A pharmacoeconomic evaluation method, system, device, and storage medium for health insurance decision-making.

Abstract

This application discloses a pharmacoeconomic evaluation method, system, device, and storage medium for medical insurance decision-making, relating to the field of medical data processing technology. Starting from multi-source heterogeneous real-world raw data, it goes through patient-level correlation, standardization processing, confounding factor correction, missing data processing, cost-effectiveness calculation, and dynamic simulation prediction, and finally outputs a regulatory decision report containing incremental cost-effectiveness ratio and budget impact range. This enables the pharmacoeconomic evaluation process to achieve integrated automatic processing from raw data to decision output in a real-world data environment, reducing data bias and efficiency loss caused by manual processing. At the same time, the internal validity of the evaluation results is improved through confounding factor correction and missing data processing, and the predictive ability of the evaluation results for future medical insurance decisions is enhanced through dynamic simulation models.

Description

Technical Field

[0001] This application relates to the field of medical data processing technology, and in particular to a pharmacoeconomic evaluation method, system, device and storage medium for health insurance decision-making. Background Technology

[0002] Pharmacoeconomic evaluation is a core evidence-based tool for medical insurance access assessment, rational drug use in clinical practice, and optimal allocation of health resources. Traditional evaluation methods mainly rely on clinical trial data and idealized models, and their evaluation results have significant limitations in extrapolating to real-world clinical environments, making it difficult to accurately reflect the cost-effectiveness of drugs in complex and heterogeneous real-world populations. Current technologies for pharmacoeconomic evaluation using real-world data still face challenges such as data integration difficulties, non-standardized analysis processes, and a lack of dedicated tools for medical insurance regulatory decision-making, resulting in low utilization of real-world data and insufficient reliability of evaluation results.

[0003] Therefore, how to fully explore the economic value of real-world data, establish standardized evaluation processes, and provide accurate and reliable evaluation results to meet the needs of medical insurance regulatory decision-making are technical problems that urgently need to be solved in this field. Summary of the Invention

[0004] To address the aforementioned issues, this application provides a pharmacoeconomic evaluation method, system, device, and storage medium for healthcare decision-making, including the following: Firstly, this application provides a pharmacoeconomic evaluation method for healthcare decision-making, which includes: Acquire multi-source, heterogeneous, real-world raw data for medical insurance access assessment; The raw data is subjected to patient-level association and standardization processing to generate an analytical dataset; The analytical dataset is subjected to confounding factor correction and missing data processing. Based on the processed analytical dataset, standardized cost-effectiveness calculation results are generated. A dynamic simulation model is constructed based on the cost-effectiveness calculation results, and cost-effectiveness prediction data is generated through the simulation model. Based on the cost-effectiveness prediction data, a regulatory decision output report is generated, which includes the incremental cost-effectiveness ratio and the budget impact range.

[0005] Optionally, obtaining multi-source heterogeneous real-world raw data for medical insurance access assessment includes: Obtain medical insurance claim settlement data, disease registry data, and electronic medical record data from different medical institutions.

[0006] Optionally, the step of performing patient-level association and standardization processing on the original data to generate an analytical dataset includes: Using the patient's unique identifier as the key field, outpatient records, inpatient records, medication records, examination and test results, and expense details of the same patient from different data sources are linked together to form a complete patient medical treatment sequence that includes different times and different medical institutions; The data in the patient's medical treatment sequence is mapped to the national standard coding system to obtain a standardized analytical dataset.

[0007] Optionally, the process of correcting for confounding factors and handling missing data in the analytical dataset includes: The baseline characteristics of patients in the analytical dataset in terms of age, sex, disease severity, and comorbidities were adjusted by using any one of the propensity score matching method, inverse probability weighting method, or instrumental variable method to complete the correction for confounding factors. The corrected data is then processed using a missing data handling strategy to obtain a processed analytical dataset. The missing data handling strategy includes removing samples with missing values ​​or estimating and filling in missing values ​​based on existing information.

[0008] Optionally, generating standardized cost-effectiveness calculation results based on the processed analytical dataset includes: Based on the processed analytical dataset, the direct medical costs of each patient are collected. Health output data are calculated based on clinical outcome indicators, including life years or quality-adjusted life years. Based on the collected cost data and calculated health output data, the incremental cost-effectiveness ratio is calculated and the calculation results are output. The calculation results include the mean cost, mean health output, incremental cost-effectiveness ratio and confidence interval of each drug group.

[0009] Optionally, the collection of direct medical costs for each patient includes: The actual direct medical costs for each patient are collected according to a pre-defined cost structure; wherein, the direct medical costs include drug costs, hospitalization costs, outpatient costs, and examination and testing costs. Alternatively, a unit price mapping method based on standard treatment pathways can be used to collect theoretical direct medical costs. The unit price mapping method determines the standard treatment plan by analyzing clinical guidelines for the target disease, obtains the unit dose price from the drug procurement platform or medical insurance payment standard, and accumulates the theoretical treatment cost item by item according to the standard treatment plan.

[0010] Optionally, generating a regulatory decision output report based on the cost-effectiveness prediction data, which includes the incremental cost-effectiveness ratio and the budget impact range, includes: The incremental cost-effectiveness ratio in the cost-effectiveness prediction data is compared with a preset willingness-to-pay threshold to generate an economic assessment conclusion. If the economic assessment conclusion meets the preset access conditions, based on the expected user population size, market penetration rate and price assumptions of the target drug, the impact range of the drug on medical insurance fund expenditure in the future preset period after being included in the medical insurance reimbursement catalog is calculated. The economic assessment conclusion and the impact range are integrated according to a preset decision report template to generate a regulatory decision output report.

[0011] Secondly, this application provides a pharmacoeconomic evaluation system for health insurance decision-making, the system comprising: The acquisition unit is used to acquire multi-source heterogeneous real-world raw data for medical insurance access assessment. A standardization processing unit is used to perform patient-level correlation and standardization processing on the raw data to generate an analytical dataset. The correction unit is used to perform confounding factor correction and missing data processing on the analytical dataset; The measurement unit is used to generate standardized cost-effectiveness measurement results based on the processed analytical dataset; The simulation prediction unit is used to construct a dynamic simulation model based on the cost-effectiveness calculation results, and generate cost-effectiveness prediction data through the simulation model. The report generation unit is used to generate a regulatory decision output report based on the cost-effectiveness prediction data, which includes the incremental cost-effectiveness ratio and the budget impact range.

[0012] Optionally, the acquisition unit is specifically used to acquire medical insurance claim settlement data, disease registry data, and electronic medical record data from different medical institutions.

[0013] Optionally, the standardized processing unit is specifically used to connect the outpatient records, inpatient records, medication records, examination and test results and expense details of the same patient from different data sources, using the patient's unique identifier as the key field, to form a complete patient medical treatment sequence that includes different times and different medical institutions; The data in the patient's medical treatment sequence is mapped to the national standard coding system to obtain a standardized analytical dataset.

[0014] Optionally, the correction unit is specifically used to adjust the baseline characteristics of patients in the analytical dataset in terms of age, sex, disease severity and comorbidities among different drug groups using any one of propensity score matching, inverse probability weighting or instrumental variable methods, in order to complete the correction of confounding factors. The corrected data is then processed using a missing data handling strategy to obtain a processed analytical dataset. The missing data handling strategy includes removing samples with missing values ​​or estimating and filling in missing values ​​based on existing information.

[0015] Optionally, the calculation unit is specifically used to collect the direct medical costs of each patient based on the processed analytical dataset; Health output data are calculated based on clinical outcome indicators, including life years or quality-adjusted life years. Based on the collected cost data and calculated health output data, the incremental cost-effectiveness ratio is calculated and the calculation results are output. The calculation results include the mean cost, mean health output, incremental cost-effectiveness ratio and confidence interval of each drug group.

[0016] Optionally, the collection of direct medical costs for each patient includes: The actual direct medical costs for each patient are collected according to a pre-defined cost structure; wherein, the direct medical costs include drug costs, hospitalization costs, outpatient costs, and examination and testing costs. Alternatively, a unit price mapping method based on standard treatment pathways can be used to collect theoretical direct medical costs. The unit price mapping method determines the standard treatment plan by analyzing clinical guidelines for the target disease, obtains the unit dose price from the drug procurement platform or medical insurance payment standard, and accumulates the theoretical treatment cost item by item according to the standard treatment plan.

[0017] Optionally, the report generation unit is specifically used to compare the incremental cost-effectiveness ratio in the cost-effectiveness prediction data with a preset willingness-to-pay threshold to generate an economic assessment conclusion. If the economic assessment conclusion meets the preset access conditions, based on the expected user population size, market penetration rate and price assumptions of the target drug, the impact range of the drug on medical insurance fund expenditure in the future preset period after being included in the medical insurance reimbursement catalog is calculated. The economic assessment conclusion and the impact range are integrated according to a preset decision report template to generate a regulatory decision output report.

[0018] Thirdly, this application provides an apparatus comprising a memory and a processor, the memory for storing instructions or code, and the processor for executing the instructions or code to cause the apparatus to perform the pharmacoeconomic evaluation method for healthcare decision-making described in any of the implementations of the first aspect.

[0019] Fourthly, this application provides a computer-readable storage medium storing code, wherein when the code is executed, a device running the code implements the pharmacoeconomic evaluation method for healthcare decision-making described in any of the implementations of the first aspect.

[0020] This application provides a pharmacoeconomic evaluation method for healthcare insurance decision-making. When implementing the method, firstly, multi-source heterogeneous real-world raw data for healthcare insurance access assessment is acquired. Then, the raw data undergoes patient-level correlation and standardization processing to generate an analytical dataset. Next, the analytical dataset undergoes confounding factor correction and missing data processing. Based on the processed analytical dataset, standardized cost-effectiveness calculation results are generated. Finally, a dynamic simulation model is constructed based on the cost-effectiveness calculation results. Cost-effectiveness prediction data is generated through the simulation model. Based on the cost-effectiveness prediction data, a regulatory decision output report including incremental cost-effectiveness ratio and budget impact range is generated.

[0021] In this way, by starting with multi-source, heterogeneous real-world raw data, and through patient-level correlation, standardization, confounding factor correction, missing data handling, cost-effectiveness calculation, and dynamic simulation prediction, the final output is a regulatory decision report containing incremental cost-effectiveness ratios and budget impact ranges. This enables the pharmacoeconomic evaluation process to achieve integrated automated processing from raw data to decision output in a real-world data environment, reducing data bias and efficiency losses caused by manual processing. Simultaneously, confounding factor correction and missing data handling improve the internal validity of the evaluation results, and the dynamic simulation model enhances the predictive ability of the evaluation results for future medical insurance decisions. Thus, economic evidence required for medical insurance access assessment can be quickly generated based on real-world data, reducing the time cost of data cleaning and integration in traditional pharmacoeconomic evaluations. Furthermore, the standardized evaluation process improves the comparability of evaluation results among different drugs, providing quantitative economic basis for medical insurance catalog access decisions. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in this embodiment or the prior art, the drawings used in the description of the embodiment or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart illustrating a pharmacoeconomic evaluation method for healthcare decision-making, provided as an embodiment of this application; Figure 2 This is a schematic diagram of the structure of a pharmacoeconomic evaluation system for health insurance decision-making, provided as an embodiment of this application. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0025] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0026] Figure 1 A flowchart illustrating a pharmacoeconomic evaluation method for healthcare decision-making, provided as an embodiment of this application. (Combined with...) Figure 1 As shown in the embodiments of this application, the pharmacoeconomic evaluation method for health insurance decision-making may include: S101. Obtain multi-source heterogeneous real-world raw data for medical insurance access assessment.

[0027] Multi-source heterogeneous real-world raw data refers to medical-related data originating from different business systems and possessing different data structures and storage formats. Specifically, this includes: electronic medical record data from medical institution information systems, which records patients' diagnoses, medications, examinations, and tests during outpatient and inpatient periods; medical insurance settlement data from medical insurance settlement platforms, i.e., medical insurance claim settlement data, which reflects the medical expenses incurred by patients each time they seek medical treatment and their medical insurance payment status; and disease-specific follow-up data from disease registration systems, i.e., disease registry data, which typically involves long-term tracking and recording of specific disease populations.

[0028] When these raw data are actually accessed, they may be transmitted to the system in unstructured text formats such as doctors' handwritten medical records, structured relational database tables such as the expense details table of the medical insurance settlement system, or semi-structured exchange document formats such as medical data exchange files in XML or JSON format. By acquiring the above-mentioned multi-source heterogeneous data, the limitations of traditional pharmacoeconomic evaluation relying solely on clinical trial data are overcome. This allows for a comprehensive reflection of the drug's effectiveness and cost consumption in real clinical practice, providing a more realistic data foundation for medical insurance access decisions.

[0029] S102. Perform patient-level association and standardization processing on the original data to generate an analytical dataset.

[0030] Patient-level association refers to the process of identifying and connecting various medical records of the same patient scattered across different data sources, using the individual patient as the basic unit. Standardization processing refers to converting medical data from different data sources with varying coding systems into a unified expression format that conforms to national standards and specifications. The analytical dataset generated after these two processes serves as the foundational data source for subsequent pharmacoeconomic evaluation calculations.

[0031] Specifically, the system first uses the patient's unique identifier as the key field to link outpatient records, inpatient records, medication records, examination and test results, and expense details of the same patient from different data sources, forming a complete patient medical treatment sequence that includes different times and medical institutions. The patient's unique identifier can be an ID card number or a medical insurance card number. These two identifiers have nationwide uniqueness and stability, making them suitable as a basis for cross-institutional and cross-time data association.

[0032] In terms of implementation, a patient master index technology is adopted. This technology creates a master index table that maps different identifiers of the same patient across various business systems to a single globally unique internal identifier. For example, a patient with a Class A disease may have undergone chemotherapy at Hospital A, with corresponding medication and cost information recorded. Six months later, they may have transferred to Hospital B for further treatment and received new test results. Through the patient master index technology, it can be identified that the two hospitals recorded the same patient, thus linking these scattered records to form a complete medical timeline from initial diagnosis to subsequent treatment. This linking operation breaks down data silos between medical institutions, allowing previously isolated medical records to be aggregated by patient dimension, providing data support for reconstructing the patient's true and continuous clinical treatment process.

[0033] After completing the patient-level association, the module maps the data in the patient's medical treatment sequence to the national standard coding system, resulting in a standardized analytical dataset. The mapping operation is implemented as follows: a coding lookup table is established, creating a one-to-one correspondence between the local codes actually used in each data source and the national standard codes. Then, through programmatic field conversion, the original codes are batch-replaced with the standard codes. For example, for diagnostic codes, the in-hospital disease codes used by different hospitals are uniformly mapped to the corresponding national standard version; for surgical codes, they are uniformly mapped to the corresponding national standard version; for drug codes, they are uniformly mapped to the unified drug codes in the National Medical Insurance Drug Catalog; and for expense items, the expense item names defined by each hospital, such as treatment fees, material fees, and drug fees, are uniformly collected under the standard expense classification system stipulated in the medical insurance settlement system.

[0034] While performing encoding mapping, the module also identifies and removes abnormal records that clearly do not conform to clinical logic. Examples include records with negative costs resulting from irregular system refund operations, and records with reversed hospitalization and discharge times due to data entry errors. This removal operation is automatically completed through preset data quality verification rules, such as setting constraints that the cost field must be greater than or equal to zero, and the discharge time must be later than or equal to the admission time. This standardization and cleaning process eliminates encoding heterogeneity issues between different data sources, allowing data from different institutions to be compared and summarized under the same encoding system. Simultaneously, it removes obviously erroneous data records, preventing these corrupted data from interfering with subsequent analysis results, thus ensuring the consistency, accuracy, and usability of the analytical dataset.

[0035] S103. Perform confounding correction and missing data processing on the analytical dataset, and generate standardized cost-effectiveness calculation results based on the processed analytical dataset.

[0036] Confounding factors refer to those factors that are both relevant to the studied drug intervention and can independently affect patient clinical outcomes, such as patient age, sex, disease severity, and comorbidities. In real-world data, patients receiving different drug treatments often exhibit differences in these baseline characteristics. Directly comparing clinical outcomes under different drug regimens may lead to results distorted by these differences, rather than accurately reflecting the differences in the efficacy of the drugs themselves. Missing data refers to situations where certain variable values ​​for some patients in the dataset were not recorded. This step systematically processes confounding factors and missing data, ultimately outputting standardized cost-effectiveness calculation results. This eliminates common biases and data incompleteness issues in real-world data, making pharmacoeconomic evaluation results based on real-world data more reliable and robust, thereby providing higher-quality evidence-based support for health insurance access decisions.

[0037] Specifically, the baseline characteristics of patients in the analytical dataset regarding age, sex, disease severity, and comorbidities are first adjusted using any one of the following methods: propensity score matching, inverse probability weighting, or instrumental variable method, to correct for confounding factors between different drug groups. This confounding factor correction operation can best simulate the randomization effect in randomized controlled trials, reduce the interference of selection bias and confounding bias on drug efficacy assessment, and make the treatment group and control group more comparable, thereby improving the accuracy and reliability of pharmacoeconomic evaluation conclusions.

[0038] After correcting for confounding factors, the corrected data is processed using a missing data handling strategy to obtain the processed analytical dataset. This missing data handling strategy includes removing samples with missing values ​​or estimating and imputing missing values ​​based on existing information. This missing data handling operation avoids sample size loss and information waste caused by incomplete data, while reducing bias that may be introduced by improper missing data handling, thus ensuring the integrity and representativeness of the analytical dataset.

[0039] Next, based on the processed analytical dataset, the direct medical costs for each patient are aggregated. In terms of aggregation methods, the actual direct medical costs for each patient can be aggregated according to a preset cost composition criterion; these direct medical costs include drug costs, hospitalization costs, outpatient costs, and examination and testing costs. Alternatively, theoretical direct medical costs can be aggregated based on a unit price mapping method using standard treatment pathways; this method determines standard treatment plans by analyzing clinical guidelines for the target disease, obtains unit dose prices from drug procurement platforms or medical insurance payment standards, and accumulates these costs item by item according to the standard treatment plan to obtain the theoretical treatment cost. This cost aggregation operation provides a flexible cost calculation scheme, adaptable to the needs of different data quality and usage scenarios. It can utilize real-world cost data for real-world assessments and perform simulation predictions through standardized pathway calculations, providing medical insurance decision-makers with multi-faceted cost evidence.

[0040] Simultaneously, health output data is calculated based on clinical outcome indicators. This health output data includes life years or quality-adjusted life years. This health output calculation uses commonly used health output indicators in pharmacoeconomics, ensuring comparability of evaluation results across different drugs and disease areas. The quality-adjusted life years incorporate both survival time and quality of life dimensions, providing a more comprehensive reflection of the overall health benefits the drug brings to patients.

[0041] Finally, based on the collected cost data and calculated health output data, the incremental cost-effectiveness ratio is calculated, and the results are output. These results include the mean cost, mean health output, incremental cost-effectiveness ratio, and confidence interval for each drug group. This output provides an intuitive and quantitative economic basis for medical insurance access decisions. Decision-makers can determine whether a drug has cost-effectiveness based on a preset willingness-to-pay threshold. Furthermore, the confidence interval allows decision-makers to fully understand the uncertainty of the estimated results, thus enabling more prudent and scientific decisions.

[0042] S104. Based on the cost-effectiveness calculation results, construct a dynamic simulation model, and generate cost-effectiveness prediction data through the simulation model.

[0043] A dynamic simulation model is a computational model used to simulate the evolution of a disease over time. Cost-effectiveness prediction data refers to cost-effectiveness estimates over a longer future period obtained through model extrapolation.

[0044] This step receives the cost-effectiveness calculation results from step S103 and extracts time-series data on patient disease progression from the standard analysis dataset as the parameter basis for the dynamic simulation. This dynamic simulation model can be constructed using decision trees or Markov models. Based on real-world observations of patient disease evolution patterns and treatment pathway transitions, it builds a dynamic model encompassing multiple health states, such as disease stability, disease progression, complication onset, and death, and estimates the transition probabilities between each state based on the data. The transition probability refers to the likelihood of a patient transitioning from their current healthy state to another healthy state. This dynamic model construction operation quantifies real-world disease evolution patterns into computable mathematical parameters, giving the simulation results clinical credibility.

[0045] Monte Carlo simulation technology is used to generate large-scale virtual patient cohorts, simulating the migration of each patient between different health states over a long timeframe (e.g., lifetime or 20 years), and accumulating and recording the costs and health outputs generated in each simulation cycle. This approach enables the generation of statistically stable long-term cost-effectiveness predictions through computer simulation when long-term real-world follow-up is not feasible.

[0046] Simultaneously, sensitivity analysis is conducted by iteratively changing the values ​​of key parameters to observe their impact on the evaluation results, quantifying the uncertainty of the simulation results. The final output is a dynamic simulation results dataset containing long-term cost-effectiveness predictions, probabilistic sensitivity analysis results, and cost-effectiveness acceptable curves, thus providing healthcare decision-makers with long-term, uncertainty-considered pharmacoeconomic information.

[0047] S105. Generate a regulatory decision output report based on the cost-effectiveness prediction data, which includes the incremental cost-effectiveness ratio and the budget impact range.

[0048] This step receives the dynamic simulation result dataset from step S104 and uses it as the raw material for generating decision information. First, the incremental cost-effectiveness ratio calculated through simulation is compared with a preset willingness-to-pay threshold. This automatically generates a judgment on whether the target drug is economically viable compared to the control drug. The willingness-to-pay threshold refers to the maximum amount that health insurance decision-makers are willing to pay to obtain one unit of health output. This comparison provides a clear standard for evaluating the economic viability of drugs, making the decision-making process more objective and transparent.

[0049] If the economic feasibility assessment meets the preset inclusion criteria, the budget impact analysis function is activated. Based on the expected user population size, market penetration rate, and price assumptions of the target drug, the impact range on medical insurance fund expenditures over a certain period after the drug is included in the medical insurance reimbursement catalog is calculated. This calculation, assuming the drug is economically viable, further assesses the affordability of the medical insurance fund, providing decision-makers with information on the balance between drug inclusion and fund security.

[0050] The above analysis results are organized according to the standardized format of a medical insurance access assessment report, generating a complete decision report that includes an abstract, research background, research methods, basic analysis results, sensitivity analysis results, budget impact analysis, and conclusions. Simultaneously, a visual interface transforms complex simulation data into intuitive charts such as cost-effectiveness scatter plots, acceptable curves, and budget impact waterfall plots, allowing decision-makers to interactively adjust key parameters and observe real-time trends. Finally, the decision report and visual analysis interface are delivered to medical insurance regulatory authorities via web browser or encrypted document format, completing the entire transformation process from raw data to regulatory decision support.

[0051] In one implementation of this application, the functional units corresponding to the above steps can adopt different organizational schemes at the system architecture level. One implementation encapsulates the four core functions—data fusion and governance, confounding factor correction and missing data processing, dynamic simulation and deduction, and regulatory decision output—into a single system through an integrated solution. Data flows directly between the functional units without external interface interaction. This solution improves data processing security and operational efficiency. Another implementation uses a distributed service approach, where each function is broken down into independent microservices, and data interaction and function calls are performed through standardized application programming interfaces (APIs). This approach enhances system flexibility and scalability, supporting multi-agency collaboration and distributed computing. In practical applications, any of the above architectural schemes can be selected for implementation based on the specific deployment environment, data security requirements, and organizational collaboration needs.

[0052] To make the pharmacoeconomic evaluation method provided in this application more comprehensive and sufficient in practical applications, this application also provides some preparatory and auxiliary operational steps. These steps belong to the basic configuration work at the system level and do not need to be repeated in each evaluation for a specific drug. By pre-constructing standardized processing rules and output templates, they effectively enrich the method's adaptability to multi-source heterogeneous data and the presentation of decision results.

[0053] In one implementation of this application, the definition of coding mapping rules in data fusion and governance is a preparatory step. This work involves mapping various diagnostic codes, surgical codes, drug codes, and expense items from different data sources to the national standard coding system during the initial stage of system construction. Once this mapping relationship is established, it can be reused for a long time. Subsequent newly accessed data only needs to be automatically converted according to the established mapping rules, without the need to redefine the coding correspondence each time an evaluation is conducted.

[0054] In one implementation of this application, the configuration of data quality control rules in data fusion and governance is a preparatory step. This work involves setting up anomaly identification logic based on clinical expertise during the system initialization phase. For example, rules such as costs cannot be negative, hospitalization dates cannot be later than discharge dates, and age and diagnosis cannot contradict each other are set. Once these rules are configured, they will continue to function, and the system will automatically call the existing rule base for anomaly identification each time data is processed, without the need for repeated configuration.

[0055] In one implementation of this application, the selection of a confounding factor correction strategy in standardized evaluation analysis is a preparatory step. This involves researchers, based on their expertise, pre-determining the set of key covariates requiring correction and the specific correction methods used when designing a particular class of drugs or a particular study. Once determined, this strategy can be reused in multiple evaluations of similar drugs without requiring re-selection each time.

[0056] In one implementation of this application, the definition of cost composition in standardized evaluation analysis is a preparatory step. This work is carried out during the system construction phase, based on the needs of medical insurance regulatory decisions, to pre-determine which expense items are included in the scope of direct medical cost statistics, such as cost collection strategies. Once this definition is determined, it serves as a standard template for continuous use, ensuring comparability between different drug evaluations.

[0057] In one implementation of this application, the customization of the report template in the regulatory decision output is a follow-up or preparatory step. This involves designing a report format and chart style that conforms to the access assessment standards according to the requirements of the medical insurance regulatory department after the system is built. Once this template is customized, it can be used long-term, and subsequent evaluation results can be directly filled into the established template to generate reports without needing to redesign the report format each time.

[0058] Through the above embodiments, this application provides a pharmacoeconomic evaluation method for medical insurance decision-making. Driven by real-world data, it uses dynamic simulation analysis algorithms to simulate the long-term cost-effectiveness evolution of drugs in real clinical environments. Combining probability sensitivity analysis and scenario analysis, it quantifies the uncertainty of evaluation results. This provides medical insurance departments with refined calculation results covering key indicators such as incremental cost-effectiveness ratio, cost-effectiveness acceptable curve, and budget impact range. Furthermore, it fills the application gap of existing technologies in macro-regulatory decision-making fields such as medical insurance access assessment and drug price negotiation, enabling medical insurance decision-makers to optimize the allocation of limited health resources based on evidence closer to real clinical environments, significantly improving the scientific rigor and accuracy of decision-making.

[0059] The above are some specific implementations of a pharmacoeconomic evaluation method for health insurance decision-making provided in the embodiments of this application. Based on this, this application also provides a corresponding system. The system provided in the embodiments of this application will be described below from the perspective of functional modularization.

[0060] Figure 2 This is a schematic diagram illustrating the structure of a pharmacoeconomic evaluation system for healthcare decision-making, provided as an embodiment of this application. (Combined with...) Figure 2 As shown in the embodiments of this application, the pharmacoeconomic evaluation system 200 for medical insurance decision-making includes: Acquisition unit 210 is used to acquire multi-source heterogeneous real-world raw data for medical insurance access assessment; The standardization processing unit 220 is used to perform patient-level correlation and standardization processing on the raw data to generate an analytical dataset. Correction unit 230 is used to perform confounding factor correction and missing data processing on the analytical dataset; The measurement unit 240 is used to generate standardized cost-effectiveness measurement results based on the processed analytical dataset; The simulation prediction unit 250 is used to construct a dynamic simulation model based on the cost-effectiveness calculation results and generate cost-effectiveness prediction data through the simulation model. The report generation unit 260 is used to generate a regulatory decision output report based on the cost-effectiveness prediction data, which includes the incremental cost-effectiveness ratio and the budget impact range.

[0061] In one implementation of this application, the acquisition unit is specifically used to acquire medical insurance claim settlement data, disease registry data, and electronic medical record data from different medical institutions.

[0062] In one implementation of this application, the standardized processing unit is specifically used to connect the outpatient records, inpatient records, medication records, examination and test results and expense details of the same patient from different data sources, using the patient's unique identifier as the key field, to form a complete patient medical treatment sequence that includes different times and different medical institutions. The data in the patient's medical treatment sequence is mapped to the national standard coding system to obtain a standardized analytical dataset.

[0063] In one implementation of this application, the correction unit is specifically used to adjust the baseline characteristics of patients in the analytical dataset in terms of age, gender, disease severity and comorbidities among different drug groups using any one of propensity score matching, inverse probability weighting, or instrumental variable method, so as to complete the correction of confounding factors. The corrected data is then processed using a missing data handling strategy to obtain a processed analytical dataset. The missing data handling strategy includes removing samples with missing values ​​or estimating and filling in missing values ​​based on existing information.

[0064] In one implementation of this application, the calculation unit is specifically used to collect the direct medical costs of each patient based on the processed analytical dataset. Health output data are calculated based on clinical outcome indicators, including life years or quality-adjusted life years. Based on the collected cost data and calculated health output data, the incremental cost-effectiveness ratio is calculated and the calculation results are output. The calculation results include the mean cost, mean health output, incremental cost-effectiveness ratio and confidence interval of each drug group.

[0065] In one implementation of this application, the collection of direct medical costs for each patient includes: The actual direct medical costs for each patient are collected according to a pre-defined cost structure; wherein, the direct medical costs include drug costs, hospitalization costs, outpatient costs, and examination and testing costs. Alternatively, a unit price mapping method based on standard treatment pathways can be used to collect theoretical direct medical costs. The unit price mapping method determines the standard treatment plan by analyzing clinical guidelines for the target disease, obtains the unit dose price from the drug procurement platform or medical insurance payment standard, and accumulates the theoretical treatment cost item by item according to the standard treatment plan.

[0066] In one implementation of this application, the report generation unit is specifically used to compare the incremental cost-effectiveness ratio in the cost-effectiveness prediction data with a preset willingness-to-pay threshold to generate an economic assessment conclusion. If the economic assessment conclusion meets the preset access conditions, based on the expected user population size, market penetration rate and price assumptions of the target drug, the impact range of the drug on medical insurance fund expenditure in the future preset period after being included in the medical insurance reimbursement catalog is calculated. The economic assessment conclusion and the impact range are integrated according to a preset decision report template to generate a regulatory decision output report.

[0067] This application also provides corresponding devices and computer storage media for implementing the solutions provided in this application.

[0068] The device includes a memory and a processor. The memory stores instructions or code, and the processor executes the instructions or code to cause the device to perform the method described in any embodiment of this application.

[0069] The computer storage medium stores code, and when the code is run, the device running the code implements the method described in any embodiment of this application.

[0070] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that all or part of the steps in the methods of the above embodiments can be implemented by means of software plus a general-purpose hardware platform. Based on this understanding, the technical solution of this application can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as a read-only memory (ROM) / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, a server, or a network communication device such as a router) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0071] It is understood that in the specific embodiments of this application, the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved need to obtain user permission or consent when the above embodiments of this application are applied to specific products or technologies, and the collection, use and processing of related data need to comply with the relevant laws, regulations and standards of relevant countries and regions.

[0072] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0073] It should also be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for the device and system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments. The device and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components indicated as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0074] The above description is merely one specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A pharmacoeconomic evaluation method for health insurance decision-making, characterized in that, The method includes: Acquire multi-source, heterogeneous, real-world raw data for medical insurance access assessment; The raw data is subjected to patient-level association and standardization processing to generate an analytical dataset; The analytical dataset is subjected to confounding factor correction and missing data processing. Based on the processed analytical dataset, standardized cost-effectiveness calculation results are generated. A dynamic simulation model is constructed based on the cost-effectiveness calculation results, and cost-effectiveness prediction data is generated through the simulation model. Based on the cost-effectiveness prediction data, a regulatory decision output report is generated, which includes the incremental cost-effectiveness ratio and the budget impact range.

2. The method according to claim 1, characterized in that, The acquisition of multi-source heterogeneous real-world raw data for medical insurance access assessment includes: Obtain medical insurance claim settlement data, disease registry data, and electronic medical record data from different medical institutions.

3. The method according to claim 1, characterized in that, The step of performing patient-level association and standardization processing on the original data to generate an analytical dataset includes: Using the patient's unique identifier as the key field, outpatient records, inpatient records, medication records, examination and test results, and expense details of the same patient from different data sources are linked together to form a complete patient medical treatment sequence that includes different times and different medical institutions; The data in the patient's medical treatment sequence is mapped to the national standard coding system to obtain a standardized analytical dataset.

4. The method according to claim 1, characterized in that, The confounding correction and missing data processing for the analytical dataset includes: The baseline characteristics of patients in the analytical dataset in terms of age, sex, disease severity, and comorbidities were adjusted by using any one of the propensity score matching method, inverse probability weighting method, or instrumental variable method to complete the correction for confounding factors. The corrected data is then processed using a missing data handling strategy to obtain a processed analytical dataset. The missing data handling strategy includes removing samples with missing values ​​or estimating and filling in missing values ​​based on existing information.

5. The method according to claim 1, characterized in that, The standardized cost-effectiveness calculation results generated based on the processed analytical dataset include: Based on the processed analytical dataset, the direct medical costs of each patient are collected. Health output data are calculated based on clinical outcome indicators, including life years or quality-adjusted life years. Based on the collected cost data and calculated health output data, the incremental cost-effectiveness ratio is calculated and the calculation results are output. The calculation results include the mean cost, mean health output, incremental cost-effectiveness ratio and confidence interval of each drug group.

6. The method according to claim 5, characterized in that, The collected direct medical costs for each patient include: The actual direct medical costs for each patient are collected according to a pre-defined cost structure; wherein, the direct medical costs include drug costs, hospitalization costs, outpatient costs, and examination and testing costs. Alternatively, a unit price mapping method based on standard treatment pathways can be used to collect theoretical direct medical costs. The unit price mapping method determines the standard treatment plan by analyzing clinical guidelines for the target disease, obtains the unit dose price from the drug procurement platform or medical insurance payment standard, and accumulates the theoretical treatment cost item by item according to the standard treatment plan.

7. The method according to claim 1, characterized in that, The regulatory decision output report generated based on the cost-effectiveness prediction data, which includes the incremental cost-effectiveness ratio and the budget impact range, includes: The incremental cost-effectiveness ratio in the cost-effectiveness prediction data is compared with a preset willingness-to-pay threshold to generate an economic assessment conclusion. If the economic assessment conclusion meets the preset access conditions, based on the expected user population size, market penetration rate and price assumptions of the target drug, the impact range of the drug on medical insurance fund expenditure in the future preset period after being included in the medical insurance reimbursement catalog is calculated. The economic assessment conclusion and the impact range are integrated according to a preset decision report template to generate a regulatory decision output report.

8. A pharmacoeconomic evaluation system for health insurance decision-making, characterized in that, The system includes: The acquisition unit is used to acquire multi-source heterogeneous real-world raw data for medical insurance access assessment. A standardization processing unit is used to perform patient-level correlation and standardization processing on the raw data to generate an analytical dataset. The correction unit is used to perform confounding factor correction and missing data processing on the analytical dataset; The measurement unit is used to generate standardized cost-effectiveness measurement results based on the processed analytical dataset; The simulation prediction unit is used to construct a dynamic simulation model based on the cost-effectiveness calculation results, and generate cost-effectiveness prediction data through the simulation model. The report generation unit is used to generate a regulatory decision output report based on the cost-effectiveness prediction data, which includes the incremental cost-effectiveness ratio and the budget impact range.

9. A computing device, characterized in that, The computing device includes: a memory and a processor; The memory is used to store computer programs; The processor is configured to implement the steps of the method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1 to 7.

Images