From February to May 2022, nurses from PC pilot medical institutions in Hangzhou, China, were enrolled as the study population. Based on sample size estimation, 723 nurses were included in the quantitative study, and 8 participated in qualitative interviews to evaluate their current PC practice competence and to identify factors affecting it.

Inclusion criteria for the quantitative survey included: (1) employment at a PC pilot institution in Hangzhou, China; (2) at least one year of clinical work experience; and (3) possession of a valid nursing practice license. However, eligibility criteria for the qualitative survey included: (1) currently working as a dedicated PC nurse in hospitals, community settings, or institutions implementing the pilot program; (2) at least two years of palliative care experience; and (3) strong verbal communication skills and willingness to cooperate with the researchers.

This study adopted Miller’s Pyramid as the theoretical model [8]. Developed by American medical educator George Miller, this model offers a structured framework for evaluating clinical competence. This model conceptualizes the progression from knowledge to practice in four hierarchical levels: “knows” (acquisition of knowledge), “knows how” (understanding how to apply knowledge), “shows how” (demonstration of skills), and “does” (performance in clinical settings).

This study adopted both quantitative and qualitative survey approaches. In the former method, previously established questionnaires, including Palliative Care Self-Reported Practices Scale (PCPS), Palliative Care Difficulties Scale (PCDS), and Palliative Care Quiz for Nursing (PCQN), were used. In the latter method, an objective sampling approach was applied, where semi-structured interviews were conducted. These approaches aimed to understand the practical performance of nurses in PC and to determine how various factors limit or support their capabilities to deliver high-quality nursing care.

The sample size was calculated based on the standard requirement for multifactorial analysis, which suggests including 5 to 10 participants per questionnaire. To maximize the reliability and stability of the results, an upper limit of 10 participants per time was adopted. The survey included 26 items on general information, 18 items on practice competence, and two additional scales (difficulty and knowledge), each considered as a single item, resulting in a total of 46 items. Based on this, the initial required sample size was 460. To account for potential non-response and incomplete data, a 20% increase was applied, yielding a minimum required sample size of 552. Ultimately, 723 nurses participated in the survey, exceeding the estimated sample size and meeting the predetermined sample size criterion.

Data were collected using various questionnaire types as follows:

Internal consistency of the scale was assessed using Cronbach’s $\alpha$ for both the overall scale (PCPS) and each subdimension (End-of-Stage Care, Patient- and Family-Centered Care, Pain, Delirium, Dyspnea, and Communication). The overall Cronbach’s $\alpha$ was 0.969, while the values for the individual dimensions were 0.886, 0.937, 0.935, 0.864, 0.893, and 0.875, respectively, indicating good internal consistency.

The Chinese versions of the PCPS, PCDS, and PCQN scales, whose psychometric properties (including reliability and validity) have been validated in previous studies [12, 13, 14], were employed in this survey. In the present sample, all instruments demonstrated satisfactory internal consistency, confirming their reliability for use in this study.

Data were analyzed using SPSS 26.0 (IBM Corp., Armonk, NY, USA). Qualitative variables were presented as frequencies and percentages. Normality across data was evaluated using the Kolmogorov-Smirnov test, and homogeneity of variance was examined using the Levene test. Quantitative data that met the normality and homogeneity of variance were presented as mean $\pm$ standard deviation. Item-level, dimension, and total scores for PC practice ability, knowledge, and difficulty were calculated. Between-group comparisons were performed using independent sample t-tests or one-way analysis of variance (ANOVA) as appropriate.

Univariate analysis, Pearson correlation analysis, and multiple linear regression analysis were used to examine the associations among demographic variables, knowledge, difficulty, and practical ability scores. A p-value 0.05 was considered statistically significant. For the multiple linear regression analysis, all variables with significant associations in univariate analyses (p 0.05) were entered into the model using the enter method. Multicollinearity was evaluated using the Variance Inflation Factor (VIF) and Tolerance. Variables with a VIF greater than 10 or a Tolerance less than 0.1 were considered to indicate significant multicollinearity and were excluded from the model, prioritizing the removal of those with lower statistical significance.

Qualitative interview data were analyzed thematically using NVivo 12 (QSR International Pty Ltd., Doncaster, Victoria, Australia). Transcribed interviews were coded, categorized into nodes, and comparatively reviewed to identify relationships, which were then consolidated and integrated into thematic categories.

This study enrolled 723 nurses. The participants were predominantly female (93.91%), with the majority aged between 21–40 years (87.55%) and holding a bachelor’s degree or higher (82.02%). Furthermore, most participants were employed in tertiary hospitals (58.23%). Professional titles and years of work experience varied across the cohort, with senior nurses comprising the largest subgroup (48.69%, Table 1).

Thematic analysis of interview data yielded three primary themes and their respective sub-themes: (1) PC knowledge: comprising clarity regarding the service content and scope, existing knowledge gaps, lower public awareness, and nurses’ personal willingness and competency. (2) PC practices: encompassing skill deficiencies, challenges in symptom control, variability in institutional commitment to PC training, and the impact of family member behaviors. (3) Practice challenges: including the lack of standardized clinical protocols and limited access to analgesics, environmental barriers (e.g., bed shortages), limited governmental and insurance support, limited resources and time, challenges in transitioning from curative to palliative goals, and the lack of well-structured and systematic training.

Multivariate regression analysis identified gender, age, and willingness to participate in PC-related volunteer services as three independent factors influencing practice ability scores.

Age demonstrated the strongest impact on palliative care practice ability. Compared with nurses aged 21–30 years, those aged 31–40 scored 4.40 points higher (B = 4.40, t = 2.20, p = 0.028), and those aged 41 and above scored 6.93 points higher (B = 6.93, t = 2.01, p = 0.045). With increasing age, nurses usually gain more work experience and advance in professional title; accumulated clinical exposure and social experience are likely to contribute to enhanced competency. Given that PC is a highly specialized field that requires substantial practical experience and interpersonal understanding, it is possible to explain the positive correlation between age and practice ability. This suggests that administrators should promote PC training widely across pilot sites, enabling more nurses to be introduced to PC concepts and practice earlier in their careers and refine their abilities through ongoing clinical practice. Team composition should also consider age distribution to enhance the quality of PC services.

Second, female nurses scored significantly higher in practice ability than male nurses, which may be associated with greater patience, attentiveness, and detail orientation often observed among female practitioners. As the number of male nurses increases, they are playing increasingly important roles across specialties, particularly in operating rooms, emergency departments, intensive care units (ICUs), and psychiatric units, while remaining less represented in general wards. Evidence indicates that traditional gender stereotypes may lead some male nurses to perceive nursing as having limited career prospects, which can reduce professional commitment and contribute to variability in performance and competency. Therefore, administrators should provide targeted guidance and support for male nurses, implement evidence-based management approaches, enhance their professional identity in palliative care, and promote continuous improvement in service quality to support the development of these services.

Finally, personal willingness also plays an essential role in enhancing PC competency. Nurses who showed willingness to participate in palliative care scored significantly higher in practice ability than those who were unwilling. Such willingness is often driven by emotional motivation, job satisfaction and emotional engagement function as intrinsic drivers of performance. Given that palliative care involves long-term contact with patients at the end of life, interest in this field is typically shaped by the broader social and cultural environment rather than emerging spontaneously. These observations highlight the need for managers to respect nurses’ personal preferences, foster internal motivation, enhance subjective initiative, and strengthen public awareness and understanding of PC.

In this study, the nurses scored lowest in the community coordination domain, consistent with the results reported by An et al. [15]. This study involved nurses from four tertiary general hospitals in Beijing. The findings may reflect the relatively recent introduction of palliative care in China; for instance, Hangzhou’s pilot program was commenced just over two years ago, and managerial emphasis on community-based and interdepartmental coordination remains limited. The significant difficulties nurses reported in symptom management and communication align with findings from international studies. Many nurses have experienced considerable challenges in managing symptoms, communicating with multidisciplinary professionals, and obtaining support from palliative care specialists [16]. Multidisciplinary teamwork can help address the multidimensional needs of terminally ill patients. Evidence has shown that team-based interventions can alleviate symptoms, such as pain, dyspnea, weakness, fatigue, nausea, vomiting, and edema, indicating that effective communication and strong multidisciplinary support have positive effects in palliative care [17]. These findings suggest that in the future, coordinated measures should be implemented at both the nursing and governmental levels.

This study further revealed a negative association between palliative care knowledge and perceived difficulty, indicating that higher knowledge levels are linked with lower levels of difficulty in practice. Thus, knowledge and behavior are substantially correlated. These findings are consistent with those of Zhou et al. [18]. As one researcher has pointed out, “The problem facing the nursing profession is that theory does not guide practice, nor are theory and practice combined.” Bökberg et al. [19] further reported that nurses’ PC knowledge and perceived difficulties improved significantly after interventions. These results suggest that managers should strengthen education and training and better combine theory with clinical practice to effectively enhance nurses’ practice ability in PC.

Palliative care is a holistic model that addresses the physical, psychological, social, and spiritual needs of patients and their families. It involves multiple providers and is limited by funding, staffing, and ethical risks. To support its development, the government should emphasize the public welfare nature of PC, improve systemic mechanisms, increase financial investment, and refine medical insurance reimbursement policies to enhance institutional pilot programs.

This study found that high service costs and insufficient funding significantly hinder effective PC service delivery. Moreover, due to incomplete insurance policies, providers, particularly those delivering home-based care, are unable to charge for services beyond basic treatment costs. As workload has increased while reimbursement has decreased, most pilot institutions struggle to remain financially sustainable [20]. Furthermore, support from ancillary medical institutions is also critical. Thus, the authority for palliative care institutions to prescribe necessary anesthetic and analgesic drugs should be appropriately expanded at all levels to enable the implementation of home-based and community day-care services.

Evidence shows that although palliative care has made some advances in recent years, overall social participation is generally low, and patients and their families often lack an accurate understanding of its value [21]. Most family members equate PC to giving up treatment and therefore favor active interventions, often neglecting patients’ quality of life and resulting in unnecessary suffering [22].

In the current medical environment, patients and families tend to place greater trust in doctors than in nurses. Therefore, enhancing interprofessional collaboration and communication is essential in PC practice. Enhanced collaboration can help them better understand their illness, accept the clinical reality, and make informed decisions aligned with their values and preferences [22].

Interviewees reported opportunities for annual internal and external training. They further stated that knowledge acquisition is fragmented, primarily sourced from clinical case summaries and departmental lectures, and lacks systematic professional instruction. Consequently, managers should enhance nursing education by developing structured curricula that strengthen application and establish a comprehensive palliative care knowledge system. Incorporating simulation-based and experiential teaching could help develop a new educational model for palliative care training, enriching current strategies [23, 24].

Research indicates that a lack of palliative care knowledge leads to unfamiliarity with work processes and limited practical experience, ultimately decreasing clinical practical abilities [25]. Optimizing education and training methods is therefore crucial to enhancing nurses’ PC knowledge and preparing them to provide comprehensive and professional support to end-of-life patients and their families.

Consistent with previous evidence [26], many nurses experienced considerable difficulties in symptom control, communication with multidisciplinary professionals, and obtaining support from palliative care specialists. It has been reported that multidisciplinary teamwork can address the complex, multidimensional needs of terminally ill patients, with team-based interventions in symptoms, such as pain, dyspnea, weakness, fatigue, nausea, vomiting, and edema, thereby reflecting the beneficial impact of multidisciplinary support in palliative care [27]. Enhancing interprofessional collaboration and enabling access to expert input should therefore be a key managerial priority.

In many high-income countries, palliative care has been integrated into healthcare systems, supported by well-established policies, standardized clinical practices, and robust multidisciplinary collaboration. For instance, in the United Kingdom and Australia, palliative care is supported by comprehensive national policies, adequate funding, and extensive community involvement. These systems emphasize the synergy between medical institutions and communities, ensuring seamless transitions for patients across different levels of care. Additionally, palliative care education and training are systematically incorporated into nursing and medical curricula, thereby enhancing the professional competencies of healthcare providers.

In contrast, palliative care in China remains at a nascent stage of development and is characterized by uneven progress and significant urban-rural variations. The policy support is fragmented, funding is insufficient, and PC is not yet fully integrated into mainstream medical education. Moreover, public awareness and acceptance of palliative care remain low, often hindered by cultural biases and misconceptions about death and end-of-life care. Learning from international experience, China should prioritize the formulation of uniform national policies, increase financial investment, and promote public education to foster a more supportive environment for palliative care development.

Funding and human resources (staffing) limitations restricted this study to the Hangzhou region, which may affect the generalizability and representativeness of the findings. Future studies should increase the sample size and extend geographical coverage to comprehensively capture the broader landscape of PC development. Expanding the scope in this way would offer a strong empirical basis for policy formulation and for designing educational and training programs.