This study aims to develop an MCDM model and apply it to the community hospitals in the present study. The HHNJS instrument is an evaluation tool that includes eight subscales. The dimensions and criteria weights were then constructed using the CFPRs method. This model was ultimately used to evaluate and improve the satisfaction of HHNs in the community hospitals in the present study. The study flow chart is shown in Figure 1.

The HHNJS is the only scale specifically designed to measure HHNs job satisfaction [20]. It was developed by Ellenbecker and her colleagues and has been revised several times to enhance its psychometric properties [17, 20–22]. Previous empirical studies regarding job satisfaction and Neal’s theory of HHNs practice form the theoretical framework of the instrument [23–27]. The HHNJS has been proved to have adequate validity and reliability. The reliability coefficient of most subscales, except for “stress and workload,” was 0.70 or above in the latest version [17, 22].

The HHNJS comprises 30 items to measure the level of job satisfaction in eight components: “relationship with patients (C ₁),” “relationship with peer (C ₂),” “professional pride (C ₃),” “salary and benefit (C ₄),” “relationship with physician (C ₅),” “relationship with organization (C ₆),” “autonomy and control (C ₇)” and “stress and workload (C ₈)” [14]. Each of the 30 items is rated on a 5-point Likert-type scale from 1 to 5 (“strongly disagree” to “strongly agree”). The components of the HHNJS instrument are listed in Table 1.

AHP is a commonly used MCDM method to evaluate the relative attribute weights by performing pairwise comparisons [28]. The decision-makers compare a pair of attributes at a time. While there are k attributes in the evaluation system, it needs to be compared c 2 K = k ( k − 1 ) / 2 times. Too many comparisons between attribute pairs may result in inconsistencies, including primacy, recency, and the bandwagon effect [28, 29]. Hence, Herrera-Viedma et al. developed the CFPRs method to cope with the many pairwise comparisons and inconsistencies in AHP [30]. This approach requires only k − 1 pairwise comparisons when k attributes are involved in the evaluation system [31]. The remaining ( k − 1 ) ( k / 2 − 1 ) comparisons can be computed using the CFPR method, which only refers to simple calculative procedures and ensures consistent comparison results [32]. CFPRs are easy to use and offer a practical way to obtain the attribute weights in the MCDM method [33]. Thus, they have been applied to hazard assessment [34], supplier selection [31], logistic outsourcing [35], and many other topics. The calculation procedure for this method is shown in the Supplementary Appendix [30, 31, 36].

The IPA method was first proposed by Martilla et al. in 1977 in marketing [37] and has been applied to many academic fields, such as dental practice [38], hospital quality [39], education [40], and tourism [41]. The IPA method is a useful research technique to help decision-makers evaluate customer satisfaction with products or services [42]. IPA provides valuable information by measuring the importance and performance of each attribute and dividing the attributes into four quadrants, as shown in Figure 2.

In quadrant I—keep up the good work—the attributes are both highly important and high performance. This means that these attributes are organizational strengths, and decision-makers should maintain resource input. In other words, nursing department decision-makers should maintain these attributes to maintain acceptable levels of HHNs job satisfaction.

Attributes in the “possible overkill” quadrant have high customer satisfaction with performance but are unimportant. Resources invested in these attributes can be reduced and should be applied elsewhere. In the short term, nursing decision-makers can temporarily reduce their attention to these attributes and use related resources for Quadrant IV attributes to improve job satisfaction.

Attributes in the “low priority” quadrant have low importance, poor performance, and can be temporarily ignored. Compared with other attributes, nursing department decision-makers can directly identify the resources related to these and other attributes that can improve job satisfaction.

Attributes located in “concentrate here” are highly important but have poor performance, which indicates that the attributes require immediate improvement, and decision-makers should focus on them. Nursing decision-makers should increase their satisfaction with these attributes and divert resources from Quadrants II or III to these attributes.

Thirty-one HHNs from three community hospitals in Zhejiang Province participated in the study and completed a questionnaire survey from February to March 2022. The questionnaire was comprised of three parts: the first evaluated the relative importance of the attributes, the second was a self-evaluation of job satisfaction with home healthcare, and the third covered participant demographics. 100% of participants were female, and 48% had a bachelor’s degree. 58% percent of nurses were younger than 30 years old, and most were senior nurses. 61% of participants had been nursing for less than 10 years, while 81% had worked in home healthcare nursing for no more than 5 years. Table 2 shows the participant information.

Table 3 lists the attribute weights of the 31 HHNs based on the HHNJS scale. The confidence level of the weighted results was 0.7%, lower than 5%.

Overall, the “Relationship with patients (C ₁)” dimension was the most important, with an average weight level of 0.1731, followed by “Salary and benefit (C ₄)” and “Relationship with peer (C ₂)”. Relative to the other dimensions, “Relationship with organization (C ₆)” was the least important. “I am treated as a professional colleague by physicians (C ₅₂)” was the most important criterion, with an average global weight of 0.0662, while “At times I am overwhelmed by all the work I have to do (C ₈₁)” was the least important, with an average global weight of 0.0134.

The IPA results of HHNs job satisfaction are summarized in Table 4 and illustrated in Figure 3.

Quadrant I includes “I have helped patients maintain or improve their quality of life (C ₁₃),” “My work is important and worthwhile (C ₁₄),” “The patient care that I provide adheres to my professional standards (C ₁₅),” “There is a good amount of collegiality among the nurses I work with (C ₂₃),” “I have peers whom I can rely on and turn to if necessary (C ₂₄),” “An upgrading of the pay scales at this agency is needed (C ₄₂),” “Physicians value my input on the status of their home healthcare patients (C ₅₁)”, and “I am treated as a professional colleague by physicians (C ₅₂).”

Quadrant II includes “Patients are satisfied with the care that I provide (C ₁₁),” “The relationships that I have established with patients are rewarding (C ₁₂),” “The support I have from my nursing peers is a positive aspect of my job (C ₂₁),” “I can communicate comfortably with the nurses I work with (C ₂₂),” “If I had to do it over again, I would choose home healthcare as my area of practice (C ₃₁),” and “I am able to adjust the hours of my work if needed (C ₇₁).”

Quadrant III includes “My present salary is satisfactory (C ₄₁),” “I am satisfied with the professional relationship that I have with nursing administration at this agency (C ₆₁),” “I have more flexibility in my hours of work than nurses in other practice settings (C ₇₂),” “At times I am overwhelmed by all the work I have to do (C ₈₁),” “I am able to meet the demands of my job (C ₈₃),” “I am able to cope with the increased demands for documentation in home care (C ₈₄),” and “Sometimes I get frustrated because all my activities are programmed for me (C ₈₅).”

Quadrant IV includes “I would commend my job to another health care professional (C ₃₂),” “I am proud to talk to people about the work I do (C ₃₃),” “Nursing salaries at other agencies are better than salaries at this agency (C ₄₃),” “The benefit package at this agency is satisfactory to me (C ₄₄),” “I have the power to generate change in organizational policy at this agency (C ₆₂),” “I have the opportunity to grow and develop as a professional nurse within this agency (C ₆₃),” “I have sufficient control over scheduling my time (C ₇₃),” and “I have the independence to make important decisions in my day-to-day work (C ₇₄).”

HHNs provide skilled patient care and spend most of their time interacting with them. A relationship is established when nurses care for patients [43], and direct patient care can be rewarding and give a sense of value to nurses [44]. Nurses’ job satisfaction is positively correlated with the quality of care they provide. Improving the health status of patients and having a positive impact on patients are related to higher job satisfaction [45]. The degree to which a job reflects an individual’s values affects job satisfaction. Therefore, a sense of value and job identity plays an important role in nurses’ job satisfaction [46]. In addition, interactions with peers and physicians—teamwork, group cohesion, respect, and acknowledgment by physicians—are important job satisfaction-related factors [47, 48]. Remuneration is an important extrinsic factor affecting job quality and satisfaction [46], and many nurses recommend increasing remuneration as the most relevant financial incentive [49]. Therefore, these attributes (C₁₃, C₁₄, C₁₅, C₂₃, C₂₄, C₄₂, C₅₁, and C₅₂) have higher relative importance and priority than those in quadrant III. This indicates that home-healthcare-agency administrators should focus primarily on these attributes as they are the critical factors influencing HHNs job satisfaction.

Compared to other attributes, the attributes in quadrant IV (C₃₂, C₃₃, C₄₃, C₄₄, C₆₂, C₆₃, C₁₃, C₇₃, and C₇₄) take precedence in improving job satisfaction performance in home healthcare nursing. Professional pride is key to nursing job satisfaction due to the positive image of nursing and pride in nursing skills. Threatening professional pride may lead to nurses choosing not to continue with nursing and not recommending the field to others [50]. Although remuneration was considered an important influencing factor for job satisfaction, the focus was on increasing salary and benefits to recruit and retain nurses. This may be a short-sighted approach as administrators must be mindful of pay equity [46, 51]. The opportunity for professional growth and influence on changes in organizational policy can help increase nurses’ job satisfaction [47, 52]. However, administrators limited by hospital budgets seldom provide sufficient time or opportunities for nurses seeking professional development [44]. Autonomy was defined as the ability to arrange one’s own schedule by prioritizing tasks, working alone, and having the freedom to make decisions within the nursing scope. It was considered a key factor affecting job satisfaction [46, 53]. Nurses in home healthcare settings have a higher level of autonomy than those in hospital-based settings [54]. However, because nurses in China do not exercise autonomous practices, the performance of the attributes C ₇₃ and C ₇₄ is poor [55].

The main findings of the study are as follows. First, we obtained the attribute weights of HHNJS and found that C ₁ is the most important factor influencing HHNs’ job satisfaction. Second, we evaluated the job satisfaction of HNNs and found that the HHNs were not satisfied with the attributes in quadrants III (C ₄₁, C ₆₁, C ₇₂, C ₈₁, C ₈₃, C ₈₄, C ₈₅) and IV (C ₃₂, C ₃₃, C ₄₃, C ₄₄, C ₆₂, C ₆₃, C ₁₃, C ₇₃, C ₇₄). Combined with the attribute weights, home-healthcare-agency administrators should focus their attention and resources on the attributes in quadrant IV.

This study has a few limitations. First, the study used the optimized CFPR method based on the AHP method to explore the relative weights of attributes and identify key factors. Although CFPRs can overcome the shortcomings of AHP and improve its practical value effectively, they are still based on the independent relationship between attributes. Second, this study was a cross-sectional survey, limited by its self-report nature, and might not reflect the long-term HHNs job satisfaction performance. Finally, the study only included 31 HHNs in one prefecture-level city in China; therefore, the findings cannot be widely generalized. Future research using random selection over a wider group of nurses from different regions is recommended.

This study used the hybrid MCDM model to investigate the critical factors and performance of HHNs job satisfaction based on nurse experiences in three community hospital. The findings can help home-healthcare-agency administrators better understand the critical factors influencing the HHNs job satisfaction and the current job satisfaction performance. This could enable them to establish reasonable strategies to improve job satisfaction and retain the nursing workforce.