Research Of Influence On Income To Medical Expense As Well As The Optimization Of The New Rural Cooperative Medical System Reimbursement Scheme

BackgroundNew Rural Cooperative Medical System(also referred to as NRCMS) has been put into force for over ten years, with 98.3 percent of participation rate and fund spending per year reaching 240.8 billion(as shown by data in year 2012).But the reality is far from optimistic, the Fourth Health Service Survey indicated that 44% of inpatients were not satisfied with medical service in hospital, with the claim of the excessive medical costs falling into the least satisfied one, and this phenomenon was especially true in rural areas. "Financial difficulties" could account for 70.3 percent of the reasons why rural residents who should be hospitalized fail to be hospitalized, which suggests that unbalanced beneficiaries are faced by NRCMS, resulting in the fulfillment of goals being constrained. Meanwhile, between the year 2005 and 2010, the medical expenditures of participants in NRCMS were growing rapidly with an annual increase of 19.0 percent, which is more than the social economy could sustain (the growth rate of GDP per capita was only 16.1percent in the corresponding period).From the perspectives of the achievement of goals, the functions of demand guidance and cost control, scientific and reasonable clarification of NRCMS compensation plan and its well-grounded adjustment is the key to solving problems. However, studies has showed that the present NRCMS in our country fails to consider the heterogeneity with regards to the economic state of the participating groups which, on one hand, causes that compensation levels only depend on the magnitude of absolute risk (medical expenses) with poor sensitivity in identify high risk group and undefined beneficiaries, on the other hand,compensation plan provides no cure for the reasonable and effective control over medical supply and demand due to unclear beneficiaries, the NRCMS lose restriction and control function, without intervention in the unchecked growth of demand, leading to the rapid increase of medical expenses Hence, to figure out the influence of individual’s socio-economic status, income as the main indicator, on medical expenses is the key to solving problems.Therefore, our study based on the relevant studies at home or abroad, delves into the specific mechanism of income affecting medical expenditure and extracts and deepens the existing research results and theories and methodologies. Our study, based on the quantitative relationship between income and medical expenditure, endeavors to achieve the optimization and improvement of the compensation payment system, address the problem of unbalanced participating groups, recover the control functions of NRCMS, and restrict unreasonable growth of rural residents’ medical expenses.Goals and contentsGoals:demonstrated systematically by econometric model, we found out the explicit influencing mechanism of income on medical expenses, which provides technical support for the optimization and improvement of the compensation payment system, thus promoting the realization of the system goals and the implementation of functions.Content:Firstly, our study was based on qualitative methods such as literature argument, making explicit the research overview and existing theories of the influencing mechanism of income on medical expenses and establishing research hypotheses by logical deduction.Secondly, on the one hand, our study used the data from the macro level and adopted methods such as time series model, demonstrating causational relationships between income and medical expenses; on the other hand, we used household survey data from the micro level to build the specific influencing mechanism of income on medical expenses at individual level through methods such as regression model.In the end, we adjusted and improved the key elements of NRCMS compensation plan such as compensation section, top line, etc. based on the quantified influencing mechanism.Methodologies1. Our study was guided by methodologies like health economics and health demand theory and adopted methods such as literature deductive and literature boundary analysis, making the influencing mechanism of income on medical expenses explicit and establishing research hypotheses by logical deduction. There were 372 articles concerning NRCMS compensation plans and 347 articles about influencing factors of medical costs.2. Using released data from macro level, and the data from household survey conducted five times per year, and through the construction of general linear model and and the model for review, we quantified the influencing mechanism of income on medical expenses. Effective sampling size of household survey was 2998, including household survey conducted in Jiading, Shanghai, Changzhou, Jiangsu province, Zhangqiu, Shandong province in the year 2000and 2008; household survey conducted in Jiading, Shanghai, in 2005, in Songjiang in 2001 and 2007, and Yanbian, Jilin province in 2008, with emphasis on hospitalization expenses and income data.3. Based on the actual compensating database from some area in eastern China, we established mock compensation plan and used comparative analysis method,clarifying strategies to optimize the compensation plans and testing the feasibility and rationalization of the theories.ResultsThe theoretical assumptions of income influencing medical expensesThrough literature analysis, we decide to classify the influencing factors into 38 categories, the variable "income" representing socio-economic status is the most important determinant of health spending (the mention rate is 58.12 percent); through literature deduction, and we found that many studies have formed certain theory on the influencing mechanism of income on the medical expense. Our study develop two hypotheses through logical deduction:① There does not exist two-way covariant relationship between income and medical consumption, with income being the cause of medical consumption.② we can find a standardized reference (we decided to take local average per capita income as standardized "reference" to facilitate regional comparison) to define the level medical costs. "Income effect" of income on medical expenses would dominate in a/some regions.The examination of relationship between income and medical expenditure1. Based on macro statistics, we make causality reasoning between income and medical expenses. With per capita disposable income and per capita medical spending as the variables, the original time series after logarithmic transformation, on the basis of 1 order difference stationary series counteraction analysis, the results showed that per capita income is the cause of medical spending per capita (stage two lag granger causality test F= 6.14, P= 0.046), further logarithm linear model fitting showed steady residual error sequence, suggesting there existed counteraction and causal relationship was established.2. We analyzed the specific effect of income on medical expenses based on the household survey data. Given that NRCMS is a poverty-stricken system,90% of the funds expenditure in the reality goes into hospitalization costs compensation, and the annual outpatient data accessibility, this study selected the total annual hospitalization costs per capita as indicators of medical expenses; Considering that the phenomena of family risk sharing is prevalent in the rural areas of China, personal medical consumption affordability is usually indicative of the family’s medical consumption affordability. Our study chose household annual disposable income as an index of income. After proper data processing, based on the results of theoretical assumptions 1,we took hospitalization expenses as the dependent variable, household disposable income as the independent variable, and based on theoretical assumption 2, we tried piecewise linear model, which was verified through the virtual regression model, identified the specific interval of "income effect" dominating medical consumption was 5 to 10 times the local income per capita. Within the interval, there existed obvious linear relationship between rural residents’ household disposable income and the total hospitalization costs per capita. For NRCMS, it need to consider the interval when developing compensation plans, make appropriate reduction in compensation rate to weaken the influence of the income effect, so as to adjust the distribution benefits and guide the demand.This study further pointed out when hospitalization was less than 5 times or more than 10 times of the local income per capita, the relationship between household disposable income and hospitalization costs conformed to exponential distribution with no intercept.The qualitative influencing mechanism of income on medical cost facilitates the improvement of NRCMS compensation plansBased on a whole set of methodologies in the improvement of NRCMS and the income effect interval defined by this study, we focused on the optimization of two key technologies, the classification of people and plans development ideas and steps. Comparing the original plan with preliminary achievements of our teaching and research section, this study puts forward two control points with regards to the optimization of compensation plans:①compensation section need to set up two endpoints, five times and 10 times of the local per capita income respectively, and reduce compensation ratio in the cost range;② considering the transformation between "health effect" and the "income effect", when medical costs are more than ten times of the local average per capita income, people may have high degree of risk aversion to health, characterized by high medical costs, therefore, we suggest that top line is set no less than 10 times the local average per capita income.DiscussionOur study targets solving practical hurdles faced by the present NRCMS as goals, takes the quantitative relationship between income and health care costs as basis, achieves an targeted and well-grounded optimization of the NRCMS compensation plans, thus adjusting the compensation focuses with purpose and addressing problems like unbalanced beneficiaries, inappropriate use of funding, unreasonable demand and excessively rapid growth of rural residents’medical expenses. The values embodied by this study when solving practical problems are listed as follows:①The combination of econometrics and policy sciences, findings through construction of econometrics model transformed into theories of policy sciences, are applied into the optimization of the NRCMS compensation plans. It delves into the fairness of compensation benefits at individual level when exploring the group equity. It integrates the external supporting optimization strategy into the NRCMS;② completing the deepening of existing study theories, we study comprehensive causal relationships at macro level and specific influencing mechanism from micro level.By increasing the sample size, we use macro statistics to complete causal reasoning between income and medical expense, laying a foundation for studying the effect of income on medical costs at individual level.Meanwhle, a large amount of household data is used to analyze the effect of household disposable income on total hospitalization cost, which is a further quantitative deepening of economics theories such as the "income effect", etc.③ a focused exploration of the influencing factors of medical expenses,it revolves around the effect of income on medical expenses, focusing on the literature review and in-depth study of the influencing factors of medical expenses itself.At the same time, duo to the limitations of data availability, there are still some points to be improved① failed to obtain the latest household survey data, there may exist some deviation from influencing mechanism of the present income on medical expenses;② our study used only the annual hospitalization costs to investigate the effect of income on medical expenses;③ the NRCMS compensation data lacks the information of income, which makes it hard to make strict analysis on benefit fairness after compensation between simulated plans and original plans.