Abstract
Background
China’s aging population will escalate palliative care (PC) needs in the next decade. Scalable and culturally-adapted training is necessary to equip practicing clinicians with essential PC skills. The objective of this study is to evaluate a culturally-adapted basic PC training course in mainland China.
Methods
A total of 29 practicing physicians from Zhejiang Province, China, were selected to participate in an in-person training program that spanned six days. We analyzed pre- and post-course quantitative surveys on knowledge, self-efficacy, and behavior using descriptive statistics. We also thematically analyzed post-course semi-structured participant interviews.
Results
The majority of participants were aged 41–50 (51.7%), trained in internal medicine (55.2%), worked at tertiary medical centers (93.1%), and did not have clinical PC experience (65.5%). After the course, participants’ PC knowledge (p < 0.01) and self-efficacy (p < 0.01) increased, especially in the domains of PC philosophy and physical symptom management. Although statistically significant, changes in participants’ self-perceived behaviors were less profound. Thematic analysis of the participant interviews revealed concordant themes, including recognition of cohesiveness between PC principles with traditional Chinese philosophy, and acquisition of actionable clinical knowledge. Key points that expanded beyond the meta-inferences were: 1) emotional resonance with the teaching team is necessary to create a transformational learning experience; and 2) a longitudinal, relationship-centered mentorship process may aid in participants’ implementation of PC skills.
Conclusion
The results of this study indicate that our culturally adapted PC training can increase practicing physicians’ PC knowledge and self-efficacy. Scalable basic PC training should preserve and facilitate emotional resonance between participants and instructors to ensure uptake of PC principles. Efforts to understand and overcome the implementation challenges of new PC champions should also be prioritized.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12909-025-07665-y.
Keywords: Palliative care, Education, Middle-income country
Introduction
From 2015 to 2021, China’s ranking on the international Quality of Death and Dying index climbed from the 71 st to the 53rd place [1]. Despite this progress, patients in mainland China continue to experience unmet end-of-life care needs [2, 3] and catastrophic health expenditure at the end-of-life [4]. One major driver of the current state of care is the lack of “integration of palliative care (PC) into the broader health system” [1, 5]. In Chinese medical settings, PC implementation faces a multitude of barriers ranging from limited awareness of PC [5, 6], to evolving attitudes toward death and truth telling [7], to family-based decision-making [8]. Therefore, a culturally-rooted, effective, and scalable training program is urgently needed to equip the healthcare workforce for addressing PC needs [9].
Contextualized PC training based on the Massachusetts General Hospital (MGH) global PC program has been successfully carried out in Asian and other low and middle-income settings [10–13]. In 2022, the Palliative Medicine Center at Peking Union Medical College Hospital (PUMCH) and the Program in Global Palliative Care at MGH/Harvard Medical School (HMS) jointly developed a culturally-adapted basic PC training program for practicing physicians in mainland China [14]. While preliminary findings revealed that the course was well received [14], a robust evaluation of the course is needed to demonstrate participants’ experiences during and changes in knowledge, self-efficacy, and behavior following the training and to inform the content and delivery of future training [15].
In our study we will employ a convergent mixed-method approach to characterize the impact of PC training for practicing physicians in mainland China. Because no culturally and linguistically validated tools exist, we developed Chinese language tools informed by prior literature and the New World Kirkpatrick Model [16–21]. Whereas the quantitative data provides insights on the effect of PC training on knowledge acquisition, self-efficacy, and behavior change, the qualitative data provides complementary insights on the uptake and implementation of the training. By integrating quantitative with qualitative data, we derive insights for scaling and implementing culturally-adapted PC training for the greater China region and other middle-income settings.
Methods
Study design
We conducted a convergent parallel mixed-methods evaluation of a culturally-adapted PC course for practicing physicians in mainland China [22]. A convergent parallel design is characterized by independent collection and analysis of quantitative data and qualitative data, followed by an assessment of the areas of concordance, expansion, and contradiction between the quantitative results and the qualitative findings [23] (Fig. 1). We consecutively recruited physician participants who attended the weeklong basic PC training in Zhejiang, China [14]. We aimed to recruit 30 participants to allow adequate assessment of effect sizes for a pilot study [24] and to allow thematic saturation in the qualitative analysis [25]. This study received approval by the Institutional Review Board of Peking Union Medical College Hospital (I-23PJ925).
Fig. 1.
Flow diagram of the study design
Quantitative phase
Data collection
The quantitative evaluation included pre-course and post-course evaluation of participants’ PC knowledge, self-efficacy, and behavior informed by the New World Kirkpatrick Model (NWKM) [26] and existing literature [12, 14, 16, 19, 27]. On the day immediately before the course, we collected participants’ informed consent, socio-demographics, clinical background, and self-assessments of PC knowledge and behavior via electronic surveys. We re-administered the knowledge and behavior surveys on the final day of the course. We also administered a self-efficacy survey on the final day using a single assessment retrospective pre-post design [12].
The knowledge assessment comprised 20 true or false questions from the Palliative Care Knowledge Test [19] and 3 multiple-choice questions [17] covering four major domains: Principles of PC, symptom management, psychosocial, and communication (Supplemental Table 1). The self-efficacy assessment comprised 12 items that measured clinicians’ proficiency across 5 domains: Principles of PC, symptom management, psychosocial and ethical, and communication. The behavior assessment comprised 15 items that measured PC activities across four domains: Principles of PC, symptom management, psychosocial and ethical, and communication. Participants provided ratings across a seven-point Likert scale for both self-efficacy (1-no skill to 5-competent to 7-expert, can teach others) and behavior (ranging from 1-never to 7-multiple times a day) assessments.
Data analysis
We used descriptive statistics to describe participant demographics and clinical background (Table 1). Paired t-tests were used to compare the participants'total scores on the knowledge assessment before and after the course. Fisher’s exact tests were used to compare the proportion of correct answers for individual knowledge assessment items before and after training. Since self-efficacy and behavior ratings were ordinal and not normally distributed, we used the nonparametric Wilcoxon signed-rank test to compare participants'ratings before and after the training. Differences were considered statistically significant if the p-value was less than 0.05. Statistical analysis was conducted using SPSS 26.0 software.
Table 1.
Participant demographic and training background (n = 29)
| Participants Characteristics | |
|---|---|
| Age [years, (n, %)] | |
| 20–30 y | 1 (3.5%) |
| 31–40 y | 5 (17.2%) |
| 41–50 y | 15 (51.7%) |
| 50+ y | 8 (27.6%) |
| Female Gender (n, %) | 21 (72.4%) |
| Degree (n, %) | |
| Bachelor degree | 18 (62.1%) |
| Master degree or above | 11 (37.9%) |
| Background of clinical training (n, %) | |
| Internal medicine | 16 (55.2%) |
| Oncology | 12 (41.4%) |
| Geriatrics | 13 (44.8%) |
| Surgery | 1 (3.5%) |
| Hospice and Palliative care | 2 (6.9%) |
| Infectious Disease | 1 (3.5%) |
| Other | 2 (6.9%) |
| Level of hospital (n, %) | |
| Medical center/tertiary hospital | 27 (93.1%) |
| Regional hospital/secondary hospital | 1 (3.5%) |
| Community health center | 1 (3.5%) |
| Prescribed oral opioids in the past year (n, %) | 25 (86.2%) |
| Prescribed subcutaneous or intravenous opioids in the past year (n, %) | 21 (72.4%) |
| Prescribed opioids for dyspnea in the past year (n, %) | 13 (44.8%) |
| Prior Palliative Care clinical experience (n, %) | 10 (34.5%) |
| Format of prior palliative care training (n, %) | |
| Short Conferences/Courses | 25 (86.2%) |
| Other | 4 (13.8%) |
| Number of prior training events completed [median, (IQR)] | 2 [1, 3] |
| Length of prior palliative care training [median, (IQR)] | 4 [1, 8] |
IQR interquartile range
Qualitative phase
Data collection
We consecutively recruited participants who completed the pre- and post-course surveys for semi-structured post-course interviews to understand their lived experiences and perceptions of the training. The NWKM and a critical realist epistemology informed the development of a semi-structured interview guide [14] (Supplemental Table 2). The guide was iteratively revised with an interdisciplinary team involving qualitative researchers (RP, ZJ), medical anthropologists (FL), clinician educators (IS, BF) and education scientists (TL). One author (FL) trained research volunteers to conduct the interviews via phone or web-conference. Interviews were recorded and transcribed by research volunteers.
Data analysis
We analyzed the Mandarin interview transcripts using directed thematic analysis [28, 29]. The first step involved two authors (HL and TD) familiarizing themselves with the primary data through reading and rereading. The second step involved developing a codebook using deductive (i.e. NWKM and our prior framework) and inductive (i.e. open-coding 18% of the transcripts) approaches. The third step involved two authors (HL and TD) independently coding the transcripts using the finalized codebook (Supplemental Table 3). Any discrepancies were resolved in consultation with a third author (ZJ). The final step involved developing descriptive themes through constant comparison. Qualitative data were analyzed in NVivo 12.
We took several additional steps to ensure a rigorous analysis [29]. Two authors (HL and ZJ) held weekly peer debriefing sessions, the primary author recoded the primary data twice, the primary author recorded coding memos, and research meeting minutes. Finally, we documented the study population to establish transferability.
Mixed methods analysis
In the mixed-methods analysis phase, the quantitative and qualitative data are tabulated and compared to examine connections [23] (Tables 2, 3, and 4). Through this integrative process, we identified qualitative themes and quantitative trends that converged, expanded upon, or refuted each other. This comparative analysis allowed authors to draw meta-inferences that explicated the phenomenon of receiving basic PC training among practicing physicians in mainland China.
Table 2.
Joint display of knowledge assessment and exemplary quotes
| Quantitative findings (n = 29) Participants with correct answers n (%) |
Qualitative findings (n = 26) | ||||
|---|---|---|---|---|---|
| Domain | Pre- | Post- | P value | Quotes | Meta-inferences |
| Total score | 62.2 ± 11.6 | 73.9 ± 8.9 | < 0.01 |
✓ I was particularly impressed by what Professor He Renfu taught us about Chinese death culture on the first day. I felt deeply touched by how China has considered the serious topic of death from ancient times to the present… (Participant 10) ✓ Palliative care is actually a concept which can be applied to every patient everywhere, not only in a certain ward, not only to the end-stage patients. (Participant 18) ✓ This is where I improved the most, where I gained the most. When I feel I can’t do anything, I can alleviate suffering. I think it’s also what patients really want and need. (Participant 8) ✓ Then, for example, such as difficult breath, patients with end-stage COPD and shortness of breath can be treated with morphine. This left a strong impression on me. I recall several cases in my own practice. (Participant 13) ✓ I learned that, for example, if our terminal patient has intestinal obstruction and unstable blood pressure, we can use steroids to relieve their symptoms (Participant 21) |
Expansion - Participants recognized concordance between palliative care philosophy and Chinese cultural values - 学员认识到缓和医疗理念与中华文化价值观的一致性。 Concordance - Participants gained in their management of common palliative care symptoms and syndromes - 学员收获了缓和医疗中常见症状管理的相关知识。 |
| Principles of Palliative Care | |||||
| Item 01 | 17 (58.6%) | 25 (86.2%) | 0.04 | ||
| Item 02 | 24 (82.8%) | 27 (93.1%) | 0.42 | ||
| Symptom Management | |||||
| Item 03 | 19 (65.5%) | 21 (72.4%) | 0.78 | ||
| Item 04 | 23 (79.3%) | 25 (86.2%) | 0.73 | ||
| Item 05 | 20 (69.0%) | 21 (72.4%) | > 0.99 | ||
| Item 06 | 12 (41.4%) | 13 (44.8%) | > 0.99 | ||
| Item 07 | 15 (51.7%) | 20 (69.0%) | 0.28 | ||
| Item 08 | 26 (89.7%) | 28 (96.6%) | 0.61 | ||
| Item 09 | 19 (65.5%) | 29 (100.0%) | < 0.01 | ||
| Item 10 | 22 (75.9%) | 28 (96.6%) | 0.05 | ||
| Item 11 | 5 (17.2%) | 14 (48.3%) | 0.02 | ||
| Item 12 | 27 (93.1%) | 28 (96.6%) | > 0.99 | ||
| Item 17 | 12 (41.4%) | 25 (86.2%) | < 0.01 | ||
| Item 18 | 23 (79.3%) | 25 (86.2%) | 0.73 | ||
| Item 19 | 25 (86.2%) | 28 (96.6%) | 0.35 | ||
| Item 20 | 17 (58.6%) | 23 (79.3%) | 0.16 | ||
| Psychosocial | |||||
| Item 13 | 11 (37.9%) | 14 (48.3%) | 0.59 | ||
| Item 14 | 18 (62.1%) | 17 (58.6%) | > 0.99 | ||
| Item 15 | 29 (100.0%) | 29 (100.0%) | > 0.99 | ||
| Item 16 | 14 (48.3%) | 25 (86.2%) | < 0.01 | ||
| Communication | |||||
| Item 21 | 5 (17.2%) | 0 (0%) | 0.05 | ||
| Item 22 | 9 (31.0%) | 11 (37.9%) | 0.78 | ||
| Item 23 | 23 (79.3%) | 28 (96.6%) | 0.10 | ||
Table 3.
Joint display of self-efficacy and exemplary quotes
| Quantitative findings (n = 29) | Qualitative findings (n = 26) | ||||
|---|---|---|---|---|---|
| Domains | Pre- | Post- | P value | Quotes | Meta-inferences |
| Principles of Palliative Care (PC) | |||||
| Recognize benefits of PC | 2.00 (2.00,4.00) | 5.00 (3.50,5.00) | < 0.01 |
✓ The most profound experience is that the course elevated my approach to advanced cancer care, from focusing on symptom management to focusing on human dignity. (Participant 6) ✓ This comfort that you feel, when a physician speaks to you softly, gazes at you warmly, with good body language, and listens to you intently, this is (palliative) care. (Participant 9) ✓ I think compared with my prior superficial understanding, I believe palliative care to be more proactive now. I used to think that it’s to dissuade patients from ineffective and futile treatments. (Participant 17) |
Expansion - Palliative care actively addresses suffering holistically through effective communication and attention to dignity - 学员认识到缓和医疗通过有效的沟通和对尊严的关注, 积极全面地解决痛苦 |
| Palliative care promotion | 2.00 (1.00,3.00) | 4.00 (3.00,5.00) | < 0.01 | ||
| Symptom Management | |||||
| Pain Management | 4.00 (3.00,4.50) | 5.00 (4.00,5.50) | < 0.01 |
✓ In terms of interventions, standardizing management of pain, nausea, anticipating side effects of specific treatments, etc. I think it is actionable, these narrower tasks. We can work on these things. (Participant 9) ✓ Perhaps I think this course should be the most comprehensive palliative care course in China. …The biggest gain may be the guidance in my approach to specific palliative care problems. This gives me confidence in caring for similar cases and patients in the future. (Participant 12) |
Concordance - Participants feel confident implementing new knowledge in symptom management - 学员自信能够应用新知识进行症状管理 |
| Dyspnea Management | 3.00 (2.50,4.00) | 5.00 (4.00,5.50) | < 0.01 | ||
| Nausea Management | 4.00 (3.00,4.00) | 5.00 (4.00,5.00) | < 0.01 | ||
| Psychosocial and Ethical | |||||
| Psychosocial Distress Management | 3.00 (2.00,4.00) | 5.00 (4.00,5.00) | < 0.01 |
✓ I feel that if I am on duty, I will be very confident. I can help the family understand that sedating medications isn’t meant to speed up death, nor is it to prolong things. It’s to help him/her rest, to spare him/her from suffering and discomfort. (Participant 10) ✓ One of the new insights is to put patient first and not family members first, right? Because in Shenzhen, the patient’s wishes, the patient’s thoughts, and the patient’s arrangements for the things after his death, we must respect the patient's ideas. (Participant 6) ✓ At this time, I will start to reflect on some previous cases. I now think that if we can’t extend a patient’s survival… and it will also increase his pain, it feels unnecessary. (Participant 17) ✓ I feel that if there’s promotion of advance care planning, that would be good. It will help them achieve their wishes. If I can help them, I should do my best. (Participant 16) |
Concordance - Participants appreciate the importance of considering psychological and spiritual dimensions of suffering - 学员认识到考虑心理、精神层面痛苦的重要性 - Participants gained awareness of ethical dilemmas surrounding autonomy, non-maleficence, principle of double effect - 学员意识到了围绕自主原则、不伤害原则和双重效应原则的伦理困境 |
| Spiritual Distress Management | 2.00 (1.50,4.00) | 4.00 (3.00,5.00) | < 0.01 | ||
| Recognizing ethical dilemmas | 3.00 (2.00,4.00) | 4.00 (3.00,5.00) | < 0.01 | ||
| Resolving ethical challenges | 3.00 (2.00,3.50) | 4.00 (3.00,5.00) | < 0.01 | ||
| Communication | |||||
| Communication about Prognosis | 4.00 (2.50,4.00) | 5.00 (4.00,5.00) | < 0.01 |
✓ Previously, I felt the need to deliver bad news, directly and transparently to benefit the patient… It was Dr. Ning that taught me to consider whether there’s discordance in illness understanding. If I deliver the content in such detail, the actions may in fact be harmful to him/her. Maybe we don’t need to always unload it all onto him? (Participant 1) ✓ During the consultation process, they (teachers) say that they can spend a lot of time with a patient, chat in slow and measured pace, to listen to the patient… In the process of listening, and subsequent line of questioning, caring, they’re able to elicit patient’s concerns. (Participant 20) ✓ But how do we comfort the patient and the family members? I think it is possible to obtain the patient’s hopes and wishes through family members. (Participant 15) |
Concordance - Participant gained critical insights into pacing prognostic disclosure and listening for distressing emotions - 学员对逐步交代预后与倾听痛苦情绪方面有了关键见解。 |
| Responding to emotions | 3.00 (2.00,4.00) | 5.00 (4.00,5.00) | < 0.01 | ||
| Managing ambivalence | 3.00 (2.00,4.00) | 5.00 (4.00,5.00) | < 0.01 | ||
Table 4.
Joint display of Behavioral self-assessment and exemplary quotes
| Quantitative findings (n = 29) (7-point Likert scale) |
Qualitative findings (n = 26) | ||||
|---|---|---|---|---|---|
| Domains | Pre- | Post- | P value | Quotes | Meta-inferences |
| Principles of Palliative Care (PC) | |||||
| Talking with PC leaders about PC development | 2.00 (2.00,3.00) | 3.00 (2.00,4.00) | 0.02 |
✓ I think in the future, institutional leaders should participate in PC training, like the president of our Zhejiang Hospital, this will greatly aid our efforts. (Participant 11) ✓ If the department team mature, I may gradually spread it to the whole hospital to attract these individuals who are interested in hospice palliative care. (Participant 24) ✓ I hope to spread this concept, especially to the community and primary level of the Taizhou Healthcare system. (Participant 16) ✓ We wonder if the teaching hospitals can make some brochures to community hospitals… It may be easier to promote PC in this way. (Participant 25) ✓ I think we do not lack of enthusiasm or motivation but the support of protective policy for both medical teams and patients. (Participant 23) |
Expansion - Although participants were motivated to promote PC, feelings of inadequacy lingers - 尽管学员们被鼓励在各自的机构促进缓和医疗的发展, 但力不从心感觉仍然存在 |
| Talking with non-PC leaders about PC development | 2.00 (2.00,3.00) | 4.00 (2.50,5.00) | < 0.01 | ||
| Symptom Management | |||||
| Pain | 3.00 (2.00,4.00) | 3.00 (2.00,4.00) | 0.96 |
✓ Because some of the medicines involved in palliative care are a risk, after communication, we’re able to address his dyspnea with morphine. (Participant 14) ✓ When I came back, I taught my colleagues that for end-of-life patients who are suffering, we really should be daring in our efforts to reduce suffering. (Participant 3) ✓ I also spend time training younger doctors during rounds. (Participant 14) ✓ I’m going to make all these handouts into slides and teach the doctors in our department…. I also ask all the medical staff in our department to master the contents and methods of palliative care. (Participant 24) |
Expansion - Participants see more opportunities to teach than to manage burdensome symptoms - 学员看见了更多的教学复杂症状管理的机会而不是处理复杂症状本身 Expansion - Participants experienced the fulfilment of recentering care on the experiences of patients and families - 学员们在患者和家属身上体验到了以患者为中心照护的成功实施 |
| Other Symptoms | 3.00 (2.00,4.00) | 3.00 (2.00,5.00) | 0.09 | ||
| Teaching Pain Management | 3.00 (2.00,5.00) | 4.00 (3.00,5.00) | 0.02 | ||
| Teaching other symptoms | 3.00 (2.00,4.00) | 4.00 (3.00,5.50) | < 0.01 | ||
| Psychosocial and Ethical | |||||
| Psychosocial Issues | 2.00 (1.00,3.00) | 3.00 (2.00,4.00) | 0.60 |
✓ Through learning, I became more willing to help patients achieve unfulfilled wishes. I want to find a social worker or a volunteer to help this individual, to synthesize his memories and put them into words. (Participant 21) ✓ My focus will be on the family experience. There are some aspects of death education I can do through the family, and they can support the patient. (Participant 7) ✓ After training, I’ve encountered several patients, where we didn’t need to intubate and allowed them to pass away peacefully. (Participant 2) ✓ One patient had a nasogastric tube, and he asked the doctor to remove it. If before, I would certainly refuse him… But that day when he said, I immediately agreed with him… Then he is very comfortable, very grateful… I satisfied him at the last stage… this is a big change from this palliative care training. (Participant 18) |
|
| Ethical Issues | 2.00 (1.00,3.00) | 3.00 (2.00,4.50) | < 0.01 | ||
| Teaching Psychosocial | 3.00 (2.00,4.00) | 4.00 (3.00,5.00) | < 0.01 | ||
| Teaching Ethics | 3.00 (2.00,3.50) | 4.00 (3.00,5.00) | < 0.01 | ||
| Communication | |||||
| Illness understanding | 3.00 (2.00,5.00) | 4.00 (3.00,5.00) | 0.06 |
✓ I may be able to identify the patient’s emotional changes regarding the test results. (Participant 4) ✓ During training, I was quite moved. I had tears streaming down my face. But when I return to clinical work, I felt my empathy wane. (Participant 3) ✓ To talk about these issues with patients and family, we may need to consider practices that are easy to understand and aligned with local cultural practices. (Participant 17) ✓ On the first day, we had role play simulation. The teachers from PUMCH let us to play different roles in the medical case and experience the mood of different identities, I think in that kind of practice, more empathy can be achieved, and communication can be strengthened. (Participant 11) ✓ Teachers are very enthusiastic and caring. What impressed me most was these teachers, I felt truly respected. I felt this. This impressed me. (Participant 5) ✓ What impressed me deeply was the emotional expression of participants as they shared their stories. (Participant 26) |
Expansion - Communication skills may require adaptation to local culture and norms - 沟通技能可能需要适应当地文化与规范 - Participants’ were moved to emulate the teaching team’s compassionate approach - 学员因教师团队的专业和热情而感动, 效仿了相同的教学方法。 |
| Delivering prognostic information | 3.00 (2.00,4.00) | 4.00 (2.50,5.00) | 0.08 | ||
| Treatment preferences | 3.00 (2.50,4.00) | 4.00 (2.50,5.00) | 0.03 | ||
| Goals of care | 3.00 (2.00,4.00) | 3.00 (2.50,5.00) | 0.02 | ||
| Teaching communication | 4.00 (2.00,4.50) | 5.00 (3.00,5.50) | < 0.01 | ||
Results
Participant characteristics
All 30 physicians who attended the training course were recruited into the study. We excluded one physician from analysis due to her incomplete post-course questionnaire. Of the 29 participating physicians, the majority were female (n = 21,72.4%), between 41 and 50 years old (n = 15, 51.7%), trained in Internal Medicine (n = 16, 55.2%), Oncology (n = 12, 41.4%), or Geriatrics (n = 13, 44.8%), and worked in a tertiary medical center (n = 27, 93.1%). Most physicians had prescribed oral (n = 25, 86.2%) and intravenous or subcutaneous (n = 21, 72.4%) opioids in the past year. Only some physicians prescribed opioids for dyspnea (n = 13, 44.8%) and included PC as part of their clinical practice (n = 10, 34.5%). All physicians had prior PC training, with most training taking the form of short conferences or courses (n = 25, 86.2%) (Table 1).
Quantitative changes in self ratings of knowledge, self-efficacy, and behavior
Comparing between the pre- and post- surveys, participant rating of self-efficacy had the highest proportion of items achieving statistically significant differences (12/12, 100%), followed by self-rating of behavior change (10/15, 66.7%) and knowledge test (5/23, 21.2%). Across the surveys, the subdomain “Principles of PC” had the highest proportion of items achieving statistically significant differences (5/6, 83.3%), followed by “Psychosocial, spiritual, or ethical” (8/12, 66.7%), then “Communication” (6/11, 54.5%), and finally “Symptom management” (8/21, 38.1%).
Overall, participant knowledge scores increased post-course (11.7%, 95% CI: 9.18–17.51) with participants either continuing to answer the question correctly or showing improvements in all but 1 of the 23 knowledge items. The items with the largest improvement were Item 1 in “Principles of PC” (58.6% to 86.2%) and Item 9 in “symptom management” (65.5% to 100%). Item 21 in “communication” demonstrated a decrease in proportion of correct responses (17.2% to 0%) (Table 1).
Although all 12 items in the self-efficacy survey experienced a statistically significant improvement, the effect sizes differ across subdomains. Whereas median ratings of all items in “symptom management” and “communication” improved to “5 or competent,” items in “Principles of PC” (i.e. palliative care promotion) and “Psychosocial, spiritual and ethics” domains (i.e. spiritual distress, recognizing ethical dilemmas, resolving ethical challenges) did not (Table 1).
There was a statistically significant increase in self-reported PC behaviors in 10 out of 15 items post-course. Pre-course, the only item that achieved a median frequency rating of “once per week” (i.e. self-rating of 4) was in the domain of communication education. Post-course, self-ratings of 9 behavioral items improved to at least “once per week.” Among the 9 items that improved, most belonged to education (n = 5), followed by communication (n = 3), followed by leadership (n = 1) (Table 1).
Qualitative findings
We recruited 26 (89.6%) physicians for post-course interviews (Supplemental Table 4). All interviews were completed online by videoconference within one month of course completion. The duration of the interviews ranged between 20 to 70 min.
In accordance with quantitative changes in PC knowledge, participants identified that “this is where I improved the most, where I gained most.” Many participants were shocked to learn that “shortness of breath can be treated with morphine” as they had historically avoided the medication for its effects on respiratory depression. Other participants identified the use of steroids in intestinal obstruction and protocols for palliative sedation as important takeaways. Learning more about PC led to the recognition of cohesiveness between PC principles and traditional Chinese philosophy. For example, participants expressed that the principles of palliative care “can be applied to every patient everywhere” and that it shares roots with traditional views on a “Chinese death.”
In accordance with the ratings of self-efficacy, participants shared that the training “gave me confidence in the future” and “elevated my approach to advanced cancer care.” These self-identified improvements in the provision of PC were attributed to several different factors. The new PC knowledge improved participants’ “confidence in caring for similar cases.” Furthermore, participants’ stance toward comprehensive assessment of serious illness changed from “dissuading patients from ineffective and futile treatments” to “focusing on symptom management and human dignity” and helping patients as a “whole person.” Participants also reported gaining skills in pacing their delivery of prognostic information by actively inviting emotional expression (e.g. frequent pauses, warm gaze, active listening, and inviting body posture). Finally, participants identified developing skills in managing ethical dilemmas by “starting from the patient’s perspective,” considering the principle of double effect, and striving for shared decision-making.
Expanding beyond the ratings of self-behavior, participants described challenges in implementing their new knowledge and skills at their respective institutions. First, participants were moved to propagate PC education, inspired by the teaching team’s professional and empathic approach. Participants described sprinkling PC knowledge through their daily clinical routine and building “a platform and a team.” Second, participants noted that the implementation of their new communication skills might require “alignment with local cultural practices to make communication more effective.” Finally, participants described moments of powerlessness when their new PC skills were inhibited by interpersonal and institutional barriers. Participants emphasized the need for continued support and hoped for sustained contact with the teaching team to overcome implementation barriers post-course.
Meta-inferences
As depicted in Tables 2 and 3, quantitative findings were concordant with and expanded upon one another. By evaluating these meta-inferences and the associated concordance and expansion between qualitative and quantitative findings, we synthesized the following take-away points:
Emotional resonance with the teaching team is necessary to create a transformational learning experience. During the intensive course, the teaching team built an immersive teaching environment that dually emphasized PC knowledge delivery and emotional engagement with participants. The teaching team’s commitment to and consideration of learners’ needs inspired respect among participants. In turn, participants were better able to internalize PC principles and a compassionate approach.
A longitudinal, relationship-centered process may aid in participants’ implementation of PC skills. Participants anticipated challenges to implementing and advocating for PC upon returning to their home institutions. Participants perceived the importance of maintaining connection with the teaching team to overcome implementation challenges and to hone clinical skills. Participants also proposed case discussions and continuing professional development opportunities as important tools to foster.
Discussion
Effective and scalable PC training for physicians is needed to meet the serious illness-related suffering among patients and families in mainland China. In our theory-driven, mixed-methods evaluation of a culturally adapted PC training program, we found the training to be relevant and effective for a diverse group of experienced physicians. Through integrating quantitative survey data with qualitative experiential data, we identified two key elements defining successful PC training in mainland China: 1) Emotional resonance with the teaching team is necessary to create a transformational learning experience and 2) A longitudinal, relationship-centered process may aid in participants’ implementation of PC skills. Future scalable basic PC training should employ caring pedagogy which refers to strategies that foster care and nurturance in teaching and enable longitudinal investment in learners to ensure uptake and implementation of palliative care skills.
Our quantitative findings demonstrating improvement in PC knowledge and self-efficacy are consistent with prior PC training in low and middle-income countries [12, 13]. Our study participants had similar baseline knowledge scores as physicians from other middle income countries [30], and the magnitude of effect on palliative care knowledge was similar to prior studies using the Palliative care knowledge test (PCKT) [31]. While self-efficacy scores across all domains demonstrated statistically significant improvements post-training, the median scores in the psychosocial and ethics domains remained below “competent” [12]. This may be due to inadequate didactic content or pedagogical methods to address clinically relevant ethical and social dilemmas. Future training should consider crafting training content catering to relational autonomy and family-based care planning in this Asian context [32–34]. Furthermore, a combination of theoretical sessions, empirical case analysis, and observed consultations are needed to bolster trainees’ skills in these domains [31].
Our theory-driven mixed-methods evaluation builds on the literature by triangulating critical elements defining impactful PC training [11, 35]. Specifically, our meta-inferences highlight the emotional and pedagogical considerations for PC training. By investing in the teacher-learner relationship, participants are better able to build an emotional connection with PC and absorb its principles. These findings underscore the importance of demonstrating a compassionate stance toward PC learners [36–38], which aligns with the praxis of caring pedagogy. Caring pedagogy views medical education as a relational process unfolding with nurturance as its core ideal [39]. Future PC training should consider strategies that foster a sense of nurturing to facilitate uptake of PC principles (e.g. learning rules, suitable sized groups, more dialogue).
Our quantitative findings on clinician behavior revealed mixed findings across symptom management and communication domains. Outside the educational and communication domains, very few items improved beyond a median rating of 4 or “once a week.” This may be due to anticipated barriers to implementing PC skills when trainees return to their home institutions. Consistent with prior studies, our course participants also identified lack of awareness of PC among local providers, challenges advocating for PC with institutional leadership, and a lack of a trained interprofessional team as key barriers to implanting new PC skills [10]. To meet the needs of future PC trainees, implementation research is urgently needed to identify test strategies to address contextualized barriers to PC implementation.
We identified several limitations regarding our research. Firstly, the lack of culturally adapted and validated survey/interview tools may result in misclassification bias. Future work is needed to prospectively validate quantitative tools to evaluate PC training. Secondly, the generalizability of our study findings may be limited to practicing physicians in tertiary academic hospital settings with institutional support to develop PC expertise. Future work is needed to understand the PC needs and impact of training in primary and community settings. Finally, due to the small sample size of the study, we were unable to adjust for possible residual confounding stemming from sociodemographic and clinical background. Future hybrid effectiveness implementation studies with longitudinal measurements of behavior change, implementation processes, and clinician reported experiences can uncover modifiable elements of PC training to guide further adaptations.
Conclusion
There is a shortage of clinicians with PC skills and institutions to support them in China. Our research found that a culturally adapted PC training program improved practicing physicians’ PC knowledge and self-efficacy. Our study suggests that future scalable PC training requires emotional resonance between participants and instructors to ensure uptake of PC principles. Furthermore, participants should be supported longitudinally to overcome PC implementation barriers in their unique clinical settings.
Supplementary Information
Authors’ contributions
All authors contributed to conception and design or acquisition of data. HJ Liu, ZM Jia and XH Ning contributed to analysis and interpretation of data; all authors contributed to the writing and revising of the manuscript; and all authors provided final approval of the manuscript prior to publication.
Funding
This work was supported by 2023 Education and Teaching Reform Project of Peking Union Medical College (No. 2023zlgl 021), Social Technology for Global Aging Research Initiative, and the Mount Sinai Hospital AMO Sabbatical Grant.
Data availability
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Declarations
Ethics approval and consent to participate
This study received approval by the Institutional Review Board of Peking Union Medical College Hospital (I-23PJ925).
Competing interests
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Finkelstein EA, Bhadelia A, Goh C, Baid D, Singh R, Bhatnagar S, Connor SR. Cross country comparison of expert assessments of the quality of death and dying 2021. J Pain Symptom Manage. 2022;63(4):e419–29. [DOI] [PubMed] [Google Scholar]
- 2.Zhang NJ, Guo M, Zheng X. China: awakening giant developing solutions to population aging. Gerontologist. 2012;52(5):589–96. [DOI] [PubMed] [Google Scholar]
- 3.Li X, Fan L, Leng SX. The aging tsunami and senior healthcare development in China. J Am Geriatr Soc. 2018;66(8):1462–8. [DOI] [PubMed] [Google Scholar]
- 4.Leng A, Jing J, Nicholas S, Wang J. Catastrophic health expenditure of cancer patients at the end-of-life: a retrospective observational study in China. BMC Palliat Care. 2019;18(1):43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Yan Y, Zhang H, Gao W, Liu D, Endo M, Deshpande GA, Uehara Y, Watanabe D, Yoshikawa S, Mizushima A. Current awareness of palliative care in China. Lancet Glob Health. 2020;8(3):e333–5. [DOI] [PubMed] [Google Scholar]
- 6.Xie Y, Xu Y, Yang S, Yan J, Jin XQ, Liu C. Investigation of the awareness of and demand for hospice care and attitudes towards life-sustaining treatment at the end of life among community residents in Hangzhou. BMC Palliat Care. 2020;19(1):128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Tang Y. Death attitudes and truth disclosure: a survey of family caregivers of elders with terminal cancer in China. Nurs Ethics. 2019;26(7–8):1968–75. [DOI] [PubMed] [Google Scholar]
- 8.Yi T-w, Deng Y-t, Chen H-p, Zhang J, Liu J, Huang B-y, Wang Y-q, Jiang Y. The discordance of information needs between cancer patients and their families in China. Patient Educ Couns. 2016;99(5):863–9. [DOI] [PubMed] [Google Scholar]
- 9.Willemsen AM, Mason S, Zhang S, Elsner F. Status of palliative care education in mainland China: a systematic review. Palliat Support Care. 2021;19(2):235–45. [DOI] [PubMed] [Google Scholar]
- 10.Daubman B-R, Stoltenberg M, Kane K, Tulsky JA, Krakauer EL, Greco L, Jackson VA. Analyzing the unique needs of international palliative care learners attending a United States-based palliative care education and practice course. J Palliat Med. 2021;24(11):1721–4. [DOI] [PubMed] [Google Scholar]
- 11.Daubman B-R, Stoltenberg M, Spence D, Krakauer EL, Farrell S, Traeger L, Thomas R, Bromfield B, Sharpe N, Toppin PJ, et al. Implementing palliative care training in the Caribbean: development and assessment of a basic palliative care training course in Jamaica. J Pain Symptom Manage. 2021;62(6):1145–53. [DOI] [PubMed] [Google Scholar]
- 12.Daubman B-R, Stoltenberg M, Spence D, Krakauer EL, Farrell S, Traeger L, Thomas R, Bromfield B, Sharpe N, Toppin PJ, et al. Implementing palliative care training in the Caribbean: development and assessment of a basic palliative care training course in Jamaica. J Pain Symptom Manage. 2021;62(6):1145–53.Daubman B-R, Stoltenberg M, Spence D, Krakauer EL, Farrell S, Traeger L, Thomas R, Bromfield B, Sharpe N, Toppin PJ, et al. Implementing palliative care training in the Caribbean: development and assessment of a basic palliative care training course in Jamaica. J Pain Symptom Manage. 2021;62(6):1145–53. [DOI] [PubMed] [Google Scholar]
- 13.Tsao L, Kwete XJ, Slater SE, Doyle KP, Cuong DD, Khanh QT, Mauer R, Thy DNM, Thinh DHQ, Tuan TD, et al. Effect of training on physicians’ palliative care-related knowledge and attitudes in Vietnam. J Pain Symptom Manage. 2023;66(2):146–59. [DOI] [PubMed] [Google Scholar]
- 14.Dai X, Ning X, Lin J, Jing J, Daubman B-R, Seow H, Krakauer EL, Jia Z: Cultural adaptation and pilot testing of a basic palliative care curriculum for practicing physicians and nurses in mainland china. J Palliat Med 2025. [DOI] [PubMed]
- 15.Thomas PA, Kern DE, Hughes MT, Chen B: Curriculum development for medical education: a six-step approach. In: 1998; 1998. [DOI] [PubMed]
- 16.Nakazawa Y, Miyashita M, Morita T, Umeda M, Oyagi Y, Ogasawara T. The palliative care knowledge test: reliability and validity of an instrument to measure palliative care knowledge among health professionals. Palliat Med. 2009;23(8):754–66. [DOI] [PubMed] [Google Scholar]
- 17.Yamamoto R, Kizawa Y, Nakazawa Y, Morita T. The palliative care knowledge questionnaire for PEACE: reliability and validity of an instrument to measure palliative care knowledge among physicians. J Palliat Med. 2013;16(11):1423–8. [DOI] [PubMed] [Google Scholar]
- 18.Li WW, Chhabra J, Singh S. Palliative care education and its effectiveness: a systematic review. Public Health. 2021. 10.1016/j.puhe.2021.02.033. [DOI] [PubMed] [Google Scholar]
- 19.López-García M, Rubio L, Martin-de-Las-Heras S, Suárez J, Pérez-Cárceles MD, Martin-Martin J. Instruments to measure skills and knowledge of physicians and medical students in palliative care: a systematic review of psychometric properties. Med Teach. 2022;44(10):1133–45. [DOI] [PubMed] [Google Scholar]
- 20.Bainbridge D, Bishop V, Myers J, Marshall D, Stajduhar K, Seow H. Effectiveness of training programs about a palliative care approach: a systematic review of intervention trials for health care professionals. J Palliat Med. 2023;26(4):564–81. [DOI] [PubMed] [Google Scholar]
- 21.Seow H, Bishop VC, Myers J, Stajduhar KI, Marshall DI, Incardona NK, Levine OH, Steinberg LJ, Bainbridge DB. Outcome measures in palliative care training interventions: a systematic review of trial-based studies. Ann Palliat Med. 2023;12(2):399–417. [DOI] [PubMed] [Google Scholar]
- 22.Fetters MD: The Mixed methods research workbook: activities for designing, implementing, and publishing projects. In: 2020; 2020.
- 23.Creswell JW, Clark VLP: Designing and conducting mixed methods research. In: 2006; 2006.
- 24.Totton N, Lin J, Julious S, Chowdhury M, Brand A. A review of sample sizes for UK pilot and feasibility studies on the ISRCTN registry from 2013 to 2020. Pilot Feasibility Stud. 2023;9(1):188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hennink M, Kaiser BN. Sample sizes for saturation in qualitative research: a systematic review of empirical tests. Soc Sci Med. 2022;292: 114523. [DOI] [PubMed] [Google Scholar]
- 26.Moreau KA. Has the new Kirkpatrick generation built a better hammer for our evaluation toolbox? Med Teach. 2017;39(9):999. [DOI] [PubMed] [Google Scholar]
- 27.Tsao L, Slater SE, Doyle KP, Cuong DD, Khanh QT, Maurer R, Minh Thy DN, Quoc Thinh DH, Tuan TD, Van Dung D, et al. Palliative care-related knowledge, attitudes, and self-assessment among physicians in Vietnam. J Pain Symptom Manage. 2019. 10.1016/j.jpainsymman.2019.08.001. [DOI] [PubMed] [Google Scholar]
- 28.Hsieh H-F, Shannon SE. Three approaches to qualitative content analysis. Qual Health Res. 2005;15(9):1277–88. [DOI] [PubMed] [Google Scholar]
- 29.Nowell LS, Norris JM, White DE, Moules NJ: Thematic analysis. Int J Qual Methods 2017, 16(1).
- 30.Martín-Martín J, López-García M, Medina-Abellán MD, Beltrán-Aroca CM, Martín-de-Las-Heras S, Rubio L, Pérez-Cárceles MD. Physicians’ and nurses’ knowledge in palliative care: multidimensional regression models. Int J Environ Res Public Health. 2021. 10.3390/ijerph18095031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Nakazawa Y, Kato M, Miyashita M, Morita T, Kizawa Y. Changes in nurses’ knowledge, difficulties, and self-reported practices toward palliative care for cancer patients in Japan: an analysis of two nationwide representative surveys in 2008 and 2015. J Pain Symptom Manage. 2018;55(2):402–12. [DOI] [PubMed] [Google Scholar]
- 32.Dove ES, Kelly SE, Lucivero F, Machirori M, Dheensa S, Prainsack B. Beyond individualism: is there a place for relational autonomy in clinical practice and research? Clin Ethics. 2017;12(3):150–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Cheng S-Y, Lin C-P, Chan HY-L, Martina D, Mori M, Kim S-H, Ng R. Advance care planning in Asian culture. Jpn J Clin Oncol. 2020;50(9):976–89. [DOI] [PubMed] [Google Scholar]
- 34.Killackey T, Peter E, Maciver J, Mohammed S. Advance care planning with chronically ill patients: a relational autonomy approach. Nurs Ethics. 2020;27(2):360–71. [DOI] [PubMed] [Google Scholar]
- 35.Allen LM, Hay M, Palermo C. Evaluation in health professions education-is measuring outcomes enough? Med Educ. 2022;56(1):127–36. [DOI] [PubMed] [Google Scholar]
- 36.Suurmond J, Lanting K, de Voogd X, Oueslati R, Boland G, van den Muijsenbergh M. Twelve tips to teach culturally sensitive palliative care. Med Teach. 2021;43(7):845–50. [DOI] [PubMed] [Google Scholar]
- 37.Maybee A, Winemaker S, Howard M, Seow H, Farag A, Park H-J, Marshall D, Pereira J. Palliative care physicians’ motivations for models of practicing in the community: a qualitative descriptive study. Palliat Med. 2022;36(1):181–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Motoki K, Kinoshita Y, Nakano R, Hosokawa M, Nakazaki T. Quantitative analysis of florigen for the variability of floral induction in cabbage/radish inter-generic grafting. Plant Cell Physiol. 2022;63(9):1230–41. [DOI] [PubMed] [Google Scholar]
- 39.Velasquez A, West R, Graham C, Osguthorpe R. Developing caring relationships in schools: a review of the research on caring and nurturing pedagogies. Rev Educ. 2013;1(2):162–90. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.