HRP dynamic planning optimization model construction method based on DIP medical insurance payment mode

Abstract

The invention belongs to the technical field of model construction, and particularly relates to an HRP dynamic planning optimization model construction method based on a DIP medical insurance payment mode, which comprises the following specific steps: determining a target of a model and defining key variables; collecting various data indexes, and cleaning and preprocessing the data; selecting variables having significant influence on the target, and defining constraint conditions of the model; a multi-stage decision model is established, long-term planning is decomposed into short-term decision points, a mathematical model is constructed by applying a dynamic planning theory, and the relationship between variables is quantified; and selecting a dynamic programming algorithm to solve the model so as to realize algorithm writing program codes for calculation. The method achieves a remarkable effect in the aspect of improving the operation efficiency of the hospital. Resource allocation is more reasonable, cost control is enhanced, and service efficiency is remarkably improved. In addition, patient satisfaction investigation shows that the overall evaluation of the patient on hospital services is improved, especially in the aspects of service processes and medical quality.

Description

Technical Field

[0001] The present invention relates to the technical field of model construction, and in particular to a method for constructing an HRP dynamic programming optimization model based on a DIP medical insurance payment method. Background Art

[0002] It is of great significance to improve the efficiency of medical services, ensure the quality of medical care, and meet the health needs of the people. According to the "Guiding Opinions of the State Administration of Traditional Chinese Medicine on Strengthening the Construction of TCM Service Capacity" (State Administration of Traditional Chinese Medicine

[2018] No. 1) [3], public hospitals of traditional Chinese medicine need to strengthen internal management and improve service efficiency and quality while maintaining the unique advantages of traditional Chinese medicine. However, the current TCM hospitals have poor medical service quality and patient satisfaction during operation. Therefore, a method for constructing an HRP dynamic programming optimization model based on the DIP medical insurance payment method is invented. Summary of the invention

[0003] In view of the above-mentioned and / or existing problems in the HRP dynamic programming optimization model construction method based on the DIP medical insurance payment method, the present invention is proposed.

[0004] Therefore, the purpose of the present invention is to provide a method for constructing an HRP dynamic programming optimization model based on the DIP medical insurance payment method, which can solve the above-mentioned existing problems.

[0005] To solve the above technical problems, according to one aspect of the present invention, the present invention provides the following technical solutions:

[0006] The HRP dynamic programming optimization model construction method based on the DIP medical insurance payment method includes the following specific steps:

[0007] Step 1: Problem definition and goal setting: clarify the model's objectives and define key variables;

[0008] Step 2: Data collection and preprocessing: collect various data indicators, and clean and preprocess the data;

[0009] Step 3: Variable selection and constraint definition: Select variables that have a significant impact on the target and define the constraints of the model;

[0010] Step 4: Model construction: Establish a multi-stage decision-making model, decompose long-term planning into short-term decision points, and use dynamic programming theory to build a mathematical model to quantify the relationship between various variables;

[0011] Step 5, algorithm selection and implementation: select a dynamic programming algorithm to solve the model, implement the algorithm and write program code for calculation;

[0012] Step 6: Model verification and optimization: Verify the effectiveness of the model through simulation and empirical analysis, and adjust the model parameters to optimize the model according to the verification results;

[0013] Step 7: Save the optimization suggestions proposed by the model so that when the same data appears next time, it can be determined whether the proposed optimization suggestions are consistent to test the accuracy of the model.

[0014] As a preferred solution of the method for constructing the HRP dynamic programming optimization model based on the DIP medical insurance payment method described in the present invention, wherein: the goals of the model defined in step one include improving operational efficiency, reducing costs, and optimizing resource allocation; the key variables defined in step one include patient waiting time, bed utilization, and drug and material costs.

[0015] As a preferred solution of the HRP dynamic programming optimization model construction method based on the DIP medical insurance payment method described in the present invention, wherein: the data cleaning and preprocessing in the step 2 includes processing missing values ​​and outliers to ensure data quality.

[0016] As a preferred solution of the method for constructing the HRP dynamic programming optimization model based on the DIP medical insurance payment method described in the present invention, the constraints of the model defined in step three include resource limitations, policy regulations, and service quality standards.

[0017] As a preferred solution of the method for constructing the HRP dynamic programming optimization model based on the DIP medical insurance payment method described in the present invention, wherein: the expression formula of the multi-stage decision model in the step four includes a state transfer equation, a benefit function and an optimal value function.

[0018] As a preferred solution of the HRP dynamic programming optimization model construction method based on the DIP medical insurance payment method described in the present invention, the calculation formula of the state transfer equation is:

[0019] S t+1 =T(S t ,d t )

[0020] Among them, S t is the state of stage t, d t is the decision of the tth stage, and T is the state transition function.

[0021] As a preferred solution of the HRP dynamic programming optimization model construction method based on the DIP medical insurance payment method described in the present invention, the calculation formula of the profit function is:

[0022] r t =R(s t ,d t)

[0023] Among them, r t is the profit in the tth stage, and R is the profit function.

[0024] As a preferred solution of the HRP dynamic programming optimization model construction method based on the DIP medical insurance payment method described in the present invention, the calculation formula of the optimal value function is:

[0025] Vt(st)=md t ax{R(st,dt)+Vt +1 (T(st,dt))}

[0026] Among them, V t (s t ) is the state s from the tth stage t Initially, the maximum total benefit that can be obtained through the optimal decision sequence.

[0027] As a preferred solution of the method for constructing the HRP dynamic programming optimization model based on the DIP medical insurance payment method described in the present invention, the dynamic programming algorithm in step five includes the Bellman equation.

[0028] As a preferred solution of the method for constructing the HRP dynamic programming optimization model based on the DIP medical insurance payment method described in the present invention, the results of the empirical analysis in step six include the validity verification and optimization suggestions of the model.

[0029] Compared with existing technologies:

[0030] The invention has achieved remarkable results in improving hospital operation efficiency. Resource allocation is more reasonable, cost control is strengthened, and service efficiency is significantly improved. In addition, patient satisfaction surveys show that patients' overall evaluation of hospital services has improved, especially in terms of service process and medical quality. DETAILED DESCRIPTION

[0031] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be described in further detail below.

[0032] The present invention provides a method for constructing an HRP dynamic programming optimization model based on the DIP medical insurance payment method, which includes the following specific steps:

[0033] Step 1: Problem definition and goal setting: clarify the model's goals and define key variables; the model's goals include improving operational efficiency, reducing costs, and optimizing resource allocation; key variables include patient waiting time, bed utilization, and drug and material costs;

[0034] Step 2, data collection and preprocessing: collect various data indicators, and clean and preprocess the data; data cleaning and preprocessing include processing missing values ​​and outliers to ensure data quality; and various data indicators include: patient waiting time, outpatient visits, inpatient visits, surgical volume, bed utilization rate, operating room utilization rate, medical equipment utilization rate, human resource allocation (number of doctors and nurses), drug costs, material costs, labor costs, operating costs, income, support, profit margin, medical accident rate, patient satisfaction, readmission rate, medical quality indicators (such as cure rate, improvement rate), medical insurance settlement amount, medical insurance payment ratio, and medical insurance refusal;

[0035] Step 3: Variable selection and constraint definition: Select variables that have a significant impact on the target and define the constraints of the model; the constraints of the model include resource limitations, policy regulations, and service quality standards;

[0036] Step 4: Model construction: Establish a multi-stage decision model, decompose long-term planning into short-term decision points, and use dynamic programming theory to build a mathematical model to quantify the relationship between variables; the expression formula of the multi-stage decision model includes the state transfer equation, the benefit function and the optimal value function;

[0037] The calculation formula of the state transfer equation is:

[0038] S t+1 =T(S t ,d t )

[0039] Among them, S t is the state of stage t, d t is the decision of the tth stage, T is the state transition function;

[0040] The calculation formula of the profit function is:

[0041] r t =R(s t ,d t )

[0042] Among them, r t is the profit in the tth stage, R is the profit function;

[0043] The calculation formula of the optimal value function is:

[0044] Vt(st)=md t ax{R(st,dt)+Vt +1 (T(st,dt))}

[0045] Among them, V t (s t) is the state s from the tth stage t Initially, the maximum total benefit that can be obtained through the optimal decision sequence;

[0046] The multi-stage decision model is defined as:

[0047] Stage: The problem is decomposed into a number of sequential sub-problems, each stage corresponds to a decision point;

[0048] State: The condition or situation of the system at the beginning of each stage. The state variable describes the specific situation of the system at a certain stage.

[0049] Decision-making: The choices made at each stage based on the current state;

[0050] Transfer equation: an equation that describes how the state of a system changes from one stage to the next; Payoff function: at each stage, the payoff obtained based on the current state and the decisions made;

[0051] Optimal value function: the maximum (or minimum) total benefit that can be obtained through the optimal decision sequence at a certain stage starting from the current state;

[0052] The rules for using the stage decision model are:

[0053] When using a multi-stage decision model, follow these rules:

[0054] (1) Define stages and states: clarify the stage division of the problem and the state variables of each stage;

[0055] (2) Determine the decision and transition equations: determine the decisions that can be made in each state and how these decisions affect the state of the next stage;

[0056] (3) Construct a profit function: define the profit generated by each decision under the current state;

[0057] (4) Recursively solve the optimal value function: starting from the last stage, gradually calculate the optimal value function of each stage;

[0058] (5) Record the optimal decision path: While calculating the optimal value function, record the optimal decision at each stage so that the optimal decision sequence can be obtained by backtracking;

[0059] (6) Verify and adjust the model: verify the effectiveness of the model through actual data and adjust the model parameters or structure as needed;

[0060] The detailed calculation process of the stage decision model is as follows:

[0061] Process 1, initialization:

[0062] Determine the total number of stages N of the problem;

[0063] Define the state set St of each stage;

[0064] Define the decision set Dt(st) of each stage, where st∈St;

[0065] Process 2, construct the state transfer equation:

[0066] For each stage t and each state St, define the state transition equation T(st,dt), which means making a decision dt from state st to a new state;

[0067] Process three, construct the profit function:

[0068] For each stage t and each state st, define the profit function R(st,dt), which represents the profit obtained after making decision dt in state st;

[0069] Process 4, calculate the optimal value function:

[0070] Starting from the last stage N, the optimal value function Vt(st) of each stage is calculated step by step;

[0071] For the last stage N, the optimal value function VN(sN) is usually equal to the profit function R(sN, dN) of that stage;

[0072] For other stages t, the optimal value function is calculated using the recursive formula:

[0073]

[0074] Process 5: Record the optimal decision path:

[0075] While calculating the optimal value function, record the optimal decision at each stage In order to backtrack to obtain the optimal decision sequence;

[0076] Process 6: Verify and adjust the model:

[0077] Verify the effectiveness of the model through actual data and adjust the model parameters or structure as needed;

[0078] In practical applications, multi-stage decision-making models need to be appropriately adjusted and optimized according to specific problems to ensure the accuracy and practicality of the model;

[0079] Step 5, algorithm selection and implementation: select a dynamic programming algorithm to solve the model, implement the algorithm and write program code for calculation; the dynamic programming algorithm includes the Bellman equation;

[0080] Step 6: Model verification and optimization: Verify the effectiveness of the model through simulation and empirical analysis, and adjust the model parameters to optimize the model according to the verification results; the results of the empirical analysis include the effectiveness verification and optimization suggestions of the model; through simulation and empirical analysis, the effectiveness of the model can be verified, and the hospital's operating strategy can be adjusted and optimized according to the results. The expected analysis results include the following aspects:

[0081] Improve operational efficiency: Improve hospital operational efficiency by optimizing resource allocation and process management;

[0082] Cost control effect: By optimizing the cost structure, reducing the cost of drugs and materials, and improving the cost control effect;

[0083] Quality management improvement: Improve quality management by increasing patient satisfaction and reducing medical error rates;

[0084] Medical insurance payment optimization: Improve medical insurance payment efficiency and reduce refusal of payment by optimizing medical insurance settlement strategies;

[0085] Step 7: Save the optimization suggestions proposed by the model so that when the same data appears next time, it can be determined whether the proposed optimization suggestions are consistent to test the accuracy of the model.

[0086] Based on the above, its data collection includes but is not limited to the following processes:

[0087] Process 1: Use the literature review method to sort out the relevant theories and practical experiences on DIP medical insurance payment method and public hospital operation performance management at home and abroad, so as to provide a theoretical basis and reference framework. Through in-depth analysis of existing literature, the correspondence between DIP logical structure and the logical structure of national examination performance indicators is clarified, laying the foundation for subsequent data collection and analysis;

[0088] Process 2: Conduct an in-depth on-site inspection of the TCM hospital using the method of field research, and collect first-hand information on hospital operation management, service processes, patient satisfaction, etc. through face-to-face communication with hospital management, medical staff and patients; field research not only helps to understand the actual operation status of the hospital, but also captures some detailed information that is difficult to obtain through literature;

[0089] Process 3: Use questionnaire surveys to collect data in a quantitative manner for hospital management, medical staff and patients; the questionnaires cover multiple dimensions such as the hospital's organizational structure, service efficiency, cost control, and patient experience, ensuring the breadth and representativeness of the data; by statistically analyzing the questionnaire data, an objective evaluation of the hospital's operating performance can be made;

[0090] Process 4: Invite hospital management experts, medical insurance policy researchers, and scholars in related fields to conduct in-depth interviews using the interview method; through expert interviews, we can gain in-depth insights into the DIP medical insurance payment method and operational performance management model to provide theoretical support and practical guidance.

[0091] Data analysis includes but is not limited to the following processes:

[0092] Process 1: Data collection and cleaning:

[0093] Collect data related to DIP payment methods and performance indicators;

[0094] Clean the data and handle missing values, outliers and duplicate records;

[0095] Process 2, descriptive statistical analysis:

[0096] Calculate the mean, standard deviation, minimum value, maximum value, etc. of each performance indicator;

[0097] Use charts (e.g. bar charts, box plots) to visually display the distribution of performance indicators;

[0098] Process 3, correlation analysis:

[0099] Calculate the Pearson correlation coefficient or Spearman rank correlation coefficient between the performance indicator and the DIP payment method;

[0100] Use scatter plots to visualize the results of the correlation analysis;

[0101] Process 4, regression analysis:

[0102] A multiple linear regression model was established with the DIP payment method as the independent variable and the performance indicator as the dependent variable;

[0103] Check the goodness of fit of the regression model (such as R 2 value) and significance (such as p-value);

[0104] Process 5, analysis of variance (ANOVA):

[0105] If there are multiple categories of DIP payment methods, use ANOVA to compare the mean differences of performance indicators under different categories;

[0106] Perform post hoc tests (such as Tukey HSD test) to determine which categories differ significantly.

[0107] Process 6, Logistic Regression Analysis:

[0108] If the performance indicator is a categorical variable, logistic regression is used to assess the impact of the DIP payment method on the categorical performance indicator;

[0109] Process 7, Structural Equation Model (SEM):

[0110] Use SEM to simultaneously consider the complex relationships between multiple performance indicators and DIP payment methods;

[0111] Process 8, sensitivity analysis:

[0112] Sensitivity analyses were performed to assess the robustness of the model results;

[0113] Process 9: Data Analysis Tools

[0114] Statistical software: such as SPSS, R, Stata, SAS, etc.;

[0115] Data visualization tools: such as Tableau, Power BI, Excel, etc.;

[0116] Programming languages: such as Python, R, etc., used for advanced data analysis and building machine learning models.

[0117] The empirical analysis includes but is not limited to the following processes:

[0118] Process 1: Conduct literature review and construct theoretical framework. By combing through relevant domestic and foreign literature, clarify the theoretical basis of the DIP medical insurance payment method and its relationship with the operational performance management of public Chinese medicine hospitals; based on the underlying logic of value = effect / cost, construct a theoretical framework to provide theoretical support for subsequent empirical analysis;

[0119] Process 2: Carry out data collection and preprocessing; collect data related to DIP payment methods and national examination performance indicators for traditional Chinese medicine hospitals, including indicators such as cost, effect, efficiency, finance, patient flow, human resources and medical insurance payment; clean the collected data, process missing values ​​and outliers, and ensure data quality;

[0120] Process 3: Conduct descriptive statistical analysis; calculate the mean, standard deviation, minimum value, maximum value, etc. of each performance indicator, and use charts (such as bar charts and box plots) to intuitively display the distribution of performance indicators, providing a basis for in-depth analysis;

[0121] Process 4: Conduct correlation analysis and regression analysis; calculate the correlation coefficient between performance indicators and DIP payment methods, and use scatter plots to visualize the correlation; establish a multivariate linear regression model, using DIP payment methods as independent variables and performance indicators as dependent variables, and evaluate the degree and significance of the impact of DIP payment methods on performance indicators;

[0122] Process 5: Conduct ANOVA and logistic regression analysis; if there are multiple categories of DIP payment methods, use ANOVA to compare the mean differences of performance indicators under different categories and conduct post hoc tests; for categorical performance indicators, use logistic regression to analyze the impact of DIP payment methods;

[0123] Process 6: Construct a structural equation model (SEM); consider the complex relationship between multiple performance indicators and DIP payment methods, use SEM to conduct path analysis and causal inference, and evaluate the model's fit and path coefficients;

[0124] Process seven: Conduct sensitivity analysis and interpret the results; conduct sensitivity analysis by changing model parameters or excluding certain data points to evaluate the robustness of the model results; based on the results of empirical analysis, explain the relationship between the DIP medical insurance payment method and the public Chinese medicine hospital operation performance management model, and put forward optimization suggestions.

[0125] Although the present invention has been described above with reference to the embodiments, various modifications may be made thereto and parts thereof may be replaced by equivalents without departing from the scope of the present invention. In particular, as long as there is no structural conflict, the various features in the embodiments disclosed in the present invention may be used in combination with each other in any manner, and the fact that these combinations are not exhaustively described in this specification is only for the sake of omitting space and saving resources. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for constructing a HRP dynamic programming optimization model based on the DIP medical insurance payment method, characterized in that: The specific steps are as follows: Step 1: Problem definition and goal setting: clarify the model's objectives and define key variables; Step 2: Data collection and preprocessing: collect various data indicators, and clean and preprocess the data; Step 3: Variable selection and constraint definition: Select variables that have a significant impact on the target and define the constraints of the model; Step 4: Model construction: Establish a multi-stage decision-making model, decompose long-term planning into short-term decision points, and use dynamic programming theory to build a mathematical model to quantify the relationship between various variables; Step 5, algorithm selection and implementation: select a dynamic programming algorithm to solve the model, implement the algorithm and write program code for calculation; Step 6: Model verification and optimization: Verify the effectiveness of the model through simulation and empirical analysis, and adjust the model parameters to optimize the model according to the verification results; Step 7: Save the optimization suggestions proposed by the model so that when the same data appears next time, it can be determined whether the proposed optimization suggestions are consistent to test the accuracy of the model.

2. The method for constructing a HRP dynamic programming optimization model based on the DIP medical insurance payment method according to claim 1, characterized in that: The objectives of the model defined in step one include improving operational efficiency, reducing costs, and optimizing resource allocation; the key variables defined in step one include patient waiting time, bed utilization, and drug and material costs.

3. The method for constructing a HRP dynamic programming optimization model based on the DIP medical insurance payment method according to claim 1, characterized in that: The data cleaning and preprocessing in step 2 include processing missing values ​​and outliers to ensure data quality.

4. The method for constructing a HRP dynamic programming optimization model based on the DIP medical insurance payment method according to claim 1, characterized in that: The constraints of the model defined in step three include resource limitations, policy regulations, and service quality standards.

5. The method for constructing a HRP dynamic programming optimization model based on the DIP medical insurance payment method according to claim 1, characterized in that: The expression formula of the multi-stage decision model in step 4 includes a state transfer equation, a benefit function and an optimal value function.

6. The method for constructing a HRP dynamic programming optimization model based on the DIP medical insurance payment method according to claim 5, characterized in that: The calculation formula of the state transfer equation is: S t+1 =T(S t ,d t ) Among them, S t is the state of stage t, d t is the decision of the tth stage, and T is the state transition function.

7. The method for constructing a HRP dynamic programming optimization model based on the DIP medical insurance payment method according to claim 5, characterized in that: The calculation formula of the profit function is: r t =R(s t ,d t ) Among them, r t is the profit in the tth stage, and R is the profit function.

8. The method for constructing a HRP dynamic programming optimization model based on the DIP medical insurance payment method according to claim 5, characterized in that: The calculation formula of the optimal value function is: Among them, V t (s t ) is the state s from the tth stage t Initially, the maximum total benefit that can be obtained through the optimal decision sequence.

9. The method for constructing a HRP dynamic programming optimization model based on the DIP medical insurance payment method according to claim 1, characterized in that: The dynamic programming algorithm in step five includes the Bellman equation.

10. The method for constructing a HRP dynamic programming optimization model based on the DIP medical insurance payment method according to claim 1, characterized in that: The results of the empirical analysis in step six include model validity verification and optimization suggestions.

Images