Hospice Care Providers’ Knowledge, Attitudes and Practices in China: A Cross-sectional Study

Background: In recently years, China has been growing awareness about the hospice service movement although hospice care was Initially introduced 30 years ago. Hospice care providers’ knowledge, attitudes, self-eciency, and practices status in China should be investigated. This study aims to survey the general actuality of hospice care providers’ knowledge, attitudes, and practices of hospice care (KAPHC) in ve cities in China, and explore relevant inuencing factors. Method: Relying on our KAPHC scale, we randomly surveyed 3647 valid health care providers in ve sampled cities of China in 2019, which respectively represent the north, northeast, east, central, and southwest area. In each city, 14 institutions were selected and 50 health providers were surveyed in each institution. Increment of voluntary participation was accepted. The scaling outcomes were compared among each sampled city-groups with Chi-square test or ANOVA. Multiple correspondence analyses were also performed for further results. Results: Of all, 41.46% providers had real experience in providing hospice care, and 58.68% conrmed willingness of providing. The overall knowledge correct rate was 57.13%. The average scores of threats, benets, barriers, subjective norms, condence, and self-report behaviour items were 15.40, 41.56, 16.75, 15.65, 41.56, and 38.61. Providers from eastern China, willing to provide hospice care, or experienced death witness in providing hospice care, proved better score in knowledge, threats, benets, barriers, subjective norms, condence, and self-report behaviour (P<0.05). Conclusions: The investigation reected urgent need for systematic training of professional knowledge and skills on hospice care for health care providers. Governments’ attention and policy measures are crucial as most Chinese hospice care providers practiced as “organization actors”. More attention was need on balanced development of different areas.


Study design and setting
The study planned to survey 14 institutions in each ve cities, and to choose randomly 50 health care providers in each institution. The ve sampled cities in China in this study refer to as City A to City E, respectively represent the area of north (City-A), northeast (City-B), east (City-C), central (City-D), and southwest (City-E). And if some institutions were newly approved pilot of hospice care in 2019, they also can be voluntary to take part in the survey. Thus, the total samplings were estimated to be more than 3500. The anonymous, cross-sectional questionnaire survey for particular group was conducted, through the SO JUMP, an online Chinese questionnaire platform. The inclusion criteria were health providers working in health care institution in the cities had voluntarily signed up for the survey in prenotice enrolment. And the exclusion criteria were other care providers in health care institutions or not in voluntary registration book, or those staff who were unable to complete the questionnaire independently. To minimise the probably bias, we conducted a concentrated training in Beijing for all quality controllers from each city and investigate delegates from primary service institutions one week ahead. Scale A structured self-report questionnaire consisting of ve parts administered to all subjects. It was health providers' Knowledge, Attitudes, and Practices of Hospice Care (KAPHC) constructed more suitable for Chinese [15] . The four parts of the questionnaire included questions on demographic characteristics, knowledge (15items), attitudes (24items with 4 sub-concepts) and practices (22items with 2 subconcepts) on providing hospice care.
The knowledge sheet included questions about concept and philosophy, psychosocial and spiritual care, management of pain and other symptoms, opioid use, policies and localization problems, etc. The four sub-concepts of attitudes were threats (5 items), bene ts of life quality promotion and death preparation (10 items), barriers (5 items) and subjective norms (4 items) for providing hospice care. The two sub-concepts of practices were con dence and self-reported practices.

Statistics
As scaling outcomes, the scores of knowledge, attitude and practices compared by deferent groups in different cities. Statistical analyses were performed using commercial software IBM SPSS Statistics 24.0 (IBM Corporation, Armonk, NY), the threshold of statistical signi cance set at P<0.05 (2-tailed). The Chi-square test or ANOVA were used to compare the characteristics and scores of the 5 citygroups. Multiple correspondence analyses were also performed. The inference in correspondence analyses is whether certain levels of one characteristic (eg, city) are associated with some levels of another characteristic (eg, knowledge).

Results
From March 18th to 31th in 2019, total 3647 valid out of 3653 questionnaires were returned (valid rate 99.8%). Participants were recruited from 80 grass medical institutes of 5 sampled cities in China. Of all valid questionnaires, 756(20.73%) were from City-A, 593(16.26%) were from City-B, 687(18.84%) from City-C, 580(15.90%) from City-D and 1031(28.27%) were from City-E. The average age was 35.65±9.072 years old. The social and demographic characteristics of study participants see Table 1. For each item in knowledge scale, the detail choices comparisons of respondents in deferent ve cities see Table 2. And the multiple comparisons showed that the total scores among 5 cities were statistically different (F=80.612, P<0.01), and ranks in 5 cities were C>D>E>B>A (Tamhane, P<0.05).  Table 3 showed attitudes and practices scaling for respondents from different cities. Also with the results of multiple comparisons, the ranks of attitude in ve cities were C>D, E>B, A (Tamhane, P<0.01), of con dence were C, D>E>A, B (Tamhane, P<0.01), and of self-report practice were C, E>D>B>A (Tamhane, P<0.05). The mean scores of four attitude sub-concepts were 15.40, 41.56, 16.75 and 15.65, respectively (Table 3). These indicated positive attitudes towards providing hospice care. The most percent respondents agree about the threats of providing hospice care were: "I feel guilty when a patient of mine dies" (51.0%), "advanced cancer patient is hopeless for cure" (44.3%). About the barriers toward providing hospice care, 59.0% responders agree it is a barrier that "Advanced cancer patients have many di cult symptoms". On the other hand, the higher-score items in subjective norms included: "It is meaningful" (4.25), and "It is a part of my duty" (3.94). The multiple correspondence analyses indicated the most possible in uencing factors of Chinese hospice care providers' knowledge, attitude, con dence and self-report practice, details see Table 4. Hospice care providers in City-C, providers who willing to provide hospice care, who have the experience of death witness or experience in providing hospice care, proved better in knowledge, attitude, con dence and self-report behaviour. And those who work in CHC or THC did worse in knowledge, attitude, con dence and self-report behaviour. Moreover, a) of knowledge, providers with the higher education level, professional title, or as a doctor, would get higher scores and providers in rehabilitation nursing scored lower; b) of attitude, providers more aged or working in rehabilitation nursing scored lower; c) providers from specialized hospital showed more con dence in practicing; d) of self-report behaviour, doctor and nurse scored higher and men scored higher than women.
The correlation analysis showed high pair wise correlations (P<0.01) between every two measurement dimensions among knowledge, attitude, con dence and self-report practice, details see Table 5.

Overall Situation
In this investigation, there were 1512(41.46%) providers had real experience in providing hospice care, and 58.68% of all con rmed their willingness of providing. It proved those grassroots health service providers' supports in the national hospice movement, from the aspects of their emotional and practical involvement. Furthermore, ndings showed that providers who willing to provide hospice care, who have the experience of death witness or experience in providing hospice care, proved better in knowledge, attitude, con dence and self-report practice, which was partly proved by some previous researches [6,11,16,17] . We could know from that the exploration and practice of service provider is the key to break through the bottleneck. Primary medical institutions had accumulated certain manpower reserves and practical experience through institutional investment and talent training. As the country attached great importance to it and governments at all levels took active measures, the cause of hospice care in China has developed as if it woke up in winter and waited for the spring to blossom [18] . Hospice care in China has a distinct national characteristic of government leadership and policy support [3,19] . What about the other side of the coin? This investigation also showed a negative correlation between primary medical institutions and knowledge, attitude, con dence and practice, which probably indicated a relatively poor service quality and most urgent need for systematic training in the main force. Furthermore, systemic barriers to provision of hospice care among GPs need to be identi ed and addressed. [20,21] Among the ve sampled area, hospice care providers in City-C proved better in knowledge, attitude and behavior. This re ects the uneven development of hospice services across the country. Hospice services in the east of China have a long history. As early as October 1988, Shanghai Nanhui Elderly Nursing Hospital was established, taking the lead in carrying out hospice care services [22] . And till the year of 2018, Shanghai was the only city in the nationwide where the local government has issued a statement and special funds to promote hospice care [3,23] . It also played a great role in stimulating the development of surrounding eastern cities.

Knowledge
The overall correct rate of hospice care providers in the investigation was 57.13% (8.57/15), which was close to 60% as the calibrating in original scale design. We thought the result is patchy at best. Seeing into each item, hospice care providers were lack of cognition in philosophy and principles of hospice care, pain and symptom management, opioid use, and psychosocial and spiritual care.
The rst item "the provision of hospice care requires emotional detachment" was quoting from PCQN developed by Ross M M et al. The di culty index of this item was 0.67~0.93 for sampled nurses in her stud y [24] . In this study, it was the most di cult item with the di culty index of 0.15, and from 0.1 to 0.24 in the different ve sampled areas, also as low as the di culty of 0.19 in the pre-testing (u= 1.11, P=0.133>0.05). Apart from possible confounders of the scale translated into Chinese, it does re ect the lack of understanding among the respondents. Health service providers often must face the death in routine clinical work, and it is even more common for hospice caregivers in taking care for dying patients. It is undoubtedly a great challenge to meet all needs including emotional needs of the dying patients, thus hospice caregivers need to be prepared well and have a positive attitude. Chinese hospice providers practiced worse on item No.7, No.9 and No.10 in this investigation than those in PCQN as the correct rate were 0.31, 0.39, and 0.28 respectively. The sixth item was about hospice care for special people, with a high di culty and low correct rate of 0.36. If hospice care givers never be systemically well trained before, they would be more likely to answer wrong in uenced by their traditional ideas and unique bereavement culture in China [25] . The eighth item was for Traditional Chinese Medicine (TCM) in hospice care. Its total correct rate was 0.47, lower than pretesting, and furthermore in City-A the rate was as low as 0.28. It showed a weak point of knowledge for hospice caregivers and there was also an uneven among different areas. In the hospice movement in China, as a local characteristic, it is valuable and necessary to actively create conditions to explore and develop some TCM appropriate techniques, especially for the symptom alleviation and the quality of life improvement. However, in modern medicine the speed of TCM development was not so fast to match its broad and profound theory system. [26,27] To master the knowledge of TCM, to accelerate the development and implementation of TCM clinical guidelines and to apply appropriate techniques properly are both the need of TCM and hospice developments.

Attitude
Their mean of totaling score of attitude were 92.22 for all the 24 items. Across the ve cities, hospice care providers in City-C scores as high as 96.99. The four sub-concepts also indicated positive attitudes towards providing hospice care. This study showed that respondents agree about the most threats of providing hospice care were "I feel guilty when a patient of mine dies" (51.0%), and "advanced cancer patient is hopeless for cure" (44.3%), which is similar to Liu's study [28] . In general, providers in this study have a high agreement with the bene ts of hospice care. It showed that the proportion of respondents who agreed with the bene ts of improving quality of life is over 80%. Which means that they generally agree with the role of providing hospice care in improving quality of life. More than 80% of respondents also agreed that hospice care is helpful for patients and families on preparing to death, as "respect for patient's religion and burial rites", "better communication with advanced patients", and "help medical staff to take care of patients better". However, the agreements were relatively lower on "help to die at home" and "avoid the idea of euthanasia" those proportion were around 55%. It may be related to the lower service rate of hospice care at home than hospice care at institution in China. Most respondents agreed on the social value of hospice care, they thought "it is meaningful" (85.82%), "it is a part of duty" (74.39%), and 62.60% of them re ected that they chose hospice care because they got "the approval and support of department leader, colleagues, relatives and friends". It can be considered that the choice of hospice care providers in medical institutions in China to engage in hospice care service work was based on the role norm of "organization actor". Their behavioural choice was often from the perspective of the organization. They thought hospice care service as a business behaviour, and their performance was subject to the arrangement of the organization leader. Therefore, the attention and corresponding policy measures of governments are crucial. In addition, more than half of the respondents (58.63%) experienced the death of family member, which affected them to provide hospice care. It proved that the understanding of life and death, and hospice service delivery could be deepened or changed by witness or observation of the experience of important people around, which embodies the connotation of "neighbourhood effect" [29] .

Practices
The con dence of providing hospice care and their self-report practices on hospice care were to scale a caregivers' behaviour. The mean of totalling scores were 41.56 and 38.61 in their self-evaluated con dence and practice.
The con dence of providing "pain assessment of patients" and "alleviate pain and discomfort of dying patients" was 74.48% and 68.00%, respectively, while con dence to "guide the management of afterwards and funeral preparation for families" was the lowest (62.21%).
General practitioners and other health care providers in CHCs or THCs considered non-physical abilities (spiritual, cultural, ethical, and legal, etc.) almost as important as pain and symptom control, but most of them lack con dence in their non-physical abilities [30] .
Furthermore, the overall average age of the respondents was 35.65 years old, and their personal experience was relatively inexperienced. If they did not receive systematic training, it would be di cult to improve their con dence of providing such guidance services. The con dence to "coordinate the media resources of medical, social, psychological and spiritual care?" was also relatively low, re ecting that they still need to improve in mobilizing the health care system and social forces to provide coordinated care. The self-report practice is strongly related to knowledge, attitude and con dence (P<0.001). This current study didn't analyse how the con dence and self-report practice in uence each other (r=0.631), which need to be further con rmed. Respondents were more con dent to coordinate the media resources of medical, social, psychological and spiritual care (3.70) however they did practice so enough (3.42). And they practice less in guiding the body cuisine and funeral preparation for families (3.16) with also a relatively low con dent to do (3.62). Currently 41.46% of all respondents already had experience in providing hospice care, and the general mean score of self-report practice is 3.51(38.61/11) which means the frequency was between "occasionally" and "often". It indicated that for professional full-time hospice care providers, they should increase their practices frequency.

Strengths and Limitation
This study adopted a scienti c and localized developed scale to survey the general actuality of hospice care providers' knowledge, It still has some selection bias, as we conduct a strati ed random sampling. In actual operation, we investigated all of staff in those institutions which had total staff less than 50 and, if some CHCs or THCs were newly approved pilot institutions of hospice care in 2019, they also can be voluntary to take part in the survey. Secondly, in this survey, some respondents were not primarily responsible for hospice care work, and may not do it in the future according to for institutional arrangements and personal willingness. In addition, we only sampled 5 cities on behalf of 5 areas of China maybe not so strong evidence. If conditions permit, sampling should be expanded in further study. To scale a hospice care provider's behaviour, we designed the con dence of providing hospice care and self-report practice however it is still not so ideal way to evaluate a person's true behaviour. To explain better the potential meaning of data, a mix method research may be a solution.

Conclusion
Hospice care, as an integral part of human's life cycle health management, has increasingly become an important livelihood issue of common concern. The development of hospice care in China is still in its infancy, and the long march has just begun. There are still many practical problems to be solved.
The investigation of hospice caregivers' knowledge, attitudes and practices re ected the urgent need for systematic training of professional knowledge and skills in primary medical institutions. And for future formulation of training programs, the study-results of both suppliers and buyers should both be considered. The quality and quantity of hospice service should be both attached importance to and promoting it. In addition, we should pay more attention to the equity issues on balanced development of different areas, draw on the valuable experience of the eastern area in the development of hospice care, and put the work into practice. It did not involving any human material, or human data. The respondents were told in advance about the anonymous investigation and informed consent was obtained.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analysed during this study are included in this published article.

Competing interests
The authors declare that they have no competing interests.

Funding
The fund of Ministry of Education of China(20YJAZH045) supported data collection, the funds of Shanghai Plan of Philosophy and Social Science (2019BGL032) and Shanghai Pujiang Program (2019PJC099) supported article writing. Corresponding author Dr. Limei Jing is the lead and takes responsibility for above-mentioned funds.
Authors' contributions ZS made substantial contributions to data collection, analysis and interpretation and drafted the manuscript. LJ made contributions to research design and survey, and also contributed to draw a conclusion for the revised manuscript. All authors reviewed and approved the nal manuscript.