A hospital performance evaluation method based on performance management weighting

Abstract

The application relates to the field of hospital performance management, and discloses a hospital performance evaluation method based on performance management weighting, which comprises the following steps: obtaining the hospital type of an evaluation object; forming a standardized data set; calculating the factor loading of each quantitative index and the evaluation dimension to which the quantitative index belongs and the path coefficient between each evaluation dimension; correcting the preset original weight to obtain a decoupled weight; giving each quantitative index a corresponding score proportion; calculating the initial score of each evaluation dimension according to the decoupled weight and the score proportion; compensating the initial score to obtain the final score of each evaluation dimension; determining the corresponding effectiveness grade and outputting an evaluation report. By matching a differentiated quantitative index system according to the hospital type, collecting multi-dimensional index data and completing standardization processing, calculating a decoupling correction coefficient through factor loading, and giving each index a corresponding score proportion through multi-dimensional determination, the index coupling decoupling, scientific weight distribution and evaluation deviation correction are realized.

Description

Technical Field

[0001] This invention relates to the field of hospital performance management technology, specifically a hospital performance evaluation method based on performance management weighting. Background Technology

[0002] Performance evaluation of public hospitals is a core regulatory tool and management instrument for deepening the reform of the medical and health system and promoting the high-quality development of public hospitals. With the comprehensive advancement of medical informatization, the interconnection and standardized collection capabilities of medical and health data have been continuously improved. Digital performance evaluation methods based on multi-dimensional quantitative indicators have become the core technical support for health administrative departments to carry out refined industry supervision and for public hospitals to improve the quality and efficiency of internal operations.

[0003] Currently, mainstream hospital performance evaluation technologies in the industry generally adopt a linear evaluation framework with fixed weights and summation. First, based on the performance evaluation standards for public hospitals issued by the state or local governments, fixed evaluation dimensions are defined and fixed weights are preset. The actual operation data of the corresponding indicators of the hospitals to be evaluated during the evaluation period are collected through hospital information systems and regional health information platforms. Based on the total score, a preset level threshold is matched to determine the hospital's performance evaluation effectiveness level, thus completing the entire performance evaluation process. However, the linear evaluation framework with fixed weights lacks a corresponding indicator coupling and decoupling mechanism. It cannot quantify and eliminate the overlapping effects of linear and nonlinear coupling between multiple indicators. Highly coupled indicators may result in duplicate scoring during the evaluation process, which can easily lead to a mismatch between weight allocation and the independent contribution of indicators. This causes the evaluation calculation results to be distorted from the bottom layer of the model, and cannot objectively reflect the actual management effectiveness corresponding to each indicator. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a hospital performance evaluation method based on performance management weighting. This method solves the problem that existing hospital performance evaluation methods are prone to mismatches between weight allocation and the independent contribution of indicators, resulting in distorted evaluation calculation results from the bottom layer of the model and failing to objectively reflect the actual management effectiveness corresponding to each indicator.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a hospital performance evaluation method based on performance management weighting, comprising the following steps: Obtain the hospital type of the evaluation object, and match the corresponding quantitative indicator system according to the hospital type; Collect the actual values ​​of each quantitative indicator in the evaluation period, the actual values ​​in the base period, the management target values, and the median year-on-year changes of indicators of similar hospitals in the region to form a standardized dataset; A partial least squares structural equation model is constructed based on historical data in a standardized dataset to calculate the factor loadings of each quantitative indicator and its corresponding evaluation dimension, as well as the path coefficients between each evaluation dimension. Based on the factor loading, the decoupling correction coefficients of each quantitative index are calculated, and the preset original weights are corrected to obtain the decoupled weights. Based on the standardized dataset, each quantitative indicator is assessed for compliance, trend conformity, and regional median comparison, and a corresponding scoring ratio is assigned to each quantitative indicator. The initial scores for each evaluation dimension are calculated based on the decoupled weights and scoring ratios. The basic compensation factor for each evaluation dimension is calculated based on the path coefficient and the initial score. The initial score is then compensated to obtain the final score for each evaluation dimension. Calculate the performance management weighted index based on the final scores of each evaluation dimension, determine the corresponding effectiveness level, and output an evaluation report.

[0006] By adopting the above technical solution, a differentiated quantitative indicator system is matched according to hospital type. Multi-dimensional indicator data is collected and standardized to build an evaluation data foundation. Based on the partial least squares structural equation model, the factor loadings of indicators and dimensions and the path coefficients between dimensions are quantified. Decoupling correction coefficients are calculated through factor loadings, and weights are adaptively corrected. After multi-dimensional judgment, corresponding scoring ratios are assigned to indicators. Then, compensation factors are calculated based on path coefficients to offset the negative coupling effect between dimensions and complete score compensation. The performance management weighted index is calculated to determine the effectiveness level and output an evaluation report. This achieves indicator coupling decoupling, scientific weight allocation, and evaluation bias correction. It solves the problem that existing hospital performance evaluation methods are prone to mismatch between weight allocation and the independent contribution of indicators, causing the evaluation calculation results to be distorted from the bottom of the model and unable to objectively reflect the actual management effectiveness of each indicator.

[0007] Preferably, the step of obtaining the hospital type of the evaluation object and matching the corresponding quantitative indicator system according to the hospital type includes the following steps: The hospital type of the evaluation object is identified as either a Western medicine hospital or a traditional Chinese medicine hospital; When the hospital type is a Western medicine hospital, a quantitative indicator system for Western medicine is matched, which includes dimensions of medical quality and safety, operational efficiency, cost control, and asset supervision. When the hospital type is a traditional Chinese medicine hospital, a quantitative indicator system of traditional Chinese medicine is matched, which includes dimensions of medical quality and safety, operational efficiency, cost control, asset supervision, characteristics of traditional Chinese medicine, talents of traditional Chinese medicine, and income of traditional Chinese medicine.

[0008] Preferably, the formation of the standardized dataset includes the following steps: The actual values ​​of each quantitative indicator in the evaluation period, the actual values ​​of the base period, the management target values, and the median year-on-year changes of indicators of similar hospitals in the region were collected. All collected data are processed to unify the standards and remove outliers to form a standardized dataset.

[0009] Preferably, the calculation of the factor loadings of each quantitative indicator and its corresponding evaluation dimension, as well as the path coefficients between each evaluation dimension, includes the following steps: Using the evaluation dimensions as latent variables and the quantitative indicators as manifest variables, a measurement model and a structural model are constructed. The partial least squares structural equation modeling algorithm is used to estimate the parameters of the measurement model and the structural model, so as to obtain the factor loading estimates of each quantitative index and its corresponding evaluation dimension, as well as the path coefficient estimates between each evaluation dimension. The factor loading estimates and path coefficient estimates are resampled using bootstrap, and the bootstrap standard deviation of each estimate is calculated. The factor loadings and path coefficients are then determined based on the bootstrap standard deviation.

[0010] Preferably, obtaining the decoupled weights includes the following steps: Calculate the corresponding decoupling correction coefficients based on the factor loadings of each of the quantitative indicators; Obtain the national standard score for each quantitative indicator, and calculate the original weight of each quantitative indicator based on the national standard score; The original weights of each quantitative indicator are multiplied by the corresponding decoupling correction coefficients to obtain the corrected weight values ​​of each quantitative indicator. Divide the corrected weight of each quantitative indicator by the sum of the corrected weights of all quantitative indicators to obtain the decoupled weight of each quantitative indicator.

[0011] Preferably, the step of calculating the corresponding decoupling correction coefficient based on the factor loadings of each of the quantitative indicators includes the following steps: Identify the causal chain coupling relationship between each of the quantitative indicators, determine the result indicators in the causal chain coupling, and assign a preset causal strength coefficient; For quantitative indicators identified as outcome indicators, the decoupling correction coefficient is multiplied by the difference between the decoupling correction coefficient and the causal strength coefficient to obtain the causal-adjusted decoupling correction coefficient. For quantitative indicators not identified as outcome indicators, the decoupling correction coefficient remains unchanged.

[0012] Preferably, assigning corresponding scoring ratios to each quantitative indicator includes the following steps: Based on the comparison between the actual values ​​of the quantitative indicators and the management target values ​​during the evaluation period, the compliance indicators for each quantitative indicator are determined. Based on the difference between the actual value of each quantitative indicator in the evaluation period and the actual value in the base period, as well as the preset guidance of each quantitative indicator, the trend conformity indicator of each quantitative indicator is determined. For quantitative indicators that are identified as conforming to the trend, calculate their absolute year-on-year change, and compare the absolute year-on-year change with the median year-on-year change of the same type of hospital in the region to determine the regional median comparison indicator. Based on the combination of compliance indicators, trend conformity indicators, and regional median comparison indicators, the scoring ratio corresponding to each quantitative indicator is determined from the preset scoring ratio levels.

[0013] Preferably, the step of calculating the initial score for each evaluation dimension based on the decoupled weights and scoring ratios includes the following steps: For each evaluation dimension, obtain the decoupled weights and scoring ratios of all quantitative indicators under that evaluation dimension; The decoupled weight of each quantitative indicator under this evaluation dimension is multiplied by the scoring ratio to obtain the intra-dimensional contribution value of each quantitative indicator. The initial score for this evaluation dimension is obtained by summing the in-dimension contribution values ​​of all quantitative indicators under this evaluation dimension.

[0014] Preferably, obtaining the final score for each evaluation dimension includes the following steps: For each target evaluation dimension, obtain the path coefficients of other evaluation dimensions to the target evaluation dimension; Extract the negative values ​​from the path coefficients, multiply the absolute value of the negative values ​​by the initial score of the corresponding evaluation dimension, and obtain the contribution value of each negative coupling. Sum all negative coupling contribution values ​​and add 1 to obtain the basic compensation factor for the target evaluation dimension; Based on the initial score of the target evaluation dimension and the basic compensation factor, the final score of the target evaluation dimension is calculated.

[0015] Preferably, the output evaluation report includes the following steps: Multiply the decoupled weight of each quantitative indicator by the corresponding scoring ratio to obtain the weighted score of each quantitative indicator. Add up the weighted scores of all quantitative indicators to obtain the performance management weighted index. The performance management weighted index is compared with a preset performance level threshold range to determine the corresponding performance level; The quantitative indicator with the lowest scoring ratio is selected as the core weakness indicator. The output includes an evaluation report containing the performance management weighted index, the effectiveness level, and the core weakness indicators.

[0016] This invention provides a hospital performance evaluation method based on performance management weighting. It has the following beneficial effects: 1. This invention matches a differentiated quantitative indicator system based on hospital type, collects multi-dimensional indicator data for standardized processing, establishes an evaluation data foundation, quantifies the factor loadings of indicators and dimensions and the path coefficients between dimensions based on a partial least squares structural equation model, calculates decoupling correction coefficients through factor loadings, performs adaptive weight correction, assigns corresponding scoring ratios to indicators based on multi-dimensional judgment, and then calculates compensation factors based on path coefficients to offset the negative coupling effect between dimensions and complete score compensation. It calculates the performance management weighted index, determines the effectiveness level, and outputs an evaluation report, thereby realizing indicator coupling decoupling, scientific weight allocation, and evaluation bias correction.

[0017] 2. This invention achieves precise matching between the evaluation system and the evaluation object by constructing a differentiated quantitative indicator system for Western medicine and traditional Chinese medicine hospitals. At the same time, it provides a data-driven theoretical basis for the evaluation model by quantifying the indicator factor loadings and path coefficients between dimensions through a partial least squares structural equation model.

[0018] 3. This invention calculates the decoupling correction coefficient based on factor loading and completes the adaptive correction of weights by combining the causal coupling relationship of indicators. This eliminates the problem of duplicate scoring caused by overlapping indicators and achieves the matching of weight allocation with the independent contribution of indicators. At the same time, it constructs a three-in-one multi-dimensional scoring mechanism that takes into account the horizontal achievement of indicators, vertical improvement and regional benchmarking level, and realizes positive incentives for continuous improvement of hospitals, making the weight allocation and scoring logic more scientific and comprehensive.

[0019] 4. This invention constructs a negative coupling compensation mechanism based on inter-dimensional path coefficients to compensate and correct the initial scores of dimensions, thereby offsetting the evaluation bias caused by the negative coupling between dimensions and ensuring the fairness of the evaluation results. At the same time, it constructs a closed-loop performance evaluation system for the entire process, accurately identifying the core shortcomings of hospital management while outputting evaluation results. This provides a scientific basis for industry supervision and also points out the target direction for hospital quality improvement, realizing the management empowerment value of performance evaluation. Attached Figure Description

[0020] Figure 1 This is a flowchart of a hospital performance evaluation method based on performance management weighting proposed in this invention; Figure 2 This is an architecture diagram of a hospital performance evaluation system based on performance management weighting, as proposed in an embodiment of the present invention. Detailed Implementation

[0021] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1: In a first embodiment of the present invention, the present invention provides a hospital performance evaluation method based on performance management weighting, such as... Figure 1 As shown, it includes the following steps: Obtain the hospital type of the evaluation object, and match the corresponding quantitative indicator system according to the hospital type; Furthermore, the hospital type of the evaluation object is obtained, and the corresponding quantitative indicator system is matched according to the hospital type, including the following steps: The hospital type identified for evaluation is either a Western medicine hospital or a traditional Chinese medicine hospital. When the hospital type is a Western medicine hospital, a quantitative indicator system for Western medicine is matched, which includes dimensions of medical quality and safety, operational efficiency, cost control, and asset supervision. When the hospital type is a traditional Chinese medicine hospital, a quantitative indicator system of traditional Chinese medicine is matched, which includes dimensions of medical quality and safety, operational efficiency, cost control, asset supervision, characteristics of traditional Chinese medicine, talents of traditional Chinese medicine, and income of traditional Chinese medicine.

[0023] Specifically, the first step is to identify whether the hospital being evaluated is a Western medicine hospital or a traditional Chinese medicine hospital. Different types of hospitals have different focuses in performance monitoring, so different indicator systems need to be matched.

[0024] When the hospital type is a Western medicine hospital, a quantitative indicator system for Western medicine is used, which includes dimensions of medical quality and safety, operational efficiency, cost control, and asset supervision. The medical quality and safety dimension includes indicators such as the percentage of discharged patients undergoing surgery, the percentage of level 4 surgeries, and the intensity of antibiotic use; the operational efficiency dimension includes indicators such as the proportion of medical service revenue to total medical revenue; the cost control dimension includes indicators such as the increase in average cost per outpatient visit; and the asset supervision dimension includes indicators such as the debt-to-asset ratio.

[0025] When the hospital type is a Traditional Chinese Medicine (TCM) hospital, a quantitative indicator system for TCM is used. This system adds TCM characteristics, TCM talent, and TCM revenue dimensions to the Western medicine indicator system. The TCM characteristics dimension includes indicators such as the proportion of outpatient TCM prescriptions and the proportion of TCM non-drug therapies used. The TCM talent dimension includes the proportion of TCM physicians. The TCM revenue dimension includes the proportion of TCM revenue to total drug revenue.

[0026] Collect the actual values ​​of each quantitative indicator in the evaluation period, the actual values ​​in the base period, the management target values, and the median year-on-year changes of indicators of similar hospitals in the region to form a standardized dataset; Furthermore, a standardized dataset is created, including the following steps: The actual values ​​of each quantitative indicator in the evaluation period, the actual values ​​of the base period, the management target values, and the median year-on-year changes of indicators of similar hospitals in the region were collected. All collected data are processed to unify the standards and remove outliers to form a standardized dataset.

[0027] Specifically, the following steps are taken: First, the actual values ​​of each quantitative indicator in the evaluation period, the actual values ​​of the base period, the management target values, and the median year-on-year change of indicators for similar hospitals in the region are collected from the national performance monitoring platform, the regional health information platform, and the hospital's internal system. The actual values ​​of the evaluation period and the base period are used to calculate the year-on-year change of the indicators. The management target values ​​serve as the benchmark for determining compliance. The median year-on-year change of indicators for similar hospitals in the region is used for horizontal comparison of improvement. All collected data undergoes standardized processing; for example, the formula used for cost increase indicators is: ,in, For the first Dimension 1 The actual value of each indicator during the evaluation period. This corresponds to the actual value in the base period. The percentage change is the year-on-year change. For ratio indicators, the original values ​​are used directly. At the same time, outliers are removed for ratio indicators that exceed the reasonable range, such as those greater than 100% or less than 0. Linear interpolation or industry averages are used to fill in the missing data. Through the above processing, a standardized dataset with a unified format is output, providing an accurate and comparable data foundation for subsequent compliance determination, trend conformity determination, and regional median comparison.

[0028] A partial least squares structural equation model is constructed based on historical data in a standardized dataset to calculate the factor loadings of each quantitative indicator and its corresponding evaluation dimension, as well as the path coefficients between each evaluation dimension. Furthermore, the factor loadings of each quantitative indicator and its corresponding evaluation dimension, as well as the path coefficients between each evaluation dimension, are calculated, including the following steps: Using each evaluation dimension as latent variable and each quantitative indicator as manifest variable, a measurement model and a structural model are constructed. The partial least squares structural equation modeling algorithm is used to estimate the parameters of the measurement model and the structural model, and to obtain the factor loading estimates of each quantitative index and its corresponding evaluation dimension, as well as the path coefficient estimates between each evaluation dimension. The factor loadings and path coefficients are resampled using the bootstrap method. The bootstrap standard deviation of each estimate is calculated, and the factor loadings and path coefficients are determined based on the bootstrap standard deviation.

[0029] Specifically, a partial least squares structural equation model is constructed based on historical data from a standardized dataset to calculate the factor loadings of each quantitative indicator and its corresponding evaluation dimension, as well as the path coefficients between each evaluation dimension. Using each evaluation dimension as latent variables and each quantitative indicator as manifest variables, a measurement model and a structural model are constructed. The measurement model describes the relationship between latent and manifest variables, and its expression is as follows: ,in, For the first Dimension 1 The measured values ​​of each indicator, For the first Latent variables of dimension Factor loadings represent the coupling strength between the index and its corresponding dimension. For measurement error; the structural model describes the interactions between dimensions, and its expression is: ,in, Let the path coefficient represent the first Dimension to the first The coupling effect of dimensions The structural error is used; the partial least squares structural equation modeling algorithm is used to estimate the parameters of the above model, and the factor loading estimates and path coefficient estimates are obtained.

[0030] The estimated values ​​are resampled using the bootstrap method, and the bootstrap standard deviation of each estimated value is calculated. Based on this standard deviation, robust factor loadings and path coefficients are determined. This process quantifies the coupling relationship between indicators, providing a basis for subsequent weight correction and score compensation. For example, by inputting the quarterly historical data of each hospital, the factor loading values ​​corresponding to each indicator are output, realizing a quantifiable characterization of the coupling strength of indicators.

[0031] Based on the factor loading, the decoupling correction coefficients of each quantitative index are calculated, and the preset original weights are corrected to obtain the decoupled weights. Furthermore, the decoupled weights are obtained through the following steps: Calculate the corresponding decoupling correction coefficients based on the factor loadings of each quantitative indicator; Obtain the national standard score for each quantitative indicator, and calculate the original weight of each quantitative indicator based on the national standard score; The original weights of each quantitative indicator are multiplied by the corresponding decoupling correction coefficients to obtain the corrected weight values ​​of each quantitative indicator. Divide the corrected weight of each quantitative indicator by the sum of the corrected weights of all quantitative indicators to obtain the decoupled weight of each quantitative indicator.

[0032] Furthermore, the decoupling correction coefficients are calculated based on the factor loadings of each quantitative indicator, including the following steps: Identify the causal chain coupling relationship between various quantitative indicators, determine the outcome indicators in the causal chain coupling, and assign a preset causal strength coefficient; For quantitative indicators identified as outcome indicators, the decoupling correction coefficient is multiplied by the difference between one and the causal strength coefficient to obtain the causal-adjusted decoupling correction coefficient. For quantitative indicators not identified as outcome indicators, the decoupling correction coefficient remains unchanged.

[0033] Specifically, the decoupling correction coefficients for each quantitative indicator are calculated based on factor loadings, and the preset original weights are corrected to obtain the decoupled weights. First, the basic decoupling correction coefficients are calculated based on the factor loadings of each quantitative indicator. The calculation formula is as follows: ,in, The basic decoupling correction coefficient For the first Dimension 1 The factor loadings of each indicator, where the coefficient reflects the degree of coupling between that indicator and other indicators in the same dimension. The closer to 1 The closer an indicator is to 0, the greater the influence of intradimensional coupling. Its weight in the evaluation should be appropriately reduced. It is also necessary to identify the causal chain coupling relationship between each quantitative indicator, determine the result indicator in the causal chain coupling, and assign a preset causal strength coefficient.

[0034] For quantitative indicators identified as outcome indicators, the causal-adjusted decoupling correction coefficient is further calculated: ,in, The causality strength coefficient ranges from 0 to 0.3, with 0.1 for direct causality and 0.05 for indirect causality. This is the decoupling correction coefficient after causality adjustment. For quantitative indicators not identified as outcome indicators, maintain... Simultaneously, the national standard scores for each quantitative indicator are obtained, and the original weights of each quantitative indicator are calculated based on the national standard scores. The corrected weight values ​​are obtained by multiplying the original weights by the causal-adjusted decoupling correction coefficients. Then, divide the corrected weight value of each indicator by the sum of the corrected weight values ​​of all indicators to obtain the decoupled weights: In this embodiment, the input is the factor loading. and causal strength coefficient Then the basic correction factor After causal adjustment If the original weights Then adjust the weight value. After normalization, the decoupled weights are obtained. This process achieves weight suppression of highly coupled indicators, making the evaluation results more reflective of the independent contributions of the indicators.

[0035] Based on the standardized dataset, each quantitative indicator is assessed for compliance, trend conformity, and regional median comparison, and a corresponding scoring ratio is assigned to each quantitative indicator. Furthermore, assign corresponding scoring ratios to each quantitative indicator, including the following steps: Based on the comparison between the actual values ​​of the quantitative indicators and the management target values ​​during the evaluation period, the compliance indicators for each quantitative indicator are determined. Based on the difference between the actual value of each quantitative indicator in the evaluation period and the actual value in the base period, as well as the preset guidance of each quantitative indicator, the trend conformity indicator of each quantitative indicator is determined. For quantitative indicators that are identified as conforming to the trend, calculate their absolute year-on-year change, and compare the absolute year-on-year change with the median year-on-year change of the same type of hospital in the region to determine the regional median comparison indicator. Based on the combination of compliance indicators, trend conformity indicators, and regional median comparison indicators, the scoring ratio corresponding to each quantitative indicator is determined from the preset scoring ratio levels.

[0036] Specifically, based on the standardized dataset, each quantitative indicator is assessed for compliance, trend conformity, and regional median comparison, and a corresponding scoring ratio is assigned to each quantitative indicator. First, the actual values ​​of the quantitative indicators during the evaluation period are used as the basis for this assessment. With management target value The comparison results determine the compliance label. For indicators that are geared towards improvement, if but otherwise For indicators that are geared towards reduction, if but otherwise .

[0037] Then, based on the actual values ​​during the evaluation period and the actual values ​​during the base period... The difference And the indicator preset guidance to determine trend conformity indicators. If the direction of change is consistent with the guide, then otherwise .right The indicator is used to calculate the absolute value of the year-on-year change. Compare it with the median year-on-year change of indicators for hospitals of the same category in the region. Compare and determine the median comparison marker in the region. :like but otherwise Finally, the scoring ratio is determined from the preset scoring ratio levels based on the combination of the three factors. : ; For example, if the actual value of a certain indicator during the evaluation period does not meet the target, but the trend is in line with the guidance and the improvement is better than the regional median, then... This provides a positive incentive for continuous improvement of the hospital.

[0038] The initial scores for each evaluation dimension are calculated based on the decoupled weights and scoring ratios. Furthermore, the initial scores for each evaluation dimension are calculated based on the decoupled weights and scoring ratios, including the following steps: For each evaluation dimension, obtain the decoupled weights and scoring ratios of all quantitative indicators under that evaluation dimension; The decoupled weight of each quantitative indicator under this evaluation dimension is multiplied by the scoring ratio to obtain the intra-dimensional contribution value of each quantitative indicator. The initial score for this evaluation dimension is obtained by summing the in-dimension contribution values ​​of all quantitative indicators under this evaluation dimension.

[0039] Specifically, the initial score for each evaluation dimension is calculated based on the decoupled weights and scoring ratios. For each evaluation dimension, the decoupled weights and scoring ratios of all quantitative indicators under that dimension are obtained. The decoupled weights and scoring ratios of each quantitative indicator are multiplied by their respective scores to obtain the dimension contribution value of that indicator. Then, the contribution values ​​of all indicators under that dimension are summed to obtain the initial score for that evaluation dimension. The calculation formula is as follows: ,in, For the first Initial scores for each evaluation dimension, This represents the total number of quantitative indicators under this dimension. For the first Dimension 1 The decoupled weights of each indicator The weights after decoupling satisfy the following criteria: (Assignment ratio to the corresponding indicator) , Between 0.5 and 1.0, therefore The value range is from 0 to 1. In this embodiment, the medical quality and safety dimension includes 6 indicators, and the decoupled weights of each indicator are 0.15, 0.10, 0.12, 0.18, 0.25, and 0.20, respectively, with corresponding scoring ratios of 1.0, 0.9, 0.8, 1.0, 0.5, and 0.9. The initial score for this dimension is then... =0.15×1.0+0.10×0.9+0.12×0.8+0.18×1.0+0.25×0.5+0.20×0.9=0.15+0.09+0.096+0.18+0.125+0.18=0.821; This initial score serves as the basis for subsequent dimensional coupling compensation, reflecting the preliminary evaluation results of each dimension without considering the mutual influence between dimensions.

[0040] The basic compensation factor for each evaluation dimension is calculated based on the path coefficient and the initial score. The initial score is then compensated to obtain the final score for each evaluation dimension. Furthermore, the final scores for each evaluation dimension are obtained, including the following steps: For each target evaluation dimension, obtain the path coefficients of other evaluation dimensions to the target evaluation dimension; Extract the negative values ​​from the path coefficients, multiply the absolute value of the negative values ​​by the initial score of the corresponding evaluation dimension, and obtain the contribution value of each negative coupling. Sum all the negative coupling contribution values ​​and add 1 to obtain the basic compensation factor for the target evaluation dimension; Based on the initial score and basic compensation factor of the target evaluation dimension, the final score of the target evaluation dimension is calculated.

[0041] Specifically, the basic compensation factor for each evaluation dimension is calculated based on the path coefficients and initial scores to compensate for the initial scores, resulting in the final scores for each evaluation dimension. For each target evaluation dimension, the path coefficients of other evaluation dimensions for that target evaluation dimension are obtained. A positive path coefficient indicates positive collaboration, while a negative one indicates negative competition. The negative values ​​in the path coefficients are extracted, and the absolute value of the negative value is multiplied by the initial score of the corresponding evaluation dimension to obtain the negative coupling contribution value. All negative coupling contribution values ​​are summed and then incremented by 1 to obtain the basic compensation factor for the target evaluation dimension. The calculation formula is as follows: ,in, For the first The basic compensation factors for each evaluation dimension For the first The evaluation dimension for the first Path coefficients for each evaluation dimension This indicates that only the absolute values ​​of the negative path coefficients are extracted. For the first Initial scores for each evaluation dimension.

[0042] Then, the initial score of the target evaluation dimension is multiplied by the basic compensation factor to obtain the final score of that dimension: ,in, The final score after compensation. This is the initial score. The function ensures that the score does not exceed the maximum value of 1. In this embodiment, the path coefficient of the medical quality and safety dimension to the operational efficiency dimension is set to -0.32, and the initial score of the operational efficiency dimension is... The negative coupling contribution value is Basic compensation factor If the initial score for operational efficiency is 0.70, then the final score will be... This compensation mechanism offsets the undue losses caused by negative coupling, making the evaluation results more fairly reflect the actual management effectiveness of the hospital.

[0043] Calculate the performance management weighted index based on the final scores of each evaluation dimension, determine the corresponding effectiveness level, and output an evaluation report.

[0044] Furthermore, the evaluation report is generated, including the following steps: Multiply the decoupled weight of each quantitative indicator by the corresponding scoring ratio to obtain the weighted score of each quantitative indicator. Add up the weighted scores of all quantitative indicators to obtain the performance management weighted index. The performance management weighted index is compared with the preset performance level threshold range to determine the corresponding performance level; The quantitative indicator with the lowest scoring ratio is selected as the core weakness indicator. The output includes an evaluation report that includes a performance management weighted index, effectiveness level, and key weakness indicators.

[0045] Specifically, the performance management weighted index is calculated based on the final scores of each evaluation dimension to determine the corresponding effectiveness level and output an evaluation report. First, the decoupled weight of each quantitative indicator is multiplied by its corresponding scoring ratio to obtain the weighted score for each quantitative indicator. Then, the weighted scores of all quantitative indicators are summed to obtain the performance management weighted index. The calculation formula is as follows: ,in, This is a performance management weighted index, with a value ranging from 0 to 1. The total number of quantitative indicators; For the first The decoupled weights of each indicator; For the first The scoring ratio for each indicator. This index comprehensively reflects the hospital's overall performance management effectiveness in both horizontal standard achievement and vertical improvement.

[0046] Then The corresponding performance level is determined by comparing the results with a preset performance level threshold range. To achieve significant results, For good results, The results were mediocre. The results are considered poor. The quantitative indicator with the lowest assigned score is selected as the core weakness indicator; this indicator, with a score of 0.5, indicates that it has not met the standard and its trend is contrary to the guidance. Finally, an evaluation report is output, including the performance management weighted index, performance level, and core weakness indicator. In one possible implementation, the weights and assigned scores of each indicator are input, and the results are calculated... The results were deemed satisfactory, with the core weakness being the "intensity of antimicrobial drug use." The report was output for use by regulatory authorities and hospitals, realizing the transformation of evaluation results from numerical output to management application.

[0047] Example 2: In a second embodiment of the present invention, the present invention provides a hospital performance evaluation system based on performance management weighting, such as... Figure 2 As shown, it includes the following modules: Acquisition module: Used to acquire the hospital type of the evaluation object and match the corresponding quantitative indicator system according to the hospital type; Standardization module: Used to collect the actual values ​​of each quantitative indicator in the evaluation period, the actual values ​​of the base period, the management target values, and the median year-on-year changes of indicators of the same type of hospitals in the region, forming a standardized dataset; The calculation module is used to construct a partial least squares structural equation model based on historical data in a standardized dataset, and to calculate the factor loadings of each quantitative indicator and its corresponding evaluation dimension, as well as the path coefficients between each evaluation dimension. Correction module: used to calculate the decoupling correction coefficients of each quantitative index based on factor loadings, correct the preset original weights, and obtain the decoupling weights; Judgment module: Used to determine the compliance, trend conformity and regional median comparison of each quantitative indicator based on the standardized dataset, and to assign a corresponding scoring ratio to each quantitative indicator; Scoring module: Used to calculate the initial score for each evaluation dimension based on the decoupled weights and scoring ratios; Compensation module: Used to calculate the basic compensation factor for each evaluation dimension based on the path coefficient and the initial score, to compensate for the initial score, and to obtain the final score for each evaluation dimension. Output module: Used to calculate the performance management weighted index based on the final scores of each evaluation dimension, determine the corresponding effectiveness level, and output the evaluation report.

[0048] In scenarios where local health commissions conduct routine performance evaluations of public Western medicine and traditional Chinese medicine hospitals within their jurisdictions, traditional evaluation systems rely on manually setting weights, fail to eliminate interference from indicator coupling, and do not consider the negative coupling effects between evaluation dimensions. This leads to distorted scoring, insufficient reflection of traditional Chinese medicine-specific indicators, an inability to accurately identify hospital management shortcomings, and evaluation results that are insufficient to support differentiated supervision and hospital quality improvement. To address these issues, this invention employs a hospital performance evaluation system based on performance management weighting, the architecture of which is as follows: Figure 2 As shown. The specific implementation process of this system is as follows: First, the acquisition module automatically identifies the Western medicine / traditional Chinese medicine type of the hospital to be assessed and matches it with the corresponding exclusive quantitative indicator system to ensure that the assessment dimensions of Western medicine and traditional Chinese medicine hospitals have different focuses. The standardized module is synchronously connected to the regional health information platform and hospital management system to collect evaluation period and base period data, management target values ​​and median values ​​of hospitals of the same type in the region for each indicator, and to complete the unification of data standards and the removal of outliers to form a standardized dataset. Subsequently, the calculation module constructs a partial least squares structural equation model based on historical assessment data to calculate the path coefficients between the factor loadings of each indicator and the evaluation dimensions. The correction module calculates the decoupling correction coefficient based on the factor loading, and completes the weight correction and normalization by combining the causal coupling relationship of the indicators to obtain the decoupled weights. Next, the judgment module sequentially completes three judgments: indicator compliance, trend conformity, and comparison with the regional median, and assigns corresponding scoring ratios to each indicator according to preset levels; The scoring module calculates the initial score for each evaluation dimension based on the decoupled weights and scoring ratios; the compensation module extracts the negative path coefficients between dimensions, calculates the basic compensation factor to compensate for the initial scores, and obtains the final score for each dimension. Finally, the output module calculates the performance management weighted index based on the final scores of each dimension, matches the preset threshold to determine the assessment effectiveness level, simultaneously identifies the core weakness indicators, and outputs a complete evaluation report that includes assessment results, weakness analysis and improvement suggestions.

[0049] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A hospital performance evaluation method based on performance management weighting, characterized in that, Includes the following steps: Obtain the hospital type of the evaluation object, and match the corresponding quantitative indicator system according to the hospital type; Collect the actual values ​​of each quantitative indicator in the evaluation period, the actual values ​​in the base period, the management target values, and the median year-on-year changes of indicators of similar hospitals in the region to form a standardized dataset; A partial least squares structural equation model is constructed based on historical data in a standardized dataset to calculate the factor loadings of each quantitative indicator and its corresponding evaluation dimension, as well as the path coefficients between each evaluation dimension. Based on the factor loading, the decoupling correction coefficients of each quantitative index are calculated, and the preset original weights are corrected to obtain the decoupled weights. Based on the standardized dataset, each quantitative indicator is assessed for compliance, trend conformity, and regional median comparison, and a corresponding scoring ratio is assigned to each quantitative indicator. The initial scores for each evaluation dimension are calculated based on the decoupled weights and scoring ratios. The basic compensation factor for each evaluation dimension is calculated based on the path coefficient and the initial score. The initial score is then compensated to obtain the final score for each evaluation dimension. Calculate the performance management weighted index based on the final scores of each evaluation dimension, determine the corresponding effectiveness level, and output an evaluation report.

2. The hospital performance evaluation method based on performance management weighting as described in claim 1, characterized in that: The process of obtaining the hospital type of the evaluation object and matching the corresponding quantitative indicator system according to the hospital type includes the following steps: The hospital type of the evaluation object is identified as either a Western medicine hospital or a traditional Chinese medicine hospital; When the hospital type is a Western medicine hospital, a quantitative indicator system for Western medicine is matched, which includes dimensions of medical quality and safety, operational efficiency, cost control, and asset supervision. When the hospital type is a traditional Chinese medicine hospital, a quantitative indicator system of traditional Chinese medicine is matched, which includes dimensions of medical quality and safety, operational efficiency, cost control, asset supervision, characteristics of traditional Chinese medicine, talents of traditional Chinese medicine, and income of traditional Chinese medicine.

3. The hospital performance evaluation method based on performance management weighting as described in claim 1, characterized in that: The process of forming a standardized dataset includes the following steps: The actual values ​​of each quantitative indicator in the evaluation period, the actual values ​​of the base period, the management target values, and the median year-on-year changes of indicators of similar hospitals in the region were collected. All collected data are processed to unify the standards and remove outliers to form a standardized dataset.

4. The hospital performance evaluation method based on performance management weighting as described in claim 1, characterized in that: The calculation of the factor loadings of each quantitative indicator and its corresponding evaluation dimension, as well as the path coefficients between each evaluation dimension, includes the following steps: Using the evaluation dimensions as latent variables and the quantitative indicators as manifest variables, a measurement model and a structural model are constructed. The partial least squares structural equation modeling algorithm is used to estimate the parameters of the measurement model and the structural model, so as to obtain the factor loading estimates of each quantitative index and its corresponding evaluation dimension, as well as the path coefficient estimates between each evaluation dimension. The factor loading estimates and path coefficient estimates are resampled using bootstrap, and the bootstrap standard deviation of each estimate is calculated. The factor loadings and path coefficients are then determined based on the bootstrap standard deviation.

5. The hospital performance evaluation method based on performance management weighting according to claim 1, characterized in that: Obtaining the decoupled weights includes the following steps: Calculate the corresponding decoupling correction coefficients based on the factor loadings of each of the quantitative indicators; Obtain the national standard score for each quantitative indicator, and calculate the original weight of each quantitative indicator based on the national standard score; The original weights of each quantitative indicator are multiplied by the corresponding decoupling correction coefficients to obtain the corrected weight values ​​of each quantitative indicator. Divide the corrected weight of each quantitative indicator by the sum of the corrected weights of all quantitative indicators to obtain the decoupled weight of each quantitative indicator.

6. The hospital performance evaluation method based on performance management weighting as described in claim 5, characterized in that: The calculation of the corresponding decoupling correction coefficients based on the factor loadings of each of the quantitative indicators includes the following steps: Identify the causal chain coupling relationship between each of the quantitative indicators, determine the result indicators in the causal chain coupling, and assign a preset causal strength coefficient; For quantitative indicators identified as outcome indicators, the decoupling correction coefficient is multiplied by the difference between the decoupling correction coefficient and the causal strength coefficient to obtain the causal-adjusted decoupling correction coefficient. For quantitative indicators not identified as outcome indicators, the decoupling correction coefficient remains unchanged.

7. The hospital performance evaluation method based on performance management weighting according to claim 1, characterized in that: Assigning corresponding scoring ratios to each quantitative indicator includes the following steps: Based on the comparison between the actual values ​​of the quantitative indicators and the management target values ​​during the evaluation period, the compliance indicators for each quantitative indicator are determined. Based on the difference between the actual value of each quantitative indicator in the evaluation period and the actual value in the base period, as well as the preset guidance of each quantitative indicator, the trend conformity indicator of each quantitative indicator is determined. For quantitative indicators that are identified as conforming to the trend, calculate their absolute year-on-year change, and compare the absolute year-on-year change with the median year-on-year change of the same type of hospital in the region to determine the regional median comparison indicator. Based on the combination of compliance indicators, trend conformity indicators, and regional median comparison indicators, the scoring ratio corresponding to each quantitative indicator is determined from the preset scoring ratio levels.

8. The hospital performance evaluation method based on performance management weighting according to claim 1, characterized in that: The calculation of the initial score for each evaluation dimension based on the decoupled weights and scoring ratios includes the following steps: For each evaluation dimension, obtain the decoupled weights and scoring ratios of all quantitative indicators under that evaluation dimension; The decoupled weight of each quantitative indicator under this evaluation dimension is multiplied by the scoring ratio to obtain the intra-dimensional contribution value of each quantitative indicator. The initial score for this evaluation dimension is obtained by summing the in-dimension contribution values ​​of all quantitative indicators under this evaluation dimension.

9. A hospital performance evaluation method based on performance management weighting as described in claim 1, characterized in that: Obtaining the final score for each evaluation dimension includes the following steps: For each target evaluation dimension, obtain the path coefficients of other evaluation dimensions to the target evaluation dimension; Extract the negative values ​​from the path coefficients, multiply the absolute value of the negative values ​​by the initial score of the corresponding evaluation dimension, and obtain the contribution value of each negative coupling. Sum all negative coupling contribution values ​​and add 1 to obtain the basic compensation factor for the target evaluation dimension; Based on the initial score of the target evaluation dimension and the basic compensation factor, the final score of the target evaluation dimension is calculated.

10. A hospital performance evaluation method based on performance management weighting according to claim 1, characterized in that: The output evaluation report includes the following steps: Multiply the decoupled weight of each quantitative indicator by the corresponding scoring ratio to obtain the weighted score of each quantitative indicator. Add up the weighted scores of all quantitative indicators to obtain the performance management weighted index. The performance management weighted index is compared with a preset performance level threshold range to determine the corresponding performance level; The quantitative indicator with the lowest scoring ratio is selected as the core weakness indicator. The output includes an evaluation report containing the performance management weighted index, the effectiveness level, and the core weakness indicators.

Images