Developing supportive policy environments for hospice care in china: a quantitative policy evaluation based on the PMC-Index model

Abstract

Background

Public policies play a crucial role in enhancing practical hospice care services and have attracted significant attention in China in recent years. However, there is a lack of research systematically evaluating hospice care policies from an empirical perspective.

Objectives

To address this gap, we aimed to define and assess overall and individual indicators of hospice care policy importance, coherence, and performance in central, provincial, and local government policy guidelines in China.

Methods

We applied content analysis and text mining to 112 hospice care policy documents. Using the Policy Modeling Consistency Index (PMC-Index) model, we developed a comprehensive evaluation framework comprising 10 primary and 47 secondary indicators. Subsequently, we quantitatively evaluated 18 selected policy samples through both multidimensional holistic and individual-sample analyses, and examined improvement trends in three national-level stand-alone hospice care strategies.

Results

The average PMC-Index score of the 18 selected policy samples was 7.20 (out of 10.00), indicating a Good level of consistency. Two policies achieved an Excellent rating, eight were rated as Good, six Acceptable, and two Low. In terms of the 10 policy dimensions, Policy nature (X₁) and Policy tools (X₃) exhibited excellent performance, whereas Policy equity (X₆) and Policy guarantee (X₈) showed unsatisfactory performance. Notably, all three national-level stand-alone strategies were rated at the Good level and demonstrated improvements across different pilot periods.

Conclusions

Hospice care policies in China are considered comprehensive and rational, yet they have the potential for further improvement, especially in the areas of policy equity, guarantee of implementation, timeliness coordination, and multi-agent collaboration. Moreover, a series of practical national strategies play a significant role in the policy system. These findings offer valuable insights into the strengths and limitations of China’s hospice care policies, particularly from perspectives of policy equity and policy innovation. This serves as a reference for establishing a supportive policy environment conducive to the high-quality development of hospice care.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12904-025-01877-1.

Introduction

In response to the global trend of aging populations, addressing the significant gap in end-of-life care availability has emerged as a critical issue in healthcare [1]. Hospice care policies (HCPs) are widely recognized as crucial, and the World Health Organization (WHO) urges all countries and regions to create supportive policy environments, highlighting that policy reform is essential for improving hospice care practices [2]. Despite the growing focus on HCPs, a survey released in 2020 revealed that only 28% of countries and regions have established national strategies or plans, primarily in high-income and upper-middle-income countries [3]. This indicates that policy gaps remain a significant barrier to the widespread adoption of hospice care. As one of the world’s most populous countries, China faces a rapidly growing demand for hospice care, yet its service capacity remains severely limited, resulting in a comparatively low quality of death [4]. A crossnational survey by Finkelstein et al. ranked China 54th of 81 jurisdictions worldwide on the quality of death and dying [5]. To address this gap, China introduced its first national hospicecare programme and service guidelines in 2017 and initiated pilot projects in five districts [6]. To date, there have been three batches of national trials, covering a total of 185 cities across the country. This has led to the introduction of a series of widespread policies, marking a significant milestone in the institutional development of hospice care in China.

As hospice care is now formally institutionalized in China, a systematic review of the emerging policy framework is both timely and necessary. Official guidelines define hospice care as a multidimensional service for terminally ill patients—typically those with a life expectancy of six months or less—and their families, covering symptom control, comfortfocused nursing, psychological counselling, linking patients with social services, family support, and bereavement care [7]. As part of the national publichealth strategy, the central policy advocates a multitiered servicedelivery system that integrates hospice care across hospital, community, and home settings [8]. Consistent with China’s pilotdriven approach to policy innovation, these national guidelines serve chiefly as a strategic compass and legal mandate, while the design and implementation of concrete measures are largely left to local governments [9, 10].

However, a crucial question remains largely unanswered: to what extent can China’s current HCPs effectively support the development of hospice care practices? Key findings of existing studies reveal critical links between policy constraints and practice barriers. First, because no dedicated medical insurance payment scheme exists for hospice care, the prevailing fee-for-service model leaves many essential services outside the reimbursement system, indicating the absence of policy-making in key areas [11]. Second, although numerous training initiatives seek to alleviate workforce shortages, studies suggest their influence is limited without stronger backing from the public health and education sectors [12], reflecting structural hindrance to policy effectiveness caused by the lack of systematic coherence. Finally, fragmented policy regimes and divergent regional regulations create wide disparities in service accessibility, both between urban-rural areas and among provinces [13, 14].

Given the significance of a consistent and systematic policy environment for supporting real-world practices, it is imperative to conduct a rigorous, evidencebased assessment of China’s existing HCPs framework—something largely absent in current research. Policy evaluation is the key to unlocking the “black box” of governmental decisionmaking [15]. By applying appropriate theoretical frameworks and quantitative methods, policy evaluation provides empirical evidence for assessing policy implementation [10, 11]. The Policy Modeling Consistency Index (PMC-Index) constructs a unified evaluation index system based on policy texts, enabling quantitative assessments of both individual and multiple policies. This model has the significant advantage of not only evaluating the inherent consistency in policy making, but also comprehensively and intuitively reflecting the strengths and weaknesses of policies across various temporal, geographical, and stakeholder dimensions. It addresses the limitations of many policy evaluation models, which often lack representative indicators and overlook key elements of policy texts [16]. The PMC-Index model has been extensively applied in evaluating various policy issues, particularly in public health, including traditional Chinese medicine policies [17], health promotion policies [18], and emergency response policies for public health events [19], demonstrating its applicability to HCPs. Policy evaluation research in hospice care has garnered some attention, with studies in the United States and Europe establishing specific policy evaluation index systems through content analysis and cross-sectional surveys [19, 20]. However, to our knowledge, no research has yet applied the PMC-Index approach to evaluate HCPs.

To bridge this academic gap, we aim to assess the extent to which HCPs in China demonstrate measurable coherence, with the goal of providing evidence to support the development of a more enabling policy environment. This study follows a three-step process. First, we formalize key elements of the policy system through an inductive, semi-automated text mining analysis of HCPs issued by central, provincial, and local governments in China. Second, we construct a comprehensive evaluation index system based on the PMC-Index model, enabling the quantitative assessment of both overall policy performance and individual policy dimensions. Finally, we generate index scores for representative sub-samples of Chinese HCPs, identifying policy elements that exhibit high consistency across documents and those where coherence remains weak. We further discuss how these findings may inform evidence-based pathways for future policy refinement and optimization.

Materials and methods

Theoretical basis and model construction

The PMC-Index model, introduced by Estrada (2010), provides an evidence-based, multidisciplinary framework designed for comprehensive policy evaluation. Rooted in the Omnia Mobilis assumption, this model highlights that the absence of non-economic indices in policy modeling significantly increases the vulnerability of any policy [21]. Therefore, effective policy modeling should encompass diverse economic, social, political, technological, and environmental dimensions, dynamically reflecting changes across temporal and spatial contexts. This model refrains from ranking indicators according to subjective importance, as this evaluation primarily concerns the integrity of a holistic outcome and its internal equilibrium [22, 23].

The PMC-Index model was originally proposed to assess the consistency level of policy modeling, constructing a general index system with fifty sub-indicators distributed in ten main dimensions [16]. Multiple studies have leveraged analysis of empirical contexts and policy texts to enhance indicator design, expanding the comprehensiveness and extensiveness of policy evaluation while improving analytical specificity [24, 25]. Its capability to systematically quantify textual completeness and coherence makes it particularly suitable for the Chinese governance context, characterized by policy-driven administration and hierarchical document-based management.

In this study, the PMC-Index model was specifically adapted to evaluate Chinese hospice care policies through a structured methodological approach comprising three key stages: (1) data preparation; (2) classification of indicators and identification of parameters; and (3) measurement of the PMC-Index. The construction framework is illustrated in Fig. 1.

Fig. 1.

Construction framework for the PMC-Index model

Data collection

We searched policy documents on keywords including “hospice care”, “palliative care”, and “endoflife care” published up to 5 April 2024. Sources included the official websites of national, provincial, and municipal government departments, the PKU Law database, and the CNKI PolicyText database. To ensure thematic relevance, we retained only documents whose primary objective was the development of hospice care and in which the search keywords appeared at least twice. We further limited the corpus to programmatic documents that explicitly promote hospice care—namely plans, outlines, opinions, measures, and notices with clear administrative authority—while excluding reports, letters, approvals, and any texts in which hospice care was mentioned only tangentially.

After applying these criteria, a total of 112 HCP documents were obtained, comprising 19 national-level policies and 93 provincial-local-level policies. For specialized policies, the full text was included, while only highly relevant parts of non-specialized policies were selected.

Data preparation

Text coding, segmentation, and high-frequency word statistics were conducted on all HCP texts in the database using ROST CM6 text mining software. Following secondary processing, high-frequency words lacking content differentiation or research significance, such as “hospice care” and “according to,” were filtered out. Additionally, words with similar meanings within the study, such as “evaluation” and “assessment,” were merged. Ultimately, the top 60 effective keywords by frequency are shown in Table 1, providing a reference for the subsequent indicator-setting process.

Table 1.

Statistics of the top 60 effective words of HCPs

NO.	Word	Frequency	NO.	Word	Frequency	NO.	Word	Frequency
1	Service	1673	21	Staff	227	41	Publicity	116
2	Agency	997	22	Principle	221	42	Inpatient	116
3	Medical	860	23	Education	211	43	Ward	116
4	Patients	828	24	Community	201	44	Bed	112
5	Nursing	607	25	Home-based	198	45	Team	112
6	Pilot	601	26	Family	193	46	Facility	110
7	Health	530	27	Training	192	47	Insurance	108
8	Implement	484	28	Treatment	189	48	Pricing	107
9	Center	421	29	Development	181	49	Senior care	99
10	Management	354	30	Organization	170	50	Promotion	96
11	System	352	31	The elderly	170	51	Research	95
12	Strength	323	32	Incentive	167	52	Demonstration	92
13	Construction	318	33	Mechanism	166	53	Payment	92
14	Hospital	290	34	Medicine	157	54	Supervision	91
15	Instruction	275	35	Need	153	55	Rural	91
16	Evaluation	259	36	Government	145	56	Financial	90
17	Quality	258	37	Technology	138	57	TCMa	90
18	Standard	258	38	Disease	129	58	Market	90
19	Social	251	39	Referral	127	59	Innovation	89
20	Coordination	238	40	Reform	126	60	Funding	88

^aTCM Chinese Traditional Medicine

Indicator design and scoring criteria

To rigorously evaluate the quality and internal coherence of hospice care policies, we developed an objective, systematic, and quantifiable indicator system. Following the design logic of the PMC-Index model, which emphasizes the inclusion of multidisciplinary and even “unforeseen” factors, we assigned equal weight to all secondary indicators and adopted a binary rule: 1 if the policy text satisfies the criterion, 0 otherwise.

The index system was constructed using three sources: (1) an inductive, semi-automated text mining of HCPs issued at central, provincial, and local levels; (2) prior applications of the PMC-Index in social policy evaluation [23, 24, 26, 27]; and (3) adaptation to the institutional logic of China’s hospice care system. Based on these inputs, the two authors independently drafted the initial indicator set and refined it through iterative discussion.

To ensure scientific validity and policy relevance, an expert advisory panel of three senior professors in palliative care and health policy reviewed the framework from the angles of theory, practice, and evaluation methods. After incorporating their feedback, we finalized a multi-dimensional index comprising 10 primary and 47 secondary indicators, covering dimensions including Policy nature (X₁), Policy timeliness (X₂), Policy tools (X₃), Policy fields (X₄), Policy objects (X₅), Policy equity (X₆), Policy content (X₇), Policy guarantee (X₈), Policy structure (X₉), and Policy innovation (X₁₀) (see Table 2).

Table 2.

Indicators setting of quantitative evaluation of HCPs

Primary indicators	Secondary indicators	Evaluation criterion
X1 Policy nature	X1−1 forecast	Does the policy reflect predictions? If it does, the value is 1; if not, the value is 0.
	X1−2 supervise	Does the policy include provisions for supervision? If it does, the value is 1; if not, the value is 0.
	X1−3 instruct	Does the policy include instructional elements? If it does, the value is 1; if not, the value is 0.
	X1−4 describe	Does the policy provide a description of the current status? If it does, the value is 1; if not, the value is 0.
	X1−5 suggest	Does the policy provide suggestions? If it does, the value is 1; if not, the value is 0.
	X1−6 support	Does the policy include supportive initiatives? If it does, the value is 1; if not, the value is 0.
X2 Policy timeliness	X2−1 long Term	Does the policy include plans for more than 5 years? If it does, the value is 1; if not, the value is 0.
	X2−2 medium-term	Does the policy include plans for 3–5 years? If it does, the value is 1; if not, the value is 0.
	X2−3 short term	Does the policy include plans for less than 3 years? If it does, the value is 1; if not, the value is 0.
X3 Policy tools	X3−1 demand-oriented	Does the policy apply demand-oriented policy tools? If it does, the value is 1; if not, the value is 0.
	X3−2 supply-oriented	Does the policy apply supply-oriented policy tools? If it does, the value is 1; if not, the value is 0.
	X3−3 environment-oriented	Does the policy apply environment-oriented policy tools? If it does, the value is 1; if not, the value is 0.
X4 Policy fields	X4−1 politics	Does the policy involve the political field? If it does, the value is 1; if not, the value is 0.
	X4−2 economy	Does the policy involve the economic field? If it does, the value is 1; if not, the value is 0.
	X4−3 technology	Does the policy involve the technological field? If it does, the value is 1; if not, the value is 0.
	X4−4 society	Does the policy involve the social field? If it does, the value is 1; if not, the value is 0.
	X4−5 medical service	Does the policy involve the field of medical service? If it does, the value is 1; if not, the value is 0.
X5 Policy objects	X5−1 government sections	Does the policy have an impact on government sections? If it does, the value is 1; if not, the value is 0.
	X5−2 public medical service agencies	Does the policy have an impact on public medical service agencies? If it does, the value is 1; if not, the value is 0.
	X5−3 enterprises	Does the policy have an impact on enterprises? If it does, the value is 1; if not, the value is 0.
	X5−4 social organizations	Does the policy have an impact on social organizations? If it does, the value is 1; if not, the value is 0.
	X5−5 research institutions	Does the policy have an impact on research institutions? If it does, the value is 1; if not, the value is 0.
	X5−6 the public	Does the policy affect communities, families, and individuals? If it does, the value is 1; if not, the value is 0.
X6 Policy equity	X6−1 rural-urban coordination	Does the policy involve initiatives to promote rural and suburban services? If it does, the value is 1; if not, the value is 0.
	X6−2 service Affordability	Does the policy involve affordable cost control and medical assistance? If it does, the value is 1; if not, the value is 0.
	X6−3 service accessibility	Does the policy involve initiatives to develop home-based and community-based services? If it does, the value is 1; if not, the value is 0.
	X6−4 specific group support	Does the policy involve initiatives to support specific groups? If it does, the value is 1; if not, the value is 0.
X7 Policy content	X7−1 service system	Does the policy content involve service system development? If it does, the value is 1; if not, the value is 0.
	X7−2 institution construction	Does the policy content involve the construction of service institutions? If it does, the value is 1; if not, the value is 0.
	X7−3 working mechanism	Does the policy content involve assessment and referral mechanisms? If it does, the value is 1; if not, the value is 0.
	X7−4 medicine and equipment	Does the policy content involve medicine and equipment management? If it does, the value is 1; if not, the value is 0.
	X7−5 pricing and payment	Does the policy content involve pricing and the social insurance payment system? If it does, the value is 1; if not, the value is 0.
	X7−6 standard specification	Does the policy content involve standard specifications? If it does, the value is 1; if not, the value is 0.
	X7−7 education and training	Does the policy content involve education and practitioner training? If it does, the value is 1; if not, the value is 0.
	X7−8 social propaganda	Does the policy content involve social propaganda? If it does, the value is 1; if not, the value is 0.
	X7−9 quality monitoring	Does the policy content involve service quality monitoring? If it does, the value is 1; if not, the value is 0.
	X7−10 needs and rights	Does the policy content involve the needs and rights of patients and their families? If it does, the value is 1; if not, the value is 0.
X8 Policy guarantee	X8−1 leading group	Does the policy establish a leading group? If it does, the value is 1; if not, the value is 0.
	X8−2 execute plan	Does the policy include an executive plan? If it does, the value is 1; if not, the value is 0.
	X8−3 financial support	Does the policy include financial support? If it does, the value is 1; if not, the value is 0.
	X8−4 period schedule	Does the policy clarify the schedule of implementation phases? If it does, the value is 1; if not, the value is 0.
	X8−5 performance assessment and incentives	Does the policy include performance assessment and incentives? If it does, the value is 1; if not, the value is 0.
	X8−6 team development	Does the policy include team development? If it does, the value is 1; if not, the value is 0.
X9 Policy structure	X9−1 reliable basis	Does the policy have a reliable legal basis? If it does, the value is 1; if not, the value is 0.
	X9−2 specific goals	Does the policy have specific goals? If it does, the value is 1; if not, the value is 0.
	X9−3 detailed planning	Does the policy have the detailed implementation planning? If it does, the value is 1; if not, the value is 0.
	X9−4 scientific strategies	Does the policy have evidence-based strategies? If it does, the value is 1; if not, the value is 0.
X10 Policy innovation	—	Does the policy include innovative measures beyond existing and superior policies? If it does, the value is 1; if not, the value is 0.

For parameter assignment, we applied a dual-coding procedure. Two researchers independently coded each policy using the binary scheme, yielding a mean Cohen’s κ of 0.84 across all indicators—indicating strong inter-coder reliability. Any discrepancies were resolved through structured discussion with the expert panel until the consensus was reached.

In sum, our indicator development process combines data-driven design with expert-informed validation, ensuring methodological rigor, reliability, and practical applicability for real-world policy assessment.

PMC-Index calculation

The calculation of the PMC-Index in this study follows steps below. First, the values of primary indicators are calculated by the sum of secondary indicator scores divided by the total number of subordinate secondary indicators according to Eq. (1). In this formula, 𝑡 represents the primary indicator, 𝑗 represents the secondary indicator, and 𝑇 is the number of secondary indicators within each primary indicator. The differences in each primary indicator reflect the internal consistency level of each policy.

Second, Eq. (2) is used to sum the values of the primary indicators to obtain the overall PMC-Index. Based on the PMC-Index scores, HCPs are categorized into four levels, as defined in previous literature [28]: Excellent (9.00–10.00), Good (7.00-8.99), Acceptable (5.00-6.99), and Low (0.00-4.99).

1

2

Results

The inter-annual analysis of the number of HCPs issued

As illustrated in Fig. 2, the years 2017, 2020, and 2023 represent significant milestones for the issuance of HCPs in China. Accordingly, we divided the development process of HCPs into four stages. First Period (before 2017): The number of HCPs issued remained relatively low. During this stage, the issuance of HCPs was primarily driven by the independent exploration of a few districts, as national strategies had not yet been introduced. Second Period (2017–2019): With the National Health and Family Planning Commission issuing “The Notice on the pilot work of hospice care,” 2017 marked the first critical milestone in the development of HCPs in China. Following national policy guidance, pilot areas formulated provincial-local-level practical strategies, resulting in a significant increase in HCP publications. Third Period (2020–2022): With the release of the second batch of national pilot programs in December 2019, the scope of pilot areas continuously expanded, culminating in a record-high number of HCPs issued in 2020. Fourth Period (2023 to present): The release of the third batch of national pilot plan in April 2023 further expanded the scope of pilot areas, resulting in another peak in the number of HCPs issued in 2023. In summary, the inter-annual changes in the volume of HCPs issued exhibit a growth trajectory, reflecting a significant policy diffusion path dominated by national practice plans, followed by provincial and local policies.

Fig. 2.

Inter-annual variation curve of the number of HCPs issued

Empirical study on policy evaluation

Selection of policy samples

The PMC-Index model aims to objectively consider all indicators without imposing any special requirements on the evaluation object. To comprehensively reflect the differences between the periods and administrative levels of issued HCPs, we adopted the following sample selection strategy. Firstly, based on the findings of the inter-annual analysis, three national pilot work strategies (P₁₈, P₃₇, and P₈₁) were prioritized as policy samples. Secondly, one national-level policy and three provincial-local-level policies were randomly selected from each period. Notably, since no national policies meeting our criteria were issued before 2017, only three provincial- and local-level policies were selected for the first period. The final selection of 18 HCPs samples is shown in Table 3.

Table 3.

Sample policies for PMC-Index model evaluation

Code	Policy name	Issuing agency	Date
P1	Notice on the 2012 Municipal government practical palliative care (hospice care) project	Shanghai Municipal Health Bureau, Shanghai Municipal Finance Bureau, Shanghai Municipal Human Resources and Social Security Bureau	2012.04.27
P5	Notice of Qingdao Civil Affairs Bureau and Municipal Health and Family Planning Commission on the development of hospice Care	Qingdao Civil Affairs Bureau, Qingdao Health and Family Planning Commission	2015.04.16
P6	Notice on the implementation plan of the pilot work of Inpatient palliative care (Hospice care) for advanced Cancer Patients	Jilin Human Resources and Social Security Bureau	2015.06.25
P8	Notice on the issuance of Basic Standards and Management Norms for Hospice Centers (trial)	the National Health and Family Planning Commission	2017.01.25
P14	Notice of the General Office of the Hebei Health and Family Planning Commission on the implementation of Hospice Care Practice Guidelines (trial)	the General Office of the Hebei Health and Family Planning Commission	2017.03.16
P18	Notice of the General Office of National Health and Family Planning Commission on the pilot work of hospice care	the General Office of National Health and Family Planning Commission	2017.10.27
P24	Notice of the General Office of the Kunming Health and Family Planning Commission on the implementation plan of hospice care	the General Office of the Kunming Health and Family Planning Commission	2018.07.23
P26	Notice of the General Office of the Fujian Provincial Health Commission on the pilot work of hospice care	the General Office of the Fujian Provincial Health Commission	2018.11.13
P36	Implementation opinions of Lianyungang Municipal Health Commission on the pilot work of hospice care	Lianyungang Municipal Health Commission	2019.12.03
P37	Notice of the General Office of the National Health Commission on the second batch of the pilot work of hospice care	the General Office of the National Health Commission	2019.12.05
P52	Notice of Shanghai Municipal Health Commission on the issuance of Shanghai Hospice Service Standards	Shanghai Municipal Health Commission	2020.08.05
P55	Notice of Dalian Medical Security Bureau on hospice settlement by average cost of beds-based payment	Dalian Medical Security Bureau	2020.10.03
P61	14th Five-Year Plan of national old-age development and old-age service system planning	the State Council	2021.12.30
P78	Notice of Qingdao Municipal Health Commission on the issuance of Qingdao Hospice Service Standards	Qingdao Municipal Health Commission	2023.01.09
P81	Notice of the General Office of the National Health Commission on the third batch of the pilot work of hospice care	the General Office of the National Health Commission	2023.04.11
P85	Notice of the General Office of the Qingyang Municipal People’s Government on the implementation plan of the pilot work of hospice care	the General Office of the Qingyang Municipal People’s Government	2023.06.09
P94	Notice of the General Office of the Yichang Municipal People’s Government on the implementation plan of hospice care	the General Office of the Yichang Municipal People’s Government	2023.09.01
P99	Notice on the issuance of guidelines for Home and Community Integrated Medical and Nursing Services (trial)	the General Office of the National Health Commission; State Bureau of Traditional Chinese Medicine General Department; National Disease control Bureau General Department	2023.11.01

The PMC-Index for each policy was then calculated, and the policies were ranked and classified (see Additional file 1). Results were organized into a score table and presented through a heatmap to visually represent the data (see Fig. 3). According to the classification in relevant studies [23], each primary indicator is divided into the following five levels: “excellent performance” (0.90-1.00), “good performance” (0.70–0.89), “acceptable performance” (0.50–0.69), “non-satisfactory performance” (0.30–0.49), and “poor performance” (0.00-0.29), which forms the basis of designing specific policy improvement pathways.

Fig. 3.

The PMC-Index values and grades of HCPs samples

Holistic evaluation of policy samples

The total mean value of the PMC-Index for the 18 HCPs is 7.20, which falls within the “Good” grade, indicating general positive consistency. Policy examples were evaluated across all five grades, with the following distributions: two policies at the Excellent level, eight at the Good level, six at the Acceptable level, and two at the Low level. This distribution reflects commendable overall performance, although there are significant differences among individual policies.

The primary indicators of Policy nature (X₁) and Policy tools (X₃) are rated as “excellent performance.” The PMC-Index value for X₃ is 0.93, the highest score among all the primary indicators. 14 policy samples adopted all three types of policy tools simultaneously, indicating that HCPs employ a combination of multiple policy tools, integrating demand satisfaction, service supply, and supportive environment construction. Meanwhile, the score for X₁, Policy Nature, is 0.90, reflecting the sufficient policy functions adopted by HCPs, which promote comprehensive strategies based on the policy context and reality.

The primary indicators of Policy fields (X₄), Policy structure (X₉), Policy content (X₇), and Policy innovation (X₁₀) are rated as “good performance.” Policy fields (X₄): The mean value of X₄ is 0.89. However, the secondary indicator economy (X₄₋₂) has the lowest value, suggesting that HCPs prioritize policy design in the areas of politics, technology, society, and medical services, while relatively overlooking economic initiatives. Policy structure (X₉): The PMC-Index value of X₉ is 0.83. This indicates that clear objectives, sufficient legal support, and scientific justification are consistently included in HCPs. However, the lack of feasible implementation plans remains a limitation. Policy content (X₇): The score of X₇ is 0.75. Among the 10 identified HCP content themes, four secondary indicators scored below average: medicine and equipment (X₇₋₄), pricing and payment (X₇₋₅), social propaganda (X₇₋₈), and quality monitoring (X₇₋₉). These areas are crucial for ensuring the promotion of services, and their deficiencies compromise the comprehensiveness of policy content. Policy innovation (X₁₀): The average value of X₁₀ is 0.78. 14 selected HCPs implemented effective innovative measures that promote the optimization of policies and their practical applicability. Conversely, 4 policies were mere transmissions of superior policies or summaries of past policies. These evaluations illuminate strengths and weaknesses within HCPs, offering an evidence-based foundation for developing more comprehensive and effective strategies.

Policy objects (X₅) and Policy timeliness (X₂) are rated as “acceptable performance,” with average values of 0.68 and 0.59, respectively, leading to relatively low consistency. Policy objects (X₅): Among the secondary indicators of X5, research institutes (X₅₋₄) display the lowest score, indicating a lack of involvement from research institutions in the policy formulation process. Additionally, the scores for enterprises (X₅₋₃) and social organizations (X₅₋₄) are below average, suggesting a scarcity of collaborative participation from private service agencies, social work organizations, and charitable institutions in hospice care. Policy timeliness (X₂): The analysis of X₂ reveals that most HCPs primarily focus on long-term guidance and short-term work plans but lack effective transitions and connectivity. This leads to the lowest average score for mid-term planning (X₂₋₂). These evaluations suggest areas where HCPs could improve by enhancing collaboration with research institutions and private organizations and by developing more coherent mid-term planning strategies.

Policy guarantee (X₈) and Policy equity (X₆), rated as “non-satisfactory,” are significant barriers to the consistency of selected HCPs. Policy guarantee (X₈): With a PMC-Index value of 0.45, this indicator highlights the absence of financial support, operational advancement, and performance assessment. These deficiencies hinder the effective implementation of HCPs and their translation into tangible health services. Policy equity (X₆): The PMC-Index score of 0.40 reveals a systematic omission in incorporating equity into current HCPs. Service accessibility (X₆₋₃): Attains the highest mean score of 0.72 among the four secondary indicators, suggesting that HCPs widely emphasize supporting home-based services, broadening paths for hospice care delivery. Specific group support (X₆₋₄): Scores of 0.33 indicate that while elderly individuals are predominantly targeted by most HCPs, other specific patient groups receive inadequate attention. Rural-urban coordination (X₆₋₁) and service affordability (X₆₋₂): Both scores are 0.28. Most HCPs focus on urban areas, disregarding significant disparities in medication availability, infrastructural development, human resources, and societal perceptions in rural areas. Specific rural support measures are largely ignored. Additionally, while numerous policies emphasize enhancing the availability of social insurance, few advocate for hospice assistance strategies for economically disadvantaged groups. These evaluations identify key areas for improvement, emphasizing the need for better financial support, operational strategies, and a more equitable approach to policy formulation and implementation.

Comparative analysis of national stand-alone strategies

National strategies constitute the fundamental backbone and guiding principles of HCPs, exerting a normative and directive influence on provincial-local policies. Therefore, we analyzed three national stand-alone strategies (P₁₈, P₃₇, and P₈₁) issued during different national pilot programs.

Among the three policies, P₁₈, Notice of the General Office of National Health and Family Planning Commission on the pilot work of hospice care, represents the inaugural national pilot plan for hospice care in 2017. It scored 7.67 on the PMC-Index and ranked ninth among the 18 policy samples. Subsequently, P₃₇, Notice of the General Office of the National Health Commission on the second batch of the pilot work of hospice care issued in 2019, scored 8.25 on the PMC-Index, ranking seventh. Finally, P₈₁, Notice of the General Office of the National Health Commission on the third batch of the pilot work of hospice care issued in 2023, attained a PMC-Index score of 8.67, ranking third. All three policies were classified within the Good grade, collectively demonstrating a commendable degree of consistency. The data show a notable enhancement in policy quality across the three pilot periods.

In exploring the optimization paths among the three strategies, initial improvements were observed in Policy timeliness (X₂) and Policy structure (X₉). Compared to P₁₈, which primarily outlined the operational framework for the inaugural pilot phase, P₃₇ established a sustainable guidance mechanism, thereby improving the score for X₂. Additionally, P₃₇ provided more detailed execution plans for the various policy designs introduced in P₁₈. For instance, while P₁₈ proposed the inclusion of hospice care within the social insurance system, it lacked a concrete implementation plan. Accordingly, P₃₇ clarified the average cost of bed-based payment approach, providing a specific roadmap for payment reform, and thus enhancing the value of X₉. Building on this foundation, P₈₁ further refined its Policy objects (X₅) and Policy equity (X₆). Specifically, P₈₁ formally proposed setting up hospice beds in country health centers, promoting a comprehensive service system that includes both urban and rural areas, thereby improving the consistency index of X₆. Furthermore, P₈₁ engaged colleges and research associations in the education and training program, reflecting the continued evolution of the coordination mechanism and enhancing the PMC-Index score of X₅. Consequently, the evolutionary path of the national stand-alone strategies across the three national pilot programs can be summarized as X₂-X₉-X₆-X₅.

Specific evaluation of single policies

We conducted a micro-analysis of representative policies from four PMC-Index grades and devised respective optimization paths. We selected P₈₅, which had the highest PMC-Index score, and P₂₆, which had the lowest PMC-Index score, while also randomly drawing one policy each from the Good level (P₅) and the Acceptable level (P₅₅). Policies P₁₈, P₃₇, and P₈₁ were excluded from the sampling frame. All selected policies are provincial-local policies.

We visually presented outcomes from a multi-dimensional perspective using the PMC-Surface [16]. According to Formula (3), the PMC-Index scores for each policy were organized into a surface matrix. Notably, Since primary indicator Policy innovation (X₁₀) is endorsed to capture any innovative measures taken by policymakers, it does not contain any secondary indicators, and it is eliminated to maintain matrix symmetry. As depicted in Figs. 4, 5, 6 and 7, the lower the depression degree of the PMC-surface, the higher the internal consistency of the policy across all content dimensions; meanwhile, the higher the horizontal position of the surface, the more comprehensive of the policy text.

Fig. 4.

The PMC-Surface of P₈₅

Fig. 5.

The PMC-Surface of P₅

Fig. 6.

The PMC-Surface of P₅₅

Fig. 7.

The PMC-Surface of P₂₆

3

P₈₅, Notice of the General Office of the Qingyang Municipal People’s Government on the implementation plan of the pilot work of hospice care, represents a local-level implementation plan of Qingyang City during the third national pilot phase. With a PMC-Index value of 9.58, it attained the highest rating of Excellent among all policies, indicating remarkable consistency across various policy dimensions, as all indicators surpassed the average. Among the ten indicators, only Policy equity (X₆) and Policy guarantee (X₈) fell short of the maximum score, resulting in a moderate degree of surface depression. This is primarily attributable to the absence of targeted relief mechanisms serving disadvantaged and uninsured populations, along with the lack of specific evaluation criteria for implementation outcomes. Future policy revisions could prioritize the pathway of X₆-X₈, strengthening equity mechanisms and refining practical implementation strategies.

P₅, Notice of Qingdao Civil Affairs Bureau and Municipal Health and Family Planning Commission on the development of hospice Care, introduced by the Qingdao Municipal Government as an independently formulated local development project in 2014, achieved a consistency index of 8.40, categorized as Good and ranking fifth among all samples. As a locally initiated strategy, P₅ exhibits commendable policy innovation and comprehensive policy design considerations, displaying robust consistency. Notably, P₅ is the only individual policy among the 18 sampled policies to attain a perfect score in Policy equity (X₆). By advocating programs such as hospice-involved rural elderly care centers, hospice assistance systems, home-based mutual assistance hospice care projects, and end-of-life care security systems for disabled elderly, it comprehensively addresses the equity perspective across four secondary indicators in the study. Among other dimensions, Policy nature (X₁) and Policy structure (X₉) lagged below the average, while Policy timeliness (X₂) and Policy assurance (X₈) also contributed to significant surface depression. The primary reason is the emphasis on macro-level institutional design for hospice care development instead of clear implementation plans, short-term work agendas, and supervisory mechanisms. Moreover, having been issued prior to the national framework, it lacks policy alignment and legal anchoring. Future policy updates should integrate national directives and specify actionable implementation pathways to improve policy consistency and applicability. Thus, this policy could follow an optimization pathway of X₉-X₁-X₂-X₈.

P₅₅, Notice of Dalian Medical Security Bureau on hospice settlement by average cost of beds-based payment, is a dedicated payment-reform policy of Dalian City during the second pilot phase. With a PMC-Index consistency score of 6.07, classified as Acceptable, it ranks 14th among the reviewed policies. The policy follows national reform guidelines while setting localized payment standards tailored to regional economic conditions, thereby achieving above-average scores in Policy fields (X₄), Policy structure (X₉), and Policy innovation (X₁₀). Nevertheless, it scores lower for primary indicators X₁, X₂, X₃, X₅, X₆, X₇, and X₈. This is primarily due to the micro-orientation of the policy, which limits the range of content, target audience, and functionality while lacking long-term standard adjustment plans. Notably, as it exclusively specifies inpatient service payment standards for urban service providers and restricts eligibility to locally insured individuals, the absence of equity-enhancing provisions may exacerbate regional disparities. To improve policy coherence and effectiveness, the recommended optimization pathway for P₅₅ is X₆-X₇-X₂-X₃-X₅-X₈-X₁. This pathway is designed to strengthen equity-orientation measures, broaden content coverage, and enhance timeliness and inclusiveness.

P₂₆, Notice of the General Office of the Fujian Provincial Health Commission on the pilot work of hospice care, a pilot work plan of Fujian Province during the first batch of national pilots, received a PMC-Index score of 4.73, categorized as Low, and ranks 18th. As evident from the PMC-Surface chart, P₂₆ exhibits considerable surface depression, indicating poor consistency across multiple dimensions. Content analysis reveals that the policy adopts a task-oriented rather than service-oriented approach, largely replicating central directives without contextual adaptation or practical implementation strategies. Consequently, its influence on hospice care practices remains limited. Future enhancements should prioritize fostering local-level innovation, followed by a comprehensive refinement and elaboration of the policy. The recommended refinement roadmap is X₁₀-X₆-X₅-X₉-X₈-X₂-X₇-X₄. This approach emphasizes policy innovation, equity enhancement, objective clarity, structural robustness, assurance measures, timeliness, content richness, and field inclusivity.

Discussion

Policy consistency and coherence are critical for translating intentions into effective implementation. Using the PMC-Index model, this study provides a quantitative evaluation of China’s hospice care policies, identifying both facilitating factors and structural weaknesses, and proposing directions for refinement grounded in empirical patterns.

Firstly, the overall consistency of China’s HCPs is relatively strong, with an average indicator score of 7.20 across 18 policies. Most documents reflect thoughtful design, with only two showing notably low consistency. Coherence has generally improved over time, especially across the three national pilot phases, aligning with previous findings that policy-driven service expansion follows a phased evolution from experimentation to institutionalization [10, 12].

Secondly, policy consistency varies significantly across dimensions. High scores in Policy nature (X₁), Policy tools (X₃), and Policy fields (X₄) suggest solid technical design, while persistent weaknesses in Policy timeliness (X₂), Policy objects (X₅), and Policy guarantees (X₈) reveal gaps in support mechanisms. As robust enabling conditions are vital to implementation success [29], addressing these deficits is essential for turning design into delivery. Equity remains a particularly underdeveloped area. Consistent with prior studies, this research confirms rural–urban disparities in hospice access [30, 31], financial burdens among low-income and uninsured populations [32], and institutional inadequacies in caring for children and non-cancer patients [33, 34]. Additionally, only a limited number of equity-related terms were identified via text mining, with critical structural factors—such as gender, cognitive impairment, and intersectional disadvantage—largely absent [35]. These findings suggest that the neglect of equity may be more profound than currently measurable. We strongly recommend embedding equity as a foundational principle throughout the policy lifecycle.

Thirdly, national-level planning has provided the backbone for hospice policy development, but local innovation has been indispensable for contextual adaptation [36]. Key national directives—such as the 2017 national pilot guidelines and the integration into the National Basic Public Service Standards (2021 Edition)—established the overarching framework. Subnational policies followed these central strategies while introducing context-specific innovations, reinforcing the dynamic interplay between top-down coordination and bottom-up experimentation [37, 38]. Meanwhile, our findings suggest that subnational policies incorporating innovative elements systematically outperformed conventional ones in policy coherence. This confirms that local experimentation not only addresses contextual implementation bottlenecks but also informs national policymaking through institutional learning. We thus recommend incentivizing local innovation and enhancing horizontal and vertical diffusion mechanisms to support system-level adaptability.

This study contributes to the literature in several important ways. Unlike most prior research that has relied on qualitative case studies [39–41], this study applies the PMC-Index as a standardized, data-driven framework to assess policy consistency in a systematic manner. In addition, building on Kuang’s argument [25] regarding the need for context-specific indicators, we introduce two novel dimensions—Policy equity and Policy innovation—to better capture implementation gaps and governance adaptability in the Chinese hospice context, where disparities in access and the need for institutional innovation are widely acknowledged [42, 43]. These enhancements improve both the diagnostic precision and the applicability of the evaluation framework for inclusive and forward-looking policy design.

Building on the diagnostic insights, the study moves beyond evaluation to demonstrate how these findings can inform practical optimization. Structural evaluation using the PMC-Index can support targeted, scalable improvement strategies. By mapping recurring deficiencies—especially in equity articulation (X₆), objective clarity (X₅), policy structure (X₉), and assurance mechanisms (X₈)—and examining variations in PMC-Surface topographies (e.g., from uniform consistency in P₈₅ to fragmentation in P₂₆), we highlight the multidimensional nature of coherence. Rather than treating such deficits as isolated flaws, we propose a structured pathway logic. For underperforming policies, refinement typically begins with strengthening innovation (X₁₀), followed by adjustments in equity design, target specification, structural coherence, and implementation support. This generalized pathway—abstracted from representative cases—offers a replicable roadmap for iterative policy revision and institutional learning across jurisdictions. This approach enhances the methodological value of structured policy appraisal. It also improves policy relevance by converting diagnostic insights into actionable reform strategies. Through such evidence-informed translation, policy evaluation not only benchmarks current quality but also supports the development of adaptive and resilient hospice care governance.

Finally, this study lays the groundwork for multi-level, temporal, and interregional policy comparisons. It reveals latent dynamics in central-local and inter-local interactions. For example, while recent payment system reforms appear as centrally-led initiatives, our findings suggest they originated in local pilot efforts before being formalized as national policy [44]. This insight helps illuminate how policy innovations emerge, diffuse, and become institutionalized across governance levels and actors [45].

Limitations

This study has several limitations. First, while the PMC-Index model effectively evaluates the internal consistency and structural completeness of policy texts, it does not directly reflect the actual outcomes of policy implementation, such as improvements in service accessibility, patient satisfaction, or quality of hospice care. Therefore, the evaluation conducted here focuses primarily on policy design rather than real-world effectiveness or impact. Second, the model assigns equal weights to all indicators, potentially overlooking interactions among them. Third, despite systematic text mining techniques, important yet less explicitly represented indicators may have been omitted due to their limited presence in existing policy documents. Lastly, the current study is limited to policy text analysis and lacks integration with empirical data from policy implementation, such as healthcare service utilization rates, medical insurance reimbursement records, and stakeholder experiences. Future studies should combine textual analysis with empirical impact assessments and qualitative stakeholder feedback, providing a more comprehensive and practical evaluation framework that bridges policy design and actual implementation outcomes.

Conclusion

This study utilized text mining and content analysis to examine 112 Chinese hospice care policy documents, identifying critical policy characteristics and elements. By integrating targeted dimensions of policy equity and innovation, a tailored PMC-Index model was constructed to comprehensively evaluate hospice care policies (HCPs) in China. Quantitative analysis of 18 representative policy samples was conducted, encompassing an overall policy assessment, comparative evaluations of national-level strategic policies, and detailed analyses of selected individual policies. Based on these analyses, specific pathways for policy enhancement were identified. The key findings indicate an average PMC-Index score of 7.20 for evaluated HCPs, reflecting a generally robust and well-structured policy framework. Specifically, the three national-level standalone policies demonstrated strong internal consistency and exhibited continuous improvement across successive implementation phases.

However, notable disparities exist among individual policies in terms of consistency levels and identified improvement pathways. Despite these strengths, the analysis highlights critical areas for future improvement. Policy recommendations emphasize the need to enhance strategic coordination, particularly regarding policy timeliness and planning continuity; encourage active and collaborative participation from diverse policy stakeholders; provide more substantial guarantees and resource allocation to ensure effective policy implementation; and systematically integrate equity considerations across all stages of policy formulation and execution.

In conclusion, applying the PMC-Index model with an emphasis on equity and innovation provides a robust and practical framework for evaluating and optimizing hospice care policies. This study enriches existing literature by offering a structured analytical approach, thereby providing valuable tools and actionable insights for policymakers, researchers, and healthcare practitioners committed to fostering supportive environments for high-quality hospice care development.

Abbreviations

HCPs

Hospice Care Policies

WHO

World Health Organization

PMC

Policy modeling consistency

TCM

Traditional Chinese medicine

Authors' contributions

LL and XG wrote the manuscript, interpreted the results, and edited each draft. LL designed the study, collected and analyzed the data. XG revised the language and provided administrative support. All authors reviewed and approved the final submitted version.

Funding

This study was funded by the National Natural Science Foundation of China (72574071).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.