Medical insurance management system and management method

Abstract

The invention discloses a medical insurance management system and management method, and belongs to the field of medical insurance management, and the method comprises the following steps: collecting all pharmacy sales data, patient medical insurance medicine purchasing records, patient feedback data and medicine enterprise circulation data of a region, and carrying out the correlation integration processing of the data; and analyzing a price strategy and a prescription recommendation behavior of drug sales of each pharmacy, evaluating an influence degree of the pharmacy on a drug purchase decision of the patient in combination with a drug purchase record of the patient, and finally generating an individual behavior portrait of each pharmacy. According to the method, the multi-source data is collected, associated and integrated, the pharmacy behavior portraits are generated, the pharmacy portraits of the same specification are compared to find the abnormal points, the abnormal standards are formulated according to the abnormal points, regulation and settlement are performed on the illegal pharmacies, and the suspected pharmacies are manually screened, so that the medicine purchasing behavior can be accurately identified and induced, and the loss of medical insurance funds is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical insurance management, in particular to a medical insurance management system and method. BACKGROUND

[0002] In the field of medical insurance fund supervision, it is common to see that drugstores induce patients to purchase high-priced discount medicines and other irregular behaviors, resulting in prominent annual fund loss problems, and the traditional supervision mode has obvious short boards; the existing mode mainly relies on manual spot checks, which has limited coverage and low efficiency, and it is difficult to systematically monitor the operating behaviors of drugstores; due to differences in operating environment, service population and drug types, the single standard supervision adopted by drugstores in different regions is prone to misjudgment or omission, and it is difficult to accurately identify hidden drug purchase inducing behaviors; At the same time, the traditional means lack in-depth analysis of the correlation between the price strategies of drugstores, prescription recommendation behaviors and patient drug purchase decisions, and it is difficult to form a scientific abnormality judgment standard, resulting in a lag in the disposal of irregular behaviors; in addition, the existing system lacks the integration capability of multi-source data, and the information such as drugstore sales data, patient drug purchase records and pharmaceutical circulation data is scattered, which cannot provide comprehensive support for supervision, and an urgent need exists for a medical insurance management scheme that can adapt to regional differences, accurately identify irregularities and respond quickly to solve the above problems.

[0003] At present, no effective solution has been proposed for the problems in the related art. SUMMARY

[0004] In view of the problems in the related art, the present application proposes a medical insurance management system and method to overcome the technical problems of extensive supervision, single standard and difficulty in accurately identifying drugstore-induced high-priced drug purchase behaviors existing in the prior art.

[0005] Therefore, the specific technical solutions adopted by the present application are as follows: A medical insurance management system and method, the method comprising the following steps: S1, collecting all drugstore sales data, patient medical insurance drug purchase records, patient feedback data and pharmaceutical circulation data in the region, and performing correlation and integration processing on the data; S2, analyzing the price strategies and prescription recommendation behaviors of each drugstore, and combining with the patient drug purchase records, evaluating the influence degree of the drugstore on the patient drug purchase decision, and finally generating a separate behavior portrait of each drugstore; S3, comparing the behavior portraits of all drugstores in multiple dimensions, analyzing the differences between the behavior portraits of different drugstores, and setting abnormality judgment standards according to the analysis results to screen out the specific abnormal points of abnormal drugstores; S4, specific analysis and comparison of abnormal points of abnormal pharmacies, comparison of selling price, cost price, market price and medical insurance payment standard price, and retrieval of electronic prescriptions to determine whether there is an induced change of prescription to push high-priced drugs behavior; S5, when the pharmacy has induced high-priced drug purchase violation behavior, adjust the medical insurance settlement process of the involved pharmacy according to the results, and organize professional personnel to conduct on-site investigation for the pharmacy which is not found to have violation but still suspicious.

[0006] As a preferred embodiment, all pharmacy sales data in the region, patient medical insurance drug purchase records, patient feedback data, and pharmaceutical circulation data are collected and processed by correlating and integrating the data, including the following steps: S11, obtaining drug sales data from pharmacy system, specifically including drug inventory, sales flow, and electronic prescriptions, obtaining user's medical insurance drug purchase records from medical insurance settlement system, and obtaining drug production and transportation, drug pricing information from pharmaceutical circulation data; S12, cleaning, converting and standardizing the collected multi-source data; taking drug as the core, establishing correlation between pharmacy sales data, user's medical insurance drug purchase records and pharmaceutical circulation data.

[0007] Unifying drug name, specification and other expressions, standardizing date and amount format, and eliminating data differences.

[0008] As a preferred embodiment, the price strategy and prescription recommendation behavior of each pharmacy are analyzed, combined with patient purchase records, and the influence degree of pharmacy on patient purchase decision is evaluated, and finally a separate behavior portrait of each pharmacy is generated, including the following steps: S21, for each pharmacy, calculate the average price, price fluctuation range and deviation rate of the sold drugs compared with market price and medical insurance payment standard, identify the price strategy mode, and establish a time series model to evaluate the impact of price change on sales volume; S22, review the prescription records of the pharmacy, count the distribution of recommended drugs, analyze the relationship between recommendation behavior and patient disease type, and detect whether there is an abnormal recommendation mode; S23, correlate patient purchase records with price strategy and recommendation behavior, compare the change of patient's purchase selection before and after recommendation, and quantify the contribution of pharmacy factors to patient's final purchase decision; S24, based on the cross-analysis results, calculate the net influence score of each pharmacy on patient's purchase decision, distinguish between active and passive influence, and correct according to patient feedback data to improve accuracy; S25, integrate the quantitative indicators and high-priced drug preference types to create a comprehensive behavior portrait for each pharmacy.

[0009] In a preferred embodiment, the step of comparing the behavioral profiles of all pharmacies from multiple dimensions, analyzing the differences in the behavioral profiles of different pharmacies, and setting anomaly judgment criteria based on the analysis results to screen out the specific anomalies of abnormal pharmacies includes the following steps: S31. Collect behavioral profile data of all pharmacies and group and classify them according to dimensions such as region, scale, and product category structure; S32. Based on the profiling indicators, pharmacies are divided into several behavioral pattern groups; and the variances of core indicators among the groups, including the recommendation rate of high-priced drugs and the deviation of medical insurance prices, are calculated to identify the mainstream behavioral patterns and significant outliers. The specific formula is as follows: ; in, For the number of groups, For the first The number of pharmacies in each group For the first Average recommendation rate for high-priced drugs in each group This represents the average recommendation rate for high-priced drugs across all pharmacies. Variance of recommendation rate for high-priced drugs; ; in, For the variance of medical insurance price deviation, No. Groups For groups A single pharmacy in the country, For pharmacies The total number of drugs covered by medical insurance sold. For pharmacies The Deviation of the actual selling price of individual medical insurance drugs For groups The deviation of the average medical insurance price; S33. By comparing the mean within a group with the distance between groups, the key feature dimensions that cause the differences are identified, and in conjunction with medical insurance policies, indicators that significantly deviate from the normality of the group are marked as outliers to be tested. S34. Using the standard deviation multiple method, set the difference tolerance threshold for each dimension; automatically mark pharmacies that exceed the threshold as initial screening abnormal objects, and associate them with their specific abnormal indicators.

[0010] As a preferred implementation, the specific analysis and comparison of the abnormal points of abnormal pharmacies with the pharmacy's selling price, cost price, market price, and medical insurance payment standard price, and the retrieval of electronic prescriptions to determine whether there is any behavior of inducing the prescription to increase the price of drugs, includes the following steps: S41, according to the abnormal pharmacy and specific abnormal points, automatically the real-time selling price of the corresponding medicine, the circulation cost price of the medicine enterprise, the market average price of the provincial bidding platform, the medical insurance payment standard price; S42, calculate the abnormal medicine in the pharmacy selling price and cost price margin, selling price and market price deviation, selling price and medical insurance payment standard over-standard rate, and locate the price difference source; S43, call the original electronic prescription record related to the abnormal medicine, compare the final sales prescription of the pharmacy end, and identify the unnecessary modification in the prescription; S44, establish the prescription modification-drug price increase-profit correlation rule for verifying the induced drug promotion behavior; S45, set the violation level according to the price difference over-standard degree and the prescription modification frequency, specifically: single induction and price difference < 50% is judged as first-level violation, three times of modification in a month and price difference > 100% is judged as second-level violation, and generate the prescription modification comparison graph and price difference radar graph.

[0011] As a preferred embodiment, the obtained to-be-retrieved feature text is matched with the feature text library for similarity, the prescription modification-drug price increase-profit correlation rule is established for verifying the induced drug promotion behavior, and the specific content of the correlation rule is: S441, ①statistical modification of single price increase rate; ②cross analysis of modification behavior and clerk recommendation record; ③verify the agreement of high-priced medicine and pharmacy profit sharing, three times of verification of induced drug promotion behavior.

[0012] As a preferred embodiment, when the pharmacy has induced high-priced drug purchase violation behavior, adjust the medical insurance settlement process of the pharmacy involved, and for the pharmacy which is not found to have violation but still suspicious, organize professional personnel to conduct on-site investigation including the following steps: S51, obtain the violation data, and analyze the violation level, the list of medicines involved, and the violation transaction period, when the pharmacy has first-level violation, notify the pharmacy to automatically reduce the medical insurance payment amount of the medicine to the medical insurance standard price, and set the settlement delay; when the pharmacy has second-level violation, freeze all medical insurance transaction settlement permissions of the pharmacy, and generate a list of pending deductions; S52, calculate the investigation priority of the pharmacy which is determined to be suspicious, and select the top 20% pharmacies to enter the verification queue, and organize professional personnel to conduct on-site investigation.

[0013] A medical insurance management system, which adopts the medical insurance management method of any one of the above, comprising a data acquisition module, an image generation module, an abnormal screening module, an abnormal analysis module, and an abnormal processing module; The data acquisition module is used to acquire the sales data of all pharmacies in the region, the medical insurance drug purchase records of patients, patient feedback data, and pharmaceutical circulation data, and to pre-process the data, and after processing, to associate and integrate the data; The image generation module analyzes the price strategy of drug store drug sales and prescription recommendation behavior, and generates a separate behavior image for each drug store; The abnormality screening module compares the behavior images of drug stores of the same type, analyzes the differences between the behavior images of different drug stores, and sets an abnormality determination standard to screen out specific abnormal points of abnormal drug stores; The abnormality analysis module analyzes the abnormal points of the abnormal drug stores to determine whether there is an induced high-price drug behavior The abnormality processing module automatically notifies the drug store to adjust the medical insurance settlement process when the drug store has an induced high-price drug purchase irregularity behavior, and notifies professional personnel to conduct on-site investigation for the drug store which is not found to have irregularities but still suspicious.

[0014] The beneficial effects of the present application are: 1. The present application collects drug store sales data, patient medical insurance drug purchase records, pharmaceutical enterprise circulation data and other multi-source information, establishes a behavior image for each drug store covering price strategy, prescription recommendation behavior and influence degree on patient drug purchase decision, compares and analyzes the images of drug stores of the same specification, and accurately locates specific abnormal points. This method fully considers the differences between different regions, avoids the limitations of a single standard, can effectively identify whether a drug store has an induced high-price drug purchase irregularity, solves the problem of annual medical insurance fund loss caused by induced drug purchase from the source, makes the supervision more suitable for the actual operating conditions of each region, and improves the identification accuracy of irregularities; 2. The present application formulates an abnormality determination standard based on the comparison and analysis results of the drug store behavior images, so that the standard is more scientific and targeted. The standard is formed based on actual operating data and can adapt to the operating characteristics of drug stores of different types and different regions, reduces the misjudgment or omission caused by unreasonable standards, and also provides a reliable basis for subsequent accurate supervision and disposal.

[0015] 3. The present application judges the specific abnormal points analyzed, adjusts the medical insurance settlement process for the drug store confirmed to have irregularities in time, can quickly block unreasonable medical insurance fund expenditure, and maximally reduces fund loss. Artificial screening is organized for the drug store which is not found to have irregularities but still suspicious, which can make up for the possible blind spots of intelligent analysis, avoid missing potential risks due to data limitations or algorithm bias, and effectively maintain the safety of medical insurance fund. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0017] Figure 1 This is a flowchart of a medical insurance management system and management method according to an embodiment of the present invention. Detailed Implementation

[0018] To further illustrate the various embodiments, the present invention provides accompanying drawings, which are part of the disclosure of the present invention. These drawings are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these drawings, those skilled in the art should be able to understand other possible implementation methods and the advantages of the present invention. The components in the drawings are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0019] According to embodiments of the present invention, a medical insurance management system and management method are provided.

[0020] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 As shown, according to an embodiment of the present invention, a medical insurance management method includes the following steps: S1. Collect sales data from all pharmacies in the region, patient medical insurance drug purchase records, patient feedback data, and pharmaceutical company distribution data, and perform correlation and integration processing on the data; Furthermore, the process of collecting and integrating sales data from all pharmacies in the region, patient medical insurance purchase records, patient feedback data, and pharmaceutical company distribution data, and performing related and integrated data processing includes the following steps: S11. Obtain drug sales data from the pharmacy system, including drug inventory, sales flow, and electronic prescriptions; obtain users' medical insurance drug purchase records from the medical insurance settlement system; and obtain drug production, transportation, and pricing information from pharmaceutical company distribution data. S12. Clean, transform, and standardize the collected multi-source data; establish a correlation between pharmacy sales data, users' medical insurance drug purchase records, and pharmaceutical company distribution data, with pharmaceuticals as the core. S2. Analyze the pricing strategies and prescription recommendation behaviors of each pharmacy, and combine them with patients' medication purchase records to assess the degree of influence of pharmacies on patients' medication purchase decisions. Finally, generate a separate behavioral profile for each pharmacy. Furthermore, the pricing strategies and prescription recommendation behaviors of each pharmacy are analyzed, and combined with patient purchase records, the degree of influence of pharmacies on patients' medication purchase decisions is assessed. Finally, a unique behavioral profile of each pharmacy is generated, including the following steps: S21. For each pharmacy, calculate the average price of the drugs it sells, the range of price fluctuations, and the deviation rate from the market price and medical insurance payment standard. Identify the pricing strategy pattern and establish a time series model to assess the impact of price changes on sales volume. S22, audit the prescription records of the pharmacy, count the distribution of recommended drugs, analyze the relationship between recommendation behavior and patient disease type, and detect whether there is an abnormal recommendation pattern; S23, associate the patient's drug purchase record with the price strategy and recommendation behavior, compare the patient's drug purchase selection change before and after the recommendation, and quantify the contribution degree of the pharmacy factor to the patient's final drug purchase decision; S24, based on the cross-analysis result, calculate the net influence score of each pharmacy on the patient's drug purchase decision, distinguish between active influence and passive influence, and correct according to the patient feedback data to improve accuracy; S25, integrate the quantitative indicators and high-price drug preference types to create a comprehensive behavior portrait for each pharmacy.

[0021] It should be noted that by analyzing the price strategy, prescription recommendation behavior of each pharmacy, and combining the patient's drug purchase record to evaluate its influence on the drug purchase decision, a single behavior portrait is finally generated, which can accurately capture the operating characteristics and risk inclination of each pharmacy; This way not only establishes a separate behavior profile for the pharmacy, allowing the regulatory department to clearly understand its operating mode, but also identifies the early signs of hidden violations such as inducing high-priced drug purchases through the details of the portrait, providing standards for subsequent multi-dimensional comparison and abnormal judgment; S3, compare the behavior portraits of all pharmacies in multiple dimensions, analyze the differences between different pharmacy behavior portraits, and set abnormal judgment standards according to the analysis results to screen out specific abnormal points of abnormal pharmacies; Further, the behavior portraits of all pharmacies are compared in multiple dimensions, the differences between different pharmacy behavior portraits are analyzed, and specific abnormal points of abnormal pharmacies are screened out according to the analysis results, including the following steps: S31, collect the behavior portrait data of all pharmacies and group them according to dimensions such as region, size, and category structure; S32, divide the pharmacies into several behavior mode groups according to the portrait indicators; and calculate the variance of core indicators including high-price drug recommendation rate and medical insurance price deviation between groups to identify mainstream behavior patterns and significantly outlying groups, the specific formula is: ; Wherein, is the number of groups, is the number of pharmacies in the th group, is the average high-price drug recommendation rate of the th group, is the average high-price drug recommendation rate of all pharmacies, is the variance of the high-price drug recommendation rate; ; Wherein, the variance of the medical insurance price deviation, the first group, the total number of medical insurance drugs sold by the single pharmacy in the group the total number of medical insurance drugs sold by the pharmacy the actual selling price deviation of the first medical insurance drug of the pharmacy the average medical insurance price deviation of the group S33, by comparing the mean value within the group and the distance between the groups, positioning the key feature dimension causing the difference, and combining the medical insurance policy, marking the indicators significantly deviating from the group normal as abnormal points to be tested; S34, using the standard deviation multiple method, setting the difference tolerance threshold for each dimension; automatically marking the pharmacies exceeding the threshold as preliminary screening abnormal objects, and associating their specific abnormal indicators.

[0022] It should be noted that by comparing the pharmacy behavior portrait in multiple dimensions, setting abnormal judgment standards and screening specific abnormal points, precise supervision can be achieved driven by data. This method can identify abnormal behaviors deviating from the norm from the overall business trend by comparing the operating behavior characteristics of different pharmacies horizontally, and can locate specific irregularities such as inducing high-priced drug purchase and unreasonable prescription recommendation, significantly improving the pertinence and efficiency of supervision. At the same time, according to the analysis structure, the abnormal judgment standard can adapt to the operating characteristics of different types of pharmacies in different regions, effectively reducing the risk of misjudgment, ensuring the fairness of medical insurance fund supervision, and reducing the loss of fund caused by pharmacy irregularities; Further, the specific analysis of the abnormal points of the abnormal pharmacy includes the following steps: S41, according to the abnormal pharmacy and the specific abnormal point, automatically comparing the real-time selling price of the corresponding drug of the abnormal pharmacy, the circulating cost price of the pharmaceutical enterprise, the market average price of the provincial bidding platform, and the medical insurance payment standard price; S42, calculating the profit margin of the selling price and the cost price of the abnormal drug in the pharmacy, the deviation of the selling price and the market price, and the over-standard rate of the selling price and the medical insurance payment standard, positioning the main source of price difference; S43, retrieving the original electronic prescription records associated with the abnormal drug, comparing the final sales prescription on the pharmacy side, and identifying unnecessary modifications in the prescription; ​​​​​S44, establish a prescription modification-drug price increase-profit correlation rule for verifying the induced drug promotion behavior; Further, the obtained to-be-searched feature text is matched with the feature text library for similarity, and a prescription modification-drug price increase-profit correlation rule is established for verifying the induced drug promotion behavior, wherein the specific content of the correlation rule is: S441, ①statistical modified drug unit price increase rate; ②cross analysis of modification behavior and clerk recommendation record; ③verify high-priced drugs and pharmacy profit sharing agreement, three times to verify the induced drug promotion behavior.

[0023] S45, set the violation level according to the price difference and the prescription modification frequency, specifically: single induction and price difference < 50% is judged as a first-level violation, and monthly cumulative 3 times of modification and price difference > 100% is judged as a second-level violation, and a prescription modification comparison chart and a price difference radar chart are generated; S5, when the pharmacy has induced high-priced drug purchase violation behavior, adjust the medical settlement process of the involved pharmacy according to the results, and for the pharmacies that are not found to have violations but still have doubts, organize professional personnel to conduct on-site investigation.

[0024] Further, when the pharmacy has induced high-priced drug purchase violation behavior, adjust the medical settlement process of the involved pharmacy, and for the pharmacies that are not found to have violations but still have doubts, organize professional personnel to conduct on-site investigation including the following steps: S51, obtain the violation data, and analyze the violation level, the involved drug list, and the violation transaction period, when the pharmacy has a first-level violation, notify the pharmacy that the medical payment amount of the drug is automatically reduced to the medical standard price, and set a settlement delay; when the pharmacy has a second-level violation, freeze all medical settlement permissions of the pharmacy, and generate a list of pending deductions; S52, calculate the investigation priority of the pharmacy that is determined to be suspicious, and select the top 20% pharmacies to enter the verification queue, and organize professional personnel to conduct on-site investigation.

[0025] It should be noted that this combination of intelligent settlement adjustment and on-site investigation mechanism can not only quickly intervene in the pharmacies that have clear violations through the automatic dynamic settlement system, timely block unreasonable medical fund expenditure, and improve the response speed of supervision and the efficiency of fund protection, but also can carry out on-site verification on the pharmacies that are suspicious but have not yet found violations, make up for the blind spots that may exist in intelligent analysis, avoid missing potential risks due to data limitations or algorithm bias, and form a closed-loop supervision of intelligent screening and manual review; A medical insurance management system adopts a medical insurance management method according to any one of the above, comprising a data acquisition module, an image generation module, an abnormality screening module, an abnormality analysis module, and an abnormality processing module. A data collection module is configured to collect all the drugstore sales data, patient medical insurance drug purchase records, patient feedback data, and drug enterprise circulation data in the region, and to pre-process the data. After the processing, the data is associated and integrated. An image generation module is configured to analyze the price strategy of drugstore drug sales and prescription recommendation behavior, and to generate an individual behavior image for each drugstore. An abnormality screening module is configured to compare the behavior images of drugstores of the same type, analyze the differences between the behavior images of different drugstores, and set abnormality determination standards to screen specific abnormal points of abnormal drugstores. An abnormality analysis module is configured to specifically analyze the abnormal points of abnormal drugstores, and to determine whether there is an induced high-price drug behavior of changing prescriptions. The abnormality processing module is configured to automatically notify the drugstore involved to adjust the medical insurance settlement process when the drugstore has an induced high-price drug purchase irregular behavior. For a drugstore that is not found to have a violation but is still suspicious after analysis, a professional is notified to conduct an on-site investigation.

[0026] The above is only a preferred embodiment of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A medical insurance management method, characterized in that, The method includes the following steps: S1. Collect sales data from all pharmacies in the region, patient medical insurance drug purchase records, patient feedback data, and pharmaceutical company distribution data, and perform correlation and integration processing on the data; S2. Analyze the pricing strategies and prescription recommendation behaviors of each pharmacy, and combine them with patients' medication purchase records to assess the degree of influence of pharmacies on patients' medication purchase decisions. Finally, generate a separate behavioral profile for each pharmacy. S3. Compare the behavioral profiles of all pharmacies from multiple dimensions, analyze the differences between the behavioral profiles of different pharmacies, and set anomaly judgment criteria based on the analysis results to screen out the specific anomalies of abnormal pharmacies. S4. Conduct a detailed analysis and comparison of the abnormal points of abnormal pharmacies with the pharmacy's selling price, cost price, market price, and medical insurance payment standard price, and retrieve electronic prescriptions to determine whether there is any behavior of inducing the prescription to increase the price of drugs. S5. When a pharmacy is found to have engaged in illegal practices such as inducing customers to purchase drugs at high prices, the medical insurance settlement process of the pharmacy in question will be adjusted based on the results. For pharmacies where no violations were found in the analysis but doubts remain, professional personnel will be organized to conduct on-site investigations.

2. The medical insurance management method according to claim 1, characterized in that, The process of collecting and integrating sales data from all pharmacies in the region, patient medical insurance purchase records, patient feedback data, and pharmaceutical company distribution data, and then performing the associated and integrated data processing includes the following steps: S11. Obtain drug sales data from the pharmacy system, including drug inventory, sales flow, and electronic prescriptions; obtain users' medical insurance drug purchase records from the medical insurance settlement system; and obtain drug production, transportation, and pricing information from pharmaceutical company distribution data. S12. Clean, transform, and standardize the collected multi-source data; with pharmaceuticals as the core, establish a correlation between pharmacy sales data, users' medical insurance drug purchase records, and pharmaceutical company distribution data.

3. The medical insurance management method according to claim 1, characterized in that, The analysis of each pharmacy's drug pricing strategies and prescription recommendation behavior, combined with patient purchase records, assesses the pharmacy's influence on patients' drug purchase decisions, and finally generates a unique behavioral profile for each pharmacy, including the following steps: S21. For each pharmacy, calculate the average price of the drugs it sells, the range of price fluctuations, and the deviation rate from the market price and medical insurance payment standard. Identify the pricing strategy pattern and establish a time series model to assess the impact of price changes on sales volume. S22. Review the prescription records of pharmacies, statistically analyze the distribution of recommended drugs, analyze the relationship between recommendation behavior and patient disease type, and detect whether there are abnormal recommendation patterns; S23. Link patients' medication purchase records with pricing strategies and recommendation behaviors, compare changes in patients' medication purchase choices before and after recommendations, and quantify the contribution of pharmacy factors to patients' final medication purchase decisions. S24. Based on the results of cross-analysis, calculate the net impact score of each pharmacy on patients' medication purchase decisions, distinguish between active and passive impacts, and make corrections based on patient feedback data to improve accuracy. S25. Integrate quantitative indicators and high-priced drug preference types to create a comprehensive behavioral profile for each pharmacy.

4. The medical insurance management method according to claim 3, characterized in that, The process of comparing the behavioral profiles of all pharmacies from multiple dimensions, analyzing the differences between the behavioral profiles of different pharmacies, and setting anomaly judgment criteria based on the analysis results to screen out the specific anomalies of abnormal pharmacies includes the following steps: S31. Collect behavioral profile data of all pharmacies and group and classify them according to dimensions such as region, scale, and product category structure; S32. Based on the profiling indicators, pharmacies are divided into several behavioral pattern groups; and the variances of core indicators among the groups, including the recommendation rate of high-priced drugs and the deviation of medical insurance prices, are calculated to identify the mainstream behavioral patterns and significant outliers. The specific formula is as follows: ； in, For the number of groups, For the first The number of pharmacies in each group For the first Average recommendation rate for high-priced drugs in each group This represents the average recommendation rate for high-priced drugs across all pharmacies. Variance of recommendation rate for high-priced drugs; ； in, For the variance of medical insurance price deviation, No. One group, For groups A single pharmacy in the country, For pharmacies The total number of drugs covered by medical insurance sold. For pharmacies The Deviation of the actual selling price of individual medical insurance drugs For groups The deviation of the average medical insurance price; S33. By comparing the mean within a group with the distance between groups, the key feature dimensions that cause the differences are identified, and in conjunction with medical insurance policies, indicators that significantly deviate from the normality of the group are marked as outliers to be tested. S34. Using the standard deviation multiple method, set the difference tolerance threshold for each dimension; automatically mark pharmacies that exceed the threshold as initial screening abnormal objects, and associate them with their specific abnormal indicators.

5. A medical insurance management method according to claim 1, characterized in that, The specific analysis and comparison of the abnormal points of abnormal pharmacies with their selling price, cost price, market price, and medical insurance payment standard price, and the retrieval of electronic prescriptions to determine whether there is any behavior of inducing prescriptions to increase the price of drugs, includes the following steps: S41. Based on the abnormal pharmacy and the specific abnormality, automatically generate the real-time selling price, pharmaceutical company distribution cost price, provincial bidding platform average market price, and medical insurance payment standard price of the corresponding medicine; S42. Calculate the gross profit margin of abnormal drugs at pharmacies, the deviation between the selling price and the market price, and the rate of exceeding the medical insurance payment standard, and determine the main sources of price difference; S43. Retrieve the original electronic prescription records associated with the abnormal drugs, compare them with the final sales prescription at the pharmacy, and identify unnecessary modifications in the prescriptions; S44. Establish a rule linking prescription modification, drug price increase, and profit to verify the behavior of inducing drug sales. S45. Set violation levels based on the degree of price difference exceeding the limit and the frequency of prescription modification. Specifically: a single inducement with a price difference of <50% is judged as a Level 1 violation, and three cumulative modifications in a month with a price difference of >100% are judged as a Level 2 violation. Generate a comparison chart of prescription modifications and a price difference radar chart.

6. The medical insurance management method according to claim 5, characterized in that, The obtained feature text to be retrieved is matched with the feature text library for similarity matching to establish a prescription modification-drug price increase-profit association rule, which is used to verify the inducement of drug promotion behavior. The specific content of the association rule is as follows: S441. ① Statistical analysis of the increase rate of drug unit price after modification; ② Cross-analysis of modification behavior and store staff recommendation records; ③ Verification of profit-sharing agreements between high-priced drugs and pharmacies, triple verification of drug-inducing behavior.

7. A medical insurance management method according to claim 6, characterized in that, When a pharmacy is found to have engaged in illegal practices such as inducing customers to purchase overpriced medications, the pharmacy's medical insurance settlement process will be adjusted. For pharmacies where no violations were found in the analysis but questions remain, professional personnel will be organized to conduct on-site investigations, including the following steps: S51. Obtain violation data and analyze the violation level, the list of drugs involved, and the period of violation transactions. If a pharmacy has a Level 1 violation, notify that the medical insurance payment amount for the drug will be automatically reduced to the medical insurance standard price and set a settlement delay. If a pharmacy has a Level 2 violation, freeze all medical insurance transaction settlement permissions of the pharmacy and generate a list of deductions to be made. S52. Calculate the investigation priority for pharmacies that are deemed questionable, select the top 20% of pharmacies to enter the verification queue, and organize professional personnel to conduct on-site investigations.

8. A medical insurance management system, characterized in that, The system adopts a medical insurance management method as described in any one of claims 1-7, including a data acquisition module, a profile generation module, an anomaly screening module, an anomaly analysis module, and an anomaly handling module; The data acquisition module is used to collect sales data from all pharmacies in the region, patient medical insurance drug purchase records, patient feedback data, and pharmaceutical company distribution data, and to preprocess the data and then integrate and correlate the data after processing. The profile generation module analyzes the pricing strategies and prescription recommendation behaviors of pharmacies' drug sales and generates a unique behavioral profile for each pharmacy. The anomaly screening module compares the behavioral profiles of pharmacies of the same type, analyzes the differences in the behavioral profiles of different pharmacies, and sets anomaly judgment criteria to screen out the specific anomalies of abnormal pharmacies. The anomaly analysis module performs specific analysis on the anomalies of abnormal pharmacies to determine whether there is any behavior that induces the prescription to increase the price of drugs. When a pharmacy is found to be engaging in illegal practices such as inducing customers to purchase drugs at inflated prices, the exception handling module will automatically notify the pharmacy to adjust its medical insurance settlement process. For pharmacies where no violations are found in the analysis but there are still doubts, professional personnel will be notified to conduct on-site investigations.

Images