Content area
Aim
To explore the effects of a theory-based Multi-media Experiential Advance Care Planning (MEACP) training programme for nurses in acute care settings, which aims to improve their decision-support skills, attitudes towards advance care planning (ACP), knowledge of ACP and level of confidence in conducting ACP.
BackgroundNurses in acute care settings are often less engaged in ACP discussions due to insufficient training and time constraints.
DesignA single -blinded, cluster randomised controlled trial with a repeated-measures, two-arm, parallel-group design was employed.
MethodsA total of 230 registered nurses were recruited from 23 acute care wards across eight public-funded hospitals. Participants were randomly assigned to either a control group (n = 112) or an intervention group (n = 118) at the ward level. Informed by the theory of planned behaviour and the experiential learning model, the MEACP programme included a 10-module mobile application (app) and a 2-hour nurse-led, face-to-face, skill-building workshop.
ResultsThe participants in the MEACP group showed significantly greater improvements in their decision-support skills (95 % CI:2.47–3.34, p < 0.001), knowledge of ACP (95 % CI: 0.7–1.61, p < 0.001) and confidence in conducting ACP (95 % CI: 0.37–0.79, p < 0.001) than the control group post-intervention.
ConclusionThe intervention has the potential to empower nurses with better decision-support skills and knowledge, which lay the groundwork for future efforts to promote ACP engagement in acute care settings.
According to the World Health Organization, palliative care is an approach that enhance the quality of life and relieves the suffering of patients and their families who have life-limiting illnesses ( World Health Organization, 2020), including Advance care planning (ACP). ACP involves continuous and proactive communication among patients, their families and healthcare professionals about patient’s preference for end-of-life care before they become too frail to make medical decisions ( Sudore et al., 2017). During ACP discussions, a mentally competent adult can choose to sign an advance directive (AD), which allows the patient to specify their preference for medical care in the future ( Hong et al., 1996). A recent systematic review of randomised controlled trials (RCTs) examining the efficacy of ACP provided consistent evidence that ACP significantly enhanced proximal patients outcomes including patient-physician communication, preference for comfort care, decisional conflict, alignment of preferences between patients and caregivers and ACP documentation ( Malhotra et al., 2022). However, inadequate knowledge of the care options, unclear values concerning their care goals and feelings of pressure in the decision-making process are common factors hindering patients’ ability to make decisions on end-of-life care ( Stacey et al., 2008).
In Hong Kong China, over 90 % of deaths occur in acute care hospitals, which is the highest rate in the world ( Chung, 2017; Woo et al., 2009). Anecdotal data shows that the ACP discussion rate is less than 1 % even in palliative care wards. A systematic review of the level of ACP knowledge among healthcare professionals in Asia found that only 57 % of physicians and 49 % of nurses in Hong Kong were familiar with ADs ( Martina et al., 2021). Addressing this deficit will require a cultural shift towards nurses’ engagement in end-of-life care in various settings, especially in the acute care setting.
Various ACP training interventions have been developed for healthcare professionals. However, there has been a shortage of high-quality randomised controlled trials (RCTs) assessing the effectiveness of ACP training programmes for nurses in acute care settings. For example, there has been no objective measurement of the improvement to their decision-support skills in previous studies. Moreover, most tested interventions are very lengthy, intensive and difficult to implement in the acute care setting, which is a high-pressure and fast-paced environment. Furthermore, not all of the reviewed training programmes addressed the unique challenges that nurses may encounter in ACP discussions, such as dealing with ethical dilemmas when patients’ wishes conflict with their family preferences. Our team conducted qualitative interviews with nurses working in acute care settings, asking about their training needs for the provision of ACP ( Ng and Chan, 2018). The findings suggested that the use of multi-media and experiential learning was more effective than merely didactic teaching. To build their confidence in engaging in ACP conversations, nurses are encouraged to engage in role-play, learn about case studies and practise using decision aids.
Nurses are considered better positioned to facilitate end-of-life conversations than other allied health professionals because they usually spend more time at the bedside and have closer interactions with patients while providing daily care ( Gazarian et al., 2019; Hasdianda et al., 2021). Therefore, an innovative ACP training programme specifically designed for nurses working in busy acute care settings is essential. The current study addressed this training gap by examining the effects of a Multi-media Experiential Advance Care Planning (MEACP) training programme.
1.1 Conceptual frameworkGuided by the theory of planned behaviour and the experiential learning model, an MEACP training programme was developed to encourage nurses to participate in ACP discussions. The theory and model were used to develop the focus, content and activities of the intervention. The theory of planned behaviour suggests that behaviour is mutually affected by three interrelated factors: behavioural attitudes, perceived norms and perceived behavioural control ( Ajzen, 1985). ‘Behavioural attitude’ refers to an individual’s disposition towards and perceptions of the value of a certain form of behaviour. Perceived norms include features of an individual’s knowledge and social environment. Generally, doctors maintain the ultimate responsibility for documenting and deciding patients’ end-of-life treatment plans and this is a crucial factor influencing nurses’ perceptions of their professional role in ACP discussions. Lastly, perceived behavioural control involves an individual’s confidence in having the necessary resources and capacity to carry out a certain task. The MEACP programme was designed to enhance nurses’ attitudes towards conducting ACP, introduce support from administrators and boost their behavioural control via simulated practical experience.
The MEACP training programme is also based on Kolb’s experiential learning model, which emphasises the importance of experience in the learning process ( Kolb, 2014). Learning occurs when knowledge is constructed through a transformative reflection of experience ( Schnepp and Rogers, 2017). It is essentially a cycle of concrete experience, reflective observation, abstract conceptualisation and application. Hence, the proposed MEACP programme consists of a mixed model of web tutorials and skill-building workshops to maximise learners’ direct experience with ACP role-play involving standardised patients. The MEACP tools provide nurses with a basis for learning by teaching them how to implement ACP and practise the use of decision aids (concrete experience) with standardised patients. They discuss examples of controversial cases and proper and improper care skills through video demonstrations (reflective observation). Nurses use logic to understand the situations and reflect on ways to improve their professional skills and knowledge (abstract conceptualisation) and practise their newly acquired skills in role-play sessions with standardised patients (applying with active experimentation) ( Schnepp and Rogers, 2017).
1.2 ObjectiveThis study aimed to explore the effect of an MEACP training programme in enhancing nurses’ skills, knowledge, confidence, attitudes and actual ability to support patients’ decision-making on ACP. In this context, 'effect' refers to the specific changes or outcomes resulting from the MEACP, which may include positive changes, no change, or negative changes. We hypothesised that nurses working in acute care settings who received the MEACP training programme would experience improvements in their decision-support skills, knowledge, confidence and attitudes and gain actual experience of supporting patients’ decision-making on ACP from baseline to after intervention, compared with those who did not receive the ACP education intervention (control group).
2 DesignThis study adopted a single-blinded, cluster randomised controlled trial with a repeated-measures, two-arm, parallel-group design. The protocol was prospectively registered at ClinicalTrials.gov (ID: NCT05625906). The trial is reported in accordance with the Consolidated Standards of Reporting Trials (CONSORT) guidelines for RCTs. The privacy rights of human subjects have been observed and that informed consent was obtained for experimentation with human subjects. Ethical approval for this study was granted by the Joint Clinical Research Ethics Committee of the Chinese University of Hong Kong (CREC-2018.610).
3 Materials and methods3.1 Participants
Eligible subjects were recruited from 23 wards across eight local hospitals in Hong Kong. The number of wards in each hospital ranged from 8 to 52. On average, each ward has 45 beds and 20 nurses. Recruitment was performed from January 2023 to July 2024. Trained research assistants approached the potential participants in the participating hospitals and screened them according to the eligibility criteria. The inclusion criteria were: (1) registered nurses; and (2) no prior specialty training in ACP communication. To minimise the impact of confounding variables on the effectiveness of the programme, participants who had completed any post-registration courses in palliative care were excluded from the study.
Sample size was estimated based on the results of a previous RCT assessing the effectiveness of a decision-support skills training programme for nurses, as assessed using the Decision Support Analysis Tool (DSAT) ( Murray et al., 2010). Murray et al. found an effect size of 0.91 for the DSAT, with mean score changes of 3.75 (SD = 3.79) and 0.67 (SD = 2.90) in the intervention and control groups, respectively. By allowing for a more conservative effect size of 0.6, it was estimated, using the power analysis software GPower 3.1, that 45 participants in each of the intervention and control groups would be required to achieve 80 % power at a two-sided 5 % level of significance for the primary outcome of skill performance, as assessed by the DSAT. The cluster RCT was originally planned to recruit from 26 wards. To account for an attrition rate of up to 20 % and a clustering design effect with an intra-cluster correlation coefficient of up to 0.1 ( Donner and Klar, 1994), eight nurses were recruited from each of the 26 wards, yielding a total of 208 nurses.
3.2 Randomisation‘Acute care wards’ in this paper refers to general medical, surgical, or non-palliative care wards. Randomisation was performed at the cluster (ward) level instead of the individual participant level to avoid contamination between the intervention and control groups. The wards were randomised to either the intervention or the control group at a 1:1 ratio. Eligible participants recruited from the same ward were all allocated to either the intervention or the control group accordingly. The participating wards were sequentially numbered by a research assistant and a computer-generated random allocation sequence of group identifiers was prepared by an independent statistician who was not involved in the study. The group allocation of the participating wards was made sequentially according to the generated random sequence of group identifiers. At recruitment, the group allocation was concealed from the research assistants.
3.3 InterventionEnrolled wards were randomly assigned to two groups: the intervention group (who received MEACP training) and the control group (no MEACP training). The MEACP training consisted of a mobile application (app) and a 2-hour nurse-led, face-to-face, skill-building workshop. The application of the theory of planned behaviour and the experiential learning cycle to develop the MEACP programme is illustrated in Supplementary Material 1.
The mobile app included 10 modules of ACP training materials. At the end of each module, the participants were required to complete a quiz relating to the content of that module. The app would display a checkmark icon for participants to track their learning progress. The app also featured a range of audiovisual aids to enhance the learning experience, including informative PowerPoint slides and high-fidelity videos produced by our team. The videos described real case studies shared by patients and controversial scenarios that may occur in the implementation of ADs. The layout of the mobile app is demonstrated in Supplementary Material 2. The participants were required to complete all 10 modules of training within 1 month before attending the workshop.
The 2-hour training workshop included a 1-hour enhanced lecture, a 30-minute role-play exercise and a 30-minute reflective session for a group of 4–5 participants. Each workshop session was facilitated by an experienced nurse with a palliative care speciality qualification and all of the sessions were held at the hospitals where the participants worked. An enhanced lecture is defined as a series of mini-lectures supplemented by active learning between each session ( Bonwell, 1996), which offers an action-based, team-based and inquiry-based learning environment to facilitate in-depth learning ( Bonwell, 1996).
In the role-play session, each participant was instructed to initiate a one-on-one ACP conversation with a standardised patient based on a designated scenario (Scenario 2). The ‘standardised patients’ were trained actors who portrayed patients’ characteristics or reactions in given scenarios. This method has been used in the teaching and evaluation of students’ clinical examination skills ( Luck and Peabody, 2002) and recently to assess nurses’ decision-making support skills ( Stacey et al., 2008). After the role-play exercise, feedback was provided by the workshop intervener on the participants’ performance. The participants were also tasked with writing a journal, reflecting on their strengths and weaknesses in their role-play performance and what they had learned in the workshop. Finally, the participants completed a post-intervention questionnaire.
3.4 Outcome measuresThe primary outcome was the nurses’ decision-making support skills, as measured by the Brief DSAT (DSAT 10) (0–10) ( Stacey et al., 2008). Decision support is defined as efforts to prepare patients for decision-making by providing information, clarifying values and enhancing implementation ( Stacey et al., 2008). The Brief DSAT 10 is a 10-item objective assessment tool with five domains: (1) decision-making status, (2) knowledge, (3) values/preferences, (4) others’ involvement in the decision and (5) next steps. The tool has demonstrated a satisfactory internal consistency, with a Cronbach’s alpha of 0.79, high validity and the ability to discriminate between different kinds of decision support, such as trained versus untrained practitioners’ performance ( Stacey et al., 2008). Each item was scored as 1 point, with the summed DSAT score ranging from 0 to 10. A higher DSAT score indicated better decision-making support skills.
There were six secondary outcomes: (1) Nurses’ attitudes towards ACP, as measured on a nine-item scale (9–45) developed and validated by Putman-Casdorph et al. (2009). The scale included items related to the perceived helpfulness of ACP and perceptions of life-sustaining treatment. Each item was rated on a 5-point Likert scale (1 = totally disagree to 5 = totally agree). A higher score indicated a more positive attitude towards ACP. The tool demonstrated a satisfactory internal consistency of 0.69 in a previous study evaluating nurses' knowledge, attitude, confidence and experience regarding advance directives ( Putman-Casdorph et al., 2009). (2) Nurses’ knowledge of ACP and AD, as assessed by a 10-item true or false knowledge test developed by ( Siu et al., 2010; Yee et al., 2011). The questions in the knowledge test were reviewed by an expert panel of two academics and eight clinicians, who assessed their relevance using a 10-item, 4-point scale. The results demonstrated that the knowledge test had a good item-level content validity of 0.97. (3) Nurses’ level of confidence in conducting ACP, as assessed by a single-item, 5-point Likert scale (1 = not confident at all to 5 = extremely confident). The scale was adopted to assess nurses’ confidence in discussing ADs with patients in a previous study ( Putman-Casdorph et al., 2009). (4) Nurses’ satisfaction with the MEACP programme, as assessed by an eight-item, 5-point Likert scale adapted from So et al. (2019) ( So et al., 2019) (1 = strongly disagree to 5 = strongly agree), where higher scores indicate higher satisfaction. This scale was adopted to evaluate health workers’ satisfaction with a breast and cervical cancer prevention training programme ( So et al., 2019). (5) Nurses’ actual experience with ACP discussions, as measured by the responses to an open-ended question, ‘ how many times have you participated in ACP discussions with your patients in the last 12 weeks?’ (6) Nurses’ level of involvement in the MEACP programme, including (a) the utility of the online training component, as measured by the cumulative time spent on the app and the completion rate of the 10 modules of education materials in the app and (b) the intervener’s observations and records of the nurses’ levels of interest, participation and responses to the activities in the workshop.
3.5 Data collectionThe primary and secondary outcomes were collected at recruitment (T0) and immediately (T1) and 12 weeks (T2) after the training by four trained research assistants. At recruitment (T0), after providing informed consent, all participants, including those in the intervention and control groups, received the same set of pre-test questionnaires. The participants were instructed to initiate a one-on-one ACP conversation with a standardised patient based on a designated scenario (Scenario 1). The conversation between the participant and the standardised patient was audiotaped for further DSAT analysis.
Immediately after the intervention (T1, 4 weeks after recruitment), the participants in the intervention group received the post-test questionnaire assessing their attitudes, knowledge, level of confidence and satisfaction with the MEACP programme. During the role-play, the nurses’ decision-making support skills were assessed based on an audiotape recording of their ACP conversation with a standardised patient in Scenario 3. The details of scenarios 1–3 used in the role-play sessions are illustrated in Supplementary Material 3. At 12 weeks, the participants received an online questionnaire assessing their knowledge of ACP, level of confidence in conducting ACP and actual frequency of ACP discussions. Fig. 1 illustrates the flow of the data collection process. The primary outcome, the nurses’ decision-support skills, was scored by four trained DSAT coders who were blinded to the group allocation and did not participate in the data collection. A meeting was held with all coders to discuss any inconsistencies and issues in the ratings.
3.6 Data analysisDescriptive statistics were used to summarise the study data, including the baseline characteristics of the participants and their outcomes across the study time points. The normality of the continuous variables was assessed using skewness and kurtosis statistics and a normal probability plot. The outcome variable ‘number of ACP discussions’ was found to deviate from normality and thus was square root transformed before being subjected to inferential analysis. As the study outcome data were hierarchical in nature with three levels, namely the repeated measures outcomes within each participant (level 1), individual participants clustered in each ward (level 2) and the participating wards (level 3), we adopted a mixed-effects model to analyse the multilevel data to account for intra-correlated repeated measures and inter-correlation within wards (clusters) ( Singer, 1998). Specifically, a (three-level) random intercept mixed-effects model was used to compare the differential change in each of the repeated-measures outcomes at T1 (and/or T2 whenever appropriate) with respect to T0, accounting for the intra-correlated clustered and repeated-measures data. In particular, the fixed-effect group-by-time interaction terms (Group*T1 and/or Group*T2) were included in the model to assess the between-group mean differences in changes in each outcome at T1 and/or T2 relative to T0. The significance of the interaction terms was assessed based on the Wald test ( Singer, 1998). The intention-to-treat principle was adopted in the outcome analysis. The maximum likelihood method was used in the mixed-effects models to handle missing data, of which the effect estimates were not threatened by attrition bias, provided that the data were missing at random. The mixed-effects modelling was performed using PROC MIXED (release 9.4, SAS Institute, Cary, NC, USA) with significance set at 0.05 (2-sided).
4 Results4.1 Participant flow
Nurses from 23 wards at eight local hospitals in Hong Kong were recruited for the study, with 287 potential participants screened for eligibility. Fifty-seven of the participants were excluded from the study: five were not registered nurses and 52 refused to participate due to a lack of interest. This yielded a total of 230 nurse participants, with a recruitment rate of 80.1 %. Of the 23 participating wards, 12 (with n = 112 nurses) were randomised to the control group and 11 (with n = 118 nurses) were randomised to the intervention group. At T1, 16 of the 118 participants in the intervention group had failed to complete the MEACP training due to their busy schedules, yielding an intervention completion rate of 86.4 %. A total of 209 participants completed T1 data collection and 199 participants completed T2 data collection. Thus, the attrition rate from T1 to T2 was 4.78 %. In accordance with the intention-to-treat principle, all 230 participants were included in the outcome analysis. Fig. 2 illustrates the CONSORT diagram of the participant flow in this study.
4.2 Participant characteristicsMost of the participants were aged between 26 and 35 (48 %), were female (73 %) and had a bachelor’s degree or above (57 %). Most of the participants were registered nurses (75.7 %) with 2–10 years clinical experience (60.4 %) and were from the specialty of medicine and geriatrics (76.5 %). As shown in Table 1, the control and intervention groups were homogeneous in terms of the baseline characteristics of the participants.
4.3 Outcome evaluationThe primary and secondary outcomes between the control and intervention groups and across the study time points are presented in Table 2. To examine the changes in outcome measurements from baseline to follow-up, we employed a mixed-effects model that accounted for the intra-correlated repeated measures and inter-correlation within the cluster ( Table 3).
4.3.1 Effects on decision-making support skillsThe mean (M) score for the DSAT in the MEACP group (possible range: 0–10) increased from mean = 2.6 (SD = 1.5) at baseline to mean = 6.2 (SD = 1.7) at T1, while the increase was smaller in the control group [from mean = 2.1 ( SD = 1.5) to mean = 2.8 (SD = 1.4)]. The mixed-effects model ( Table 3) also revealed that the participants in the intervention group, on average, showed a significantly greater increase in DSAT score than their counterparts in the control group (regression coefficient for Group*T1, B = 2.90, 95 % CI: 2.47–3.34, p < 0.001). This result indicates that the MEACP programme positively improved the nurses’ decision-making support skills.
4.3.2 Effects on attitudes towards ACPBoth groups showed slight improvements in their attitudes towards ACP at both T1 and T2 relative to T0 ( Table 2). Although the increase in attitude score was larger in the intervention group at T1 (Group*T1, B = 0.11, 95 % CI: −0.66 to 0.87, p = 0.783) and at T2 (Group*T2, B = 0.18, 95 % CI: −0.73 to 1.09, p = 0.702) than in the control group, statistical significance was not reached ( Table 3). This result suggests that the MEACP programme did not have a significant impact on the nurses’ attitudes towards ACP.
4.3.3 Effects on knowledge of ACPThe means and SDs of the knowledge score across the study time points for the participants are shown in Table 2. The participants in the intervention group demonstrated significantly greater improvements in their knowledge of ACP and ADs at both T1 (Group*T1, B = 1.16, 95 % CI: 0.70–1.61, p < 0.001) and T2 (Group*T2, B = 0.72, 95 % CI: 0.27–1.16, p = 0.002) than those in the control group ( Table 3). This result suggests that the MEACP programme had a positive impact on the nurses’ knowledge of ACP.
4.3.4 Effects on confidence in conducting ACPThe means and SDs of the confidence score across the study time points for the two groups are shown in Table 2. The intervention group, on average, showed a significantly larger improvement in the score at T1 (Group*T1, B = 0.58, 95 % CI: 0.37–0.79, p < 0.001) than the control group, but the mean difference at T2 was not significant (Group*T2, B = 0.21, 95 % CI: −0.03 to 0.45, p = 0.086). The findings indicated a significant increase in nurses’ confidence in performing ACP immediately post-intervention, suggesting that the MEACP program was effective on perceived confidence in the short term. However, the improved perceived confidence was not sustained after 12 weeks.
4.3.5 Effects on actual ACP discussionsBoth groups showed a slight increase in the average number of ACP discussions in the last 12 weeks from T0 to T2 ( Table 2). However, there was no significant evidence of a between-group difference for the increase in ACP discussions (Group*T2, B = −0.19, 95 % CI: −0.48 to 0.11, p = 0.208) ( Table 3). This result suggests that the MEACP programme did not have a significant impact on the number of ACP discussions conducted by the nurses in the last 12 weeks.
4.4 Satisfaction with the MEACP trainingMost of the participants agreed or strongly agreed that the MEACP programme had helped them understand how to discuss ACP with patients (96.1 %); enhanced their skills, knowledge, confidence and attitudes in relation to ACP discussions; provided actual practice in supporting patients’ decision-making on ACP (94.1 %); and helped them understand how to enhance adherence to a patient’s expressed preference for care (92.2 %). Nearly all of the participants (99 %) reported that they understood the importance of ACP. Furthermore, 97.1 % of the participants were satisfied with the training programme and 96.1 % of the participants stated that they would recommend the MEACP training to their colleagues ( Table 4).
4.5 Nurses’ level of involvement in MEACP programmeOn average, the participants spent more than 3 h (mean = 193 min, SD 108 min) on the MEACP app, with 86.4 % of the participants completing all 10 modules. For the skill-building workshop, as rated by the workshop intervener on a scale from 0 to 10, the participants showed a high level of interest in the activities (mean = 8.43, SD = 1.07), a high level of participation in the activities (mean = 8.44, SD = 1.04) and a high level of responsiveness to the activities (mean = 8.48, SD = 1.11).
5 DiscussionThis is the first study to explore the effects of a theory-based ACP facilitator training intervention for nurses working in acute care settings. The MEACP programme is a training intervention integrating a skill-building, role-playing workshop with an easily accessible, informative mobile app. The results demonstrated that the MEACP programme enhanced nurses’ decision-making support skills, increased their knowledge of ACP and boosted confidence in conducting ACP.
Previous studies have usually focused on examining the effectiveness of ACP training interventions on the participants’ knowledge ( Jo et al., 2023; Millstein et al., 2022; Pereia-Salgado et al., 2019), attitudes ( de Campos and Polifroni, 2023), self-efficacy ( de Campos and Polifroni, 2023) and communication skills ( Chen et al., 2021). They have overlooked the importance of nurses’ actual performance of decision-making support skills for ACP discussions. Good decision-making support skills are important in helping patients clarify their values and goals by explaining the potential benefits and harms of different types of life-sustaining treatment. The key strength of our study is the selection of nurse’s decision-making support skills as the primary outcome, which were measured by an objective assessor’s rating using the Decision Support Assessment Tool (DSAT). Unlike previous studies ( Chen et al., 2021), which have usually adopted self-reported measures, the nurses’ decision-making support skills performed on standardized patients in this study were scored by several independent blinded assessors based on the DSAT, which increased the validity of the evidence for improvements in skill.
This study’s findings are in line with those of previous studies, strengthening the evidence for the effectiveness of role-play or simulation-based ACP facilitator training interventions in improving nurses’ knowledge of and self-confidence in conducting ACP. A recent study on an ACP simulation-based communication training programme also illustrated its effectiveness in improving nurses’ knowledge of ACP and confidence in communication ( Chen et al., 2021). Chen et al. suggested that the role-play scenarios reflected authentic clinical settings and allowed participants to practise one-on-one conversations with standardised patients, without worrying about harming real patients ( Chen et al., 2021). In addition, the role-play scenarios allowed them to recognise their own strengths and weaknesses in the communication process, while observing and reflecting on other participants’ performance. The feedback from the workshop intervener also helped the participants to identify the blind spots in their performance.
The MEACP programme did not lead to the expected significant increase in the number of ACP discussions in the last 12 weeks. However, the occurrence of ACP discussions is influenced by multiple factors, such as the types and stages of patients’ diseases and the readiness of patients and family members to begin discussions. The study was conducted during the COVID-19 outbreak in Hong Kong and infection control issues, such as social distancing and minimal patient contact, might have limited their opportunities to conduct ACP with patients or family members. In addition, the nonsignificant change in nurses’ attitudes towards ACP following the MEACP programme may be attributed to a ceiling effect, as the nurses in both groups had good baseline attitudes towards ACP, leaving little room for improvement. A previous cross-sectional study found that attitude was an important predictor of ACP practice behaviour among Chinese oncology nurses ( Shih and Lu, 2024). Future studies could consider using other strategies to enhance nurses’ attitudes towards ACP. It would also be worthwhile replicating the study in other populations where there is a greater need for attitude enhancement.
Compared with other ACP facilitator training interventions, the MEACP programme’s greatest strength is its ability to provide the participants with a blended, flexible mode of training that can accommodate nurses’ tight schedules and heavy workloads, allowing them to access the training materials on a mobile app at anytime and anywhere. The workshop in the MEACP programme allowed the participants to go through a cycle of experiential learning. The training materials can be reused and have a high level of sustainability and economic value. One of the most notable findings of our study was that even a brief ACP educational intervention (3 h using the mobile app and a 2-hour face-to-face group interactive workshop) enhanced nurses’ decision-making support skills, knowledge and confidence in conducting ACP. The objective assessment of their decision-support skills by DSAT provided strong evidence of their improved performance. Additionally, the findings of this study indicate that the MEACP programme is feasible and acceptable for use in acute care settings. The participants’ completion rate of the MEACP mobile app was high (>85 %) and no adverse events occurred during the MEACP programme. The participants reported a high level of satisfaction (>90 %) with the training and were willing to recommend the programme to their colleagues.
5.1 Implications for practiceGiven that a significant proportion of deaths occur in acute care wards, nurses working in such settings are expected to be involved in providing end-of-life care and to lead ACP discussions. The findings of this study will help to inform clinical managers about the implementation of such a training programme, which provides a novel, efficient and user-friendly way for nurses to improve their ACP decision-support skills in acute care settings. Mobile apps provide an ideal medium for providing access to a large audience at anytime and anywhere and the use of an app in the MEACP programme enables nurses to revisit the materials as frequently as they wish. The research team will continue to promote the integration of training strategies into clinical practice, as this is a promising way to embed ACP discussions in day-to-day nursing practice, thereby enhancing end-of-life care for patients and their families, even in non-palliative care settings.
5.2 LimitationsThis study had four limitations. First, the participants were not blinded to the group allocation due to the nature of the intervention, which may have resulted in reporting bias in self-reported assessment, such as confidence in conducting ACP. Second, there is no formal cost effectiveness analysis of the economic value of the intervention. Third, the follow-up period of 12 weeks appeared to be inadequate to assess the longer maintenance of the effects. Fourth, this study did not incorporate a framework, such as the Technology Acceptance Model (TAM) to examine the factors influencing nurses’ acceptance of the MEACP mobile apps ( Davis, 1989). Future studies are recommended to explore nurse’s perceived usefulness and perceived ease of use regarding the mobile app.
6 ConclusionMEACP is an innovative ACP facilitator training intervention for acute care nurses that incorporates a skill-building, role-playing workshop with an easily accessible, informative mobile app. It can effectively enhance nurses’ decision-making support skills, improve their knowledge of ACP and boost their confidence in conducting ACP. These positive findings indicate that the MEACP programme should be integrated into routine palliative care service training in acute care settings in the future. Further research is recommended to enhance the programme’s content and activities to foster nurses’ positive attitudes towards ACP.
CRediT authorship contribution statementCarmen W.H. Chan: Conceptualization, Methodology, Project administration, Writing – review & editing. Nancy H.Y. Ng: Investigation, Writing – review & editing. Helen Y.L. Chan: Methodology, Writing – review & editing. Kai Chow Choi: Methodology, Data analysis, Writing – review & editing. Ka Ming Chow: Methodology, Writing – review & editing. Cecilia W.M. Kwan: Investigation, Writing – review & editing. Jackie Robinson: Methodology, Writing – review & editing. Mankei Tse: Project administration, Writing – original draft, Writing– review & editing.
FundingThis study was supported by the General Research Fund ( 14602222).
Declaration of Competing InterestThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper
AcknowledgementsThis study was supported by the General Research Fund (14602222). Professional English language editing support provided by AsiaEdit (asiaedit.com).
Appendix A Supporting informationSupplementary data associated with this article can be found in the online version at doi:10.1016/j.nepr.2025.104428.
Appendix A Supplementary materialSupplementary material
| Control
(n = 112) | Intervention
(n = 118) | |
| Age (years) | ||
| 18 – 25 | 12 (10.7 %) | 22 (18.6 %) |
| 26 – 30 | 30 (26.8 %) | 23 (19.5 %) |
| 31 – 35 | 31 (27.7 %) | 28 (23.7 %) |
| 36 – 40 | 12 (10.7 %) | 7 (5.9 %) |
| 41 – 45 | 15 (13.4 %) | 12 (10.2 %) |
| 46 – 50 | 8 (7.1 %) | 14 (11.9 %) |
| > 50 | 4 (3.6 %) | 12 (10.2 %) |
| Gender | ||
| Female | 90 (80.4 %) | 78 (66.1 %) |
| Male | 22 (19.6 %) | 40 (33.9 %) |
| Educational attainment | ||
| Diploma or high diploma | 12 (10.7 %) | 13 (11.0 %) |
| Bachelor’s degree | 63 (56.3 %) | 68 (57.6 %) |
| Master’s degree or above | 37 (33.0 %) | 37 (31.4 %) |
| Current work status | ||
| Register nurse | 86 (76.8 %) | 88 (74.6 %) |
| Advanced practice nurse | 23 (20.5 %) | 27 (22.9 %) |
| Ward manager/nurse consultant | 3 (2.7 %) | 3 (2.5 %) |
| Nursing working experience (years) | ||
| ≤ 5 | 42 (37.5 %) | 42 (35.6 %) |
| > 5 – 10 | 32 (28.6 %) | 23 (19.5 %) |
| > 10 – 15 | 14 (12.5 %) | 16 (13.6 %) |
| > 15 – 20 | 10 (8.9 %) | 14 (11.9 %) |
| > 20 | 14 (12.5 %) | 23 (19.5 %) |
| Current service specialty | ||
| Medicine and geriatrics | 82 (73.2 %) | 94 (79.7 %) |
| Oncology | 9 (8.0 %) | 11 (9.3 %) |
| Pulmonary care | 11 (9.8 %) | 9 (7.6 %) |
| Surgery | 10 (8.9 %) | 4 (3.4 %) |
| | Control | Intervention | |
| | |||
| Brief Decision Support Analysis Tool (DSAT10) | T0 | 2.1 (1.5) | 2.6 (1.5) |
| [possible score range: 0 – 10] | T1 | 2.8 (1.4) | 6.2 (1.7) |
| | T0 | 31.9 (2.7) | 31.5 (2.5) |
| Attitude score | T1 | 33.0 (2.8) | 32.9 (2.9) |
| [possible score range: 9 – 45] | T2 | 33.0 (3.1) | 33.0 (3.1) |
| | |||
| Knowledge score | T0 | 7.5 (2.0) | 7.3 (1.7) |
| [possible score range: 0 – 10] | T1 | 8.1 (1.7) | 9.2 (1.3) |
| T2 | 8.1 (1.9) | 8.7 (1.2) | |
| | |||
| Confidence score | T0 | 2.2 (0.8) | 2.4 (0.8) |
| [possible score range: 1 – 5] | T1 | 2.4 (0.8) | 3.2 (0.7) |
| T2 | 2.6 (0.8) | 3.0 (0.8) | |
| | |||
| Number of discussions in the last 12 weeks | T0 | 1.2 (2.2) | 1.3 (2.2) |
| T2 | 2.2 (2.9) | 1.7 (2.1) |
| | Regression coefficients (95 % CI) # | P |
| | ||
| Brief Decision Support Analysis Tool total score (DSAT10) | ||
| Group*T1 | 2.90 (2.47, 3.34) | < 0.001 |
| | ||
| Attitude score | ||
| Group*T1 | 0.11 (−0.66, 0.87) | 0.783 |
| Group*T2 | 0.18 (−0.73, 1.09) | 0.702 |
| | ||
| Knowledge score | ||
| Group*T1 | 1.16 (0.70, 1.61) | < 0.001 |
| Group*T2 | 0.72 (0.27, 1.16) | 0.002 |
| | ||
| Confidence score | ||
| Group*T1 | 0.58 (0.37, 0.79) | < 0.001 |
| Group*T2 | 0.21 (−0.03, 0.45) | 0.086 |
| | ||
| Number of discussions in the last 12 weeks † | ||
| Group*T2 | −0.19 (−0.48, 0.11) | 0.208 |
| | Agree / strongly agree |
| 1. I gained good understanding on the topic. | 100 (98.0 %) |
| 2. The content helped me understand how to discuss ACP with the patients after a mobile app tutorial, and a skill training workshop with standardised patient encounters for nurses in acute care settings. | 98 (96.1 %) |
| 3. The content helped me to enhance my performing skills, knowledge, confidence, attitude and actual practice in supporting patients’ decision-making on ACP. | 96 (94.1 %) |
| 4. The content helped me understand how to enhance the adherence to patients’ expressed preference for care. | 94 (92.2 %) |
| 5. The speaker demonstrated expertise in teaching and learning. | 97 (95.1 %) |
| 6. The amount of the materials covered in the programme was reasonable and socio-culturally relevant. | 98 (96.1 %) |
| 7. The materials were well organised and clearly presented. My learning needs are met. | 99 (97.1 %) |
| 8. The teaching methods were appropriate. | 98 (96.1 %) |
| 9. I have understood the importance of advance care planning. | 101 (99.0 %) |
| 10. Overall, I am satisfied with the programme. | 99 (97.1 %) |
| 11. I would like to recommend this programme to my colleagues. | 98 (96.1 %) |
©2025. The Authors