Factors Influencing Knowledge, Attitudes and Behaviour Towards Engaging with Advance Care Planning: A Cross-Sectional Survey of Chinese Diaspora

Abstract

Objective

Advance care planning (ACP) is increasingly recognized as a public health priority globally, with cultural aspects influencing people's knowledge, attitudes, and behaviours toward ACP. Despite being one of the largest diaspora groups, the Chinese community remains under-researched in this area. This study aims to examine the knowledge, attitudes, and health behaviours related to ACP among Chinese diaspora within a region in the United Kingdom.

Method

A cross-sectional online survey based on the Theory of Planned Behaviour was conducted with 284 Chinese adults recruited from four social and voluntary organisations in the United Kingdom.

Results

The study found low awareness (15%) and knowledge (mean score: 2.26 ± 1.78) of ACP, with less than 5% of the participants engaging with ACP. This contrasted with participants’ attitudes (17.24 ± 2.57) and behavioural intention (14.93 ± 3.26) toward ACP, which were generally positive. Behavioural intention was the strongest predictor of engaging with ACP (OR 3.29, 95% CI: 1.35-8.02, p = 0.01). Participants with previous end-of-life care experience had a better understanding of ACP; older age and higher knowledge of ACP were associated with more positive attitudes. Cultural beliefs associated with the level of family involvement, legal aspects, and the availability of information in one's first language was significant influences on engagement with ACP.

Conclusions

Despite low awareness and engagement with ACP among the Chinese diaspora, positive attitudes and behavioural intention suggest potential for increased engagement. This could be achieved through culturally tailored interventions that address cultural influences and complexity surrounding legal requirements. Further research is needed to develop and test such interventions.

Introduction

Over the past two decades, palliative care has been increasingly recognised as a fundamental human right and an integral part of Universal Health Coverage and Primary Health Care.^1,2 Advance care planning (ACP), a process enabling individuals to communicate, identify, and document their future healthcare decisions in advance, ³ has emerged as a key strategy in promoting understanding, utilization, and goal-concordant end-of-life care. While debates on the effectiveness of ACP exist, ⁴ its potential contribution to good quality palliative and end of life care^5–7 positions ACP as a global public health priority.

It can be argued that ACP is largely viewed as a westernised concept,^8,9 which emphasises individual autonomy and decision making, despite significant international cultural differences in the acceptance and interpretation of these values.^10,11 In several Western countries, ACP is legally recognized and integrated into standard healthcare practice, allowing individuals to make legally binding decisions about their future care. ¹² However, across countries globally, there remains significant diversity, with variations in ACP legislation and practice. ¹² In the United Kingdom (UK), ACP is increasingly integrated into routine healthcare, especially for individuals with serious or life-limiting illnesses,^13,14 resulting in legally recognized outcomes such as Advance Decisions to Refuse Treatment (ADRT), Lasting Power of Attorney (LPA), and Do Not Attempt Cardiopulmonary Resuscitation (DNACPR) orders. ¹⁴

Regardless of the diversity, internationally ACP continues to evolve with a shift towards normalizing conversations that support adults of any age or health status in sharing their wishes and preferences. ¹³ Nonetheless, awareness and engagement remain low,^15–17 particularly among diverse ethnic communities.^18–20 However, effective public engagement with ACP is dependent upon how people from the general population, including diverse communities, view ACP. ⁵

Evidence indicates that a lack of cultural sensitivity may significantly affect the understanding and acceptance of ACP.^10,11 For example, the ethos underpinning ACP is centred on the concept of autonomy, reflecting westernised ideals that place emphasis on the individual's right to make decisions about their own care. ²¹ This contrasts with many Asian cultures where the cultural values of filial piety, which advocates a set of moral values and practice of respect and caring for one's parents, may lead to such discussions being avoided and decision-making being led by the family, with the individual playing a more peripheral role.^11,22,23 Other cultural factors, such as language, perceptions of authority figures leading to a preference for paternalistic decision-making with healthcare providers, and trust in providers’ expertise, have also been reported as key determinants of engagement in many Asian cultures.^11,24

While research to date would suggest that many western public health approaches significantly influence ACP awareness, understanding, and engagement,^25,26 there is criticism that existing campaigns are generic, are not tailored to multicultural societies in which they operate,^27,28 and have little consideration of the contextual and cultural variables influencing engagement. To be effective and inclusive, public engagement requires an understanding about the transcultural perspective, of how people from diverse communities’ view planning ahead for changes in health.^5,16 However, the impact of cultural identity on engagement with ACP remains understudied,^10,16 particularly among ethnic minority groups such as the Chinese diaspora in the UK.

The concept of diaspora has no set definition, but it refers to individuals and their foreign-born children who have left their country of origin but maintain links with their homeland. Hence, they are reported to have a shared identify of their host country and their background. ²⁹ The ‘Chinese diaspora’ comprises individuals born in mainland China, Hong Kong, Taiwan, and Macau, who are living elsewhere or identify based on language and/or ancestry. They represent one of largest diaspora groups globally. ³⁰ In the UK, Chinese diaspora constitutes 0.7% of the population in 2021, ³¹ yet research suggests they face an array of barriers when accessing health and palliative care. ³² This has been attributed to cultural differences, such as death-avoidance culture, family traditions and values like filial piety, dependency on authority approval, and poor language skills hindering awareness and engagement with ACP.^10,18,24

While most research to date on diaspora groups, particularly the Chinese diaspora's perspectives on ACP, originates from the United States,^18,20,33 there is a lack of comprehensive exploration of socio-cultural influences in other contexts and different health care systems such as the UK. Therefore, this study aims to examine Chinese diaspora's knowledge, attitudes and health behaviours in relation to ACP in one region of the UK and to identify recommendations for culturally tailored interventions to engage Chinese communities in ACP.

Methods

Research Design

A cross-sectional descriptive correlational design was used. While various instruments exist which explore ACP views, ¹⁸ based on a review of the evidence, no pre-existing validated instruments were identified that align with the specific focus of our study. Therefore, we developed new items by adapting elements from existing instruments,^34–36 drawing on previous research synthesis ¹⁸ and incorporating data from qualitative interviews. The questionnaire, underpinned by the Theory of Planned Behaviour (TPB), ³⁷ underwent a comprehensive development, translation, and validation process. ³⁸ A pilot study was conducted with Chinese members (N = 284) recruited from four Chinese organizations in Northern Ireland to test the questionnaire. The questionnaire (Supplemental File 1) consisted of 3 sections with 48 items. The psychometric testing indicated good reliability and validity, and a detailed description of the key measures is provided in Supplementary File 2. The Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) guidelines were applied. ³⁹

Setting and Participants

Participants were members of four Chinese organizations providing social support, cultural education, and language services in Northern Ireland. A purposive sample of 625 members aged over 18, proficient in Mandarin, English, and/or Cantonese, and identified as members of the Chinese community, were deemed eligible. Participants who had experienced bereavement in the last six months or had cognitive impairments were excluded to avoid potential emotional distress and balance ethical concerns. The minimum sample size of 239 responses was determined based on power analysis (using a 5% margin of error and 95% confidence level). ⁴⁰

Data Collection

An anonymous self-report online questionnaire was conducted using JISC, a UK-based digital platform for education and research, between October and November 2023. A gatekeeper within each Chinese Organization distributed the survey link via email and social media platforms to members available in English, Mandarin, and Cantonese. Each participant received an information letter which described the details of the study. To enhance response rates, reminders were issued 7 days later, entry into a free prize draw was offered, and paper copies were made available upon request.

Statistical Analysis

Data were exported from JISC into SPSS statistics 28. Descriptive and inferential statistics were computed. Logistic and multiple linear regression analyses were employed to identify the predictors of knowledge, attitude, behavioural intention and engagement on ACP. Prior to the regression analysis, Chi-Squared Test, Independent T-test, and One-way ANOVA were conducted to examine the relationships between demographic variables and the main variables of interest: knowledge, attitude, and engagement. Variables having both statistical and conceptual meanings were entered into the regression models. Significance was set at 0.05. ⁴¹

Ethical Considerations

The study was approved by a University Research Ethics Committee, (July 2023, FCNUR-23-060). Participants were advised that their informed consent was given implicitly by completing and returning the questionnaire. All personal data were anonymized and processed in accordance with the General Data Protection Regulation. ⁴²

Results

Of the 625 people sampled, 302 participants completed the survey. From these a total of 284 valid responses were included and analysed in this study, resulting in a response rate of 48.3%. Eighteen responses were excluded from the analysis because they were deemed invalid due to incomplete data or responses that did not meet quality control criteria, as identified during data processing in SPSS. Most participants were first-generation (64.4%) and from mainland China (81.3%) (see Table 1). Over half spoke both Mandarin and English. Participants ranged in age from 18 to 95 years. The majority were female (67.3%) and had lived in the UK for more than 6 years (44%). A large portion of participants identified as not having any religious affiliation (70.4%) and most were highly educated (59.2%), with nearly half in employment (47.9%). Most participants self-identified as being in good health (90.3%) and had no experience caring for individuals in the end-of-life stage (75%).

Table 1.

Demographic Characteristics (N = 284).

Category	n (%)
Gender
Male	65 (22.9%)
Female	191 (67.3%)
Others	28 (9.8%)
Age (years)
18–24	70 (24.6%)
25–34	82 (29.0%)
35–44	70 (24.6%)
45–54	35 (12.3%)
55–64	9 (3.2%)
65 or over	18 (6.4%)
Immigrant generation
First-generation immigrant	183 (64.4%)
Second-generation immigrant	9 (3.2%)
Third or over generation immigrant	4 (1.4%)
Don’t know	88 (31%)
Years in NI
5 and below	157 (55.3%)
6–10	39 (13.7%)
11–15	38 (13.4%)
16–20	20 (7%)
Above 20	28 (9.9%)
Don’t know	2 (0.7%)
Self-identify of home country
Mainland China	231 (81.3%)
Northern Ireland	23 (8.1%)
Hong Kong	23 (8.1%)
Others (Taiwan, Macao)	7 (2.5%)
Languages spoken
Mandarin	252 (47.9%)
English	189 (35.9%)
Cantonese	79 (15.0%)
Others (Hakka)	6 (1.1%)
Highest qualification
School exams – below university entrance level (eg, overseas equivalent to GCSE/O Level/Standard Grade)°	39 (13.7%)
School exams – university entrance level (eg, overseas equivalent to A Level or Higher Grade)	35 (12.3%)
Undergraduate degree from university or comparable institution (eg, bachelor's degree)	65 (22.9%)
Postgraduate degree (eg, Masters, Doctorate)°	103 (36.3%)
Commercial qualifications (eg, typing, shorthand, bookkeeping, commerce)	9 (3.2%)
Others	33 (11.6%)
Employment status
Full time	106 (37.3%)
Students	65 (22.9%)
Looking after the home	38 (13.4%)
Part time°	30 (10.6%)
Retired from paid work	17 (6%)
Looking for work	8 (2.8%)
Others (temporarily sick or injured)	20 (7%)
Religion
No religion	200 (70.4%)
Other Christian denomination	19 (6.7%)
Buddhist	15 (5.3%)
Others (Taoism, Muslim)	13 (4.8%)
Catholic	9 (3.2%)
Protestant	6 (2.1%)
Prefer not to say	22 (7.7%)
Care experience for someone near end of life
Yes	50 (17.6%)
No	213 (75%)
Don’t know	21 (7.4%)
Health status
Excellent	40 (14.1%)
Good	134 (47.2%)
Somewhat good	82 (28.95%)
Neither	15 (5.3%)
Somewhat poor	10 (3.5%)
Poor	3 (1.1%)
Very poor	0
ACP awareness
Yes	45 (15.8%)
No	222 (78.2%)
Not sure	17 (6%)
ACP engagement
Yes	14 (4.9%)
No	270 (95.1%)

Awareness and Knowledge Towards ACP

Findings indicated low levels of awareness and knowledge relating to “ACP”, with only 15% (n = 45) having heard of the term. The mean knowledge score was 2.26 points (standard deviation (SD) = 1.78, range 0–6), with only four (1.4%) having correctly answered questions across all 6 items and a fifth (n = 60) answered none of the items correctly (see Table 2). Analysis revealed that participants with previous end-of-life care experience had a better understanding of ACP (mean = 2.78, SD = 1.85) compared to those without such experience (mean = 2.22, SD = 1.77), with a statistically significant difference (p = 0.02) (see Table 3). Results suggest, previous end-of-life care experience was a significant predictor of a higher level of knowledge regarding ACP (β = 0.14, p = 0.02) (see Table 4).

Table 2.

Knowledge of Advance Care Planning (n = 284).

Advance Care Planning Knowledge Items	Correct n (%)	Incorrect n (%)
ACP enables individuals to make their future healthcare decisions. (T)	179 (63%)	105 (37%)
ACP is only for people when they are nearing end of life. (F)	125 (44%)	159 (56%)
Once an ACP is made, it cannot be changed. (F)	124 (43.7%)	160 (56.3%)
ACP is legally binding. (F)	18 (6.3%)	266 (93.7%)
My family must be involved in my ACP discussions. (F)	58 (20.4%)	226 (79.6%)
ACP is only for old people. (F)	137 (48.2%)	147 (51.8%)
Average total knowledge score (range 0–6 points)	2.26 (1.78)

Table 3.

Difference in Knowledge, Attitude Score Means and ACP Engagement Percentage Between Demographics.

Category	Knowledge mean (SD)	Mean difference (95% CI)	p-value	Attitude mean (SD)	Mean difference (95% CI)	p-value	ACP engagement n (%)	p-value
Gender
Female	2.39 (1.79)	Reference category	0.47 a	17.41 (2.50)	Reference category	0.18 a	5 (35.7)	0.34 b
Male	2.20 (1.86)	−0.19 (−0.70 to 0.32)		16.91 (3.00)	−0.51 (−1.25 to 0.24)		4 (28.6)
Age (years)
18–34	2.16 (1.84)	—	0.10 c	16.88 (2.74)	Reference category	0.004 c	3 (21.4)	0.02 b
35–54	2.51 (1.70)			17.42 (2.34)	0.544 (−0.09 to 1.18)		7 (50.0)
55 or over	1.78 (1.63)			18.59 (1.89)	1.718 (0.68 to 2.76)		4 (28.5)
Immigrant generation
First-generation	2.28 (1.84)	—	0.08 c	17.15 (2.58)	—	0.774 c	8 (57.1)	0.16 b
Second-generation	2.67 (1.73)			17.22 (2.17)			2 (14.3)
Third or over generation	1.25 (1.26)			17.00 (1.83)			0 (0.0)
Years in NI
5 and below	2.20 (1.81)	—	0.22 c	16.86 (2.78)	—	0.06 c	4 (28.6)	0.07 b
6–10	2.13 (1.49)			17.51 (2.15)			5 (35.7)
11–15	2.50 (1.77)			17.55 (2.39)			3 (21.4)
16–20	3.00 (1.95)			18.10 (2.63)			0 (0.0)
Above 20	1.89 (1.75)			18.00 (1.78)			2 (14.3)
Self-identify of home country
Mainland China	2.37 (1.81)	—	0.08 c	17.22 (2.61)	—	0.81 c	6 (42.9)	<0.001 b
Northern Ireland	2.04 (1.80)			17.30 (2.80)			7 (50.0)
Hong Kong	1.39 (1.31)			17.61 (1.92)			0 (0.0)
Others (Taiwan, Macao)	2.14 (1.46)			16.57 (2.82)			1 (7.1)
Highest qualification
Low educated	2.29 (1.76)	0.05 (−0.43 to 0.52)	0.85 a	17.49 (2.50)	0.40 (−0.28 to 1.09)	0.25 a	4 (28.5)	0.90 b
High educated	2.24 (1.79)	Reference category		17.09 (2.66)	Reference category		6 (42.8)
Employment status
Employed	2.46 (1.72)	—	0.19 c	17.29 (2.54)	−1.42 (0.14 to 2.70)	0.02 c	7 (50.0)	0.41 b
Retired	1.76 (1.71)			18.71 (1.21)	Reference category		2 (14.3)
Students	1.97 (1.86)			16.57 (3.17)	−2.14 (0.79 to 3.49)		2 (14.3)
Looking after the home	2.45 (1.78)			17.66 (1.77)	−1.05 (−0.40 to 2.50)		3 (21.4)
Others	1.81 (1.76)			17.69 (1.66)	−1.02 (−0.71 to 2.75)		0 (0.0)
Religion
No religion	2.11 (1.81)	—	0.25 c	17.22 (2.53)	—	0.55 c	4 (28.6)	<0.001 b
Catholic	2.33 (1.87)			17.0 (2.60)			0 (0.0)
Protestant	3.17 (1.72)			17.50 (1.76)			0 (0.0)
Other Christian denomination	3.00 (1.56)			18.37 (2.43)			5 (35.7)
Buddhist	1.93 (1.53)			17.07 (2.40)			1 (7.1)
Others (Taoism, Muslim)	2.23 (1.36)			17.38 (2.57)			3 (21.4)
Care experience for someone near end of life
Yes	2.78 (1.85)	0.56 (−0.01 to 1.11)	0.02 a	16.98 (3.04)	−0.40 (−1.32 to 0.52)	0.20 a	2 (14.3)	1.00 b
No	2.22 (1.77)	Reference category		17.38 (2.50)	Reference category		11 (78.6)
Health status
Poor health status	2.69 (2.06)	0.47 (−0.53 to 1.48)	0.35 a	16.62 (3.28)	−0.71 (−2.15 to 0.74)	0.34 a	0 (0.0)	0.82 b
Good health status	2.21 (1.78)	Reference category		17.32 (2.54)	Reference category		14 (100.0)

ACP: advance care planning; SD: standard deviation; CI: confidence interval; ANOVA: analysis of variance.

^ap-Values based on unpaired T-tests.

^bp-Values based on Chi-square tests.

^cp-Values based on ANOVA tests.

“-”no difference between groups, therefore no mean difference report in this table

Table 4.

Multivariate Analysis for ACP Knowledge and Attitude Between Demographics and co-Dependents (n = 284).

Category	Knowledge Mean (SD)	Beta coefficients (95% CI)	p-value	Attitude Mean (SD)	Beta coefficients (95% CI)	p-value
Age (years)
18–34	—			16.88(2.74)	Reference category	0.001d
35–54				17.42(2.34)	—
55 or over				18.59(1.89)	0.20 (0.73 to 2.68)
Care experience for someone near end of life
Yes	2.78 (1.85)	0.14 (0.09 to 1.18)	0.02	—
No	2.22(1.77)	Reference category
Knowledge	—			0.26e	0.27 (0.23 to 0.56)	0.001d

SD: standard deviation; CI: confidence interval.

“^d” Estimated results from the final step linear regression model with attitude score as the dependent variable and age, employment status and knowledge score as independent variables. R²= 0.102, R² change = 0.073, p < 0.001.

“^e” Estimated result from attitude and knowledge Pearson correlation coefficient.

Attitude Towards ACP

A written explanation of the concept was provided preceding the knowledge section. Despite low awareness and variable knowledge relating to ACP, findings indicated high levels of positive attitude, with the mean score of 17.24 (standard deviation (SD) = 2.57, range 3–21). Higher attitude scores were associated with being in an older age group (p = 0.004), retired (p = 0.02) and higher level of knowledge of ACP (r = 0.26, p < 0.001) (see Table 3). Analysed data revealed that being in the 55-plus age group (β = 0.20, p < 0.001) and having a high level of knowledge (β = 0.27, p < 0.001) were significant predictors of more positive attitudes toward ACP (see Table 4).

Factors Influencing ACP Behaviour

The mean perceived control score was 17.15 (standard deviation (SD) = 2.94, range 3–21), and self-efficacy was 16.00 (standard deviation (SD) = 3.43, range 3–21), both reflecting a high level of self-capacity to participate in ACP. The mean subjective norm score was 16.18 (standard deviation (SD) = 3.26, range 3–21), reflecting a high level of family and social support to participate in ACP.

There was a moderate level of intention to participate in ACP, with the mean score of 14.93 (standard deviation (SD) = 3.26, range 3–21). Only fourteen participants (4.9%) reported having engaged, however, significant differences in engagement were observed across age, religion, and home country identity (p < 0.05) (see Table 3). After adjustment, intention to engage was the most influencing predictor for engagement (OR 3.29, 95% CI: 1.35–8.02, p = 0.01) (see Table 5).

Table 5.

Results of Logistic Regression Analysis for Health Behaviour Towards ACP (n = 284).

							95% C.I.
Variables	β	S.E.	Wald	df	p	Exp(B)	Lower	Upper
Have religion	3.683	1.492	6.098	1	0.014	39.777	2.138	739.964
55 plus age group	1.257	1.762	0.509	1	0.475	3.516	0.111	111.087
Attitude	−0.533	0.321	2.762	1	0.097	0.587	0.313	1.100
Subjective norm	0.339	0.322	1.112	1	0.292	1.404	0.747	2.638
Perceived control	−0.565	0.239	5.567	1	0.018	0.569	0.356	0.909
Self-efficacy	−0.162	0.261	0.386	1	0.534	0.850	0.510	1.417
Intention	1.191	0.454	6.870	1	0.009	3.291	1.350	8.020
Knowledge	−0.717	0.395	3.304	1	0.069	0.488	0.225	1.058

Note: CI = confidence intervals. Reference groups in analysis: age = 18–34 years, religion = no religion. Significant p values in bold. Cox & Snell R Square = 0.22.

Table 6 shows results from stepwise linear regressions identifying areas for culturally tailored interventions to promote ACP engagement. Behavioural beliefs explained 23.4% of the variance in intention (F(28.456), p < 0.001), with significant predictors including “reduce uncertainty” (β = 0.29), “reduce burdens” (β = 0.24), and “bring me bad luck” (β = −0.11). Normative beliefs accounted for 34.1% of the variance (F(72.579), p < 0.001), with partners and close friends as key influencers (β = 0.33, 0.30). Control beliefs contributed to 22.8% of the variance (F(27.521), p < 0.001), with “ACP legally binding” (β = 0.25), “ACP in first language” (β = 0.23), and “lack of opportunity to discuss ACP” (β = −0.13) as significant factors.

Table 6.

Outcome of the Final Step in the Multiple Linear Regression Analysis for Beliefs and Intention (n = 284).

Models	Unstandardized Coefficients		Standardized Coefficients
	B	Std. Error	Beta	t	p value	VIF	R 2	r
Behavioural beliefs
(Constant)	6.405	1.168		5.484	0.000
A3	0.919	0.211	0.286	4.362	0.000	1.574		.425**
A4	0.709	0.199	0.235	3.566	0.000	1.585		.416**
A5	−0.193	0.091	−0.113	−2.133	0.034	1.016	23.4%	−.127*
A6								0.023
A2								.423**
Normative beliefs
(Constant)	5.247	0.823		6.379	0.000
SN5	0.902	0.192	0.331	4.691	0.000	2.120	34.1%	.547**
SN4	0.845	0.201	0.297	4.213	0.000	2.120		.538**
SN2								.421**
SN3								.526**
Control beliefs
(Constant)	8.908	1.189		7.492	0.000
PC4	0.628	0.162	0.248	3.883	0.000	1.485	22.8%	.413**
PC2	0.622	0.168	0.225	3.708	0.000	1.331		.383**
PC5	−0.312	0.138	−0.131	−2.260	0.025	1.219		−.297**
PC3								−.138*
PC6								−0.078

Note: r = Pearson correlation coefficient with intention; R² = squared multiple correlation; Beta = Standardised regression coefficients; **.Correlation is significant at the 0.01 level (2-tailed); *. Correlation is significant at the 0.05 level (2-tailed); Dependent Variable: Behavioural Intention.

A2: ACP would help me identify my future healthcare goals and preferences.

A3: Engaging in ACP would reduce uncertainty about my future healthcare decisions.

A4: Engaging in ACP would reduce burdens on my family and friends.

A5: Engaging in ACP would bring me bad luck.

A6: Engaging in ACP would cause my family conflict.

SN2: My older family members would approve of me engaging in ACP. (eg, parents, grandparents) (If not applicable, please skip this question.)

SN3: My younger family members would approve of me engaging in ACP. (eg, adult children)

SN4: My spouse/partner would approve of me engaging in ACP. (If not applicable, please skip this question.)

SN5: My close friends would approve of me engaging in ACP.

PC2: ACP provided in my first language would encourage me to engage with it.

PC3: My fear of death would prevent me from engaging in ACP.

PC4: Making ACP legally binding would encourage me to engage with it.

PC5: The lack of opportunity to talk about ACP would prevent me from engaging with it.

PC6: My expectation that my family make my future health decisions would prevent me from engaging in ACP.

Discussion

Our findings show deficits in awareness and understanding among Chinese diaspora, consistent with previous international and national research across the public^16,43 and other diverse ethnic and diaspora communities.^19,20,32,44 In our study, only 15% of participants were aware of ACP, a figure comparable to a study on Iranian American adults, where 12% had heard of ACP despite being highly educated and generally in good health. ⁴⁴ However the results of this study were notably lower compared to a UK national survey of general population, ²⁷ 28.5% of participants, most of whom were white, had heard of ACP. Several studies, however, suggest that improved knowledge and understanding on its own is not sufficient to change behaviour.^43,45 Instead, many interconnected cultural factors play a role. For example, it is increasingly recognized that ACP ideology is a Western concept, and its interpretation is based on the principles of autonomy, nonmaleficence, and beneficence, ⁴⁶ which may conflict with the values of certain cultures.

In contrast to previous research that emphasized an age and education-related increase in awareness of ACP importance within Chinese communities,^47–49 This lower level of awareness among the Chinese community may be attributed to the western ethos of ACP adopting a one size fits all approach, where individual autonomy in decision making is valued. Yet the concept of self-determination, sits at odds with Chinese diaspora cultural values, of familial or collective decision making. ⁵⁰ Moreover, such westernised concepts have been found not to be applicable to other cultures including Italian people who prefer family-based decisions ⁵¹ and Greek people ⁵² who view autonomy as dangerous and disruptive. Jimenez et al ⁵ attributes this possible ethology for such differences existing.

However, despite cultural differences, this study observed that individuals with prior end-of-life caregiving experience exhibited a better understanding, aligning with previous research on end-of-life family caregivers,^53–56 which reportedly provides them with an understanding and experience of ACP. This suggests that direct experience with end-of-life care may provide more meaningful exposure not only to ACP concepts but also to the broader decision-making processes involved in care. However, it is beyond the reach of this study to understand the context within which such discussions occurred, and if the concept of deference to clinical authority, common in Asian cultures, influenced responses.

Similar to research conducted with the general public^27,57,58 and diaspora groups,^59–61 this study found that most respondents held positive attitudes towards ACP. While previous research with Chinese diaspora reported that Chinese traditional culture relating to family relationships influenced attitudes toward family involvement in end-of-life care, ⁶² our study reveals that older age groups and those with a better understanding of ACP exhibit more positive attitudes toward it. Yet engagement remains low, aligning with previous research ¹⁸ and mirroring a global picture of low enegagement.^{15–17,27,44} In our study, only 4.9% of participants had engaged in ACP, similar to findings from a study on Iranian-American adults, the majority of whom were highly educated and generally in good health, where only 5.6% had advance directives and 12.4% had communicated their end-of-life care wishes. ⁴⁴ Although ACP was first introduced and promoted in the United States, ⁶³ where legislation mandates opportunities for the public to learn about and complete advance directives, engagement remains low even in countries with such policies.

Previous research with Chinese diaspora highlighted the influence of the diagnosis of a health-related problem and/or a diagnosis of a life-limiting condition,^64–66 acculturation related variables such as the length of time living in the host countries^36,67 and the proficiency in English were also noted as influencing factors.^47,64,68,69 However this study found an individual's intention to engage is the most significant factors in predicting ACP engagement. Therefore, a Chinese diaspora's willingness to participate was the most crucial factor in predicting engagement. This aligns with the Theory of Planned Behaviour that attitude, subjective norm, and perceived behaviour control are key determinants that combine to influence the intention, the latter acted as the strongest predictor of the actual behaviour. ⁷⁰ Studies on general population,^16,17 found attitude and knowledge to be key factors influencing engagement. However, findings from this study suggested that behavioural intention is the key predictor which may be aligned to Chinese values and beliefs.

Several cultural beliefs were found to be significant predictors of the intention to engage in ACP and may indirectly influence actual engagement. First, the provision of ACP information in culturally tailored language materials was identified as a significant predictor of intention, reflecting findings from previous research.^47,64,68,69 However, while preference for a language other than English may impact ACP knowledge and engagement, our study suggests that language alone may not be sufficient to increase engagement.

Second, the need for ACP to have an authoritarian orientation, advocated through legislation, policy, or by people in positions of authority, holds significant influence among participants. ¹¹ This is consistent with Chiang et al's findings, ⁷¹ which observed that the Chinese diaspora often aligns their behaviour with national policies, likely due to the prevalent trust and reliance on government in Chinese culture. However, while previous research^10,18,24 reports Chinese diaspora perceive healthcare professionals as authority figures in ACP discussions, this was not observed in the current study and the influence of this difference may warrant further exploration.

Finally, in this study, an important factor for engagement was strong beliefs about the importance of family who engaged in collective or deferred decision making,^11,72,73 a characteristic commonly observed in Asian studies both within^72,74 and beyond Chinese communities. ¹¹ Without such involvement, barriers to engagement existed. These findings underscores the prominence of familism in major decision-making processes, which often emphasizes communal decisions over individual autonomy and self-determination. ²¹ In our study, partners and close friends were identified as the most significant influencers supporting engagement in ACP, which aligns with findings from a national study on the Northern Ireland general population. ²⁷ Although older and younger family members did not directly predict intention in this study, strong correlations suggest they may still play a supportive role in engagement.

Regardless of location or ethnic grouping, research on the public's attitude, knowledge and engagement with ACP consistently remains low. This highlights the pressing need for novel approaches to be adopted taking cognisance of cultural diversity, societal norms and contexts.^16,52 Such an approach may require a re-orientation around how communities and health care professionals approach ACP. Further research into cultural adaptations of ACP is required.

Strengths and Weaknesses/Limitations of the Study

This study pioneers the exploration of ACP within the UK's Chinese community, offering several noteworthy benefits. Firstly, the survey used in this study was meticulously developed in alignment with rigorous methodology and thoroughly validated. This enhances the strength, reliability, and validity of the research findings. Secondly, the survey was available in multiple languages (Mandarin, English, and Cantonese) in both online and paper formats. The use of multi-dissemination strategies resulted in a higher response rate compared to other studies with a similar population. ⁷⁵ Despite these benefits, the study has certain limitations. Primarily, the findings are based on purposive sampling from four Chinese organizations, which may limit the generalizability and introduce some selection bias. Secondly, this study did not consider how acculturation influences in the process of ACP. Further investigation is warranted in the degree acculturation plays in the awareness, knowledge and engagement of Chinese diaspora. Thirdly, the correlational design restricts the establishment of causal relationships. However, the multiple regression analysis in this study identified significant determinants, paving the way for further interventional tests. Finally, while the data collection utilized a validated questionnaire based on the theory of planned behaviour, limitations in questionnaire development hindered an examination of the expectancy-value model's influence on ACP in this context. Further studies are needed for a more comprehensive understanding.

Conclusion

Recognizing cultural differences and effective cooperation in this regard may enhance efforts to meet the diverse needs of individuals from various ethnic backgrounds, ultimately ensuring equitable access to the benefits of ACP. The study strongly advocates for the use of tailored messages aligned with cultural influences to ensure the success of such initiatives. The study revealed that low levels of understanding and awareness pose a barrier to engagement. These, combined with a lack of tailored messages, may potentially lead individuals to reject care that does not align with their preferences and cultural values. This underscores the necessity for public health campaigns targeted at educating and enhancing the awareness and understanding within the Chinese community. Conversely, the positive attitudes, family support, and willingness to engage with ACP suggest that promoting such initiatives among the Chinese community is not only viable but also essential. Therefore, further research is needed to adapt ACP to different cultural contexts, allowing for more personalized and effective interventions.