Integration of rural and urban healthcare insurance schemes in China: an empirical research

Study sample

To interview the target population, a questionnaire was developed with three sections. Section A was designed to collect demographic information, including age, gender, education background, household economic status, recent health status, current health status, insurance enrollment status, etc. Section B was designed to study people’s payment intention and reimbursement expectation. Section C was designed to collect information about people’s willingness towards the integration of UEBMI, URBMI and NRCMS and the corresponding reasons. Informed consent was obtained from all participants following the protocol approved by Ethics Committee of China Medical University.

A total of 1200 questionnaires were distributed, with 300 in each city and 100 in every population group. We retrieved a total of 1187 questionnaires, among which 1178 were effective. The effective response rate is 98.17%.

Measures

Data analysis

As there were three options for the integration willingness towards the three insurance schemes in the questionnaire, namely, “support”, “oppose”, “not sure”, multinomial logistic regression was applied to examine the relationship between insurance type, age, gender, education, current health status, average family annual income and integration willingness. The dependent variable was the integration willingness, where 1 = “supporting”, 2 = “opposing” and 3 = “not sure”. Independent variables included insurance type (1 = URBMI, 2 = NRCMS, 3 = UEBMI), age (1 = 20 or younger, 2 = 41 or older, 3 = 20–40), gender (1 = male, 2 = female), education (1 = elementary or lower, 2 = college or higher, 3 = secondary to high school), current health status (1 = healthy, 2 = ill), average family annual income (1 = 10000 or 10000 below, 2 = 20000 above, 3 = 10000–20000) and integration willingness. Statistical significance was defined as P ≤ 0.05.

Furthermore, payment intention, reimbursement expectation, preferred healthcare scheme, integration willingness and the corresponding reasons were described.

Characteristics of study sample

The sample generated for analysis in our study included a total of 1178 individuals. Table 3 presented the demographic characteristics of the respondents grouped by their insurance type. The age distribution of the three groups differed significantly (χ² = 72.150, df = 4, P < 0.001). UECMS group mainly consisted of the population of 20–40 years old, while the majority of URBMI and NRCMS were of over 40 years old. Education level among the three groups also differed significantly (χ² = 272.922, df = 4, P < 0.001). Compared with NRCMS and URBMI enrollees (12.8% and 30.8%), more UEBMI enrollees have received college or higher education (60.2%).

Payment intention

When analyzed by regions, in economic developed areas like Foshan and Changshu, rural residents had equal or even stronger payment ability than urban residents, and the overall payment intention presented a scattered trend, as was shown in Figure 2. While in less economically developed areas like Changchun and Shenyang, urban residents had a stronger payment ability than rural residents and more concentrated payment intention could be observed.

Reimbursement expectation

Among the three groups of population, NRCMS enrollees presented the lowest reimbursement expectation: 27.4% chose “71%–80%”, 25.6% chose “81%–90%”, 10.7% chose “51%–60%”. A more concentrated pattern could be observed in URBMI enrollees: 31.4% chose “71%–80%”, 30.1% chose “81%–90%”, 19.6% chose “above 90%”. And UEBMI enrollees held the highest reimbursement expectation: 38.2% chose “81%–90%”, 37.5% chose “above 90%”, 18.7% chose “71%–80%”.

As was shown in Figure 3, urban employees, urban residents and rural residents in Foshan, compared with these in the other three cities, had the highest and similar reimbursement expectations. In addition, in Changshu, urban employees had higher reimbursement expectations, while urban and rural residents shared similar and lower expectations, which is consistent with the local policy of integrating rural resident healthcare insurance with urban resident healthcare insurance. In Shenyang, reimbursement expectations of the three groups were mainly distributed at three intervals. In Changchun, generally low and scattered reimbursement expectations have been witnessed: rural residents held the lowest reimbursement expectations and urban employees held the highest reimbursement expectations.

Preferred healthcare insurance package

Given the three benefit packages, with the best healthcare treatment in Package 3, followed by Package 2 and Package 1, a majority of URBMI (57.9%) and NRCMS (47.1%) enrollees preferred Package 2, and 36.7% of UECMS enrollees chose it. While the majority of UECMS enrollees took a fancy to Package 3, taking up 57.7%.

In Figure 4, the majority of Foshan and Changshu participants preferred Package 3, while the majority of Shenyang and Changchun participants preferred Package 2. Furthermore, in the same area, people’s preference differed within the same group, regardless of the economic level.

Integration willingness and the corresponding reasons

Among the 1178 respondents, 816 individuals were in support of the integration of NRCMS, URBMI and UEBMI, accounting for 69.3%. The majority of URCMS, NRCMS and UEBMI enrollees were in favor of the integration, taking up 69.6%, 80.5% and 57.7%, respectively. However, 30.6% of the UEBMI enrollees, 8.2% of the NRCMS enrollees and 18.9% of the URBMI enrollees opposed the integration.

The top reason for the integration was that it would achieve equal access to healthcare services for both rural and urban participants. Also a great number of supporters believed that they would have more choices of hospitals after integration. Other main reasons were that it would narrow the healthcare gap between urban and rural areas, and that the participants would enjoy better healthcare services.

The most common reason against the integration was that the payment ranges and benefit packages in the current insurance schemes were quite different, making it hard and unsuitable to integrate. Some participants also worried that the administration might fall behind after integration and the healthcare needs of urban and rural participants were different, as details described in Table 4.

Factors associated with integration willingness

Healthcare insurance expectations, healthcare insurance integration willingness and associated factors

To promote the participation in healthcare insurances and to smoothly integrate UEBMI, URBMI and NRCMS, the voices of mass population should be heard. Our data revealed that the expected payment range of urban employees presented a dispersed trend, while that of urban residents and rural residents presented a concentrated pattern (Figure 2). This could be explained by the fact that the insurance payment of urban employees is proportional to their salaries, while for urban and rural residents, the quota payment, which sets a payment standard for enrollees, limits their choices. The UEBMI guideline established by State Council suggests premium contributions by employers and employees be set at 6% and 2% of an employee’s salary, respectively [15]. While according to a URBMI survey conducted by State Council, the average per capita financing level of the pilot cities in 2007 reached 236 RMB for adults and 97 RMB for minors. About 36% and 50% of these amounts, respectively, were contributed by government subsidies, indicating that the majority of funding was obtained from individual contributions [7]. The total government subsidies for NRCMS enrollees were 80 RMB per rural resident in 2009, of which central and local government each contributed 40 RMB. Meanwhile, individual contributions rose to 20 RMB per enrollee and the average financing level increased to 100 RMB per enrollee in 2009 [16]. As the healthcare insurance systems of URBMI and NRCMS have been integrated in some areas, the participants shared the same reimbursement expectations (Figure 3). In addition, as indicated in our study, people in economically developed areas showed higher payment intention, higher reimbursement expectation as well as better insurance benefit package. In addition, according to the 2008 National Health Service Survey [17], income level was a major determinant of health outcomes. Therefore, the economic status of participants from different areas and different population groups should be taken as a reference when setting the payment standard and benefit package.

As noted in Table 4, rural residents were the most supportive of healthcare insurance integration, with the most common reason of achieving equal access to healthcare services. On the other hand, people who strongly opposed the integration mainly held that the gaps in the insurance coverage and healthcare benefits across different schemes remained to be so wide that it was unwise to integrate them at this moment. Despite the increasing coverage of healthcare insurance nationally [18], the access to healthcare services remained to be uneven, and the integration would not only promote the benefits for rural residents, but will also provide much convenience for rural migrants seeking medical care in cities. Jian and his colleagues have reported that even though rural Chinese with chronic disease could more easily start inpatient treatment in 2008 than they could in 2003, they were more than twice as likely to drop out of treatment as were Chinese in urban areas due to the higher hospital copayments required under insurance coverage for rural citizens [19]. Therefore, in the process of coordinating urban and rural healthcare insurance schemes, the economic background should not be neglected. The logistic regression model in Table 5 indicated that participants in Changchun were more likely to oppose the integration than the participants in the other three cities, reflecting that the enrollees in Changchun were not satisfied with the current insurance schemes. Also we found that people with better education background tended to oppose the integration, which might seem contradictory to the common sense. We proposed two reasons for it: on one hand, as no complete integration policy has been implemented yet, they might concern that the integration would drag down their healthcare benefits; on the other hand, they might be better aware of the underlying challenges of integration, involving limited financial risk pooling, inefficient purchasing and provider incentives, etc. In this case, a proper balance of benefits among each group is crucial in improving the healthcare scheme.

Policy implication

According to the feedback from the enrollees, we have come up with the following suggestions to improve the healthcare insurance policy.

Setting up a multi-grade payment and treatment standard: For developed areas, high subsidies are provided to ensure that urban and rural residents receive improved or even equivalent healthcare treatment as urban employees (such as Changshu and Foshan). While for most parts of China, where the income of urban and rural residents differ greatly and the local fund is limited, a proper fund-raising mechanism with multiple standards is needed in the transitional stage, so that participants could choose the grade according to their own economic background.

Constructing an integrated healthcare insurance information system: For areas that are immature to integrate URBMI and NRCMS, instead of leaping to the integration, a unified information platform shall be established as the first move towards the integration.

Unifying the administration: To improve the poor interconnections due to the separate healthcare administrative institutions, a unified administration will not only facilitate the coordination of healthcare insurances, but will also alleviate the financial pressure of operation. The unified policies in Foshan, Changshu and Shenyang have set successful examples for broader implementation.

Establishing a proper transferring mechanism: With the market economic development and agricultural market reforms, many rural workers transit to cities for employment, most of whom receive little NRCMS reimbursement [9]. A transferring mechanism that links with the original system could well improve the situation. But it requires a cohesive mechanism among different systems, and solving problems like how to convert payment age limitations and how to compensate for the medical funding of the local area.

Improving the coordination level: Currently, the healthcare insurance coordination remains at various levels, including county level, municipal level, and provincial level. This leads to the poor risk-resisting ability. While by improving the planning on a higher level, we could have a stronger ability to resist risks, thus increasing enrollees’ confidence in healthcare insurances.

Setting up a stable fund-raising and financial subsidy mechanism: In spite of the increasing funding level, the financing mechanism is not well regulated. We propose that the payment standard of rural residents shall be set in accordance with their net income, while that of urban residents shall be set in accordance with their disposable income, and financial subsidies from the government shall be determined partially by residents’ hospitalization cost.

Limitations

The study has several limitations. Firstly, the selection of the respondents focused on certain communities in the four cities, and the results may not be generalizable to the entire country due to differences among areas, and could not fully represent the three groups of population, namely, rural residents, urban residents and urban employees. Secondly, the differences of the same insurance scheme across cities were not considered when we studied the factors influencing participants’ willingness towards healthcare insurance integration. Thirdly, some data in out study, like household annual income per capita, were collected on the basis of personal recall and could be prone to measurement errors.