An evaluation of China’s new rural cooperative medical system: achievements and inadequacies from policy goals

Our findings show that NRCMS has achieved some positive effects on the goals of reducing
the financial burden of healthcare, mitigating impoverishment from health hazards,
and improving income equity eroded by medical expenses, which are similar to and confirm
those from other studies. Yip and Hsiao 18] showed that NRCMS could reduce the poverty headcount by 3.5–3.9 % and the average
poverty gap by 11.8–16.4 %. Sun et al. 19] found NRCMS generated a 20.4 % decrease in medical impoverishment and a 19.2 % decline
in the average poverty gap in rural areas of Shandong Province. Over the past few
years of implementation, the funding level of NRCMS has steadily increased, with per
capita annual government subsidy increasing from 40 RMB in 2003 to 120 RMB in 2010.
Meanwhile, inpatient reimbursement ratio has been raised, with its cap raised to more
than eight times the national net income of rural residents 38]. In sum, adequate funding with large risk-pools and higher reimbursement ratio are
major factors leading to the success of NRCMS so far.

Our findings also indicated that there was greater reduction in impoverishment, medical
financial risk and income inequality through reimbursements under NRCMS than RCMS.
It is probably due to several important features of NRCMS in comparison to its predecessor,
the RCMS. First, NRCMS is co-financed by the central government, provincial and local
governments, and individuals; thus the funding sources are more diversified, stable,
and reliable. In contrast, RCMS was financed mainly with individual payments plus
some township and village input, with no central, provincial and county government
subsidies; thus its funding sources were neither stable nor reliable. Second, operating
at the county level rather than the township or village level, NRCMS possesses a much
larger pool to dilute and share financial risks. Third, policies and guidelines are
promulgated by the central government, and relevant agencies are established in the
central and local governments to manage and supervise the implementation of NRCMS.
In contrast, RCMS was managed mostly by village doctors or rural residents themselves,
with no effective guidance or supervision. Finally, NRCMS gives priority to catastrophic
illnesses, the most urgent demand of the rural population; voluntary participation
with the household as the unit makes it more acceptable. These are the reasons why
NRCMS has advanced beyond RCMS 39], 40].

Despite the above mentioned successes, the outcome of NRCMS has not yet met all the
expectations, with only 30 % reduction of medical impoverishment and little effect
on improving income equity. China’s NRCMS still has a long way to go in obtaining
the policy goal of reducing, ultimately eliminating impoverishment from health hazard
in rural areas. There are several explanations for this limited outcome. First, the
focus of NRCMS so far has been limited to catastrophic diseases, mainly reimbursing
hospitalization expenses rather than all outpatient expenses. The reality is, though
outpatient expenses for each doctor’s visit are relatively low in comparison to those
of a hospital stay, the cumulative expenses can be a significant amount for a household
18]. Thus, excluding outpatient expenses would definitely have had a negative impact
on the effect of NRCMS. Second, the county-based NRCMS aims at financial break-even
of each county pool; reimbursement schemes are developed and modified on the basis
of managerial experience instead of empirical risk estimation 41], which results in low reimbursement ratios, high deductibles, low cap, and limited
benefit range of reimbursement 42], 43]. Finally, although funding has increased by several fold in recent years, it remains
difficult to keep up with the rapid rise of healthcare expenses. The above factors
have led to the overall low levels of reimbursement and security of NRCMS 14], 44], thereby less than expected impact on reducing healthcare burden on rural residents
45].

The three issues mentioned above all point to the urgent need for an optimal reimbursement
scheme that is technically sound, administratively practicable and financial sustainable.
The current implementers of NRCMS do not yet know how to make the schemes financially
and administratively more practicable. Nor do the implementers have clear ideas about
who the target reimbursement participant should be or what the main issue is for the
scheme. Lack of managerial capacity has become an important factor potentially influencing
the success of rural health insurance in China, which has been corroborated by other
studies 46]–48]. Moreover, these problems are in accord with our conclusion from the intention survey
that the primary obstacle to achieving the policy goals is to solve the technical
difficulties in implementing NRCMS.

The year-2000 surveys of government departments, healthcare administrators and health
professionals showed some differences from those in year 2008 on the major potential
difficulties. That the technical difficulties could become primary problems in implementing
NRCMS could be of various reasons. For example, the previous system (RCMS, up to 2003)
with heavy reliance on township and village financing was hard to establish and maintain
if the township government did not pay adequate attention or the township/village
did not have sufficient own-source revenue. In contrast, the new system (NRCMS, since
2003) is co-financed by three levels of government plus participant-paid premiums.
Government prioritized NCRMS with substantial outlays; besides, the central government
provides extra subsidy for each participant in the poor western and interior regions
39], 40]. Thus raising the insurance premium on NRCMS was no longer as big a barrier as in
2000 and rural residents’ enthusiasm of participation improved and their confidence
in NRCMS enhanced, which added to the motivation of implementing NRCMS. To further
promote NRCMS, it is thus important to solve the technical difficulty in terms of
actuarial funding, making appropriate reimbursement plans for implementation of the
program.

By the above analysis, our findings highlight the need to establish dynamic relations
according to actuarial calculations between medical need, financial risk, impoverishment
from medical expenses, reimbursement ratio, maximum premiums, coverage rate, and benefit
range, so as to calculate actuarially sustainable funding level and to make practicable
reimbursement schedules to achieve maximum policy goals. We would recommend several
key knobs for the implementers of NRCMS to improve their capacity building in the
design of the reimbursement schemes: (1) Identify the financial risk through an overview
of service recipients – outpatients and inpatients, patrons of county, township, and
village level healthcare providers – including the probability of participants seeking
healthcare service and their medical expenses. Identify households that are impoverished
from health hazards, which should be the main target reimbursement group when designing
the reimbursement schedules 36]. (2) Calculate the total premium level needed to eliminate the specific medical financial
risk, taking into account necessary management fee, risk reserve fund, reasonable
growth in medical costs, and rising medical demands, which will help to determine
the funding level paid by NRCMS members 49]. (3) Conduct the preliminary evaluation of the reimbursement effect after the design
or adjustment of the reimbursement schemes, including break-even of the insurance
funds, pooling of financial risks, reduction of impoverishment, and improvement of
income equity, which will contribute to choosing the optimal schemes.

Needless to say, this current study contains limitations. First, the pre- and post-implementation
comparison that we have used in evaluating the outcome of NRCMS does not control for
the endogenous effects of socioeconomic factors. Future studies should take these
factors into consideration. Second, due to data limitations, our calculation has included
only direct medical expenses but not indirect, non-medical costs such as food and
transportation, which might have affected the accuracy of our evaluation. Third, we
used three measures to evaluate the achievements of the policy goals of NRCMS in this
study; whether there are better indicators to complete this issue is worth exploring.
Finally, as the sampling cities are located in Eastern China, the results and conclusions
of this study might reflect the trends in the well-developed areas, while extrapolations
to other areas (such as the Western Region) or the entire country must be made carefully
due to the differences among the regions throughout China. These limitations need
to be addressed in subsequent studies.