The content of research paper

There is a lack of palliative care provision in the Japanese community. This can be verified by the fact that only 24% of cancer patients who died in Japan between April 2009 and May 2010 used specialized palliative care services, and most services were provided only to hospitalized patients¹⁾. Consequently, a basic policy regarding improving the social environment by enhancing cancer care and support was introduced in 2018. This policy calls for the establishment of a community support system in close collaboration with local and national governments, medical care institutions, schools, and private organizations to compensate for the lack of palliative care service provision in the community. This policy has been termed “The Future Direction of the 3rd-term Comprehensive 10-year Cancer Control Strategy.”²⁾

Community healthcare professionals need to acquire palliative care knowledge, skills, attitudes, and behaviors. Although palliative care education is being developed and delivered internationally, there are indications that patients are inappropriately cared for by physicians with inadequate training³⁾. In Japan, palliative care education for physicians has been promoted as a matter of policy. The Cancer Control Act⁴⁾ of 2007 promotes the early discovery of cancer, including the distribution of cancer treatment and research. It is part of “the Basic Plan to Promote Cancer Control Programs”⁵⁾ and includes goals such as “Conduct basic training on palliative care for all doctors and promotion of cancer registration.” In 2008, “the Palliative Care Emphasis Program on Symptom Management and Assessment for Continuous Medical Education (PEACE)” was launched nationwide to provide basic palliative care training for physicians involved in cancer treatment. PEACE has been credited with improving physicians’ knowledge and palliative care practice and difficulties with palliative care in designated cancer hospitals^6,7). However, these are mainly for physicians working in hospitals and not for community healthcare professionals.

Community palliative care training is in its developmental stages. “The Outreach Palliative Care Trial of the Integrated Regional Model (OPTIM)” was conducted as a multiple-intervention study in four regions of Japan from 2008 to 2010. The study reported improvements in the local physicians’ knowledge of palliative care and their perceptions of difficulties^8-10). However, it is not clear which training methods will improve the ability of individual community healthcare professionals to practice community palliative care. It is important to identify how community healthcare professionals apply their knowledge to practice and the training methods that influence practice to achieve high-quality palliative care delivery in the community.

Family physicians (FPs) are expected to be one of the providers of community palliative care, even though they are not physicians specializing in palliative care. FPs are primary health care (PHC)¹²⁾ professionals, and Internationally, community-oriented palliative care is referred to as primary palliative care and is provided according to PHC principles^13-15). Murray suggested that with adequate training, resources, and professional support, FPs can provide end-of-life care for most patients¹³⁾. A previous study of FPs in Taiwan found that 18.8% had experience with palliative care, and approximately 40% were engaged in home visits^16,17). A study of 52 Belgian FPs found that almost all had a receptive attitude toward palliative care; however, they were uncertain about their role in decision-making and negotiation, suggesting a lack of knowledge regarding palliative medicine¹⁸⁾. In a survey of 516 FPs in the UK, respondents felt that FPs play an important role in palliative care, indicating a need for training and professional development to enhance their palliative care knowledge, skills, and attitudes¹⁹⁾. In Europe, FPs are trained to provide palliative care during the early stages of a disease^20,21).

FPs contribute to community palliative care but lack long-term professional training in this area. It is unclear what training methods help them acquire community palliative care capacity. This study investigated the training methods used by certified FPs to enhance their professional development in palliative care and to determine the relationship between these methods and palliative care practice. The results will contribute to the development of training methods for physicians from diverse disciplines, as well as nurses and other community healthcare professionals, enabling them to acquire the necessary competencies required for community palliative care.

Design

This cross-sectional study used mailed questionnaires to collect data. The primary endpoint was the association between certified FPs’ palliative care training and palliative care practices. It was unclear what training methods the FPs had incorporated to enhance their professional development in palliative care. In addition to a systematic educational program, it was necessary to extract the methods used by the FPs individually. Therefore, a conceptual framework was created to develop this questionnaire. It related palliative care practices to FPs’ training experiences and methods and clarified the role of other factors (Figure 1). The questionnaire was modified based on a pre-test administered to 19 FPs and a small group discussion involving palliative care physicians and coresearchers.

Palliative care practice, as an outcome measure, was based on the palliative care self-reported practices scale (PCPS). The PCPS developed by Nakazawa et al.²²⁾ is a self-assessment scale used to assess palliative care practice for healthcare professionals. It quantifies recommended practices in palliative care and the degree of adherence to them, which indicates an awareness of palliative care practices. Its validity and reliability were confirmed in a development study and it has been used in previous surveys^6,7,23). The PCPS was developed to apply to a hospital setting offering palliative care for cancer patients, but it has also been used in the OPTIM Study⁸⁾ to assess community-oriented palliative care.

Fig. 1.

A conceptual framework

Participants and data collection

Questionnaires were sent to 557 Japanese Primary Care Association (JPCA) certified FPs registered with JPCA between January and March 2017. JPCA has an accreditation system for specialty training programs (JPCA-certified FPs) that has certified FPs since 2009. These FPs are not palliative care specialists but can comprehensively treat patients. They work in focus areas including community comprehensive care and palliative/end-of-life care, and case reports must be submitted. As of February 2023, there are 1,126 JPCA-certified FPs in Japan²⁴⁾. The mailing was conducted by a commissioning organization. The mail detailed the purpose of the study and the preservation of anonymity. Consent to participate was obtained by returning questionnaires. Ethics approval was obtained from the Ethics Review Committee of Fukushima Medical University (No. 2928) before the survey was distributed.

Questionnaire

The questionnaire consisted of the PCPS, training methods related to palliative care, and basic attributes and practice experience of FPs.

The PCPS consists of 18 items that reflect the five aspects of pain, dyspnea, delirium, dying -phase care, and communication (Figure 2). They are rated on a 5-point scale, with scores assigned in the following order:“always,” “usually,” “sometimes,” “not often,” and “no.”

Training in palliative care (one item) and training methods for palliative care (eight items) consisted of two options:“yes” and “no.” If the respondents had training in palliative care, they were asked to select “yes,” otherwise “no” for the following eight items:(1) Systematic lectures and training in palliative care (a few hours, a few days, 7 days to less than 1 month, more than 1 month); (2) Guidance and consultation from palliative care specialists; (3) Guidance and consultation from senior FPs; (4) Observing palliative care practice by senior FPs; (5) Self-study through literature; (6) Reflection on the practices (Reflecting intentionally on the case study); (7) Publication of case reports; and (8) Feedback in multidisciplinary teams.

Basic attributes of FPs were collected (five items):age, sex, affiliation (selected from hospital, clinic, or other), year of FP certification, and qualifications other than JPCA-certified FPs. Home-visit experience (two items) was evaluated based on the presence or absence of home-visit experience and the number of home-visit patients. Participants could choose either “yes” or “no” for the experience of home-visit treatment; if they answered “yes,” they indicated the number of patients treated in the past year using the options “less than 10,” “10 to 50,” or “50 or more.” Experience of providing palliative care to patients (three items) included the presence or absence of experience of providing palliative care, number of patients, and patients’ diseases. Options for the number of patients to whom palliative care was provided in the past year included “less than 10,” “10 to 50,” or “50 or more.” For patients’ diseases, respondents answered either “yes” or “no” to multiple diseases (cancer, dementia/frailty, neurological incurable disease, heart disease, cerebrovascular disease, pulmonary disease, kidney disease, liver disease, orthopedic disease, collagen disease) experienced in the past year.

Fig. 2.

The palliative care self-reported practices scale (PCPS)

Data analysis

Descriptive statistics for all items were obtained, followed by a Shapiro-Wilk normality test for numerical data. After testing the normality of the data and confirming that they were non-normally distributed, a univariate analysis was conducted on the association between the median PCPS score and each item. The Spearman rank-sum test was used for continuous variables, the Mann-Whitney U test was used for binary variables, and the Kruskal-Wallis test was used for variables with three or more groups, corrected using the Bonferroni method.

Quantile regression analysis (median) was performed with palliative care training as the explanatory variable and PCPS as the response variable. Model 1 shows the relationship between the eight training items and PCPS. Model 2 shows the results of the analysis with the addition of other factors. For patient diseases, owing to a large number of items, they were categorized as cancer and noncancer and were conducted using bivariate variables. Patients’ diseases were categorized as either cancer or non-cancer and used as two variables. Regression analysis was performed for “yes” results for training experience. Data analysis was performed using statistical software (Stata15 SE) with a significance level of 5%.

Characteristics of participants

Of the survey forms sent out, 311 were completed (55.8% response rate). After consulting a statistical expert, participants with missing PCPS score data were excluded (Figure 3); the participants’ characteristics are presented in Table 1. The mean age of the FPs was 38.2 years (SD ± 5.8), and they had been family medicine specialists for 5.1 years (SD ± 3.7). There was an approximate 50/50 split in the affiliations between hospitals and clinics. A total of 99.4% of the FPs had experience providing home visits and palliative care, and 92.8% had received palliative care training. The number of patients (in the past year) for whom FPs had provided palliative care was less than 10 for 42.8% of FPs, between 10 and 50 for 44.1%, and more than 50 for 13.2%.

Fig. 3.

Participant Flow Chart

Table 1.

Characteristics of participants

¹ Continuous variables are shown as mean (median) ± SD, categorical variables as No (%).

Comparison of PCPS scores

The median PCPS scores for all participants were 4.3 (min 1, max 5), and each PCPS median score was as follows:for Delirium, 3.7 (min 1, max 5); Dyspnea, 4.0 (min 1, max 5); Pain, 4.3 (min 1, max 5); Dying-phase care, 4.3 (min 1, max 5); Patient-and family-centered care, 4.4(min1, max5); and Communication, 4.4 (min 1, max 5). The comparison results for each item are shown in Table 2, which shows the PCPS scores of the “Yes” respondents, the participants who answered “No,” and the p-values resulting from these comparisons. The significant difference in PCPS scores was “Affiliation,” with clinics scoring higher than hospitals. For “Number of home visit patients” and “Number of providing palliative care patients,” an increase in the number of patients treated was associated with a higher PCPS score. For patients’ diseases for which palliative care had been provided, all disease types except for dementia and frailty showed significant differences, with higher scores overall. For patient disease, whether the clinician had experience with that disease tended to be associated with higher scores. The median PCPS score for FPs who responded “Yes” to palliative care training was 4.2 and 3.8 for those who responded “No,” a significant difference. Similarly, there were significant differences in all training items with and without training. The highest PCPS score was for “Self-study through literature.”

Table 2.

Each item relation to the PCPS score

¹Comparison with PCPS Score: Binary variables were subjected to the Mann-Whitney U test, continuous variables to Spearman’s rank-sum test, and variables with three or more groups to the Kruskal-Wallis test followed by Bonferroni correction. Categorical variables represent median values and continuous variables represent Spearman’s rho.

Palliative care training methods for palliative care practice

The results of the quantile regression analysis (median) of the association between PCPS and palliative care training experience are shown in Table 3. FPs with palliative care training had a significantly higher PCPS score (adjusted coefficient 0.4 [95% CI, 0.12-0.68];P=0.004). Table 4 shows the association between the eight palliative care training items and PCPS. In Model 1, which examined the association between PCPS and eight training methods of palliative care, “Reflection on the practices” (adjusted coefficient 0.24 [95% CI, 0.08-0.4]; P=0.004) was significant. In Model 2 after adjustment, “Self-study through literature” (adjusted coefficient 0.18 [95% CI, 0.01-0.34];P=0.03) and “Reflection on the practices” (adjusted coefficient 0.24 [95% CI, 0.08-0.4];P=0.004) were significant. To account for influences other than training effects, the analysis was conducted according to the conceptual model (Figure 1). Significant differences in other factors are shown in the footnotes of Tables 3 and 4.

Table 3.

Relationship between PCPS and palliative care training

¹The results of a quantile regression analysis (median) of “Yes” versus “No” training experience. ²Adjustment items: age, gender, affiliation (hospital, clinic, other) ***, the average number of years since obtaining JPCA-certified family physician qualification, qualifications other than JPCA-certified family physician, number of home-visit patients, number of palliative patients**, patient’s disease (cancer, non-cancer) ***P<0.001 **P<0.01

Table 4.

Relationship between PCPS and palliative care training methods

¹Each item presents the results of a quantile regression analysis (median) of “Yes” versus “No” training experience. ²Adjustment items: Palliative care 8 items of training methods. ³Adjustment items: age, gender, affiliation (hospital, clinic, other) *, the average number of years since obtaining JPCA-certified family physician qualification, qualifications other than JPCA-certified family physician, number of home-visit patients**, number of palliative patients, patient’s disease (cancer*, non-cancer). **P<0.01 *P<0.05

Palliative care delivery by FPs in Japan

A high percentage of FPs (99.4%) reported that they had experienced home visits and palliative care. This is a high value compared with the results of a survey in Taiwan^25-27). In addition, 92.8% of FPs responded “yes” to having palliative care training and their PCPS scores tended to be high (Table 2). Studies in Denmark and Norway reported that home visits by GPs were associated with home deaths and led to an improved quality of palliative care^25-27). The current study revealed many FPs had experienced some training in palliative care and actively provided palliative care in the community.

Palliative care training methods engaged by FPs

Most FPs engage in palliative care training to enhance their professional development. The most common training method was participation in structured lectures or palliative care (87.0%). Our findings suggest that palliative care practice stabilized during training periods of seven days or longer (Table 2).

Furthermore, the percentage of FPs with more than a few days of training was high (74.3%). PEACE is available for palliative care training for one–two days. In 2015, the PEACE participation rate in a randomly selected sample of physicians (from hospitals and clinics across Japan) was 51%. As of 2017, the rate among physicians involved in cancer care at cancer center hospitals was 82.1%²⁸⁾. About half were working in clinics and taking the PEACE course was not mandated by the Japanese government. Furthermore, it would not be easy for small facilities to receive training that spans several days, given the limited number of physicians available. While it is unclear whether the training attended by FPs is equivalent to PEACE, the fact that many FPs received more than a few days of palliative care training suggests that they had positive attitudes toward palliative care training.

Nearly 50-68% of FPs participated in self-study through literature and sought guidance and consultation from senior FPs and palliative care specialists. Approximately 40% of FPs observed palliative care practice by senior FPs, feedback from multidisciplinary teams, and reflection on practices. When practicing primary palliative care, FPs and palliative care specialists are encouraged to work together to provide care^16,29). In this study, FPs who learned from palliative care specialists, as well as their senior peers, enhanced their professional development in palliative care. Our findings suggest that FPs used diverse methods to enhance their professional development in palliative care.

The relationship between palliative care practice and training

Participants who reported receiving palliative care training had a total PCPS score of 4.2 with an adjusted coefficient of 0.4. Other adjusted coefficients included “Self-study through literature” at 0.18 and “Reflection on the practices” at 0.24. In the OPTIM study, which involved multiple interventions across the community, the quality of palliative care was enhanced, with physicians achieving a PCPS total score of 3.7 with an effect size of 0.17 after participating in the palliative care program¹⁰⁾. Although direct comparisons are challenging owing to differences in the survey content, the differences in PCPS scores with and without training are considered meaningful.

Among the palliative care training methods, “Reflection on the practices” and “Self-study through literature” were significantly associated with PCPS scores (Table 4). The results support the fact that knowledge is the foundation for palliative care practice, as stated in previous studies³⁾. A distinctive result of this study was that “Reflection on the practices” was implemented as part of palliative care training. This item was answered when FPs were aware of self-reflection as part of their palliative care training. Prior studies have incorporated reflection as an educational component in palliative care practice. A qualitative study in the Netherlands has shown the need for self-reflection among physicians involved in palliative care at home, and the importance of developing reflection training on end-of-life care³⁰⁾.

Reflection has gained a reputation for encouraging the integration between theory and practice within professional education. Training is conducted with the view of being in itself a skill to be mastered^31,32). Reflection skills need to be developed in palliative care education in the community. Further research is needed on how FPs perform “Reflect on the practices,” as this may lead to training methods that enhance the practice of palliative care in the community.

One limitation of this study is that it used self-reports of certified FPs, which may deviate from the actual quality of palliative care. In future research, evaluations by patients’ families should also be included. For the evaluation of non-cancer patients (especially those with dementia/frailty), it is desirable to develop a new evaluation scale.

In addition, we only targeted a population of certified FPs. Comparisons with physicians and professionals other than certified FPs who provide palliative care in the community are required. Additionally, the specialist system in Japan has changed since 2018, and the JPCA-certified FPs assessed in this study are not the same as those currently being trained. It is important that standard palliative care training be incorporated into the JPCA’s specialist training and continuing education programs to provide high-quality primary palliative care training for FPs in Japan. We hope that this new specialty system will foster high-quality FPs in the future.

Acknowledgments:We sincerely appreciate the cooperation of Masao Suzuki and Tomoko Hatakeyama in preparing the questionnaire. We would like to thank Editage for English language editing.

Authors contributions:KK was in charge of study conception and design, overall data acquisition and analysis, and writing. SK participated in the study design and made important contributions to the data interpretation and writing. RK provided guidance on the overall research plan and was a crucial intellectual part of this dissertation.

Conflict of interest:The authors declare no conflict of interest for this article.

Funding statement:This study was funded by the SASAGAWA Health Foundation.