Abstract
Background:
The global need for palliative care is increasing. To respond effectively, all healthcare professionals who care for persons with serious illnesses should be trained to provide core palliative care. Despite increasing recognition of its importance, education remains inconsistently implemented.
Objectives:
To explore and identify the essential elements that define and support palliative care education, and to propose a preliminary typology of structural, process, and outcome quality indicators.
Design:
An umbrella review was conducted according to the Joanna Briggs Institute methodology.
Data sources and methods:
We systematically searched three electronic databases (CINAHL, MEDLINE, PsycInfo) for literature reviews on palliative care education. Data were analysed using a combined deductive–inductive qualitative approach.
Results:
Twenty-four reviews were included in the analysis. Seven structural indicators were identified relating to the training and duration of palliative care exposure, instructional format/delivery, contents, structured course development, involvement of local staff and experts, multidisciplinary and context-specific learning, and organisational support. Seven process indicators were identified, encompassing experiential learning, multi-modal approaches, reflective practice and emotional processing, group-based peer learning, embedded learning, flexibility, and feedback. Six outcome indicators were identified, including the enhancement of competencies, sustainability and long-term impact, validated tools for evaluation, participant satisfaction, integration into routine practices, and impact on patients and families.
Conclusion:
This umbrella review proposes an evidence-informed framework of potential quality indicators for palliative care education, offering guidance for curriculum design and policy development while identifying key areas requiring empirical refinement and validation.
Plain language summary
Palliative care helps people with life-limiting illnesses and supports their families. The global need for palliative care is growing, and all healthcare professionals should receive education and training in its basic principles. However, such education is often delivered inconsistently, and there are few clear standards for what makes it effective. This study reviewed existing summaries of research, known as literature reviews, to identify what defines high-quality palliative care education. We searched three scientific databases for reviews focusing on palliative care education and training. Drawing on findings from 24 reviews, we proposed a set of quality indicators — features that reflect the elements of good education and training. These indicators were grouped into three categories: Structural indicators describe how education is organised and supported, such as the setting and resources. Process indicators describe how learning takes place, including teaching methods, interaction, and reflection. Outcome indicators describe the results of education, such as improved knowledge, skills, confidence, or patient outcomes. This review proposes a framework that brings these indicators together. It can help educators, researchers, and policy makers design, evaluate, and improve palliative care education. By strengthening education for all healthcare professionals, we can promote more consistent, compassionate, and high-quality care for people with serious illnesses and their families.
Background
From a public health perspective, palliative care education (PC-education) is recognised as a fundamental pillar for integrating such care into health systems. 1 Lack of education and training among healthcare professionals is one of the primary reasons for the persistence of avoidable suffering. 2 Consequently, the World Health Assembly (WHA67.19) urged Member States to incorporate palliative care into the ongoing education and training of all healthcare providers, according to their respective roles and responsibilities. 2
In alignment with this, the World Health Organization proposed two education-specific indicators for monitoring the development of palliative care: (1) the proportion of medical and nursing schools with palliative care formal education in undergraduate curricula, and (2) availability of specialisation in palliative medicine for physicians. 3 Despite these developments, the quality of education remains non-standardised and under-evaluated. Quality indicators become essential tools for educators, institutions, and policymakers to assess current performance, monitor progress, and guide improvements. Moreover, they can help define standards, support curriculum development, and inform national policies. 4 In the absence of agreed quality indicators, PC-education may be implemented inconsistently, making performance monitoring and continuous improvement difficult. Moreover, indicators are most useful when organised within a coherent framework, so they can be applied consistently and interpreted in relation to the educational system they are intended to assess.
However, identifying meaningful quality indicators in PC-education poses particular challenges. Educational approaches vary widely across contexts, shaped by pedagogical traditions, professional cultures, and differing perceptions of death, dying, and bereavement. While some countries have introduced palliative care into health professional curricula and recognise it as a (sub)specialty, the extent of implementation varies widely. Of 56 European countries surveyed, 15 mandated PC-education in medical schools, 21 recognised it as a specialty or subspecialty, and 19 required it in undergraduate nursing education. 5
The CODE-YAA@PC-EDU (COllaboratively DEveloped culturallY Appropriate and Inclusive Assessment tool for Palliative Care EDUcation) project (COST Action CA22127) seeks to define quality indicators that support the development of a gold standard for high-quality PC-education and training. Through the development and application of CODE-YAA as a culturally and ethically appropriate, evidence-based self-assessment tool, the project aims to assess, enhance, and promote access to PC-education across the Pan-European region, with a particular focus on primary healthcare as a sustainable and cost-effective model for palliative care delivery. 6
Within this context, the present umbrella review represents one important step in the conceptual groundwork for the development of the CODE-YAA tool. More specifically, this research aims to identify essential elements that define and support PC-education, to propose a preliminary typology of quality indicators.
Methods
Design
We conducted an umbrella review based on the Joanna Briggs Institute (JBI) methodology. 7 This included the Participants, Concept, Context (PCC) framework to guide the development of our research question and the aligned eligibility criteria:
Participants/population: The target population consisted of all professionals caring for persons experiencing health-related suffering, 8 as well as students in relevant disciplines.
Concept: Reviews addressing QIs, standards, competency frameworks, or other criteria that define or support the quality of PC-education.
Context: Any setting involving undergraduate, postgraduate, or continuing education in palliative care (generalist or specialist) across any region.
The review protocol was not registered or published in advance.
Search strategy
We searched the following databases in May 2024: CINAHL, MEDLINE, and PsycINFO. The search strategy combined terms related to the population (palliative care), the concept of interest (QIs), as well as the context (education and training), using Boolean operators to refine the results. The search was limited to studies published from 2014 onwards, aligning with the World Health Assembly resolution (WHA67.19) on Palliative Care. An example of the search strategy is shown in Supplemental File 1. No language restrictions were applied. All records were imported into Covidence, 9 a web-based platform designed to streamline the screening and data management processes involved in conducting systematic reviews. 9
Eligibility criteria
Eligibility criteria are listed in Table 1.
The PCC framework for eligibility of studies.
PCC: Participants, Concept, Context.
Study selection process
The initial screening of titles and abstracts was piloted by two principal researchers (A.P., T.P.) to identify potential issues in applying the eligibility criteria and to ensure a shared understanding of their interpretation. Based on the insights gained from this pilot phase, a tailored training session and instructional video were developed to promote consistency among all reviewers. In the first step, publications unrelated to PC-education or lacking study findings were excluded. In the second step, only systematic reviews with a clear focus on education were retained for full-text screening and potential inclusion. Two reviewers selected each title, abstract, and full text independently, and possible conflicts were resolved by a third reviewer.
Appraisal of reviews
To evaluate the methodological quality of the systematic reviews included in this umbrella review, we used the JBI Critical Appraisal Checklist for Systematic Reviews and Research Syntheses to assess the methodological quality of the included reviews. 10 We categorised them into low quality (0%–49% of criteria met), moderate quality (50%–74%), and high quality (⩾75%). 11
The appraisal was conducted independently by two reviewers to minimise bias and ensure the reliability of the assessments. Any discrepancies were resolved through discussion.
Data extraction
We developed and tested a standardised data extraction form. Two reviewers extracted information on reviews’ characteristics (author, year of publication, type and objective of the review, setting and databases used, time range and number of included studies, study designs, and geographical origin, appraisal tools used, methods of data synthesis), as well as identified QIs categorised by structure, process, and outcome. A comments section allowed reviewers to provide additional insights. Discrepancies in data extraction were resolved through discussion.
Data synthesis and analysis
Statistical analysis
To assess the degree of overlap among the primary studies included in the systematic reviews, we used the Graphical Representation of Overlap for OVErviews (GROOVE) tool. 12 Overlap refers to the extent to which the same primary studies are included in multiple reviews, which can bias evidence synthesis if not accounted for. Using GROOVE, we calculated the Corrected Covered Area (CCA), a quantitative measure that expresses the proportion of overlapping primary studies relative to the total number of reviews and included studies. Overlap levels were interpreted using established thresholds, ranging from slight (<5%) to very high overlap (>15%). 12
Qualitative analysis
To analyse the phenomenon of interest, we applied a descriptive qualitative approach following an abductive process. 13 In the first phase, we analysed the data deductively using a structured matrix based on Donabedian’s model. The framework was originally developed to assess healthcare quality through three interrelated components: structures, processes, and outcomes. 14 In recent years, this model has also been recognised as a coherent and logical approach for mapping educational quality.15,16
In the second phase, an inductive analysis was conducted within each of Donabedian’s domains to identify themes. No standard guidelines or analytical frameworks were followed. The analysis was conducted iteratively by multiple reviewers with discrepancies resolved through discussion and consensus.
Results
The electronic database search identified 7613 records. After removing duplicates, 5033 records remained. Following screening and eligibility assessment, 24 reviews were included in the umbrella review (cf. Figure 1).

PRISMA flow diagram.
The 24 systematic reviews covered 693 primary studies. After accounting for overlap, 546 non-overlapped primary studies were identified within the reference literature. The CCA analysis, which quantifies the extent to which primary studies are shared across reviews, indicated a slight overlap of 1.2%, suggesting minimal duplication of primary studies across the included reviews. The degree of overlap between individual reviews ranged from 0.0% to 47.8%17,18 (cf. Figure 2).

Graphic Representation of Overlap for Overviews (GROOVE tool). 12
Methodological quality
Based on the assessment tool, of the 24 included reviews, 14 were rated as high quality17,19–31 and 10 as moderate18,32–40 (Supplemental File 2).
Characteristics of reviews
The literature reviews included were published between 201432,37 and 2023,17,18,29,36 encompassing primary studies published between 1983 19 and 2023. 29 Table 2 gives an overview of all reviews included.
Summary of characteristics of included systematic review.
RCT: randomised controlled trial; UCBA: Uncontrolled Before-and-After; CBA: Controlled Before-and-After; MMAT: Mixed-Methods Appraisal Tool; ACP: advance care planning; GRADE: Grading of Recommendations Assessment, Development and Evaluation; MERSQI: Medical Education Research Study Quality Instrument; LTCFs: long-term care facilities; EoL: end of life; EoLC: end-of-life care; IPE: interprofessional palliative care education.
Four reviews conducted a meta-analysis22,24,25,37 and two applied a mixed-method synthesis approach,29,30 while the remaining were qualitative synthesis.
The number of databases searched in each review varied from 3 30 to 13. 19 MEDLINE/PubMed was used in all reviews, followed by CINAHL, which appeared in 21 reviews,17–32,35–37,39,40 EMBASE in 17,17–24,26,28–33,35,37 PsycINFO in 14,17,18,20–22,24,26,27,29,31–33,35 and ERIC in 10.19,20,23–25,29,31,38,39
The reviews addressed a range of topics, primarily focusing on educational interventions and teaching methods, training effectiveness, and areas of clinical competencies such as communication training and symptom management.
Of the 546 primary studies across 48 countries, 238 were conducted in the United States, followed by the United Kingdom (n = 103) and Australia (n = 35). North America, including Canada (n = 32) was the most represented region overall. Additional contributions came from European (n = 213) and Asian countries (n = 55). Teixeira et al. 39 included three studies from Africa (Cameroon, South Africa, Uganda), while Brighton et al. 20 included one study each from Colombia and Argentina, the only two studies representing Latin America. Although not all reviews specified the countries of origin for the included studies, we examined the original articles to retrieve this information. In six cases, however, the country of origin could not be determined with certainty.
Quality indicators for PC-education
We identified 20 quality indicators across structural, process, and outcome domains, as shown in Figure 3.

Overview of proposed quality indicators.
Structural indicators
Structural indicators refer to the organisational, human, and material resources that establish the capacity for education. Across the umbrella review, several recurring were identified as being associated with higher-quality educational outcomes.
Timing and duration of palliative care exposure
The reviews reported high variability in the timing and duration of education across settings and learner levels. Educational exposure ranged from brief, single-session interventions to longitudinal, extended over months.17,26,32,36
Short, structured interventions, such as a 2-h seminar, demonstrate statistically significant improvements in knowledge, self-efficacy, and attitudes.24,30,33 Similarly, brief but immersive experiences, such as a 48-h hospice-home stay, enhance learners’ emotional readiness and reflective capacity. 38
Several reviews reported that longer interventions22,26 were more effective in supporting attitudinal change, particularly in sensitive domains such as attitudes towards death, an outcome known to be resistant to short-term educational efforts. 22 Iida et al. 26 further noted that programs spanning 12–18 months allowed for deeper integration of learning into clinical practice. Similarly, Kirkpatrick et al. 36 highlighted the value of longitudinal interprofessional courses, supporting progressive knowledge, self-efficacy, and collaborative competence development. Courses delivered across multiple, structured sessions, rather than single workshops, reported stronger impacts for clinical outcomes and learner competence, especially in areas such as symptom management and family caregiver support.
Overall, the evidence does not support the superiority of any specific programme duration.20,33 Instead, content quality, instructional design, and delivery method appear to influence outcomes more strongly than the length of the programme itself.20,33
Instructional format/delivery
The reviews reported a broad spectrum of educational delivery methods in palliative care training. Several structural characteristics are associated with enhanced learning outcomes, learner satisfaction, and greater programme scalability.
Studies emphasised the value of voluntary, interest-driven participation. Students who enrolled in elective palliative care courses consistently demonstrated greater gains in knowledge, reduced fear of death, and a stronger commitment to patient-centred care. 25
Face-to-face teaching remains the most common and consistently favoured method, especially when using small-group formats. It is linked to greater satisfaction, deeper engagement, and improved communication skills compared to web-only programmes.25,39
Several studies identified eLearning and blended learning approaches as effective in improving knowledge and self-efficacy, particularly when grounded in educational theory and delivered over time. These formats also offer logistical advantages, including scalability and asynchronous access/delivery,29,31,33 making them particularly suitable for individual learner needs. Although not always compared directly with in-person training, eLearning and blended learning are viable and scalable options. 33 Remote learning, including virtual role-play, receives positive learner evaluations in some situations. For example, Novaes et al. 28 described a remote role-play strategy that students perceived as more effective than in-person delivery. Remote and flexible formats increase accessibility to learning in some situations (for distributed learners or where social restrictions were imposed) and reduce learner anxiety, particularly for emotionally demanding topics.
Format, and delivery or learning method superiority was not identified across reviews. These rather depend on the competencies to be acquired and the context. The design quality, contextual adaptations, and learner engagement appear more influential than delivery mode alone.32,33
Contents
PC-education includes more than clinical instruction, encompassing holistic content that integrates communication, ethics, interprofessional collaboration, and cultural safety. Educational content that includes spirituality and/or grief management is suggested to have a stronger emotional impact, especially for shifting attitudes towards death, including spiritual themes in care, and supporting attitudinal transformation. 22
Content reflecting cultural values, beliefs, and languages improves equity and learner receptivity. For example, some programmes tailored their curriculum to include indigenous language or culturally relevant practices, strengthening accessibility and inclusivity. These adaptations reflect respect for diverse communities and prepare students to provide culturally safe care. 27
Programmes co-designed with patients or caregivers, such as ABC4Nurses, brought emotional depth and real-world relevance. 29 Teaching informed by interdisciplinary scenarios and team-based care promote interprofessional skills and ethical decision-making. 25
This broader approach enhances the depth and quality of learner preparation for palliative and end-of-life care.
Structured course development
Effective education programmes often demonstrate clear theoretical grounding, intentional design aligning learning objectives, pedagogical approaches, and assessment strategies, as well as incorporation of evidence-based models. These elements contribute to both educational rigour and clinical relevance. Contextually tailored teaching and learning materials enhance relevance and learner engagement. 27
Courses that reference existing literature, prior learning, and, especially in the context of communication training, strategies such as SPIKES (Setting, Perception, Invitation, Knowledge, Emotions, Strategy/Summary) or PREPARED (Prepare, Relate, Elicit, Provide, Acknowledge, Encourage, Document), and that are grounded in learning theories and frameworks, are highlighted. 20
Involvement of local staff or experts
The inclusion of local staff or trained experts in the design and delivery of education enhances the perceived relevance and contextual applicability of the programme content. Involvement of local staff or expert input improved perceived relevance.20,35
Learners consistently perceive courses that incorporate local clinicians, cultural advisors, and community stakeholders as enhancing relevance. This contextual grounding improved both engagement and applicability to real-world clinical practice. 20
While local involvement was valued, the presence of formally trained facilitators (whether internal or external) was critical to ensure instructional quality, structured support, and continuity throughout the course. 35 Despite this, a systematic review by Collingridge Moore et al. 35 noted that only 20% of included studies reported facilitators with specific training, suggesting a significant area for development in faculty preparation.
Skilled facilitators are essential to optimise learning, no matter the learning approach or delivery medium. In simulation-based learning environments, the role of the facilitator is even more pivotal. Skilled facilitators are essential to guide debriefings, ensure emotional safety, and reinforce learning objectives. Their presence enhances the authenticity and impact of the simulation experience. 28
Multidisciplinary and context-specific learning
Palliative care is inherently team-based, requiring collaboration across medical, nursing, social, and allied health professions. Educational programmes that reflect this interdisciplinary reality by involving diverse professional groups (e.g. doctors, nurses, and students) and engaging with realistic clinical scenarios tend to yield stronger learning outcomes and promote both individual competence and broader organisational change. Reviews have identified a strong interprofessional focus, in both faculty and learner composition, as a key marker of excellence in education. 36 Combined sessions, such as those involving medical students and practicing nurses, can leverage varied perspectives, bridge theory-practice gaps, and improve students’ readiness for collaborative, ethically grounded decision-making in palliative care.22,25 Despite these advantages, only 22% of communication courses in one review were truly inter- or multiprofessional. By contrast, 75% of courses were single-profession offerings, most commonly attended only by physicians. The remaining 3% also trained a single profession, but included participants with different levels of experience in the field. 20
Programmes that adopted an inclusive approach, involving even non-clinical staff, improved overall understanding of palliative care and fostered greater staff cohesion and cultural awareness. 35 Some studies included professionals beyond healthcare. 28
Organisational support
Organisational support plays a crucial role in the implementation, sustainability, and perceived legitimacy of PC-education. Successful programmes often go beyond well-designed curricula by aligning educational efforts with institutional structures, leadership priorities, and active engagement and support from senior and managerial-level staff. 26 Visible commitment from leadership – through the allocation of time, personnel, and financial resources – is a critical enabler of sustained engagement. 35 Similarly, in the context of advance care planning, organisational infrastructure was essential to effective facilitator training and delivery. 21
Palliative care content is delivered either as a standalone course or integrated into broader medical and nursing curricula. While earlier studies 32 found no significant differences in outcomes between these two models, more recent work emphasises that formal intentional curriculum integration, leveraging opportunities for both standalone palliative care curricula and opportunities to integrate palliative care elsewhere, enhances effectiveness and consistency. Programmes that formally embed palliative care into undergraduate medical or nursing curricula rather than treat it as optional courseware or scattered in other subjects, are viewed as more effective. 34 Better alignment with academic structures improved reach and legitimacy. 34
Organisational commitment must also extend to continuing professional development through mentorship, equity, and the long-term integration of training into practice. Educational interventions are more effective when complemented by institutional systems and processes that support implementation: “Training is not enough. . .it is necessary to mentor ethnically diverse staff and integrate learning into practice.” 39
Process indicators
Process indicators refer to the methods and strategies used in delivering education that foster meaningful learning and practical competence. In PC-education, evidence supports the use of experiential, interactive, reflective, and flexible approaches to improve both clinical and emotional readiness. 34
Experiential learning
Real-world exposure provided opportunities to observe and practice care, thereby deepening clinical understanding and nurturing humanistic, person-centred care. Direct clinical exposure, such as hospice visits and interactions with patients at end-of-life, strengthened learners’ understanding of suffering, humanistic care, professional identity, and the work-related meaning 34 and the practical applications of knowledge in real-life contexts. 31
Rotations in palliative care settings improve knowledge on symptom management and practical skills, although the optimal duration of such training remains uncertain. 31
Additional experiential learning methods – such as simulation, role-play, case-based learning, and guided bedside learning – consistently improve learners’ practical competencies and emotional readiness.19,23,25,27,28,38 Experiential methods help develop emotional readiness, communication skills, and self-awareness. 29
Skill-specific simulations, particularly for communication and symptom management, have proven to be highly effective in education.25,40 Realistic, immersive environments support learner engagement and scenario authenticity. 40 These simulations bridge the gap between theoretical knowledge and practical application, enhancing learners’ understanding of clinical documentation and care scenarios. Participants in simulation-based training report increased confidence, reduced anxiety, and better preparedness for real-world clinical situations, indicating that such simulations foster both emotional resilience and practical competence. 24
Role-play is especially effective for teaching communication in emotionally charged situations, including delivering bad news and end-of-life discussions.24,28 Role-play is effective in all experimental or quasi-experimental studies as reported by Novae et al. 28
Programmes combining simulation, role-play, group discussions, and structured reflection promote empathy, self-awareness, and applied skill development.20,27,34 Simulation studies excel in pre-briefing, simulation design, and evaluation of learning. These strategies align with educational best practices and are tied to improved attitudes, role understanding, and confidence. 36
Multi-modal approaches
Multi-modal learning methods support different learning styles and improve retention, engagement, and confidence.21,25 Technology-enhanced formats such as eLearning and audio-visual tools are associated with higher educational quality, as measured by increased competence, confidence, and engagement, at least in the area of advanced care planning. 21
Programmes that combine didactic sessions with interactive formats, such as small groups, real patient exposure, and audio-visual tools, outperform lecture-only models.21,23
Web-based and game-based interventions are effective in reducing fear of death and increasing self-efficacy. 25 Similarly, online interventions (e.g. asynchronous modules and discussions) are effective and feasible, especially for working professionals with demanding schedules. 22
Humanities-based (including the use of art, literature, and theatre) methods added ethical depth, emotional resonance, and insight into topics such as grief, empathy, compassion, and person-centred care. 38 Programmes that combine lectures, films, group discussions, patient exposure, and case-based learning lead to enhanced empathy, self-awareness, and competence in communication and decision-making. 25
Reflective practice and emotional processing
Reflective practice was implemented in multiple ways across the included studies, most commonly as post-course written narrative reflections and reflective case discussions. Written reflections were predominantly evaluated using qualitative designs, with narrative texts analysed for aspects of emotional processing. In contrast, reflective case discussions and group debriefings were mostly included within multi-component evaluations and were not evaluated as a standalone component.
Structured reflection on personal values, death, and clinical experiences deepened emotional learning and attitudinal change.34,40 Relevant, emotionally grounded content encourages reflection and deeper learning. 30
Opportunities for ongoing reflection during workshops or on-the-job learning helped translate education into practice and behavioural change. 35
Ratislavová et al. 38 reported awareness of the need for a safe environment conducive to the sharing of emotions in lessons. Emotionally safe learning environments encourage reflection, deeper emotional engagement, and openness. 38
Meaning-centred teaching and reflective case discussions support deeper internalisation of palliative values. 30
Group-based, peer learning
Learning in small groups supports collaboration, mutual learning, and the exploration of ethical and emotional topics. Small-group discussions in or outside the context of problem-based or case-based learning encourage collaboration, practical applications of theory, ethical reflection, emotional processing, and critical thinking.22,23
Students value dialogue, which is inherent in small-group discussions. They help nurture an atmosphere of trust and security. Exposure to interdisciplinary teams and problem- or case-based scenarios further enhances teamwork skills, knowledge, and communication. 25
Group settings are reported to foster trust, dialogue, and peer-supported learning. 38 Face-to-face or asynchronous group-based modules increase emotional engagement and peer learning. 22 Discussion-based methods help process emotional content and build confidence in difficult conversations. 38
Embedded learning
Closely linked to experiential learning, embedded learning clarified how and when learning was applied in practice. The integration of learning tools into clinical workflows improved access and encouraged repeated, just-in-time use. These tools enabled immediate evaluation of learning experiences, including where and what learning occurred, as well as teaching methods and attitudes towards learning. 31 Embedding experiential learning and reflection into current practices supports emotional preparedness and behaviour change. In most cases, learning activities were conducted within the facility; however, in some instances, the intervention required involvement from healthcare professionals outside the facility. Adaptations are necessary to incorporate existing patterns of work into the learning process. 35
Flexibility
Flexible formats (in timing, location, and method) help overcome logistical barriers such as workforce challenges and support continual professional education. 35
Feedback
Personalised feedback (audio/video) adds educational value. 20 Active skill rehearsal and feedback promote real-world communication competence. 23
Outcome indicators
Outcome indicators are defined as the effects of education on learners, clinical practice, and health systems, reflecting the extent to which educational objectives are achieved.
Enhancement of competencies
PC-education consistently demonstrates improvements in learners’ knowledge, self-efficacy, and clinical competence. Across multiple studies, interventions lead to statistically significant gains in knowledge, especially regarding communication frameworks and symptom management.21,23,24,27,31,38,39 A meta-analysis by Donne et al. 24 reveals a large effect size for knowledge acquisition and moderate gains in attitudes. Confidence, particularly in conducting End-of-life discussions and managing complex symptoms such as pain, nausea, or delirium, is the most frequently reported outcome. 24 Although confidence increases are widely reported, 29 they are often self-assessed rather than observed in practice. 30
Many interventions contribute to positive shifts in attitudes towards caring for dying patients. Studies emphasise that changes in these specific care-related attitudes are more pronounced than shifts in abstract concepts, such as attitudes towards death. 22 Training also improves affective skills such as empathy, family-centred communication, and comfort in discussing emotional or sensitive topics. 29 However, evidence on whether these attitudinal gains consistently translate into behavioural changes in clinical practice remains mixed. For example, Tropea et al. 40 reported limited transfer of communication skills into real patient interactions, underscoring a potential gap between training and practice.
Sustainability and long-term impact
Short-term outcomes, such as pre-training versus immediate post-training, are most commonly studied and reported. Long-term impact, including knowledge retention and sustained application of what was learned, is less frequently evaluated. Short time intervals between pre-testing and post-testing31,33 and inadequate post-training intervals restrict the ability to detect long-term behavioural or system-level impact.18,19,26,28,30,32,33 In addition, it could limit meaningful measurement and the comprehensive assessment of impact, durability of learning, and clinical integration.20,28,30,31
In one study reported in Centeno and Rodríguez-Núñez’s review, learners retained and applied palliative care values and skills up to 3.5 years post-graduation, an exception that highlights the importance of longitudinal assessment. 34 Another review, however, reported a lack of sustained clinical impact, with one study assessing clinical behaviour 10 months later and found no difference in quality of care or communication, suggesting that skills were not retained or applied over time, showing a gap between training and sustained practice change. 40
Methodological design, as well as lack of resources, often limit the ability to evaluate sustained learning. 31 These types of studies often require additional resources and funding. Poor reporting, lack of control groups, and reliance on self-reported data, with limited confidence in many studies was reported in a systematic review of communication skills training for generalist palliative care providers. 20
A systematic review and meta-analysis of RCTs of palliative care teaching to health professional students concluded that there was an overall lack of impact and stressed a need for improved skill-building strategies. 24
Validated tools for evaluation
The quality of outcome measurement varies substantially across studies. Programmes using validated instruments offer more robust evidence of effectiveness.26,36,38 Studies that used validated outcome measures for knowledge, attitude, or behaviour change help measure impact bettert 20 and allow comparability across studies. 18
The lack of objective measures, such as observed communication performance or clinical outcomes, is a recurrent limitation.30,31 Incorporating validated, multi-source assessment tools is essential for improving evaluation rigour. Repeated measures would also provide a more reliable outcome assessment. 36
Self-assessment of competencies, while useful, may by themselves be insufficient to explore the full impact of education interventions.30,31 In a review of postgraduate pain and symptom management training, assessment tools often lacked reporting of psychometric properties and sometimes failed to measure application of knowledge. 31
Participant satisfaction
Learner satisfaction is a useful starting point to assess a programme, as a positive learning experience is often a pre-requisite for an impactful intervention. Learner satisfaction is generally high, particularly in programmes that use interactive, reflective, or postgraduate formats.27,29 Participants frequently report that the training is relevant, practical, and emotionally engaging. However, satisfaction, being subjective, may not necessarily correlate with improved competencies or clinical outcomes. For instance, Nevis 37 found no significant difference in satisfaction between intervention and control groups, even when other care QIs improved. Satisfaction alone may not reflect deeper educational impact. 37
Integration into routine practice
Effective integration of training into existing care structures was identified as a key facilitator of sustained impact. Embedding educational practices into routine workflows enhances sustainability and avoids duplication. 35 However, evidence of broader system-level impact remains limited. Nevis 37 and Seow et al. 18 found no consistent reduction in hospital admissions, ICU stays, or emergency visits following training. Similarly, patient and family-reported outcomes – such as communication quality or psychological wellbeing – did not consistently improve post-intervention.31,40 These findings point to a need for better alignment between educational interventions and measurable improvements in patient care and healthcare utilisation.
Impact on patients and families
Few studies assessed the impact of staff training on patient or family outcomes, although some do exist. 20 Structural and methodological design, lack of research funding, and great complexity, limit the ability to evaluate the impact on patient care. 31
Objective measurements of actual or simulated encounters (e.g. time spent, topics covered) were used, and qualitative data from patients/families were also collected to assess the real-world application of the training. 21
The quality of patient and family-reported communication did not improve; some patients had higher depression scores. Simulated training does not translate into measurable improvements in patient/family experience.21,40
Discussion
This umbrella review is the first to systematically synthesise and propose empirically derived objectives that may serve as potential quality indicators for PC-education. It presents a preliminary typology of 20 indicators across structural, process, and outcome domains, offering a foundation for further development, validation, and policy alignment. Rather than establishing consensus, this review identifies and integrates existing evidence to highlight both strengths and critical gaps in current research on educational quality.
Our analysis identifies QI tailored to the unique pedagogical demands of PC-education, which addresses complex human experiences such as suffering, existential distress, death, and bereavement. While education’s role in overcoming clinical integration barriers 42 and shaping learner competencies 27 is recognised, and while some proposed QIs may apply broadly to clinical education, 43 this work defines criteria essential for the specific context of PC-education.
Structural indicators, including timing and content, are key to enabling high-quality PC-education. However, the evidence is insufficient to establish consensus on optimal duration, intensity, or placement within various curricula. This lack of uniformity may stem from the wide variation in how education is delivered across countries, institutions, levels of training, and educational needs and contexts (e.g. non-specialist vs specialist education).44–47 These variations are not merely logistical or curricular, but may also reflect deeper cultural and systemic dimensions woven into education.48,49 Consequently, rather than striving for universal definitions of QIs on timing and structure, it may be more appropriate to emphasise the need for flexibility and considered integration within the specific context of each learning environment, healthcare system, and structural setting; and evaluated accordingly within that framework. This suggests educators should prioritise pedagogical coherence and cultural responsiveness over arbitrary duration benchmarks.
Process indicators emerged as especially salient, reflecting how learning is facilitated rather than only what is taught. Multi-modal approaches, experiential learning, including simulation, peer learning, and reflective practice, are strongly supported across the literature as QIs, reinforcing earlier research that identified these strategies as pedagogically effective.50–52 Adequate supervision, timely feedback, and structured assessment are essential to successfully achieving clinical competencies through workplace-based learning. 53 While these process indicators may also reflect principles of general clinical education, they are particularly pertinent in palliative care, given the field’s reliance on person-centred communication, 54 ethical decision-making, 55 and interprofessional collaboration. 56
Palliative care’s inherently interprofessional nature requires that process indicators support collaborative, team-based learning. Our analysis confirms that education programmes benefit significantly from interprofessional approaches integrating realistic clinical scenarios. Importantly, a distinction should be made between programmes that merely involve multiple professional groups in parallel and those that adopt a truly interprofessional and interdisciplinary approach, characterised by active interaction, shared learning objectives, and collaborative problem-solving across disciplines. Accordingly, team-based learning, grounded in constructivist principles, positively influences self-study, learning ability, decision-making, and emotional intelligence.57,58 Further, interprofessional learning enhances teamwork, patient-centred care, communication, specialised knowledge, and shared professional values, crucial for coordinated, team-based care at the end of life. 59
One included review noted that, in the pandemic context, remote teaching is increasingly common. 28 However, pandemic-related adaptations were infrequently addressed across the included reviews. Further evidence may emerge as post-pandemic evaluations are published.
The early integration of palliative care in the training enhances students’ perceived preparedness for practice and positively influences their attitudes towards care. 60 These attitudes shape how healthcare professionals engage with seriously ill patients, navigate ethical dilemmas, and collaborate, making them essential competencies to be cultivated in training in this field. Our analysis further supports the positive impact of training interventions on attitudes and emotional competencies, particularly in fostering empathy and improving communication skills. However, we also identified mixed evidence on whether attitudinal improvements translate into behavioural changes in clinical practice. While some primary studies have begun to examine long-term outcomes and retention of learning,61,62 our review finds that these remain infrequently assessed overall. Evidence from related studies illustrate that longitudinal assessments are crucial; for example, residents using spaced practice (3–6 months) plus standard curriculum had superior knowledge and care outcomes at 10-month follow-up. 63 These findings point to a pressing need for longitudinal research on how educational formats with integrated or longitudinal designs can support lasting improvements essential to high-quality palliative care, while also considering systemic barriers that may limit their implementation. Retention of learning may also vary between specialist and non-specialist education, given differences in intensity, duration, and opportunities for supervised clinical application. Notably, included reviews rarely distinguish between specialist and non-specialist PC-education, leaving it unclear whether long-term outcomes and retention differ between these training contexts.
Importantly, patient- and family-centred outcomes remain underrepresented in PC-education research. In the broader palliative care literature, well-integrated approaches may improve quality of life and symptom outcomes, although effects are typically modest and vary by intervention design and context. 64 This highlights the need for education evaluations to assess whether, and which, training components translate into measurable patient and family benefit, while recognising that downstream effects may also be influenced by human, institutional, and systemic factors beyond the scope of teaching.
Finally, as emphasised above, palliative care and education delivery vary significantly across global settings. The underrepresentation of low- and middle-income countries (LMICs) in the reviewed studies reflects a well-recognised evidence gap. 65 While the need to tailor palliative care curricula to local cultural, social, and healthcare environments has been widely acknowledged,66,67 there remains limited empirical guidance on how to adapt educational approaches systematically. This lack of geographic diversity risks skewing the proposed QIs towards the educational priorities and resource realities of high-income contexts. As such, research is needed to explore how these indicators translate to, or need to be adapted for, LMIC contexts to ensure they are meaningful and feasible across different healthcare systems.
Strengths and limitations
This umbrella review synthesises existing evidence to propose a preliminary framework of potential QIs for PC-education, relevant across professional disciplines involved in its delivery. While this research was rigorously conducted per the JBI methodology, several limitations should be acknowledged. First, the included reviews vary considerably in their aims, terminology, methodology, and rigour. In this context, the slight overlap across reviews indicates limited bias due to redundancy but also a fragmented evidence base, which may reduce cumulative knowledge building and comparability across syntheses. Although reviews were published after 2014, some included primary studies conducted prior to the World Health Assembly resolution; therefore, the findings reflect post-2014 evidence syntheses that incorporate earlier research. Greater harmonisation of review characteristics would strengthen future work.
Second, although this review aimed to develop a preliminary typology of structural, process, and outcome indicators, many identified indicators lack clear operational definitions, standardised measurement tools, or established performance benchmarks. As a result, their immediate applicability for systematic evaluation and quality assurance in educational contexts remains limited. Finally, the geographical distribution of the included studies is skewed towards high-income countries. Consequently, the findings may not adequately reflect education’s contextual, cultural, and resource-related dimensions in LMICs. Future research should address this imbalance by incorporating more regionally diverse perspectives and educational frameworks.
Conclusion
This work offers a preliminary, evidence-informed framework for developing, implementing, and evaluating PC-education programmes. Policymakers and academic institutions can use these proposed QIs to define minimum standards, inform accreditation processes, and guide resource allocation. For educators, these QIs offer practical tools for curriculum development and self-assessment.
However, many QIs lacked essential components such as operational definitions, denominators, measurement methods, and performance benchmarks, limiting their current utility for systematic evaluation and comparison. Going forward, the proposed QIs may serve as a foundation for developing more consistent, measurable, and contextually relevant approaches to evaluating PC-education at the international level and across diverse care settings.
Supplemental Material
sj-docx-1-pcr-10.1177_26323524261440959 – Supplemental material for Towards quality indicators in palliative care education: An umbrella review
Supplemental material, sj-docx-1-pcr-10.1177_26323524261440959 for Towards quality indicators in palliative care education: An umbrella review by Anna J. Pedrosa, Ulla Näppä, Minna Hökkä, Jose Pereira, Piret Paal, Jan Kaňák, Gabija Jarašiūnaitė-Fedosejeva, Michael Connolly, John Ellershaw, Sabina Antoniu, Gulcan Bagcivan, Laura Ciria-Suarez, Stephen Mason and Tania Pastrana in Palliative Care and Social Practice
Supplemental Material
sj-docx-2-pcr-10.1177_26323524261440959 – Supplemental material for Towards quality indicators in palliative care education: An umbrella review
Supplemental material, sj-docx-2-pcr-10.1177_26323524261440959 for Towards quality indicators in palliative care education: An umbrella review by Anna J. Pedrosa, Ulla Näppä, Minna Hökkä, Jose Pereira, Piret Paal, Jan Kaňák, Gabija Jarašiūnaitė-Fedosejeva, Michael Connolly, John Ellershaw, Sabina Antoniu, Gulcan Bagcivan, Laura Ciria-Suarez, Stephen Mason and Tania Pastrana in Palliative Care and Social Practice
Footnotes
Acknowledgements
We would like to thank Elif Unsal Avdal, Nadiye Barış Eren, Zorica Cvetkovic, Stylianos Katsaragakis, Aneta Kochanowicz, Juho Lehto, Victoria Llanos, Nicoleta Mitrea, Andreas-Henri Mullalli, Cathy Payne, Goksen Polat, Olga Riklikiene, Sofiia Shunkina, Karoline Skedsmo, and Ela Yılmaz Coşkun for their valuable support in screening and data extraction, and Gina Pastrana for her support with data extraction, as well as for researching and organising the data on the primary studies and contributing to the GROOVE-based overlap analysis.
Ethical considerations
Ethical approval was not required for this umbrella review, as it synthesised data from previously published reviews and did not involve the collection or analysis of primary data from human participants.
Consent to participate
Informed consent was not required for this umbrella review, as it synthesised data from previously published reviews and did not involve the collection or analysis of primary data from human participants.
Author contributions
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the COST Action CA22127 – COllaboratively DEveloped culturallY Appropriate and inclusive Assessment tool for Palliative Care EDUcation (CODE-YAA@PC-EDU), funded by COST (European Cooperation in Science and Technology).
Declaration of conflicting interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: A.P., U.N., M.H., P.P., J.K., G.J.F., M.C., J.E., S.A., G.B., and T.P. declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. As guest editors of this special collection, J.P. and S.M. recused themselves from all aspects of the editorial handling and peer-review process for this manuscript.
Data availability statement
All relevant data are provided in the manuscript and Supplemental Materials.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
