Factors affecting advance care planning models in Taiwan: Group-based versus individual-based

Study design

This study employed a retrospective secondary data analysis to compare demographic differences between participants in group-based and individual-based ACP consultations. Variables analyzed included gender, age, welfare status (based on welfare identity), religion, self-perceived health status, and consultation format. The data were organized into two primary categories: (1) type of intervention (group-based vs individual-based) and (2) key demographic characteristics from the basic medical records that were related to decision-making, including older adults’ age, self-rated health status, gender, and welfare eligibility. Given the sociocultural context of Asia, religion was also included as a relevant analytical factor. This quantitative approach was chosen because few studies have compared the demographic profiles of declarants across consultation formats, and quantitative methods provide a systematic means to identify such differences and ensure the reliability of secondary data.

Participants and data collection

This study was conducted at a public hospital in Taipei City, Taiwan. In accordance with the PRAA, ACP was implemented through a multidisciplinary team consisting of physicians, nurses, and social workers. Within this team, social workers played a central role in coordinating and facilitating the ACP process. First, individuals interested in ACP were referred for consultation, during which social workers conducted preliminary ACP discussions and explained the differences between group-based and individual-based consultation formats. Based on these discussions, individuals were supported in selecting their preferred consultation approach. Social workers were responsible for organizing and facilitating group-based consultations, addressing administrative inquiries, and serving as the primary point of contact for subsequent ACP-related communication. Their role focused on supporting understanding, clarifying values, and ensuring smooth coordination throughout the consultation process, complementing the medical decision-making roles of physicians and nurses. A total of 4890 individuals participated in ACP consultations between 2021 and 2022. At the conclusion of each ACP session, participants completed a consultation record form assessing their comprehension of the discussion, their decision regarding the signing of an AD, and their motivations for participation. These forms were completed by the facilitating social workers, integrated into participants’ medical records, and securely archived by the Department of Social Work.

The dataset for this study spans the years 2019 to 2022. However, as the ACP program was in its pilot phase during the initial years, data from earlier periods were relatively incomplete and were therefore used only for preliminary analyses. Given that the proportions of group-based and individual-based consultations became more balanced in the last 2 years, this study focused on data from 2021 to 2022 for primary analysis. All records were reviewed by responsible case administrators to ensure data completeness and minimize the risk of recording bias.

Reflexivity

All members of the research team have professional backgrounds in social work, have completed comprehensive training on the PRAA, and possess practical experience in ACP consultations. While these qualifications enhanced the researchers’ ability to understand the consultation process and contextualize the findings, they also carried the potential risk of overinterpreting the results through the lens of personal counseling experience. To mitigate this, the team adhered to an agreed analytic approach that focused solely on comparing signing rates and demographic characteristics, avoiding speculative interpretations regarding the underlying reasons for signing decisions.

Data analysis

This study employed Andersen’s Behavioral Model as the conceptual framework by incorporating operational definitions across three core dimensions: predisposing factors (gender, age, religion, and welfare status), enabling factors (consultation type: group vs individual), and need factors (self-perceived health) (Andersen, 1995). All statistical analyses were conducted using SPSS version 20.01.

Prior to analysis, cases with missing values on any of the variables included in the analysis were removed using a listwise deletion approach. No outlying or deviant cases were identified that contradicted the overall interpretation of the study findings. Descriptive statistics were first applied to examine the distribution of participants’ characteristics and their motivations for engaging in ACP consultations. Chi-square tests (χ²) were used to compare demographic and clinical differences between participants in group-based versus individual-based ACP consultations, and to assess variations in willingness to sign an AD across the independent variables. Subsequently, binary logistic regression was performed to identify the predictive effect of these variables on participants’ willingness to complete an AD. In addition, stratified analyses were conducted to further explore potential interaction effects and subgroup-specific patterns associated with consultation type and decision-making behaviors (Figure 1).

OPEN IN VIEWER

Figure 1.

Research framework for examining factors influencing AD completion among participants in group-based and individual-based ACP consultations in Taiwan (2021–2022).

Ethical considerations

The research protocol was reviewed and approved by the Institutional Review Board of Taipei City Hospital (IRB No. TCHIRB-11305017-E). This study adopted a retrospective medical record analysis, conducted in accordance with medical ethical procedures, and received permission from the data management unit to access the required data. All data were fully de-identified prior to analysis, ensuring that no individual participant could be identified. The research data were accessible only by the research team and were securely stored. As this study was retrospective in nature, it utilized only existing data and did not involve any direct intervention or risk to participants. Data collection was concluded once all necessary information had been obtained, without the need for premature termination.

Descriptive statistics

The study sample comprised 4890 participants who completed ACP consultations. The majority were female (66.7%) and non-older adults (62.9%). Over half of the participants reported having no religious belief (52.8%), while Buddhism was the most common religion (37.1%) among those with religious belief. Participants without registered welfare status accounted for 62.6% of the sample. In terms of consultation format, 53.6% attended individual-based sessions, whereas 46.4% participated in group-based consultations. Most participants (67.6%) perceived their health status as good (Table 1). The study also explored participants’ motivations for engaging in ACP consultations. The most commonly cited reason was the desire to die with dignity, followed by the intention to plan for future EOL medical arrangements (Table 2).

OPEN IN VIEWER

Table 1.

Characteristics of participants in the advance care planning (ACP) program (n = 4890; Taipei City Hospital, Taiwan, 2021–2022).

Domain	Variables	n	%
Predisposition	Gender
	Female	3264	66.7
	Male	1626	33.3
	Age
	Non-older adults (<65)	3076	62.9
	Older adults (⩾65)	1814	37.1
	Religion
	Buddhism	1819	37.1
	No religious belief	2586	52.8
	Christian and others	485	10.1
	Welfare status
	Yes	1828	37.4
	No	3062	62.6
Enabling	Consultation format
	Group-based	2271	46.4
	Individual-based	2619	53.6
Need	Self-perceived health
	Good	3308	67.6
	Not good	1582	30.2

OPEN IN VIEWER

Table 2.

Self-reported motivations for participating in advance care planning (ACP) (Taipei City Hospital, Taiwan, 2021–2022).

Variables	n	%
Reason for ACP
Wishing to die with dignity	2947	27.3
Planning EOL medical arrangements for the future	2706	25.1
Do not want family members to take responsibilities for decisions	1965	18.2
Do not want to become a family burden	1832	17.0
Self suffering from disease	720	6.7
Family member suffering from disease	312	2.9
Media publicity	152	1.4
Invitation from relatives and friends	65	0.6
Being single	38	0.4
Other reasons	39	0.4

Results on predisposing factors for ACP participation

The study’s findings revealed that a significant proportion of participants in ACP group-based consultation were female, non-older adults, individuals without welfare status and those with good self-perceived health. The higher participation rate among women aligned with previous research, which indicated that women were generally more proactive in health-related planning and discussions (Inoue, 2016; Lum et al., 2017). Similarly, the greater willingness of non-older adults to engage in ACP was consistent with prior studies. Younger adults tended to be more receptive to ACP-related knowledge and were more interested in discussing autonomous medical decision-making (Kavalieratos et al., 2015). In contrast, older adults might face barriers such as limited health literacy, restricted access to relevant information, or discomfort discussing EOL issues (Ke et al., 2017; Shen et al., 2020; Simon et al., 2015; Wang et al., 2024).

Previous studies had also identified several reasons why individuals with low income were less likely to participate in ACP, due to challenges such as a lack of family support, low self-esteem, reduced motivation, and limited access to resources (Kimpel et al., 2022). Our findings indicated that, with regard to health status, individuals who perceived themselves as healthier appear more willing to engage in ACP. Since ACP involved sensitive topics, individuals who felt anxious or unprepared might avoid discussing such issues with their families (Brighton and Bristowe, 2016). These findings suggest that ACP interventions may be more effective when initiated earlier, rather than being limited to hospital settings. Specifically, ACP discussions may be better introduced when individuals are relatively healthy, confident, and psychologically prepared to engage in future care planning. This also aligns closely with current national policies in Taiwan, particularly the “Healthy Taiwan” initiative and Long-Term Care 3.0, both of which emphasize healthy aging and the extension of healthy life expectancy as core visions (Ministry of Health and Welfare R.O.C., 2024). From an implementation perspective, ACP services should be conceptualized and delivered as an integrated continuum, encompassing community outreach, pre-ACP preparation, and formal ACP consultations. To enhance accessibility and implementation efficiency, community-based infrastructures—such as community care stations and LTC facilities—may serve as key delivery platforms and appropriate units for program implementation and subsidy allocation. In this way, ACP promotion can be integrated into existing community care systems and function synergistically with national health and LTC policies.

Motivations for ACP participation

In all variables of motivations, to die with dignity was a central motivation not only reflecting societal values but also aligned with previous research findings (Detering et al., 2010) that the promotion of ACP was primarily aimed at making arrangements for a dignified EOL. Another significant motivation for ACP participation was the wish to alleviate the decision-making burden on family members. This concern was rooted in Taiwanese and broader Asian cultural norms, where familial roles and responsibilities were highly valued. In many Asian societies, it was common for EOL decisions to be made by family members rather than the individuals themselves (Cheng et al., 2020). The findings of this study not only demonstrated the necessity of including second-degree relatives in ACP consultations of the PRAA but also reflected the cultural and social importance of familial participation in EOL care in Asia culture.

Effectiveness of group-based consultations interventions

This study indicated that the AD completion in the group-based consultation setting was significantly higher than in the individual format. Our findings were consistent with previous research, which has suggested that the group dynamics, shared disease experiences and a supportive environment are key factors encouraging patients to express their thoughts (Ahluwalia et al., 2021; Bandini et al., 2021). From a cultural perspective, discussing death with family members has traditionally been avoided in many Asian societies; however, ACP serves not only as an opportunity for transformation but also as an effective medium for family discussions (Fan and Kao, 2024; Hsu et al., 2020; Yang et al., 2020). Despite the traditional emphasis on patient self-determination in western medical decision-making, approaches that underscore the family as the fundamental social unit may facilitate decisions that better serve the client’s best interests (Ho, 2008). While there is no right or wrong answer, group-based consultations, which are often family-centered seeking collective answers in nature, may allow the Confucian or Buddhist in nature to depict death as a cyclic process in life (Lin and Lwin, 2026). As a result, the findings demonstrated group-based consultations attracted participants with diverse demographic characteristics and engaged individuals who were previously less interested in ACP-related discussions. In addition, previous studies have highlighted the need for consultation formats tailored to different demographic groups (Tsai and Yang, 2019; Yang et al., 2020). Therefore, the present findings highlight the potential of group-based ACP as a culturally adaptable model that bridges individual autonomy, social support and family-centered decision-making in diverse healthcare settings.

Factors influencing the completion of advance direction after engaging in ACP

Religious belief has been identified as a significant predisposing factor in both previous studies and the present research. Our study specifically found that Buddhists were more likely to complete an advance directive than individuals from other religious backgrounds. In Buddhism, contemplating the nature of death and what happens afterward is a central doctrinal theme, suggesting that life involves reincarnation, or the rebirth of a soul into another body (Bülow et al., 2008; McCormick, 2013). In western countries, previous studies have indicated that the Mainline Protestants show higher rates of advance directive completion than other Christian groups, such as Evangelicals and Catholics (Huang et al., 2016). Similarly, research conducted in Kenya indicated that knowledge regarding AD varied across religious groups (Oshow et al., 2024). Religious beliefs can help individuals express their care preferences and facilitate the completion of advance directives (Bülow et al., 2008; Dronamraju et al., 2022; Lum et al., 2017; Stewart et al., 2013). Religion may also help individuals clarify their values and expectations regarding life and death, thereby facilitating discussions about ACP and increasing the likelihood of completing advance directives (Miyashita et al., 2021). Future research could compare Buddhists and Mainline Protestants, which represent major religious traditions in eastern and western contexts, to further explore the core beliefs and meanings underlying religious adherence and their influence on end-of-life decision-making.

This study also found that individuals without welfare status were more likely to complete an AD than those with welfare status in the individual consultation format. This finding was consistent with previous research, indicating that individuals residing in neighborhoods with lower socioeconomic status were less likely to complete an advance directive (Nouri et al., 2020). However, prior studies had demonstrated that individuals with chronic or terminal illnesses were more likely to articulate their care preferences (Ko et al., 2016). Our study provided a more nuanced perspective. Specifically, we observed that participants with better self-perceived health were more likely to engage in ACP consultations. However, it was those with poorer self-perceived health who were more likely to complete an AD. This finding highlights the complex relationship between perceived health status and the actual decision to formalize EOL care preferences. A review of studies conducted in the United States on ACP and palliative care awareness indicated that while most individuals understood the concept of ACP, only a small proportion actually formalized their preferences through written directives. One key motivator for completing advance directives was negative past medical experiences, which served as catalysts prompting individuals to document their care preferences (Grant et al., 2021). In summary, when introducing ACP, it is essential to take into account patients’ current health and welfare status, as this approach can facilitate smoother implementation and greater public acceptance. These findings further supported the importance of introducing ACP earlier in the life course, rather than only at crisis points (Detering et al., 2010). From a social work perspective, practitioners may incorporate the 4R framework (Resources, Risks, Roles, and Relationships) in pre-ACP and ACP processes, as a structured assessment approach to promote preference-concordant decision-making. By systematically assessing the individual needs, available resources, and relational dynamics, social workers can better support clients and families in navigating complex end-of-life decisions while ensuring that care planning remains person-centered and context-sensitive.