Outpatient service payment medical insurance data monitoring analysis management system based on big data analysis

Abstract

The invention relates to the technical field of big data analysis, and discloses a big data analysis-based outpatient service payment medical insurance data monitoring analysis management system, which comprises a data acquisition module for acquiring outpatient service medical insurance settlement data before and after hierarchical diagnosis and treatment of all medical insurance fixed-point medical institutions in a target area; and the data analysis module is used for acquiring sub-data change amplitudes of the patient flow direction data, the cost benefit data and the disease matching degree data before and after the implementation of the hierarchical diagnosis and treatment in the same target area according to the outpatient medical insurance settlement data before and after the implementation of the hierarchical diagnosis and treatment, and constructing a change amplitude data set. The method comprises the steps of obtaining outpatient medical insurance settlement data before and after hierarchical diagnosis and treatment of a medical insurance fixed-point medical institution in a target area, obtaining a change amplitude data set and a difference value data set according to the outpatient medical insurance settlement data, completing correction, and performing comprehensive evaluation according to the corrected data to obtain a hierarchical diagnosis and treatment effectiveness evaluation result.

Description

Technical Field

[0001] This invention relates to the field of big data analytics technology, specifically to a big data analytics-based outpatient payment medical insurance data monitoring, analysis, and management system. Background Technology

[0002] Tiered medical services is a medical service system model for deepening the reform of the medical and health system in my country. It aims to improve the efficiency of medical resource utilization, alleviate the pressure on large hospitals, and promote the improvement of primary healthcare service capabilities by rationally allocating medical resources, clarifying the functional positioning of medical institutions at all levels, guiding patients to choose appropriate medical institutions according to the severity and urgency of their conditions. When existing technologies assess the effectiveness of the hierarchical medical system through outpatient payment medical insurance data, they often directly use the original medical insurance settlement data for evaluation. However, interference from non-hierarchical medical system factors such as adjustments to medical service prices, changes in medical insurance reimbursement policies, and fluctuations in regional population structure can cause the evaluation data to fail to accurately reflect the actual effect of the policy. To address this, the present invention proposes an outpatient payment medical insurance data monitoring, analysis and management system based on big data analysis to overcome the shortcomings of existing technologies. Summary of the Invention

[0003] The purpose of this invention is to provide an outpatient payment medical insurance data monitoring, analysis and management system based on big data analysis, in order to solve the problems mentioned in the background art.

[0004] The objective of this invention can be achieved through the following technical solutions: The outpatient payment and medical insurance data monitoring and analysis management system based on big data analytics includes the following modules: The data acquisition module acquires outpatient medical insurance settlement data before and after the implementation of hierarchical diagnosis and treatment for all medical insurance designated medical institutions within the target area; The data analysis module, based on the outpatient medical insurance settlement data before and after the implementation of hierarchical medical treatment, obtains the change range of each sub-data of patient flow data, cost-benefit data, and disease matching degree data before and after the implementation of hierarchical medical treatment within the same target area, and constructs a change range dataset. Within the same data acquisition period, obtain the differences in patient flow data, cost-effectiveness data, and disease matching data in different regions with varying levels of implementation of hierarchical medical treatment, and construct a dataset of these differences. The obtained dataset of change magnitudes and the dataset of differences are corrected. The comprehensive assessment module obtains a comprehensive assessment score based on the change magnitude correction data and the difference value correction data, and evaluates the effectiveness of the hierarchical diagnosis and treatment system.

[0005] Preferably, the data analysis module operates as follows: The outpatient medical insurance settlement data of all medical insurance designated medical institutions in the target area before and after the implementation of the hierarchical diagnosis and treatment system are classified and preprocessed. The sub-data of the patient flow data before and after the visit include the proportion of outpatient visits covered by medical insurance at primary care clinics, the rate of tiered referral, and the follow-up visit rate of patients with chronic diseases at primary care clinics. The cost-benefit data includes the average medical insurance payment per case and the growth rate of total fund expenditure for each group. The sub-data of the disease matching data includes the proportion of mild cases treated at the primary care level, the proportion of severe cases treated at tertiary hospitals in each group, and the referral rationality coefficient for each group; Based on the sub-data of each data point before and after the implementation of hierarchical diagnosis and treatment within the same target area, the change range of each data point before and after the implementation of hierarchical diagnosis and treatment is obtained, and a change range dataset is constructed. Based on the sub-data of each data point in different regions with varying levels of implementation of hierarchical medical treatment within the same data acquisition time period, the difference values ​​of each data point in different target regions are obtained and a difference value dataset is constructed. Correct for the magnitude and difference of changes in various data.

[0006] Preferably, the method for obtaining the changes in various data before and after the implementation of the hierarchical diagnosis and treatment is as follows: The percentage of outpatient visits covered by medical insurance at the primary level before and after the implementation of the hierarchical medical system, the hierarchical referral rate from primary level to tertiary hospitals in the target area, and the follow-up visit rate of patients with chronic diseases at the primary level were calculated respectively. The changes in the number of outpatient visits covered by medical insurance at the primary level, the rate of referral at the primary level, and the rate of follow-up visits for patients with chronic diseases at the primary level were obtained before and after the implementation of the hierarchical medical system. Calculate the average medical insurance payment per visit for each level of institution before and after the implementation of the hierarchical medical system, and the growth rate of total outpatient medical insurance fund expenditure in the target area. The changes in the average medical insurance payment and the growth rate of total fund expenditure before and after the implementation of the hierarchical medical system were obtained by calculating the difference. Calculate the proportion of mild cases treated at primary care facilities, the proportion of severe cases treated at tertiary hospitals, and the referral rationality coefficient before and after the implementation of the tiered medical system. The changes in the proportion of mild cases treated at primary care facilities, the proportion of severe cases treated at tertiary hospitals, and the rationality coefficient of referrals before and after the implementation of the hierarchical medical system were obtained by calculating the difference. Integrate all the variation amplitude data, remove outliers, and construct the variation amplitude dataset.

[0007] Preferably, the method for obtaining the difference values ​​of the data in different target regions is as follows: Within the same data acquisition period, select multiple different target areas and divide them into at least three groups based on the implementation strength of the hierarchical medical treatment policy; The percentage of outpatient visits covered by medical insurance, the rate of tiered referral, and the rate of follow-up visits for patients with chronic diseases at the primary level were calculated for each group. The differences in the percentage of outpatient visits covered by medical insurance, the rate of tiered referral, and the rate of follow-up visits for patients with chronic diseases at the primary level were obtained by subtracting each group from the other. Calculate the average medical insurance payment per visit for each group and the growth rate of total fund expenditure for each group, and obtain the difference between the average medical insurance payment per visit and the growth rate of total fund expenditure by subtracting each group from the groups. The proportion of mild cases treated at primary care facilities, the proportion of severe cases treated at tertiary hospitals, and the referral rationality coefficient were calculated for each group. The differences in the proportion of mild cases treated at primary care facilities, the proportion of severe cases treated at tertiary hospitals, and the referral rationality coefficient were obtained by subtracting the values ​​from each group. Integrate the various difference value data, remove outliers, and construct the difference value dataset.

[0008] Preferably, the process of correcting the acquired dataset of change magnitudes and the dataset of differences is as follows: The interference coefficients were obtained based on various data before and after the implementation of the hierarchical medical system. These interference coefficients included price change coefficients, policy adjustment coefficients, population fluctuation coefficients, and resource difference coefficients. The change magnitude dataset is corrected based on the price change coefficient and the policy adjustment coefficient to obtain the corrected change magnitude data.

[0009] Based on the resource difference coefficient and the population fluctuation coefficient, the difference value dataset is corrected to obtain the difference value corrected data.

[0010] Preferably, the comprehensive evaluation module operates as follows: Different weighting coefficients were assigned to patient flow data, cost-effectiveness data, and disease matching data. The patient flow data, cost-effectiveness data, and disease matching data were evaluated based on the data corrected for the magnitude of change and the data corrected for the difference. For the change range correction data, determine whether each data point reaches the preset threshold, and record the compliance status and the excess or deficiency range; For the difference value correction data, compare the data differences in different implementation intensity areas to determine whether the high-intensity implementation area shows a significant positive difference compared with the medium and low-intensity areas; The assessment compliance data are statistically analyzed, and a comprehensive assessment score is calculated based on the assessment compliance data; The obtained comprehensive evaluation score is compared with the set effectiveness level to obtain the effectiveness of the hierarchical diagnosis and treatment implementation in the target area.

[0011] Preferably, the method for obtaining the comprehensive evaluation score is as follows: Determine the type of each sub-data item. If it is reverse data, correct the reverse data and adjust the direction of the value. Then calculate the degree of deviation between the corrected actual value of the reverse data and the preset validity threshold. Convert the processed degree of deviation into a score for each sub-data item. The scores of individual sub-data points are grouped according to patient flow data, cost-effectiveness data, and disease matching data. The average score of the data is calculated by averaging the sum of the scores of each data point. Standardize all sub-data differences, calculate the deviation of each sub-data difference from the overall average level, and uniformly calibrate the individual deviations based on the overall fluctuation range of all sub-data differences. Obtain all calibrated deviation results and take the average value to generate the regional difference significance correction coefficient. The average scores of patient flow data, cost-effectiveness data, and disease matching data are combined with their respective weights to obtain a weighted total score. The weighted total score of the data is multiplied by the regional difference significance correction coefficient to obtain the final comprehensive evaluation score.

[0012] Preferably, the method for obtaining the validity is as follows: The overall assessment score will be compared with the validity level: A comprehensive evaluation score of 80 or above indicates effective implementation. A comprehensive evaluation score of 60 to 79 indicates successful implementation. A comprehensive evaluation score below 60 indicates poor implementation effectiveness. Based on the evaluation results of the change magnitude correction dataset and the difference value correction dataset, if all individual data meet the standards and the overall score is qualified, the hierarchical diagnosis and treatment is ultimately determined to be effective; otherwise, it is determined to be ineffective or requires optimization.

[0013] The beneficial effects of this invention are: 1. This invention obtains outpatient medical insurance settlement data of designated medical institutions in the target area before and after the implementation of hierarchical diagnosis and treatment, obtains datasets of change range and difference values ​​based on the outpatient medical insurance settlement data and completes correction, and conducts a comprehensive evaluation based on the corrected data to obtain the effectiveness evaluation result of hierarchical diagnosis and treatment, thus achieving an effective and objective evaluation of the hierarchical diagnosis and treatment effect.

[0014] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a system module block diagram of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please see Figure 1 As shown, this invention is an outpatient payment medical insurance data monitoring, analysis, and management system based on big data analysis. The system includes a data acquisition module, a data analysis module, and a comprehensive evaluation module. The data acquisition module acquires outpatient medical insurance settlement data for all designated medical institutions within the target area before and after the implementation of hierarchical medical treatment. All data is precisely divided into primary healthcare institution data, secondary hospital data, and tertiary hospital data according to the level of the medical institutions. Primary healthcare institution data includes outpatient medical insurance settlement data from community health service centers and township health centers. Then, all sub-data corresponding to patient flow data, cost-effectiveness data, and disease matching data are extracted from each category of data. The sub-data of patient flow data includes basic... The data includes the proportion of outpatient visits covered by medical insurance at the primary level, the rate of referral to higher-level hospitals, and the rate of follow-up visits for patients with chronic diseases at the primary level. Sub-data for cost-effectiveness data includes the average medical insurance payment per visit and the growth rate of total fund expenditure for each group. Sub-data for disease matching data includes the proportion of mild cases treated at the primary level, the proportion of severe cases treated at tertiary hospitals for each group, and the rationality coefficient of referral for each group. Subsequently, the extracted sub-data are preprocessed to remove abnormal data such as duplicate settlements and inconsistent statistical standards. The coding standard for disease diagnosis is unified, and the disease diagnosis names of different medical institutions are mapped to the nationally unified ICD-10 code. At the same time, the time span before and after the implementation of the hierarchical medical system policy is used as a node to obtain the time span, population structure, and seasonal factors of the implementation period before and after the implementation of the hierarchical medical system policy.

[0019] The data analysis module is used to perform horizontal and vertical comparisons of outpatient medical insurance settlement data before and after the implementation of the hierarchical medical system, thereby obtaining datasets of change magnitude and difference values. The patient flow data, cost-effectiveness data, and disease matching data acquired by the data acquisition module each include sub-data. A single data point cannot fully reflect the effectiveness of policy implementation; only by analyzing multiple data points together can the implementation effect of the hierarchical medical system be objectively evaluated. Among these, the sub-data of patient flow data includes the proportion of outpatient medical insurance visits at the primary care level, the hierarchical referral rate, and the follow-up visit rate of patients with chronic diseases at the primary care level. This is because the capacity of primary care, the smoothness of the referral mechanism, and the implementation of chronic disease management are key data for measuring the reasonable diversion of patients. The proportion of outpatient medical insurance visits at the primary care level directly reflects whether the downward flow of minor illnesses has been implemented, and the hierarchical referral rate reflects... The efficiency of the connection between primary care and tertiary hospitals, and the follow-up visit rate of chronic disease patients at the primary care level, focus on the effectiveness of primary care management for high-incidence chronic diseases such as hypertension and diabetes. This type of sub-data comes from the original outpatient medical insurance settlement records of medical institutions at all levels within the target area. Through big data analysis technology, key information such as outpatient visits, referral marks, chronic disease diagnosis codes, and the level of the medical institution are extracted from the settlement data and integrated to generate the data. For example, from the settlement data of a community health service center, the follow-up visit records and medical insurance settlement marks of hypertension patients can be extracted and combined with the total number of outpatient visits of chronic disease patients in the area to form the basic data of the follow-up visit rate of chronic disease patients at the primary care level. Using this type of sub-data, it is possible to accurately determine whether patients are receiving treatment at the corresponding level of institutions in accordance with the requirements of hierarchical medical treatment, avoiding the waste of resources such as large hospitals being overcrowded and primary care facilities being empty.

[0020] Sub-data of cost-effectiveness data includes average medical insurance payment per visit and the growth rate of total fund expenditure. Efficient utilization of medical insurance funds is an important goal of hierarchical medical treatment. Average medical insurance payment per visit can reflect the rationality of treatment costs at different levels of institutions. The average cost per visit at primary care institutions should be lower than that at secondary and tertiary hospitals, while the average cost per visit at tertiary hospitals can be reasonably higher due to the treatment of critically ill patients. The growth rate of total fund expenditure directly reflects the effectiveness of hierarchical medical treatment in controlling medical insurance fund expenditure. This type of data is obtained from the original settlement data of outpatient medical insurance payment amount, number of settlements, and total fund expenditure within the period of institutions at all levels. For example, by extracting the total outpatient medical insurance payment amount and number of settlements before and after the implementation of the system at a secondary hospital, the average medical insurance payment per visit can be calculated. By comparing the data of primary care institutions, it is possible to determine whether the cost allocation is reasonable. Using this type of sub-data can effectively identify problems such as overtreatment and fund waste.

[0021] The sub-data of disease matching includes the proportion of mild cases treated at primary care facilities, the proportion of severe cases treated at tertiary hospitals, and the referral rationality coefficient. The hierarchical medical system requires primary care facilities to focus on common and frequently occurring diseases (mild cases), while tertiary hospitals focus on complex and acute diseases (severe cases). This type of data is needed to verify the accuracy of the matching between medical resources and disease types. The sub-data is generated based on original data such as ICD-10 disease codes, the level of medical institutions, and referral records obtained from big data. By mapping the disease diagnosis names of different institutions to the national standard codes, mild and severe cases are classified. For example, the number of visits corresponding to mild case codes can be screened from the settlement data of primary care institutions, and the proportion of mild cases treated at primary care facilities can be calculated by combining the total number of outpatient visits at primary care facilities. Using this type of sub-data can avoid the resource mismatch problem of primary care facilities blindly treating severe cases and tertiary hospitals taking on mild cases.

[0022] For the three sub-data items of patient flow data, the original values ​​before and after the implementation of the hierarchical medical system were extracted to calculate the values ​​of each sub-data item. The proportion of outpatient visits covered by medical insurance at the primary care level was calculated by dividing the number of outpatient visits covered by medical insurance at the primary care level before and after the implementation of the hierarchical medical system by the total number of outpatient visits covered by medical insurance in the region during the same period. For example, if the number of outpatient visits covered by medical insurance at the primary care level in a certain region before implementation was 120,000 and the total number of outpatient visits in the region was 400,000, and the number of outpatient visits covered by medical insurance at the primary care level after implementation increased to 220,000 and the total number of outpatient visits in the region was 450,000, the calculation shows that the proportion of outpatient visits covered by medical insurance at the primary care level before implementation was 30% and the proportion of outpatient visits covered by medical insurance at the primary care level after implementation was 48.9%. The tiered referral rate is calculated by dividing the number of patients referred from primary care facilities to tertiary hospitals by medical insurance before and after the implementation of the tiered medical system by the total number of outpatient medical insurance visits to tertiary hospitals during the same period. For example, before implementation, the total number of outpatient visits to tertiary hospitals was 180,000, of which 27,000 were referred from primary care facilities, resulting in a tiered referral rate of 15%. After implementation, the total number of outpatient visits to tertiary hospitals was 120,000, of which 30,000 were referred from primary care facilities, resulting in a tiered referral rate of 25%. The follow-up visit rate for chronic disease patients at the grassroots level is calculated by dividing the number of follow-up visits for chronic diseases covered by medical insurance at the grassroots level before and after the implementation of the hierarchical medical system by the total number of outpatient visits for chronic diseases covered by medical insurance in the region during the same period. For example, before implementation, the total number of outpatient visits for chronic diseases in the region was 80,000, and the number of follow-up visits for chronic diseases at the grassroots level was 32,000, with a follow-up visit rate of 40%. After implementation, the total number of outpatient visits for chronic diseases in the region was 85,000, and the number of follow-up visits for chronic diseases at the grassroots level was 51,000, with a follow-up visit rate of 60%. Then, the change in each sub-data is obtained by calculating the difference, that is, the value after the implementation of the hierarchical medical system minus the value before the implementation of the hierarchical medical system, then divided by the value before the implementation, and finally multiplied by 100%. The change in the above three sub-data is 63%, 66.7%, and 50%, respectively.

[0023] For the two sub-data points of cost-effectiveness data, the average medical insurance payment per visit for each level of institution is calculated by dividing the total amount of outpatient medical insurance payments for a certain level of institution before and after the implementation of the hierarchical medical system by the number of settlements for that level of institution during the same period. For example, before implementation, the average cost per visit for primary care institutions was 200 yuan and for tertiary hospitals it was 600 yuan. After implementation, the average cost per visit for primary care institutions was 210 yuan and for tertiary hospitals it was 680 yuan. The difference is used to calculate the change in average cost per visit for primary care institutions (5%) and for tertiary hospitals (13.3%). The growth rate of total fund expenditure is calculated by subtracting the total expenditure before the implementation of hierarchical medical treatment from the total expenditure of outpatient medical insurance fund after the implementation of hierarchical medical treatment, dividing by the total expenditure before the implementation of hierarchical medical treatment, and finally multiplying by 100%. For example, if the total fund expenditure before implementation was 150 million yuan and after implementation it was 162 million yuan, the growth rate of total fund expenditure is 8%.

[0024] For the three sub-data points of disease matching data, the proportion of mild cases treated at the primary care level was calculated by dividing the number of mild cases treated at the primary care level with medical insurance before and after the implementation of the hierarchical medical system by the total number of outpatient medical insurance visits at the primary care level during the same period. The proportion of severe cases treated at tertiary hospitals was calculated by dividing the number of severe cases treated at tertiary hospitals with medical insurance before and after the implementation of the hierarchical medical system by the total number of outpatient medical insurance visits at tertiary hospitals during the same period. The referral rationality coefficient was calculated by dividing the number of severe cases referred from the primary care level to higher levels by the total number of referrals from the primary care level during the same period. For example, before the implementation of the hierarchical medical system, the proportion of mild cases treated at the primary care level was 65%, the proportion of severe cases treated at tertiary hospitals was 55%, and the referral rationality coefficient was 70%. After the implementation of the hierarchical medical system, these figures were 88%, 78%, and 92%, respectively. The changes were calculated using the difference, resulting in changes of 35.4%, 41.8%, and 31.4%, respectively.

[0025] Finally, the change magnitudes of all sub-data are integrated to construct a complete change magnitude dataset, presenting the changes in various data before and after the implementation of hierarchical diagnosis and treatment.

[0026] For the differential value dataset, firstly, within the same data acquisition period, multiple target areas are selected and divided into at least three groups—high-intensity, medium-intensity, and low-intensity—based on the implementation intensity of the hierarchical medical system policy. The division is based on policy implementation details, such as official documents and actual implementation data on referral incentives, primary care subsidies, implementation frequency, and resource investment. For example, areas that fully implement primary care first-visit, referral subsidies, and primary care medical staff training are classified as high-intensity implementation areas, while areas that only partially implement the primary care first-visit policy and have less resource investment are classified as low-intensity implementation areas.

[0027] Subsequently, for each group of regions, the values ​​of all sub-data were calculated separately. The calculation logic was consistent with that of the sub-data calculation logic of the change range dataset. Information was extracted and summarized from the original outpatient medical insurance settlement data of each group of regions. For example, in the high-intensity implementation area, the calculated values ​​were: 52% of outpatient medical insurance visits at the primary level, 28% of the tiered referral rate, 65% of the follow-up visit rate for chronic diseases at the primary level, 205 yuan of average medical insurance payment at the primary level, 7% of the total fund expenditure growth rate, 90% of mild cases treated at the primary level, 80% of severe cases treated at tertiary hospitals, and a referral rationality coefficient of 93%. The corresponding values ​​for the medium-intensity implementation area were 40%, 18%, 50%, 215 yuan, 12%, 75%, 68%, and 82%, respectively. The corresponding values ​​for the low-intensity implementation area were 28%, 10%, 35%, 220 yuan, 18%, 62%, 56%, and 71%, respectively. Then, the difference values ​​of each sub-data were obtained by subtracting the values ​​from each group, i.e., the difference between high-intensity and low-intensity implementation areas. By comparing the patient flow data sub-data with the medium-intensity area, the high-intensity area with the low-intensity area, and the medium-intensity area with the low-intensity area, the differences in patient flow data sub-data were obtained: the differences in the proportion of outpatient visits covered by medical insurance at the primary care level were 12%, 24%, and 12%, respectively; the differences in the tiered referral rate were 10%, 18%, and 8%, respectively; and the differences in the follow-up visit rate for chronic diseases at the primary care level were 15%, 30%, and 15%, respectively. The differences in cost-effectiveness data sub-data were: the differences in the average medical insurance payment at the primary care level were -10 yuan, -15 yuan, and -5 yuan, respectively; and the differences in the growth rate of total fund expenditure were -5%, -11%, and -6%, respectively. The differences in disease matching data sub-data were: the differences in the proportion of mild cases treated at the primary care level were 15%, 28%, and 13%, respectively; the differences in the proportion of severe cases treated at tertiary hospitals were 12%, 24%, and 12%, respectively; and the differences in the referral rationality coefficient were 11%, 22%, and 11%, respectively. Finally, all the differences between the groups were integrated to construct a complete difference value dataset.

[0028] After obtaining the difference value dataset and the change range dataset, the difference value dataset and the change range dataset are corrected. The purpose of the correction is to eliminate the interference of non-tiered medical treatment factors on the data. Specifically, based on the medical service item price adjustment documents during the evaluation period before and after the implementation of tiered medical treatment, the price changes of core service items such as outpatient visits, examinations, and drugs are sorted out. Based on the medical insurance settlement frequency of various services in medical institutions at all levels, the impact ratio of price adjustments on cost sub-data such as average medical insurance payment per visit and total fund expenditure of each level of institution is calculated. For example, in a certain evaluation period, the average price of commonly used drugs in primary medical institutions increased by 5%, and the price of outpatient examination items increased by 3%. Through big data analysis, it is found that these items account for 60% of the total medical insurance payment for primary outpatient visits, and the price change coefficient is calculated to be 4.2%. By comparing the details of the changes in medical insurance reimbursement policies before and after implementation, including reimbursement ratio, deductible, ceiling, and chronic disease reimbursement scope, the impact coefficient of policy changes on sub-data such as outpatient medical insurance visits, payment amount, and visit proportion is calculated. For example, after implementation, the medical insurance reimbursement ratio for primary outpatient visits increased by 4%. According to big data analysis, this adjustment led to an additional 3% increase in the proportion of primary outpatient medical insurance visits and the total fund expenditure. An additional 2.5% increase in expenditures leads to a policy adjustment coefficient of 3% for visitor data and 2.5% for expense data. Based on population statistics of the target and comparison regions before and after the implementation of the tiered healthcare system, data on population aging growth rate, chronic disease incidence rate fluctuations, and changes in the resident population are obtained. Using outpatient medical insurance settlement data from the same period, the impact of population structure changes on patient flow and disease matching is calculated. For example, if the aging growth rate increases by 2% and the number of chronic disease patients increases by 3% in a region before and after implementation, and if this fluctuation leads to an additional 2.5% increase in the follow-up visit rate for chronic disease patients at the grassroots level, the adjustment coefficient is determined accordingly. The fluctuation coefficient is 2.5%. By comparing the allocation of primary healthcare resources in different regions, including the number of primary healthcare institutions, number of beds, ratio of medical staff to patients, and equipment configuration, and based on outpatient medical insurance settlement data for each region, the impact of resource endowment differences on sub-data such as the proportion of outpatient visits to primary healthcare institutions and the proportion of mild cases is obtained. For example, the difference in the ratio of beds to patients in primary healthcare institutions between high-intensity implementation areas and low-intensity implementation areas is 1:0.6, and the difference in the ratio of medical staff to patients is 1:0.5. If this resource difference leads to an additional 8% increase in the difference in the proportion of outpatient medical insurance visits to primary healthcare institutions, then the resource difference coefficient is determined to be 8%.

[0029] After obtaining four types of interference coefficients, corrections were performed on the change magnitude dataset and the difference value dataset respectively. The change magnitude dataset was further corrected based on the price change coefficient and policy adjustment coefficient, resulting in corrected change magnitude data. This eliminated interference from price and policy changes, restoring the true impact of the tiered healthcare policy on data changes. For example, if the original change in the average medical insurance payment per visit at primary care institutions after the implementation of tiered healthcare in a certain region was 8%, combined with a price change coefficient of 4.2%, the original change magnitude needs to be subtracted from the price change coefficient to obtain a corrected average change magnitude of 3.8% at primary care institutions. This reflects the true impact of the tiered healthcare policy on primary care costs. For example, the change in the proportion of outpatient visits covered by medical insurance at the primary level was originally 15%. After adjusting for the policy adjustment coefficient of 3%, the change was 12%. Excluding the interference of policy adjustments, the change in the growth rate of total fund expenditure was originally 10%. After deducting the price change coefficient of 4.2% and the policy adjustment coefficient of 2.5%, the growth rate was 3.3%, thus reflecting the control effect of hierarchical medical treatment on medical insurance fund expenditure. The hierarchical referral rate and the referral rationality coefficient are less affected by prices and policies. It is necessary to check whether there is any indirect influence. If so, the coefficients should be slightly adjusted. If not, the original change should remain unchanged.

[0030] For the difference value dataset, the dataset is corrected based on the resource difference coefficient and the population fluctuation coefficient. For each group of regional comparison data, a weighted adjustment is made according to the influence of the resource difference coefficient and the population fluctuation coefficient. For example, the original difference value of the proportion of outpatient visits covered by medical insurance at the primary level between high-intensity implementation areas and low-intensity implementation areas is 24%. Combined with the resource difference coefficient of 8%, the original difference value needs to be subtracted from the resource difference coefficient to obtain a corrected difference value of 16%, which truly reflects the difference in the proportion of visits caused by the difference in implementation intensity. For example, the original difference value of the follow-up visit rate of chronic disease patients at the primary level between medium-intensity implementation areas and low-intensity implementation areas is 15%. Combined with the population fluctuation coefficient of 2.5%, the corrected difference value is 12.5%, eliminating the interference of population structure differences. For the difference value of the proportion of primary level mild cases treated in the disease matching data, the original difference value of a certain group of regions is 13%. Because the primary level medical resources in the high-intensity implementation areas are more abundant, the influence of the resource difference coefficient is 5%, and the corrected difference value is 8%, which reflects the actual impact of implementation intensity on the capacity of primary level to handle mild cases.

[0031] Meanwhile, during the correction process of the change range dataset and the difference value dataset, the correction results need to be verified one by one, and the change logic of the data before and after correction needs to be compared to ensure that the correction range is reasonable and there is no over-correction or under-correction. For example, the change range of the average cost at the base level after correction should still be lower than that of tertiary hospitals, and the referral rationality coefficient should still be within a reasonable range. If logical contradictions occur, the interference coefficient and correction logic should be readjusted until all data are corrected reasonably.

[0032] The comprehensive evaluation module is used to comprehensively evaluate the hierarchical medical system based on the sub-data of each corrected data. Specifically, it first sets different weight coefficients based on the importance differences of patient flow data, cost-effectiveness data, and disease matching data. Among them, patient flow data has the highest weight of 0.4, while cost-effectiveness data and disease matching data have equal weights of 0.3. Then, it sets an effectiveness threshold for the data with the magnitude of change correction and compares each sub-data item to determine whether it has reached the preset effectiveness threshold. It records the achievement status and the extent of exceeding or falling short of the threshold in detail. For the difference value correction data, it compares the data differences in areas with different implementation intensities to determine whether the high-intensity implementation area shows a significant positive difference compared with the medium and low-intensity implementation areas. In this way, it selects effective data that can reflect the policy effect and eliminates the interference of invalid data on the final score.

[0033] After the evaluation is completed, the overall evaluation score is obtained through a formula. :

[0034] To comprehensively evaluate the score, The number of sub-data points under each data set is as follows: patient flow data has 3 sub-data points, cost-benefit data has 2 sub-data points, and disease matching data has 3 sub-data points. , is the Sigmoid nonlinear function. , For the first The data point number The values ​​after correction of individual data. For the first The data point number Preset validity threshold for each sub-data item These are the weighting coefficients. For patient flow data weighting coefficients, For cost-benefit data weighting coefficients, The weighting coefficients for disease matching data. This represents the total number of data points with discrepancies. For the first The difference value after correction. The mean of all difference values. These represent the maximum and minimum values ​​of the difference data, respectively.

[0035] Data scoring formula middle, It is the Sigmoid non-linear smoothing function, used to reduce the excessive interference of extreme values ​​on data scores, and to prevent a single outlier from raising or lowering the evaluation result of the entire data. This standard is used to unify the measurement criteria for sub-data of different magnitudes and units, so that different types of data, such as the change in the proportion of outpatient visits covered by medical insurance at the primary level and the change in the average medical insurance payment per visit, can be evaluated equally. This represents the number of sub-data points under the corresponding data. Taking the average score of each sub-data point is to balance the differences in the number of sub-data points and prevent data with more sub-data points from having higher scores. Multiplying by 100 is to convert the 0-1 range output by the Sigmoid function into a regular scoring range of 0-100 points, which facilitates subsequent classification. This is used to calculate the average score of patient flow, cost-effectiveness, and disease matching. For example, if there are three sub-data items in the patient flow data, the deviation rate is calculated by substituting the corrected value of each sub-data item and the preset threshold, and then the score of each sub-data item is obtained by processing it with the Sigmoid function. The average score of the data can be obtained by taking the average value. middle, It refers to the standardized bias of variance data, used to eliminate differences in magnitude between different variance data, so that the degree of bias of different types of data, such as variance in the proportion of visits and variance in average cost per visit, can be measured uniformly. This is the average of the standardized deviations of all variance data, comprehensively reflecting the overall differences in inherent conditions such as medical resource endowment and population structure across different regions. Multiplying by 0.1 is to control the correction range within ±10%, preventing inherent regional differences from excessively dominating the final assessment score. The range is limited to 0.9-1.1 to prevent extreme adjustments from distorting the score. To eliminate the interference of inherent regional conditions on the assessment results and ensure that the final score only reflects the true effectiveness of the implementation of the hierarchical medical system, for example, four differential values ​​are selected after correction, their mean, maximum, and minimum values ​​are calculated, and then substituted into the formula to obtain the average value of the standardized deviation, and then the correction coefficient is calculated. These are the weighting coefficients. The weighting factor for patient flow data is 0.4. The weighting factor for cost-benefit data is 0.3. The disease matching data is weighted at 0.3. Patient flow data is given the highest weight because rational patient triage is the core objective of hierarchical medical treatment and most directly reflects the policy's implementation effect. This weight is multiplied by a correction factor. This is to correct for the influence of inherent regional conditions and make the final score more objective and fair.

[0036] For example, taking the hierarchical medical system assessment data of a certain prefecture-level city, including patient flow data... , Cost-benefit data , Disease matching data , Data weights , , Number of difference value indicators .

[0037] Corrected value of the change in the proportion of primary care outpatient visits The preset validity threshold is 12%. The value is 5%, the indicator type is positive, and the adjusted value is the change in the tiered referral rate. The preset validity threshold is 18%. The value is 10%, the indicator type is positive, and the adjusted value is the change range of the chronic disease follow-up visit rate at the primary care level. The preset validity threshold is 20%. The value is 15%, the indicator type is positive, and the base level average cost change is adjusted. The preset validity threshold is 3%. The value is 8%, the indicator type is inverse, and it is the adjusted value of the change in the total fund expenditure growth rate. The preset validity threshold is 6%. The value is 10%, the indicator type is reversed, and the adjusted value is the percentage change in the proportion of mild cases treated at the primary care level. The preset validity threshold is 15%. The percentage is 10%, the indicator type is positive, and it represents the adjusted value of the change in the proportion of severe cases treated at tertiary hospitals. The preset validity threshold is 22%. The value is 15%, the indicator type is positive, and the adjusted value is the change range of the referral rationality coefficient. The preset validity threshold is 10%. The value is 8%, and the indicator type is positive. Regional difference value after correction , Patient flow data score calculation: Calculate the deviation rate : , ,

[0038] Substitution : , 3.

[0039] Calculate the sub-data scores and take the average. Sub-data scores: , , Then the patient flow data score is .

[0040] Cost-benefit data score calculation: Handling contrarian indicators: That is, -3% and -6%; Calculate the deviation rate ,but , ; Substitution : ,

[0041] Calculate the sub-data scores and take the average. Sub-data scores: , The cost-benefit data score is: ; Disease matching score calculation: Calculate the deviation rate ,but , , ; Substitution : , , ; Calculate the sub-data scores and take the average. Sub-data scores: Sub-data scores: , , The disease matching score is: ; Calculate the mean of the differences Maximum value Minimum value , , ; ; Calculate the standardized deviation of each difference value : , , , , , ; The standard deviation mean is: ; The correction factor is: ; The overall evaluation score is: ; In summary, the comprehensive assessment score of a certain prefecture-level city's hierarchical medical system... The score was 56.83. Finally, the obtained comprehensive evaluation score was compared with the set effectiveness level: a comprehensive evaluation score of 80 or above was considered effective, a comprehensive evaluation score of 60 to 79 was considered qualified, a comprehensive evaluation score below 60 was considered poor, and 56.83 was below 60, which meant that the implementation of hierarchical diagnosis and treatment in the prefecture-level city was poor.

[0042] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in the claims, they should all fall within the protection scope of the present invention.

Claims

1. A big data analytics-based outpatient payment medical insurance data monitoring and analysis management system, characterized in that: Includes the following modules: The data acquisition module acquires outpatient medical insurance settlement data before and after the implementation of hierarchical diagnosis and treatment for all medical insurance designated medical institutions within the target area; The data analysis module, based on the outpatient medical insurance settlement data before and after the implementation of hierarchical medical treatment, obtains the change range of each sub-data of patient flow data, cost-benefit data, and disease matching degree data before and after the implementation of hierarchical medical treatment within the same target area, and constructs a change range dataset. Within the same data acquisition period, obtain the differences in patient flow data, cost-effectiveness data, and disease matching data in different regions with varying levels of implementation of hierarchical medical treatment, and construct a dataset of these differences. The obtained dataset of change magnitudes and the dataset of differences are corrected. The comprehensive assessment module obtains a comprehensive assessment score based on the change magnitude correction data and the difference value correction data, and evaluates the effectiveness of the hierarchical diagnosis and treatment system.

2. The outpatient payment medical insurance data monitoring and analysis management system based on big data analysis according to claim 1, characterized in that, The data analysis module operates as follows: The outpatient medical insurance settlement data of all medical insurance designated medical institutions in the target area before and after the implementation of the hierarchical diagnosis and treatment system are classified and preprocessed. The sub-data of the patient flow data before and after the visit include the proportion of outpatient visits covered by medical insurance at primary care clinics, the rate of tiered referral, and the follow-up visit rate of patients with chronic diseases at primary care clinics. The cost-benefit data includes the average medical insurance payment per case and the growth rate of total fund expenditure for each group. The sub-data of the disease matching data includes the proportion of mild cases treated at the primary care level, the proportion of severe cases treated at tertiary hospitals in each group, and the referral rationality coefficient for each group; Based on the sub-data of each data point before and after the implementation of hierarchical diagnosis and treatment within the same target area, the change range of each data point before and after the implementation of hierarchical diagnosis and treatment is obtained, and a change range dataset is constructed. Based on the sub-data of each data point in different regions with varying levels of implementation of hierarchical medical treatment within the same data acquisition time period, the difference values ​​of each data point in different target regions are obtained and a difference value dataset is constructed. Correct for the magnitude and difference of changes in various data.

3. The outpatient payment medical insurance data monitoring and analysis management system based on big data analysis according to claim 2, characterized in that, The method for obtaining the changes in various data before and after the implementation of the hierarchical medical system is as follows: The percentage of outpatient visits covered by medical insurance at the primary level before and after the implementation of the hierarchical medical system, the hierarchical referral rate from primary level to tertiary hospitals in the target area, and the follow-up visit rate of patients with chronic diseases at the primary level were calculated respectively. The changes in the number of outpatient visits covered by medical insurance at the primary level, the rate of referral at the primary level, and the rate of follow-up visits for patients with chronic diseases at the primary level were obtained before and after the implementation of the hierarchical medical system. Calculate the average medical insurance payment per visit for each level of institution before and after the implementation of the hierarchical medical system, and the growth rate of total outpatient medical insurance fund expenditure in the target area. The changes in the average medical insurance payment and the growth rate of total fund expenditure before and after the implementation of the hierarchical medical system were obtained by calculating the difference. Calculate the proportion of mild cases treated at primary care facilities, the proportion of severe cases treated at tertiary hospitals, and the referral rationality coefficient before and after the implementation of the tiered medical system. The changes in the proportion of mild cases treated at primary care facilities, the proportion of severe cases treated at tertiary hospitals, and the rationality coefficient of referrals before and after the implementation of the hierarchical medical system were obtained by calculating the difference. Integrate all the variation amplitude data, remove outliers, and construct the variation amplitude dataset.

4. The outpatient payment medical insurance data monitoring and analysis management system based on big data analysis according to claim 2, characterized in that, The method for obtaining the difference values ​​of data from different target regions is as follows: Within the same data acquisition period, select multiple different target areas and divide them into at least three groups based on the implementation strength of the hierarchical medical treatment policy; The percentage of outpatient visits covered by medical insurance, the rate of tiered referral, and the rate of follow-up visits for patients with chronic diseases at the primary level were calculated for each group. The differences in the percentage of outpatient visits covered by medical insurance, the rate of tiered referral, and the rate of follow-up visits for patients with chronic diseases at the primary level were obtained by subtracting each group from the other. Calculate the average medical insurance payment per visit for each group and the growth rate of total fund expenditure for each group, and obtain the difference between the average medical insurance payment per visit and the growth rate of total fund expenditure by subtracting each group from the groups. The proportion of mild cases treated at primary care facilities, the proportion of severe cases treated at tertiary hospitals, and the referral rationality coefficient were calculated for each group. The differences in the proportion of mild cases treated at primary care facilities, the proportion of severe cases treated at tertiary hospitals, and the referral rationality coefficient were obtained by subtracting the values ​​from each group. Integrate the various difference value data, remove outliers, and construct the difference value dataset.

5. The outpatient payment medical insurance data monitoring and analysis management system based on big data analysis according to claim 1, characterized in that, The process of correcting the obtained dataset of changes and the dataset of differences is as follows: The interference coefficients were obtained based on various data before and after the implementation of the hierarchical medical system. These interference coefficients included price change coefficients, policy adjustment coefficients, population fluctuation coefficients, and resource difference coefficients. The change magnitude dataset is corrected based on the price change coefficient and the policy adjustment coefficient to obtain the corrected change magnitude data.

6. Correct the difference value dataset based on the resource difference coefficient and the population fluctuation coefficient to obtain the difference value correction data.

7. The outpatient payment medical insurance data monitoring and analysis management system based on big data analysis according to claim 1, characterized in that, The comprehensive evaluation module operates as follows: Different weighting coefficients were assigned to patient flow data, cost-effectiveness data, and disease matching data. The patient flow data, cost-effectiveness data, and disease matching data were evaluated based on the data corrected for the magnitude of change and the data corrected for the difference. For the change range correction data, determine whether each data point reaches the preset threshold, and record the compliance status and the excess or deficiency range; For the difference value correction data, compare the data differences in different implementation intensity areas to determine whether the high-intensity implementation area shows a significant positive difference compared with the medium and low-intensity areas; The assessment compliance data are statistically analyzed, and a comprehensive assessment score is calculated based on the assessment compliance data; The obtained comprehensive evaluation score is compared with the set effectiveness level to obtain the effectiveness of the hierarchical diagnosis and treatment implementation in the target area.

8. The outpatient payment medical insurance data monitoring and analysis management system based on big data analysis according to claim 6, characterized in that, The method for obtaining the comprehensive evaluation score is as follows: Determine the type of each sub-data item. If it is reverse data, correct the reverse data and adjust the direction of the value. Then calculate the degree of deviation between the corrected actual value of the reverse data and the preset validity threshold. Convert the processed degree of deviation into a score for each sub-data item. The scores of individual sub-data points are grouped according to patient flow data, cost-effectiveness data, and disease matching data. The average score of the data is calculated by averaging the sum of the scores of each data point. Standardize all sub-data differences, calculate the deviation of each sub-data difference from the overall average level, and uniformly calibrate the individual deviations based on the overall fluctuation range of all sub-data differences. Obtain all calibrated deviation results and take the average value to generate the regional difference significance correction coefficient. The average scores of patient flow data, cost-effectiveness data, and disease matching data are combined with their respective weights to obtain a weighted total score. The weighted total score of the data is multiplied by the regional difference significance correction coefficient to obtain the final comprehensive evaluation score.

9. The outpatient payment medical insurance data monitoring and analysis management system based on big data analysis according to claim 1, characterized in that, The method for obtaining the validity is as follows: The overall assessment score will be compared with the validity level: A comprehensive evaluation score of 80 or above indicates effective implementation. A comprehensive evaluation score of 60 to 79 indicates successful implementation. A comprehensive evaluation score below 60 indicates poor implementation effectiveness. Based on the evaluation results of the change magnitude correction dataset and the difference value correction dataset, if all individual data meet the standards and the overall score is qualified, the hierarchical diagnosis and treatment is ultimately determined to be effective; otherwise, it is determined to be ineffective or requires optimization.

Images