • Chin M.H.
        • Humikowski C.A.
        When is risk stratification by race or ethnicity justified in medical care?.
        Acad Med. 2002; 77: 202-208
        • Realism Madill A.
        Given L.M. The SAGE Encyclopedia of Qualitative Research Methods London. Sage, 2008: 731-735
        • McMahan R.D.
        • Knight S.J.
        • Fried T.R.
        • Sudore R.L.
        Advance care planning beyond advance directives: perspectives from patients and surrogates.
        J Pain Symptom Manage. 2013; 46: 355-365
        • Hutton N.
        Palliative care, time, and core values.
        Patient Education Couns. 2005; 56: 255-256
        • Fund TKs
        Delivering better care at end of life - The next steps.
        2009 17.5.17 (Available from)
        • Tervalon M.
        • Murray-García J.
        Cultural humility versus cultural competence: a critical distinction in defining physician training outcomes in multicultural education.
        J Health Care Poor Underserved. 1998; 9: 117-125
        • Gunaratnam Y.
        Intercultural palliative care: do we need cultural competence?.
        Int J Palliat Nurs. 2007; 13: 470-477
        • Paal P.
        • Buekki J.
        If I had stayed back home, I would not be alive any more - exploring end-of-life preferences in patients with migration background.
        PLOS One. 2017; 12
        • Menon S.
        • Kars M.C.
        • Malhotra C.
        • Campbell A.V.
        • van Delden J.J.M.
        Advance care planning in a multicultural family centric community: a qualitative study of health care professionals', patients', and caregivers' perspectives.
        J Pain Symptom Manage. 2018; 56: 213-221
        • Tong A.
        • Sainsbury P.
        • Craig J.
        Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups.
        Int J Qual Health Care. 2007; 19: 349-357