Quantitative evaluation of China’s commercial health insurance policies based on the policy modeling consistency index model

Abstract

Objective

This study aims to quantitatively evaluate China’s commercial health insurance (CHI) policies and provide a reference for improving the policy.

Methods

By means of the text mining and literature research method, the Policy Modeling Consistency Index (PMC-index) model of CHI policies was constructed. The model was used to analyze the 22 policies individually and as a whole to describe the current situation of CHI policies in China.

Results

The PMC-index model consisted of 10 primary variables and 41 secondary variables. This study found that the average PMC index of the 22 policies included is 7.48, suggesting that existing policies are generally of high quality. Among them, good policies accounted for 18.18%, and excellent policies accounted for 81.82%. Among the primary variables, X10 (policy disclosure) had the highest score, and the scores of other primary variables were ranked as X3 (policy recipients) > X5 (policy content) > X6 (policy tools) > X8 (cooperation and connection) > X7 (content evaluation) > X1 (policy nature) > X9 (policy goals) > X4 (policy incentives) > X2 (policy timeliness).

Conclusions

At present, China’s CHI policies are generally at an excellent level. However, there is still room for improvement with regard to “policy incentives, policy goals, content evaluation, policy tools and policy content”. We recommend that policymakers optimize China’s CHI policies by improving the policy incentive mechanism; balancing policy goals; clarifying task distribution; optimizing policy tool structure; and developing inclusive health insurance.

Background

Chinese social health insurance, which consists of the Urban Employee Basic Medical Insurance scheme and the Urban-Rural Residents Basic Medical Insurance scheme, has covered more than 95% population by 2021 [1]. Basic medical insurance plays an important role in improving the resilience of households [2], narrowing the gap between urban and rural health service utilization, and improving regional equity [3]. However, the basic medical insurance has the problem of unbalanced and inadequate protection, especially the demand for protection of major diseases of Chinese residents is not met [4]. It is imperative to vigorously develop CHI and play its supplementary role to basic medical insurance.

The evolution of China’s CHI policy system embodies the strategic intent of the government to construct a multi-tiered medical security system. The “Decision of the State Council on Establishing the Basic Medical Insurance System for Urban Employees” in 1998 first established the supplementary role of CHI, stipulating that medical expenses exceeding the maximum payment limit of basic medical insurance could be addressed through avenues such as CHI [5]. Subsequently, the “Health Insurance Management Methods” implemented in 2006 regulated the operational behavior of market entities, laying an institutional foundation for the healthy development of the industry [6]. The “Opinions on Deepening the Reform of the Pharmaceutical and Health System” by the Central Committee of the Communist Party of China and the State Council in 2009 further clarified the functional positioning of CHI within the medical security system [7]. The issuance of the “Several Opinions on Accelerating the Development of Modern Insurance Services” by the State Council in 2014 [8] marked a shift in policy focus towards product innovation and service upgrading, driving the diversification of market supply. The “Notice on Carrying Out the Pilot Work of Personal Income Tax Policy for Commercial Health Insurance” in 2015 introduced tax preference policies, expanding the coverage of protection through incentive mechanisms, this innovation in policy tools held significant theoretical significance and practical value [9]. The inclusion of CHI into the national strategic level in the “Healthy China 2030” Plan Outline in 2016 highlighted its crucial role in achieving the goal of universal health [10]. The “Opinions on Deepening the Reform of the Medical Security System” in 2020 proposed policies to promote the convergence of CHI with basic medical insurance, demonstrating a systematic approach to institutional integration [11]. This series of policy evolutions underscore the significant role of CHI in perfecting China’s multi-tiered medical security system.

With these efforts, China’s CHI premium income reached 865.3 billion yuan [12], up 2.4% year-on-year by the end of 2022. However, according to the National Bureau of Statistics of China, CHI compensation spending was 360 billion yuan in 2022, accounting for only 4% of total healthcare spending. At the same time, China has a higher incidence of catastrophic health expenditures than countries with similar levels of economic development in the world [4]. This shows that the role of CHI in multi-level medical security is still limited. In addition, there is still a gap between the insurance density and depth of China’s CHI and that of countries implementing social medical insurance in the world [13]. Policy plays a guiding and promoting role in the development of CHI. In order to better develop CHI, two questions should be answered: What is the current status of CHI policies in China? From which dimensions do policymakers need to optimize?

The previous studies on CHI policies concentrated on evaluating the implementation effect of the policy [14,15,16], international comparison of CHI policies [17,18,19], and evaluating the policy design [20, 21]. In the study to assess the impact of policy implementation, the ARIMA model and the revenue-based method were used to measure the impact of CHI preferential tax policy implementation [15]. Another study in 2008 discussed in detail the preferential tax policies of health insurance in the developed countries, in order to provide experience for China’s relevant policy formulation [18]. One study evaluated the effectiveness of CHI policies issued nationally from 2002 to 2021 using the Policy Modeling Consistency Index Model (PMC-index model) [20]. To sum up, there are currently few studies evaluating CHI policies design, and it is only limited to evaluating it at the national level.

Policy evaluation plays an important role in understanding the pros and cons of policies and optimizing policy design. Given the critical role of CHI in supplementing basic medical insurance and addressing unmet healthcare needs, robust policy design directly influences the accessibility, affordability, and effectiveness of insurance coverage. Quantitative evaluation helps identify gaps and strengths within existing policies, guiding policymakers to develop more targeted, inclusive, and efficient strategies to enhance the role of CHI in China’s multi-tiered healthcare security system. The PMC-index model is a model for the quantitative evaluation of policy design [22]. The model is guided by the Omnia Mobilis hypothesis, which states that the role of any one related variable should not be ignored [23]. PMC-index model is suitable for quantitative evaluation of policies from the whole and various dimensions, and can also be used for comparative analysis of multiple policies. It has been used to policies’ quantitative evaluation in the fields of environmental protection, traditional Chinese medicine, health insurance and so on in China [20, 24,25,26], which have laid a foundation for applying PMC-index model to policy evaluation under China’s national conditions. Therefore, the objective of this study is to quantitatively evaluate the design of China’s CHI policies using the PMC-index model, and to propose targeted recommendations for policy optimization.

Methods

Data sources

Using “CHI” as the keyword, we searched the official websites of the national and provincial people’s governments to obtain the policies to be evaluated in this study. The retrieval time was from 1 January 2014 to 30 June 2024. The inclusion criteria were: (1) the title of the policies contains “CHI”; and (2) only one of the same policies searched on different websites is included. We excluded policies that only dealt with CHI but had no substantive content. According to the inclusion criteria, 45 policies were initially included, 23 policies with keywords but no substantive content were excluded, and a total of 22 policies on promoting the development of CHI were finally included, including 1 policy issued at the national level and 21 policies issued at the local level. Appendix 1 listed the detailed information of the policies included.

Research methods

Based on the PMC-Index model, this study quantitatively evaluated 22 core policies issued by the national and local governments. The specific steps to evaluate policy were: (1) setting variables and parameters; (2) constructing a multi-input-output table; (3) calculating and evaluating the PMC index.

The setting of variables and parameters

In this study, ROSTCM software was used for text mining of 22 policy documents included. On the basis of word segmentation analysis, common words such as “business”, “health” and “insurance” and noisy words such as “about”, “opinion” and “strengthen” were removed. In addition, words with similar meanings were merged, and the frequency of the merged words was the sum of the frequencies of all words before the merger. Finally, we extracted the top 20 most frequent words (see Table 1). These 20 high frequency words could be divided into four parts. The first part reflected the CHI policies’ recipients, such as “Urban and rural residents, Enterprises”. The second part reflected the incentives of CHI policies, such as “Individual account purchase, Information sharing, Talent team and Tax incentive”. The third part was the content of CHI policies, such as “Supervision, Risk control, Innovation, Handle and undertake, Expand supply.” The fourth part was about the cooperation and connection of CHI, such as “Cooperation and connection, Medical institution, Elderly care services.” Based on the PMC-index model proposed by Ruiz Estrada in 2011 [22], this study constructed the PMC-index model for the evaluation of CHI policies in China by referring to other relevant literatures [21, 24,25,26,27,28] and combining the results of word frequency analysis of CHI policies. The setting basis and meaning of each variable were shown in Table 2.

Table 1 The top 20 high-frequency words

Table 2 Variables setting of CHI policies

The PMC-index model assumes each variable is equally important [22]. Therefore, this study adopted the binary scoring method and set all secondary variables as binary classification variables (i.e., 0, 1 variables). The 22 included policies were analyzed individually. If the content of the policy concerns a specific secondary variable, it was assigned a score of 1; otherwise, it was scored 0.

The construction of a multi-input-output table

The construction of multi-input-output tables is the basis for calculating the PMC index, so this study constructed it about CHI policies (Table 3), including 10 primary variables and 41 secondary variables, all of which have the same weight.

Table 3 Multi-input-output table of the CHI policies

Calculation and evaluation of PMC index

First, each policy was analyzed and the scores of all secondary variables were determined for each policy (Formula 1). Then, the scores of the primary variable were calculated for each policy (Formula 2). Finally, the scores of all primary variables were added to obtain the policy’s PMC index (Formula 3).

$$\mathrm{Xna}=\left\{\begin{array}{l}1,\;\mathrm{Policy}\;\mathrm{involves}\;\mathrm{the}\;\mathrm{secondary}\;\mathrm{variable}\\0,\;\mathrm{Policy}\;\mathrm d\mathrm o\mathrm e\mathrm s'\mathrm t\;\mathrm{involve}\;\mathrm{the}\;\mathrm{secondary}\;\mathrm{variable}\end{array}\right.$$

(1)

$$\left\{\begin{array}{c}\mathrm X1=\left(\mathrm X1\mathrm a+\mathrm X1\mathrm b+\mathrm X1\mathrm c+\mathrm X1\mathrm d\right)/4\\\mathrm X2=\left(\mathrm X2\mathrm a+\mathrm X2\mathrm b+\mathrm X2\mathrm c\right)/3\\\ldots\\\mathrm X10=1\end{array}\right.$$

(2)

$$\mathrm{PMC}=\mathrm X1+\mathrm X2+\mathrm X3+\mathrm X4+\ldots+\mathrm X9+\mathrm X10$$

(3)

In this study, 10 primary variables were set, so the PMC index range was 0–10 points. Referring to Ruiz Estrada’s research [22], policy classification was carried out according to the PMC index, and policies with a score between 9 and 10 were rated as “perfect”, 7–8.99 as “excellent”, 5–6.99 as “good”, 3–4.99 as “qualified”, and 0–2.99 as “bad”.

Results

Overall PMC-index for 22 policies

By calculating the average of the primary variables for the 22 policies (see Table 4), it can be observed that the average PMC-index of these policies is 7.48. From the perspective of single-dimension, the ranking of these primary variables is X3 (policy recipients) > X5 (policy content) > X6 (policy tools) > X8 (cooperation and connection) > X7 (content evaluation) > X1 (policy nature) > X9 (policy goals) > X4 (policy incentives) > X2 (policy timeliness). Specifically, in terms of X1, all 22 policies included have proposed multiple measures to support the development of CHI, with no policies found describing the current state of CHI development. In terms of X2, only P22 has an effective period of 5 years, while the rest have a duration of more than five years. Regarding X3, the policy recipients for all policies include operating institutions and regulatory bodies, with 20 policies including consumers as recipients. In terms of X4, all policies propose “information sharing,” and only one policy involves “talent incentives.” In terms of X5, only 3 policies propose the development of inclusive CHI. Regarding X6, all policies apply supply-type and environmental-type policy tools, with only 8 policies (36%) applying demand-type policy tools. In terms of X7, all policies are rich in content and have clear goals, with only 9 policies being scientifically sound in their schemes and 10 policies having clear rights and responsibilities. In terms of X8, 21 policies suggest that CHI should be integrated with health management. Regarding X9, the goals of all policies include protecting consumers, with only 4 policies including promoting equity.

Table 4 Scores for 22 policies

Single PMC-index for 22 policies

The PMC-index for each of the 22 policies was calculated individually, as detailed in Table 5. Among them, 4 policies are at a good level, accounting for 18.18%; 18 policies are at an excellent level, accounting for 81.82%. Policy P16, issued by the General Office of the Shandong Provincial Government, achieved the highest PMC-index score. In contrast, Policy P22, issued by the Tianjin Municipal Healthcare Security Administration, had the lowest score.

Table 5 Calculation results of PMC index

Discussion

Overall analysis of 22 policies

The research findings indicate that the overall PMC-index for the 22 policies is 7.48, signifying that the current CHI policies in China are generally at an excellent level. This demonstrates that the Chinese government has provided robust policy support and given ample attention to the development of CHI. The analysis by dimensions is as follows:

The top three primary variables are X3 (policy recipients), X5 (policy content) and X6 (policy tools). Specifically, the mean of X3 (policy recipient) is 0.97, because the target audiences of the policies are diverse, including institutions that operate CHI, regulators that supervise CHI, consumers, and the insurance industry association. The mean of X5 (policy content) is 0.86, which shows that China’s CHI policies are rich in content, but only 14% of the policies emphasize the development of inclusive CHI. Inclusive CHI has the apparent advantages of “low premium, low threshold, and high insurance amount”, which effectively connects with basic medical insurance scheme [29]. Therefore, we suggest improving the top-level design of inclusive CHI to provide policy guarantees for its development. The mean of X6 (policy tools) is 0.79 points. Rothwell and Zegveld divided policy tools into supply, demand, and environment types [27]. Supply-type policies promote the development of CHI, demand-type policies stimulate the development of CHI, and environmental-type policies create a favorable external environment for the development of CHI. The currently included policies all applied supply-type and environmental-type policy instruments, but only eight policies deal with demand-type instruments. This may indicate that the development of China’s CHI is in its infancy, and there are more inputs on the supply side, such as information support, capital investment, talent incentives, and so on. The government has created a good external environment for the development of CHI, such as enhancing people’s awareness of insurance. In order to optimize the structure of policy tools, we suggest increasing the application of demand-type tools. The US government includes the premiums paid by employers to purchase CHI for employees into pre-tax expenses, giving tax incentives and not capping them. There are also tax incentives for individuals to purchase CHI, and the Affordable Care Act (ACA) provides tax incentives for taxpayers who pay for commercial insurance individually (and whose household income is at or above 100% but not more than 400% of the federal poverty line), starting in 2014 [30]. The Australian government calculates the number of premium rebates based on income and age, most enrollees are eligible for premium rebates, and a health insurance surcharge is imposed on taxpayers who do not participate in CHI and whose income exceeds a certain amount, designed to encourage individuals to buy CHI [31]. Therefore, we propose to develop demand-type policies around the following three points: The first is to suggest that the Chinese government can increase the tax incentives for CHI; The second proposal is for the government to create a publicly funded program to help low-income groups to buy CHI. The third is recommending CHI companies set premiums based on income and age to attract more people to enroll in insurance.

The three primary variables ranked in the middle are X8 (cooperative and connection), X7 (content evaluation) and X1 (policy nature). Specifically, the mean of X8 (cooperative and connection) is 0.74. We found that 82% of the policies support the combination of CHI with basic medical insurance, medical institutions, pension services and health management. Nevertheless, specific practices on how to achieve effective integration between basic medical insurance and CHI are seldom discussed. The international community promoted the integrated development of CHI and basic medical insurance by delineating their respective scopes of coverage and target populations. For instance, the Canadian government mandates that commercial insurers can only offer services beyond the coverage of public medical insurance and public supplementary medical insurance, with any deviation being considered illegal. The German government stipulates that only employees whose income exceeds a certain threshold and certain specific groups (such as self-employed individuals and public servants) are eligible to participate in CHI, while the remainder of the population is required to enroll in statutory health insurance. Drawing from the aforementioned international practices, China can delineate the respective roles of CHI and basic medical insurance to foster their synergistic development. The mean of X7 (content evaluation) is 0.72. The 22 policies included have rich content and specific goals, but only a small number of policies have clearly stipulated the implementation department of each task. Therefore, we suggest clarifying the leading unit, coordinating unit, and responsible unit for each task, and establishing a coordination mechanism among relevant government departments such as the Health Commission, Human Resources and Social Security Bureau, Medical Insurance Bureau, Finance Department, Insurance Regulatory Bureau, and so on, to effectively implement various tasks to promote the high-quality development of CHI. The mean of X1 (policy nature) is 0.72 points. The existing policies pay more attention to support, guidance and supervision, and there is no policy describing the current situation of the development of CHI.

The primary variables ranked in the bottom three are X9 (policy goals), X4 (policy incentives) and X2 (policy timeliness). The mean of X9 (policy goals) is 0.71. In the field of CHI, policymakers need to balance three conflicting goals: protecting consumers, promoting fairness, and promoting cost control [28]. Existing policy goals focus on protecting consumers and promoting cost control, with little mention of promoting fairness. To balance policy goals, multiple measures should be taken to promote fairness. Sekhri N proposed that equity could be promoted by minimizing adverse selection and risk selection [28]. European and American countries like the United States adopt methods such as Risk Adjustment, Reinsurance, and Risk Corridors for risk sharing [32]. Based on this, we recommend that the Chinese government promote equity in the formulation of CHI policies around the following: The first is to stipulate that the claim rate of commercial insurers should reach about 80% (the claim rate of China’s CHI in 2022 is 42% [12]), regularly check their claims, require insurers to lower the claim threshold for insurance types with too low claim rates, and give appropriate financial subsidies to insurance types with insufficient fund revenue. The second is to establish a risk adjustment fund system among insurers to balance the financial risks of insurance institutions and reduce the risk selection of insurers. The third is to encourage enterprises and public institutions to participate in CHI in the form of a collective, in order to reduce the adverse selection of participants. The mean of X4 (policy incentives) is 0.64 points. We found that the three incentives of “tax incentives, encourage insurance coverage, and information sharing” are well applied, while the two incentives of “talent incentives and individual account purchases” are insufficient. Therefore, we propose to carry out the pilot work of “allowing urban workers to use personal accounts to purchase CHI for themselves and their families”; At the same time, we suggest that government departments provide policy support for CHI professionals in terms of title assessment, housing, settlement, and children’s education. The mean of X2 (policy timeliness) is 0.33 points. All policies are medium and long-term, and there are no short-term policies.

Single analysis of 22 policies

To gain a deeper understanding of the current status of CHI policies, five typical policies were selected from the 22 policies for separate analysis, namely P1, P8, P15, P16 and P22. When selecting typical policies, the main considerations were: (1) policy level. The five typical policies selected in this study include both national policies (P1) and local policies (P8, P15, P16, and P22) to ensure the representativeness and coverage of policy samples. (2) PMC-index. Among the 22 policies in this study, P16 has the highest PMC-index, P22 has the lowest PMC-index, and PMC-index of P1, P8 and P15 all occupy the middle position.

P1 is a policy issued by The General Office of the State Council, and its PMC-index is 7.42 points, indicating it is at an excellent level. However, X4 (policy incentives), X7 (content evaluation) and X9 (policy goals) of P1 are all lower than the mean. In the X4 dimension, P1 implements three incentive measures, which are: learning from foreign experience and combining with China’s national conditions to improve the tax policies related to health insurance; encouraging businesses and individuals to enroll; and supporting information sharing between CHI information system, basic medical insurance information system and medical institution information system. However, there are no relevant requirements in terms of “talent incentive” and “individual account purchase”. In the X7 dimension, P1 is rich in content and specifies the stage goals of CHI, but it does not clearly specify the division of labor of each department, resulting in a low score. In the X9 dimension, P1 requires to enhance professional service capabilities, provide quality services, and strengthen supervision and management to protect the legitimate rights of the insured; Moreover, P1 requires the improvement of the cooperation mechanism between commercial insurance institutions and medical institutions to effectively reduce unreasonable medical expenses and promote cost control. However, P1 did not put forward relevant measures to decrease adverse selection and risk selection, which is insufficient in promoting fairness. Therefore, this study suggests that P1 should be optimized in the three dimensions of X4, X7 and X9 to improve the top-level design of CHI policy.

P22 has the lowest PMC-index among all policies and was issued by the Tianjin Medical Insurance Bureau. By analyzing the policy text, we found that it is deficient in that it only makes detailed regulations on CHI products and services, but does not involve other aspects that are crucial to the development of CHI, such as regulatory departments should strengthen the supervision of CHI institutions, and government departments should support innovation in the health industry. Therefore, P22 should be optimized in terms of X4, X7, X8, and X9.

The PMC-index of P8 and P15 are in the middle, and P8 and P15 were issued by the Jiangxi and Shanxi provincial governments respectively. The policy recipients of them include institutions that operate CHI, regulators that supervise CHI, consumers, and others (such as insurance industry associations, etc.), and both them adopt supply, demand and environmental type policy tools to promote the development of CHI. Therefore, P8 and P15 have full marks for X3 (policy recipients) and X6 (policy tools). In addition, P15 clearly stipulates the leading and responsible departments for each work, and it has a perfect score in X7 (content evaluation). Since X2 (policy timeliness) is a single variable option, the X2 scores of the two policies are the lowest. To refine the policy, optimize X7 for P8 and X4 for P15.

P16 has the highest PMC-index among all policies and was issued by the government of Shandong Province. This is a model of policies formulated at the provincial level to promote the development of CHI. The policy puts forward guiding opinions to promote the development of CHI in Shandong Province from five aspects, including expanding the supply of CHI, using the insurance mechanism to improve the medical security service system and so on. It clarifies the leading departments and responsible departments for various tasks. The five aspects of X3 (policy recipients), X6 (policy tools), X7 (content evaluation), X8 (cooperation and connection) and X9 (policy goals) are all full marks.

The advantage of this study is that it takes the lead in using the PMC-index model to conduct an overall quantitative evaluation of China’s national and local CHI policies, describes the status of China’s CHI policies, and provides reference suggestions for optimizing it. However, this study also has certain limitations. First, the policies included in this study are all open government policies. Some non-public policies cannot be included, which may introduce selection bias and result in a partial representation of the overall policy landscapelead to a certain deviation in the evaluation results. Second, this study only quantitatively evaluates the policy itself, and does not include the evaluation of the policies’ implementation impact. This may lead to interpretation bias, as the quality of a policy’s text does not necessarily reflect its real-world effectiveness. Therefore, in future research, more policies can be included for evaluation, and an evaluation of the policies’ implementation effect can be carried out.

Conclusion

China has established the world’s largest social and medical security system. However, the treatment level of basic medical insurance is low while the role of China’s CHI is weak, and needs of the insured could be hardly to meet. It is crucial to promote the development of CHI within the multi-level medical security system in China. Policy plays a leading role in practical activities, and determines the success or failure of practical activities. This study constructed PMC-index model to conduct a quantitative evaluation of the CHI policies to analyze the current status and problems of China’s CHI policies, and put forward suggestions for optimization. This study showed that China’s CHI policies is at an excellent level according to the overall score. However, the primary variables of “policy incentives, policy goals, content evaluation, policy tools and policy content” can be further optimized. We recommend several targeted strategies. First, policy incentives should be improved by piloting an initiative that allows urban workers to use personal accounts to purchase CHI for themselves and their families, alongside government support for CHI professionals in areas such as title assessment, housing, settlement, and education. Second, to balance policy goals and promote equity, we suggest mandating a minimum claim rate of 80% for commercial insurers, with regular audits, lowered claim thresholds for insurance types with low claim rates, and financial subsidies for those with insufficient revenue. Additionally, establishing a risk adjustment fund among insurers can balance financial risks and mitigate risk selection, while collective participation by enterprises and public institutions would reduce adverse selection and enhance equity. Third, task allocation should be clarified by defining leading, coordinating, and responsible units for each task and creating a multi-agency coordination mechanism involving the Health Commission, Social Security Bureau, Medical Insurance Bureau, Finance Department, and Insurance Regulatory Bureau to ensure effective implementation. Fourth, policy tools should be optimized by increasing tax incentives for CHI, implementing publicly funded programs to support low-income groups, and encouraging CHI providers to set premiums based on income and age to expand coverage. Finally, the development of inclusive CHI should be prioritized by improving its top-level design, clarifying its functional positioning, and establishing a robust policy framework that ensures its sustainability. This includes defining its role within the broader healthcare system, aligning it with the needs of diverse populations, and implementing measures to enhance its accessibility, affordability, and long-term viability.