Introduction
End-of-life (EOL) communication and decision-making have become central to intensive care unit (ICU) care due to technological advances and increasing survival rates (Curtis et al. Reference Curtis, Higginson and White2022). Aligning care goals with patient and family values is vital, yet transitioning to palliative care in times of critical illness remains challenging due to emotional, ethical, cultural, and psychosocial complexities (Coombs et al. Reference Coombs, Addington-Hall and Long-Sutehall2012). These challenges are shaped by broader socioeconomic, legal, and cultural factors (Kwak and Haley Reference Kwak and Haley2005; Wang et al. Reference Wang, Hsu and Huang2018), which must be addressed to ensure comprehensive support.
This review focuses on Korean American (KA) populations, broadly defined to include United States (US)-born individuals of Korean heritage, Korean individuals with naturalized citizenship, and recent immigrants from South Korea. KA populations represent the fifth-largest Asian group, now numbering nearly two million – a 17% increase over the past decade – in the US (U.S. Census Bureau 2022), which is home to the largest Korean diaspora outside of Korea (Esterline and Batalova Reference Esterline and Batalova2022). Despite this growth, their experiences with EOL care remain poorly understood. Available research revealed that Asian Americans – one of the fastest-growing and most ethnically diverse populations – receive more intensive treatments at EOL, die in hospitals frequently, and are less likely to use hospice or palliative care than non-Hispanic White Americans (Jia et al. Reference Jia, Leiter and Sanders2022; Lackan et al. Reference Lackan, Eschbach and Stimpson2009; Ngo-Metzger et al. Reference Ngo-Metzger, Phillips and McCarthy2008). However, these findings often obscure subgroup differences by collapsing over 40 distinct ethnicities under a single “Asian and Pacific Islander” umbrella – an increasingly criticized practice masking cultural and healthcare differences (Holland and Palaniappan Reference Holland and Palaniappan2012; Jin Reference Jin2021).
KA populations face unique challenges navigating the US healthcare system, where traditional family-centered and relational values from Korea may conflict with or adapt to more individualistic norms in the US (Jang et al. Reference Jang, Ko and Meischke2020). Individuals within this population often seek support from their ethnic communities and culturally concordant providers, as their challenges are compounded by the complexities of immigration status and navigating an unfamiliar system (Choi Reference Choi2013). These challenges highlight the critical need for culturally tailored care for KA populations. However, existing research has focused on community settings (Park and Hendrix Reference Park and Hendrix2018), leaving a gap in our understanding of how KAs engage in EOL care in hospital contexts.
To address the gap in the critical care literature on KA populations, we conducted a cross-context scoping review that maps ICU-based EOL literature in Korea alongside EOL literature involving KAs in the US to identify transferable cultural implications. We prioritized ICU-based studies from Korea that provide relevant contexts for US clinicians and policymakers by clearly illustrating how cultural values influence EOL decision-making during high-acuity care and nascent ICU palliative care models. With ongoing migration from South Korea to the US (Esterline and Batalova Reference Esterline and Batalova2022), evidence from Korea may inform culturally responsive care and policymaking for the KA population with critical illness. Further, exploring Korean literature in the context of recent legislation of the Hospice, Palliative Care, and Life-Sustaining Treatment Act (LST Act) in Korea (Choi et al. Reference Choi, Park and Kang2019) and the development of ICU-based palliative care is important, as approximately 75% of Koreans die at medical institutions with persistent unmet EOL needs (Cheon et al. Reference Cheon, Kim and Cho2023). This cross-cultural exploration aims to add nuance to the current understanding of Korean cultural values, which are often associated with Confucian principles wherein filial piety, emphasizing respect for parents and elders, may discourage open discussions about death or treatment withdrawal (Choi et al. Reference Choi, Rodgers and Tocher2023; Lee et al. Reference Lee, Jo and Kim2021).
As healthcare practices evolve globally, cross-cultural influences on palliative care will likely increase (Rosa et al. Reference Rosa, Male and Uwimana2018). Prior reviews have examined EOL care among KA individuals (Park and Hendrix Reference Park and Hendrix2018; Suk et al. Reference Suk, Kozachik and Cotter2021), but none have synthesized Korean and American EOL literature together to guide culturally responsive palliative care for KA populations. Furthermore, no review to date has mapped the literature on ICU-based EOL care in Korea. Therefore, this review aimed to (1) map the current state of science on EOL care for critically ill Korean individuals and all KA individuals in the US, and (2) offer culturally informed directions for future research and practice in both countries.
Method
Our scoping review follows the updated JBI methodological guidance and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) extension for Scoping Review (Peters et al. Reference Peters, Marnie and Tricco2021; Tricco et al. Reference Tricco, Lillie and Zarin2018). We chose this method for the exploratory design of this review, aiming to guide future research by mapping the literature.
Review question
Our review questions are structured using the “Patient-Concept-Context” framework (Peters et al. Reference Peters, Marnie and Tricco2021): Patient – Koreans, KAs, Korean immigrants; Concept – EOL and palliative care; Context – Critical care in Korea and all settings in the US.
(1) What is known about EOL care for critically ill Korean individuals in Korea and among KA communities – including US-born, naturalized US citizens, and immigrants – in all US settings?
(2) What culturally relevant insights can inform future research and practice to improve EOL care for these populations?
Literature search
Figure 1 shows the screening process guided by the PRISMA. A two-stage literature search was conducted to identify studies on EOL care for critically ill individuals with Korean heritage and KA populations in all US settings.
Figure 1. PRISMA flowchart diagram of the study.
In the first stage, we focused on critical care. Seven databases (PubMed, Cumulative Index to Nursing and Allied Health Literature [CINAHL], Excerpta Medica Database [EMBASE], Web of Science, Korea Citation Index, Global Index Medicus, and KoreaMed) were searched using the terms “Korean AND EOL/palliative care AND critical care.” English and Korean studies with no time restriction were included. We excluded studies focusing on pediatrics, studies not addressing patients and families (e.g., providers’ moral distress), studies outside critical care settings, and studies lacking discrete data on Koreans). Unpublished dissertations, conference proceedings, editorials, letters, and review papers were also excluded. Search strategies were validated by a university informaticist (see Appendix A). Three authors (SHK, CK, and ECH) screened the studies, resolving conflicts by consensus. Of 1018 records, we removed 328 duplicates. From 40 full-text articles and three additional studies identified from references, we included 23 studies, all conducted in Korea. None addressed KAs in critical care.
Due to this gap in KA populations, we expanded the focus to include EOL and palliative care for KAs in any care environment. Using the search terms “Korean AND EOL/palliative care” with a US geographical filter, we searched four databases (PubMed, CINAHL, EMBASE, Web of Science). From 4150 records identified, 1026 duplicates were removed. After abstract and full-text reviews, 26 US-based studies were included. Taken together, our final sample included 49 studies.
Data extraction & analysis
We conducted a thematic synthesis to identify transferable cross-pollination of cultural insights across Koreans in ICUs and KA populations in US settings (Thomas and Harden Reference Thomas and Harden2008). We juxtaposed and synthesized two literature sets to identify cultural values and patterns in EOL care that are relevant to research and care models for the KA population while mapping both literatures’ scopes and gaps. This approach aimed to conceptually translate rather than compare, as care settings and systems were not equal or directly comparable. Findings were interpreted while accounting for differences in care acuity, decisional capacity, and structural factors. Data were initially charted into three tables: Table 1 summarizes general characteristics, while Tables 2 and 3 present detailed findings. We then inductively developed thematic categories through iterative reading and open coding of the extracted data and the original authors’ interpretations (Thomas and Harden Reference Thomas and Harden2008). One author (SHK) conducted data extraction, coding, and thematic development, and the work was reviewed and validated by three co-authors (CK, ECH, RW).
Table 1. General characteristics of the included studies
a One US-based study did not report sample size.
b Only retrospective designs were used.
c This was counted, allowing for overlaps between categories.
Table 2. Detailed characteristics of Korea-based studies
Notes. AD: advanced directives, DNR: do not resuscitate, ICU: intensive care unit, LST: life-sustaining treatment, LST Act: Hospice, Palliative Care, and Life-Sustaining Treatment Act, N: number, POLST: physician order of life-sustaining treatment, WWLT: withhold or withdraw life-sustaining treatment, QOC: quality of communication, QODD: quality of death and dying.
Table 3. Detailed characteristics of US-based studies
Note. AA: Asian American, ACP: advanced care planning, AD: advanced directives, EOL: end-of-life, KA: Korean American, LST: life-sustaining treatment, N: number, NHW: non-Hispanic Whites, POA: power of attorney.
Results
General characteristics of the studies
Table 1 summarizes the characteristics of 49 studies. Since 2003, 23 Korea-based ICU studies were identified, with 14 conducted after the 2018 LST Act, which allows patients to withhold or withdraw life-sustaining treatment (WWLT). None collected data directly from patients. Instead, insights came from surrogate or staff reports and retrospective records. While Korean studies involved more sample diversity across patients, families, and providers than US studies, patients were generally critically ill older adults near EOL.
From the US, 26 studies (1996-present) focused mainly on first-generation, Korean-speaking older adults (Berkman and Ko Reference Berkman and Ko2010; Dobbs et al. Reference Dobbs, Park and Jang2015; Frank et al. Reference Frank, Blackhall and Michel1998; Hong et al. Reference Hong, Casado and Lee2019, Reference Hong, Noh and Kim2022; Jang et al. Reference Jang, Chiriboga and Allen2010; Ko and Berkman Reference Ko and Berkman2010, Reference Ko and Berkman2012; Ko and Lee Reference Ko and Lee2010; Kwak and Salmon Reference Kwak and Salmon2007; Park Reference Park2021) recruited in communities. Terminology for the population varied; most studies used the term KAs; few used Korean immigrants (Ko et al. Reference Ko, Roh and Higgins2013) or KA immigrants (Berkman and Ko Reference Berkman and Ko2009; Lee et al. Reference Lee, Noh and Choi2022a). Most studies were conducted in community settings in metropolitan areas and regions with high KA densities. None took place in hospitals. Exploration of providers’ perspectives was limited, with only one study including social workers (Ha et al. Reference Ha, Lee and Yoo2023).
Thematic synthesis of the studies
We identified four themes, each comprising commonalities and setting-specific patterns. Figure 2 maps scopes, gaps, and contextual features across both literatures. No US-based KA studies included ICU settings. Both bodies of literature shared elements, including family-centered decision-making, reluctance to engage in advance care planning (ACP), and a preference for physician-initiated EOL conversations. Korean ICU-based studies described high-acuity, near-death situations where the quality of dying was rated as inadequate. Conversely, KA studies elicited the perspectives of community-dwelling older adults through hypothetical scenarios, asking what they would prefer if they became ill. Differences in clinical context, legal structures, and social determinants alongside shared cultural traits shaped attitudes, knowledge, and perspectives.
Figure 2. Mapping across Korean ICU and community-based Korean American literatures.
EOL critical care
Quality of EOL care in Korean ICUs
Three cross-sectional studies evaluated the QODD in Korean ICUs using surrogate reports from family members (Choi et al. Reference Choi, Park and Kang2019) or staff (Cho et al. Reference Cho, Park and Kim2019; Lee et al. Reference Lee, Noh and Choi2022a). Across studies, overall QODD scores were low compared to reports from the US or Netherlands (Gerritsen et al. Reference Gerritsen, Koopmans and Hofhuis2017) and hospice patients, with poor ratings in pain control, emotional support, and dignity (Cho et al. Reference Cho, Park and Kim2019; Choi et al. Reference Choi, Park and Kang2019; Lee et al. Reference Lee, Noh and Choi2022a). However, scores were higher when patients transitioned to general wards before death and after the enactment of the 2018 LST Act, compared to the pre-enactment period (Cho et al. Reference Cho, Park and Kim2019; Choi et al. Reference Choi, Park and Kang2019; Lee et al. Reference Lee, Noh and Choi2022a).
Evidence gap for critically ill KA populations
No US study examined KA patients in ICUs or other hospital settings. One community survey examining ACP intentions in a serious illness context (dementia) found that greater disease knowledge, more positive attitudes of ACP, and stronger perceived social norms (perceptions about what others would do and expect them to do) were associated with higher ACP intentions (Hong et al. Reference Hong, Casado and Lee2019). However, this community-based design also shows gaps in how KA patients with illnesses navigate EOL or palliative care.
Decision-making and communication
EOL decision-making and communication in Korean ICUs
Ten retrospective studies from Korean ICUs indirectly illustrated EOL decision-making patterns (Baek et al. Reference Baek, Koh and Hong2016; Jang et al. Reference Jang, Ryu and Heo2023; Kim et al. Reference Kim, Choi and Park2016, Reference Kim, Hong and Cho2024; Lee et al. Reference Lee, Hong and Park2020, Reference Lee, Jang and Hong2008, Reference Lee, Jo and Kim2021, Reference Lee, Ju and Kang2024, Reference Lee, Yun and Lee2019; Shin et al. Reference Shin, Mok and Lee2014). Most do-not-resuscitate (DNR) orders, advance directives (AD), and Physician Orders for Life-Sustaining Treatment (POLST) were initiated by physicians (Lee et al. Reference Lee, Jang and Hong2008; Shin et al. Reference Shin, Mok and Lee2014) and signed by family members on behalf of patients (Baek et al. Reference Baek, Koh and Hong2016; Jang et al. Reference Jang, Ryu and Heo2023; Kim et al. Reference Kim, Choi and Park2016; Lee et al. Reference Lee, Hong and Park2020). Completion rates for AD and POLST before or during ICU admission were consistently low, ranging from 0% to 15.7% (Jang et al. Reference Jang, Ryu and Heo2023; Kim et al. Reference Kim, Choi and Park2016; Lee et al. Reference Lee, Hong and Park2020, Reference Lee, Ju and Kang2024; Shin et al. Reference Shin, Mok and Lee2014).
Another consistent finding was that DNR or WWLT orders were commonly made very late, often to prevent anticipated, non-beneficial CPR when death became imminent (Jang et al. Reference Jang, Ryu and Heo2023; Kim et al. Reference Kim, Choi and Park2016; Lee et al. Reference Lee, Hong and Park2020, Reference Lee, Ju and Kang2024; Shin et al. Reference Shin, Mok and Lee2014). In three retrospective studies, conducted between 2008 and 2024, DNR or WWLT decisions were made within three days of patient death (Kim et al. Reference Kim, Hong and Cho2024; Lee et al. Reference Lee, Jang and Hong2008; Shin et al. Reference Shin, Mok and Lee2014). A study examining DNR timing found that later DNR orders were associated with longer ICU stays and higher rates of WWLT (Baek et al. Reference Baek, Koh and Hong2016). Another study similarly challenged the assumption that care intensity decreases with DNR or POLST, reporting longer ICU stays and higher mortality among patients with such orders (Lee et al. Reference Lee, Ju and Kang2024).
The findings highlighted how POLST, designed for early, proactive EOL planning, was often used similarly to reactive approaches of WWLT. Qualitative studies interviewing nurses added contexts, indicating that discussions about life-sustaining treatment decisions rarely occur before patients become incapacitated and that both families and providers typically delay EOL conversations until patients’ conditions become critical (Chang et al. Reference Chang, Kim and Cho2024; Koh et al. Reference Koh, Ko and Park2017).
Few prospective studies addressed the perceived quality of EOL communication in ICU-based Korean studies. One study found families’ perception of the quality of EOL communication to be low regarding patients’ condition, spiritual or religious beliefs, and treatment preferences (Jo et al. Reference Jo, Song and Knafl2019). Another small-scale case study reported on the effectiveness of a multidisciplinary ICU family meeting protocol in facilitating information sharing between clinicians and families (Min et al. Reference Min, Lee and Park2016).
Culturally sensitive EOL decision-making and communication preferences of KA populations
Many KA older adults responding to hypothetical scenarios reported strong preferences for indirect EOL communication to maintain hope when discussing severe illnesses (Berkman and Ko Reference Berkman and Ko2010; Blackhall et al. Reference Blackhall, Frank and Murphy2001; Kwak and Salmon Reference Kwak and Salmon2007), with acknowledgments of the need for transparency to prepare for dying (Phipps et al. Reference Phipps, True and Murray2003) and concerns that disclosure could cause distress or hasten death. However, these US-based studies did not include clinical encounters, leaving a significant gap in understanding real-world care experiences and providers’ perspectives for the KA population within US healthcare settings.
Cross-context cultural insights: decision-making and communication
Family-centered decision-making with physician-led discussions was a common thread in both literatures. A Korea-based ICU study using focused ethnography explored the real-time decision-making experiences of families and showed that decisions often fell to the adult-child most financially or emotionally invested (Choi et al. Reference Choi, Rodgers and Tocher2023). Decisions were typically initiated by physicians, who explained the futility of treatment, and depended on families’ perceptions of the balance between the patient’s comfortable death and filial and financial obligations (Choi et al. Reference Choi, Rodgers and Tocher2023). Similarly, in US-based studies, older KA adults prioritized collective family decisions over individual autonomy (Berkman and Ko Reference Berkman and Ko2010; Blackhall et al. Reference Blackhall, Frank and Murphy1999; Frank et al. Reference Frank, Blackhall and Michel1998; Hong et al. Reference Hong, Noh and Kim2022; Ko and Berkman Reference Ko and Berkman2010, Reference Ko and Berkman2012; Kwak and Salmon Reference Kwak and Salmon2007), and expected their adult children, the emotionally closest or eldest, to make care decisions and provide the utmost care to extend life (Frank et al. Reference Frank, Blackhall and Michel1998; Hong et al. Reference Hong, Noh and Kim2022; Ko and Berkman Reference Ko and Berkman2010, Reference Ko and Berkman2012; Kwak and Salmon Reference Kwak and Salmon2007). KA participants also favored physician-initiated EOL conversations to ease emotional discomfort (Hong et al. Reference Hong, Noh and Kim2022; Ko and Berkman Reference Ko and Berkman2010; Kwak and Salmon Reference Kwak and Salmon2007) and preferred Korean providers (Berkman and Ko Reference Berkman and Ko2010; Kwak and Salmon Reference Kwak and Salmon2007). AD completions were low among older KA adults, ranging from 0 to 18%, and often viewed as guidance for adult children rather than binding legal instructions upholding autonomy (Dobbs et al. Reference Dobbs, Park and Jang2015; Hong et al. Reference Hong, Casado and Lee2019; Ko and Berkman Reference Ko and Berkman2010, Reference Ko and Berkman2012; Ko and Lee Reference Ko and Lee2010; Kwak and Salmon Reference Kwak and Salmon2007; Murphy et al. Reference Murphy, Palmer and Azen1996). As KA studies primarily included first-generation older adults, interpretation is bounded by sample composition, leaving gaps on younger and US-born KAs and intergenerational dynamics.
Perspectives and knowledge about EOL care
Perspectives and knowledge among Korean ICU family members
Five surveys of Korean ICU families assessed attitudes and perceptions regarding EOL care (Byun et al. Reference Byun, Choi and Choi2003; Kim and Kang Reference Kim and Kang2015; Kim and Tak Reference Kim and Tak2021; Lee and Kang Reference Lee and Kang2010; Oak and Kim Reference Oak and Kim2020). Many family members supported WWLT for patient dignity (Byun et al. Reference Byun, Choi and Choi2003; Lee and Kang Reference Lee and Kang2010), while some preferred to be informed about the need for DNR in patients’ imminent death (Kim and Kang Reference Kim and Kang2015). Knowledge of hospice care correlated positively with attitudes (Kim and Tak Reference Kim and Tak2021; Oak and Kim Reference Oak and Kim2020). However, these surveys only focused on family members and did not examine their education or access to information about AD or ACP.
Perspectives and knowledge among KA populations
Nine US-based studies reported generally low knowledge and awareness regarding AD, power of attorney, and ACP among KA participants (Dobbs et al. Reference Dobbs, Park and Jang2015; Ha et al. Reference Ha, Lee and Yoo2023; Hong et al. Reference Hong, Casado and Lee2019, Reference Hong, Noh and Kim2022; Jang et al. Reference Jang, Park and Chiriboga2017; Ko and Berkman Reference Ko and Berkman2012; Ko and Lee Reference Ko and Lee2010; Ko et al. Reference Ko, Roh and Higgins2013; Kwak and Salmon Reference Kwak and Salmon2007). Although KA older adults favored low-intensity care and home death, many still preferred families to make EOL decisions for them (Blackhall et al. Reference Blackhall, Frank and Murphy1999, Reference Blackhall, Murphy and Frank1995; Frank et al. Reference Frank, Blackhall and Michel1998; Ha et al. Reference Ha, Lee and Yoo2023; Ko and Berkman Reference Ko and Berkman2010, Reference Ko and Berkman2012; Kwak and Salmon Reference Kwak and Salmon2007). While general, non-personal attitudes towards leveraging and learning about palliative care and ACP were positive, actual personal engagement remained limited (Berkman and Ko Reference Berkman and Ko2009; Jang et al. Reference Jang, Chiriboga and Allen2010; Ko and Berkman Reference Ko and Berkman2012; Lee et al. Reference Lee, Noh and Choi2022a; Pan et al. Reference Pan, Abraham and Giron2015). Key factors that contributed to increased ACP intentions included greater knowledge and awareness, favorable subjective norms (how others behave and expect them to act), perceived better ability to discuss these topics, and recognition of benefits such as avoiding unwanted treatment (Hong et al. Reference Hong, Casado and Lee2019; Jang et al. Reference Jang, Chiriboga and Allen2010; Ko and Lee Reference Ko and Lee2010; Lee et al. Reference Lee, Noh and Choi2022a). Health insurance coverage and higher educational attainment were also associated with more positive attitudes and awareness of AD (Dobbs et al. Reference Dobbs, Park and Jang2015; Kim and Foreman Reference Kim and Foreman2011). Stronger spiritual and religious beliefs were linked to greater engagement in EOL communication (Ko et al. Reference Ko, Roh and Higgins2013). Reported barriers included concerns about costs, social isolation, and chronic conditions (Lee et al. Reference Lee, Noh and Choi2022a).
Cross-context cultural insights: perspectives and knowledge
Across both literatures, hesitancy around personal and family EOL planning was evident. In Korean ICUs, families advocated patient dignity yet preferred to defer DNR discussions until death was near (Kim and Kang Reference Kim and Kang2015), whereas older KA adults preferred low-intensity care, peaceful death, but at times delegated decisions to adult children (Blackhall et al. Reference Blackhall, Frank and Murphy1999, Reference Blackhall, Murphy and Frank1995; Frank et al. Reference Frank, Blackhall and Michel1998; Ha et al. Reference Ha, Lee and Yoo2023; Ko and Berkman Reference Ko and Berkman2010, Reference Ko and Berkman2012; Kwak and Salmon Reference Kwak and Salmon2007). Although greater awareness correlated with positive attitudes toward EOL topics, studies rarely specified when, where, or how education, information access, and support were provided.
Contextual and social determinants of EOL care
Familial and disease-related factors for Korean ICU patients
Two Korean ICU-based studies examined specific characteristics – comparing homeless and non-homeless individuals and cancer versus non-cancer patients (Kim et al. Reference Kim, Hong and Cho2024; Lee et al. Reference Lee, Yun and Lee2019). Although there was no difference in mortality, homeless ICU patients were reported to undergo more CPR procedures before death compared to non-homeless ICU patients in Korea, likely due to the absence of family members to make WWLT decisions on their behalf (Lee et al. Reference Lee, Yun and Lee2019). In the study conducted post enactment of the LST Act, the type of diagnosed disease was found to influence EOL decision-making outcomes, with cancer patients more likely to make decisions for themselves compared to patients with neurological or infectious diseases (Kim et al. Reference Kim, Hong and Cho2024). The results of these two studies highlighted specific aspects of social determinants and patient characteristics that pose challenges and influence EOL care decisions in the ICU, which warrant further investigation.
Sociocultural influences on EOL care preferences among KA populations
Sociocultural factors, including place of upbringing and acculturation, were key factors of EOL preferences among KA participants. Individuals with greater acculturation – through longer US residency, higher English proficiency, and younger age – tended to prefer transparency and autonomy, while less acculturated individuals often preferred traditional, family-centered, and indirect communication (Berkman and Ko Reference Berkman and Ko2009; Blackhall et al. Reference Blackhall, Murphy and Frank1995; Dobbs et al. Reference Dobbs, Park and Jang2015; Ha et al. Reference Ha, Lee and Yoo2023; Kim and Foreman Reference Kim and Foreman2011; Ko and Lee Reference Ko and Lee2010; Rhee and Jang Reference Rhee and Jang2020). However, studies acknowledged limitations in capturing a wider range of reasons behind these attitudes (Dobbs et al. Reference Dobbs, Park and Jang2015; Ha et al. Reference Ha, Lee and Yoo2023; Jang et al. Reference Jang, Chiriboga and Allen2010) or shifting preferences alongside disease progression (Berkman and Ko Reference Berkman and Ko2010).
While acculturation was frequently addressed (Dobbs et al. Reference Dobbs, Park and Jang2015; Jang et al. Reference Jang, Chiriboga and Allen2010, Reference Jang, Park and Chiriboga2017; Kim and Foreman Reference Kim and Foreman2011; Ko et al. Reference Ko, Cho and Bonilla2012), many studies lacked detailed immigration-related information, such as generational status, citizenship, or English proficiency. Most studies primarily recruited first-generation older adults and referred to them as KAs with little differentiation between naturalized citizens and recent immigrants (Berkman and Ko Reference Berkman and Ko2010; Dobbs et al. Reference Dobbs, Park and Jang2015; Frank et al. Reference Frank, Blackhall and Michel1998; Hong et al. Reference Hong, Casado and Lee2019, Reference Hong, Noh and Kim2022; Jang et al. Reference Jang, Chiriboga and Allen2010; Ko and Berkman Reference Ko and Berkman2010, Reference Ko and Berkman2012; Ko and Lee Reference Ko and Lee2010; Kwak and Salmon Reference Kwak and Salmon2007; Park Reference Park2021), limiting understanding of the nuanced effects of immigration backgrounds on EOL preferences. Six studies specified participants’ lengths of US residence (12 to 29 years) (Berkman and Ko Reference Berkman and Ko2009; Jang et al. Reference Jang, Chiriboga and Allen2010; Kim and Foreman Reference Kim and Foreman2011; Ko and Berkman Reference Ko and Berkman2012; Ko et al. Reference Ko, Cho and Bonilla2012; Kwak and Salmon Reference Kwak and Salmon2007), and one study reported that 72% of participants had limited English proficiency (Lee et al. Reference Lee, Noh and Choi2022a), highlighting the need for more granular research on the sociocultural contexts of EOL.
Cross-context cultural insights: contextual and social determinants
Across both literature sets, EOL preferences and decisions are shaped by family structures, illness characteristics, and sociocultural factors, including generational shifts and language. In Korea, the relatively recent development of palliative care, reflected in the 2018 LST Act, coincided with low formal engagement in EOL planning (Choi et al. Reference Choi, Park and Kang2019; Kim et al. Reference Kim, Hong and Cho2024). Among KAs, ongoing migration and limited English proficiency potentially add informational and linguistic hurdles (Lee et al. Reference Lee, Noh and Choi2022a; U.S. Census Bureau 2022). These observations are hypothesis-generating rather than causal and are presented as contextual features that may help interpret cross-context cultural patterns.
Discussion
Research gaps and cross-cultural insights
This scoping review mapped the state of science to draw out transferable cultural insights in EOL and palliative care across Koreans in Korean ICUs and KAs in US community settings. Notably, no US-based studies included critically ill KAs in healthcare settings; likewise, prior reviews focused on community samples (Park and Hendrix Reference Park and Hendrix2018; Suk et al. Reference Suk, Kozachik and Cotter2021). Across both bodies of literature, primary emphases were on EOL decision-making, communication, ACP, attitudes, and knowledge. Symptom management, caregiving support, and spiritual/religious care were rarely addressed. Our aim was not to clinically and statistically compare across settings but to identify cross-context implications to inform culturally responsive EOL care. Common themes included family-centered decision-making – often led by adult-children – physician-initiated conversations, reluctance to disclose and discuss serious illnesses, and low engagement and knowledge of ACP/AD. Interpretations of each literature set are shaped and bounded by its socio-legal context and the acuity and proximity to death of the samples.
This cross-context cultural analysis illustrates how heritage-country evidence can inform research and practice for immigrant and minoritized populations with sparse domestic data. Korean ICU studies highlighting family-consented DNRs and late decisions made near death (Jang et al. Reference Jang, Ryu and Heo2023; Kim et al. Reference Kim, Choi and Park2016; Lee et al. Reference Lee, Hong and Park2020, Reference Lee, Ju and Kang2024; Shin et al. Reference Shin, Mok and Lee2014) offer hypotheses for research in KA populations. Future research on KA should investigate whether similar delayed decisions reflect cultural hesitation to discuss death or systemic barriers, and conduct formative research based on KA preferences to develop culturally tailored interventions. Such interventions could involve families earlier and shift indirect communication to upfront conversations by clinicians as readiness develops. Evaluating the effects of linguistic and cultural concordance between patients and providers can identify systemic factors that impact EOL planning and help create language-concordant care models. Comparative research across immigration generations, age groups, and Asian American subgroups may uncover important within-group differences for culturally sensitive EOL care.
Rethinking metrics: access and strength-based perspectives
Existing US-based KA studies have primarily focused on knowledge, attitudes, or preferences toward EOL care, with little attention to actual access to palliative or EOL care services (Dobbs et al. Reference Dobbs, Park and Jang2015; Hong et al. Reference Hong, Casado and Lee2019; Jang et al. Reference Jang, Chiriboga and Allen2010, Reference Jang, Park and Chiriboga2017; Ko and Berkman Reference Ko and Berkman2010, Reference Ko and Berkman2012; Ko and Lee Reference Ko and Lee2009, Reference Ko and Lee2010; Ko et al. Reference Ko, Roh and Higgins2013; Kwak and Salmon Reference Kwak and Salmon2007). Assessing access to palliative care is critical to avoid misattributing disparities to cultural preferences alone, which can risk framing cultural differences as deficiencies (Abu-Saad et al. Reference Abu-Saad, Daoud and Kaplan2021; Fogarty et al. Reference Fogarty, Lovell and Langenberg2018). Instead, strength-based approaches offer a more nuanced understanding by critically examining historic and systemic factors and emphasize community assets (Abu-Saad et al. Reference Abu-Saad, Daoud and Kaplan2021; Fogarty et al. Reference Fogarty, Lovell and Langenberg2018; McEwen et al. Reference McEwen, Pasvogel and Murdaugh2019). These distinctions are summarized in Table 4.
Table 4. Deficit-based and strength-based approaches
For example, AD completion was commonly used as an indicator of EOL preparedness, which may not fully account for cultural preferences, such as Koreans’ emphasis on relativistic and family-centered values (Dobbs et al. Reference Dobbs, Park and Jang2015; Hong et al. Reference Hong, Casado and Lee2019; Jang et al. Reference Jang, Park and Chiriboga2017; Ko and Berkman Reference Ko and Berkman2012; Ko and Lee Reference Ko and Lee2010; Lee et al. Reference Lee, Hong and Park2020; Shin et al. Reference Shin, Mok and Lee2014). Thus, focusing solely on this Western-centric metric may overlook culturally specific approaches, such as ongoing family communication as a valid alternative for EOL planning (McDermott and Selman Reference McDermott and Selman2018), underscoring the need for future research to develop more inclusive and culturally sensitive palliative care metrics in today’s increasingly multicultural world (Ntizimira et al. Reference Ntizimira, Deo and Dunne2022).
Methodological innovations for EOL research in ICU
A key limitation of ICU-based studies in Korea was the reliance on surrogate reports (Cho et al. Reference Cho, Park and Kim2019; Choi et al. Reference Choi, Park and Kang2019; Lee et al. Reference Lee, Noh and Choi2022a) and medical records (Baek et al. Reference Baek, Koh and Hong2016; Jang et al. Reference Jang, Ryu and Heo2023; Kim et al. Reference Kim, Choi and Park2016, Reference Kim, Hong and Cho2024; Lee et al. Reference Lee, Hong and Park2020, Reference Lee, Jang and Hong2008, Reference Lee, Jo and Kim2021, Reference Lee, Ju and Kang2024, Reference Lee, Yun and Lee2019; Shin et al. Reference Shin, Mok and Lee2014). Innovative methods are needed to capture experiences among patients facing frailty, cognitive decline, and communication difficulties. Recruiting ICU survivors, despite possible memory limitations, may help address the lack of direct patient experience in surrogate reports (van de Leur et al. Reference van de Leur, van der Schans and Loef2004). Other innovative methods include observation of care revealing subtle aspects of patient experiences with minimal risk (Eriksson et al. Reference Eriksson, Lindahl and Bergbom2010), ICU diaries offering longitudinal documentation (Sayde et al. Reference Sayde, Stefanescu and Conrad2020), and grey literature such as patients’ and families’ social media posts providing authentic narratives (Taylor and Pagliari Reference Taylor and Pagliari2018). These approaches can expand our understanding of actual palliative care delivery in critical care, promoting more empathetic strategies while minimizing patient burden.
Practice and policy implications for EOL care in Korea
Cultural, social, and legal components shape healthcare practices. The 2018 LST Act in Korea authorized patients to refuse futile life-sustaining treatments through AD or, by the will of family members (Lee et al. Reference Lee, Jo and Kim2021), which may have reinforced families as primary decision-makers (Choi et al. Reference Choi, Rodgers and Tocher2023) and posed challenges for patients without legal immediate family (Lee et al. Reference Lee, Yun and Lee2019). Criteria for POLST orders – confirmation by two physicians of patient status, presence of hospital ethics committees, and availability of next-of-kin (Heo et al. Reference Heo, Yoo and Keam2022) – may further limit decision-making options for those without such supports. Expanding the legal definition of “next-of-kin” and improving support are crucial for vulnerable populations. In addition, EOL care disparities exist based on diagnosis (Kim et al. Reference Kim, Hong and Cho2024), reflecting how cancer-centered palliative care models have long dominated resources (Kim and Hong Reference Kim and Hong2016), leaving other patients with limited access to timely palliative referrals or discussions. To ensure fair access, palliative care in Korea should broaden its focus beyond cancer and develop advocacy for socioeconomically disadvantaged, unbefriended, and homeless individuals (Jang et al. Reference Jang, Lee and Kim2022; Park et al. Reference Park, Kim and Yoo2024).
Practice and policy implications for EOL care in the US
While disparities in Korea stem from a relatively homogeneous population, KA individuals in the US exhibit substantial internal diversity by immigration status, generation, and age cohorts. How US-borns, naturalized citizens, and recently arrived immigrants may differ in their cultural identities is insufficiently captured in research and health systems, often assuming homogeneous preferences. Integrating disaggregated data is crucial for capturing both within- and between-group differences more effectively. To promote implementation, these practices could be tied to hospital quality ratings and billing codes and inform more tailored EOL and palliative care delivery.
Limitations and implications for future research
This review is limited by the scope of Korean ICU-based studies, which may not reflect the broader Korean population or individuals with pre-ICU opportunities to access palliative care. As many patients are more cognitively intact before ICU admission, future studies should target pre-ICU patient preferences and interventions to examine the trajectories and influence on EOL outcomes in ICU settings. Additionally, we included studies with subsamples of the target population that reported relevant discrete findings, as this review aimed to broadly map the field for cultural insights. Therefore, the findings are exploratory in identifying cultural patterns and should not be considered generalizable or definitive conclusions.
Conclusion
This scoping review examined EOL care among critically ill Koreans in Korea and KA communities in the US. Findings highlight the need to explore EOL perspectives and disparities shaped by sociocultural, legal, and systemic factors and call attention to addressing immigration, generational, and acculturation-related differences within the KA populations with critical illnesses. While gaps exist regarding critically ill KA individuals, Korean ICU literature provides insights to inform culturally informed care. Culturally tailored strategies, building on the strengths of family-centered communities, are needed in EOL and palliative care research for KA populations. This cross-contextual perspective and the review design can inform research and practice for other minoritized or immigrant groups.
Funding
This research did not receive any specific grant from funding agencies, commercial, or not-for-profit sectors.
Competing interests
The authors have no disclosures.
Appendix A
Bibliographic database search strategies (search date: 07/02/2024)
1. PubMed
1) Korea
2) US
2. CINAHL (Cumulated Index to Nursing and Allied Health Literature)
1) Korea
2) US
3. EMBASE
1) Korea
2) US
4. Web of Science
1) Korea
2) US
5. KCI – Korea Citation Index
*Translation.
#2 = #1 in Korean.
#5 = #4 in Korean.
6. Global Index Medicus – West Pacific
7. KoreanMed
Databases 5–7 were searched only for Korean-based studies.
For database 5 and 7, search was conducted without including “Koreans” or “Asians” in the search queries, given that these databases were based in Korea.
Appendix B
Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) Checklist
JBI = Joanna Briggs Institute; PRISMA-ScR = Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews.
* Where sources of evidence (see second footnote) are compiled from, such as bibliographic databases, social media platforms, and Web sites.
† A more inclusive/heterogeneous term used to account for the different types of evidence or data sources (e.g., quantitative and/or qualitative research, expert opinion, and policy documents) that may be eligible in a scoping review as opposed to only studies. This is not to be confused with information sources (see first footnote).
‡ The frameworks by Arksey and O’Malley (6) and Levac and colleagues (7) and the JBI guidance (4, 5) refer to the process of data extraction in a scoping review as data charting.
§ The process of systematically examining research evidence to assess its validity, results, and relevance before using it to inform a decision. This term is used for items 12 and 19 instead of “risk of bias” (which is more applicable to systematic reviews of interventions) to include and acknowledge the various sources of evidence that may be used in a scoping review (e.g., quantitative and/or qualitative research, expert opinion, and policy document).
From: Tricco AC, Lillie E, Zarin W, O’Brien KK, Colquhoun H, Levac D, et al. PRISMA Extension for Scoping Reviews (PRISMAScR): Checklist and Explanation. Ann Intern Med. 2018;169:467–473. doi: 10.7326/M18-0850.
Open access
