A medical quality and safety risk assessment method and system based on a five-dimensional quantification model

Abstract

The application provides a medical quality and safety risk assessment method and system based on a five-dimensional quantification model, the method comprising: collecting multi-dimensional evaluation data of a target medical unit from multiple data sources; calling a corresponding five-dimensional risk index system according to the department category of the target medical unit, the five-dimensional risk index system comprising personnel, business quantity, structure quality, process quality and result quality dimensions; mapping the data to original actual scores of each index and calculating dimension standardization scores; generating a comprehensive risk score through weighted summation of the dimension weight coefficients; dynamically determining a final risk level in combination with department-specific extreme risk trigger conditions; and finally matching a closed-loop intervention instruction and generating a risk assessment report to output to a management terminal. The application can realize precise quantification, graded early warning and closed-loop intervention of medical quality and safety risks, and improve department adaptability, dynamic sensitivity of risk assessment and pertinence of management decisions.

Description

Technical Field

[0001] This invention relates to the field of medical quality management technology, and more specifically, to a medical quality and safety risk assessment method and system based on a five-dimensional quantitative model. Background Technology

[0002] With the deepening of healthcare informatization, hospital information systems have accumulated massive amounts of clinical operational data, providing a data foundation for the quantitative assessment of medical quality and safety. Currently, domestic and international medical institutions generally use the Structure-Process-Outcome (SPO) model or the Donabedian quality assessment framework as the theoretical basis for medical quality assessment, monitoring and evaluating medical services by setting Key Quality Indicators (KQIs). Regarding risk stratification, existing technologies mostly use a single comprehensive score compared to a fixed threshold to classify the assessed object into high, medium, and low risk levels, triggering corresponding management interventions accordingly. Some advanced systems have also introduced weighted scoring mechanisms, assigning differentiated weights to different indicators to reflect the degree of influence of each dimension on the overall risk. Furthermore, some studies attempt to integrate trend analysis into risk assessment, identifying the direction of risk evolution by comparing historical score changes.

[0003] Existing assessment methods typically employ a unified and fixed indicator system and scoring standards, failing to fully consider the significant differences among departments in terms of staffing, business structure, and quality characteristics. This results in assessment results lacking departmental specificity and relevance. Conventional comprehensive scoring and threshold comparison models are ill-suited to capturing complex risk scenarios such as extreme anomalies in single indicators, deterioration of multiple indicator combinations, and persistent exceedances of indicators, leading to insufficient sensitivity and accuracy in risk warnings. Existing systems often determine risk levels through static, one-off assessments, lacking a dynamic response mechanism to changing scoring trends and failing to effectively identify rapidly escalating risks. Furthermore, the connection between assessment results and subsequent intervention measures is loose, lacking standardized closed-loop management strategies based on risk levels and quantitative feedback mechanisms for intervention effects, making it difficult to form a management loop for continuous improvement. Summary of the Invention

[0004] This invention provides a method and system for medical quality and safety risk assessment based on a five-dimensional quantitative model.

[0005] In a first aspect of the present invention, a method for assessing medical quality and safety risks based on a five-dimensional quantitative model is provided, comprising the following steps: S1: Collect multidimensional evaluation data of the target medical unit from multiple data sources; S2: Based on the department category to which the target medical unit belongs, call the five-dimensional risk indicator system that uniquely corresponds to the department category from the preset differentiated indicator library. The five-dimensional risk indicator system includes personnel dimension, business quantity dimension, structural quality dimension, process quality dimension and result quality dimension. Each dimension is pre-bound with multiple indicator items as well as the assignment standard and full score associated with the department category. S3: Map the collected multidimensional assessment data to each indicator item in the five-dimensional risk indicator system to obtain the original actual score of each indicator item, and calculate the standardized score of each dimension based on the sum of the original actual scores of all indicator items under each dimension and the sum of the full score of that dimension. S4: Obtain a set of dimension weight coefficients, and perform a weighted sum of the standardized scores of each dimension and the corresponding dimension weight coefficients to generate a comprehensive risk score; S5: Obtain multiple extreme risk triggering conditions corresponding to the department category to which the target medical unit belongs, and dynamically determine the final risk level of the target medical unit based on the comprehensive risk score and the multiple extreme risk triggering conditions; S6: Based on the final risk level, match the corresponding closed-loop intervention instruction from the preset handling strategy library, generate a risk assessment report containing the comprehensive risk score, standardized scores of each dimension and the closed-loop intervention instruction, and output the risk assessment report to the corresponding management terminal.

[0006] Furthermore, the standardized scores for each dimension are calculated in S3, specifically as follows: The standardized score for the i-th dimension is calculated as follows: ; in, For the standardized score of the i-th dimension, This is the sum of the original actual scores for all indicators under the i-th dimension. This is the sum of the full scores of all indicators under the i-th dimension; i takes the values ​​1, 2, 3, 4, and 5, corresponding to the personnel dimension, business quantity dimension, structural quality dimension, process quality dimension, and result quality dimension, respectively.

[0007] Furthermore, in step S4, a comprehensive risk score is generated, specifically as follows: Obtain a set of dimensional weight coefficients, which constitute a five-dimensional weight vector and satisfy the constraint that the sum of the weight coefficients of the personnel dimension, the business quantity dimension, the structural quality dimension, the process quality dimension, and the result quality dimension is equal to 1. The comprehensive risk score is calculated as follows: ; in, For comprehensive risk scoring, For the standardized score of the i-th dimension, The preset weight coefficients are for the i-th dimension.

[0008] Furthermore, in S5, multiple extreme risk triggering conditions corresponding to the department category to which the target medical unit belongs are obtained, specifically including: Retrieve the extreme risk triggering rule set associated with the department category to which the target medical unit belongs from the department-specific threshold configuration library. The extreme risk triggering rule set includes three triggering types: single indicator exceeding threshold triggering rule, multi-indicator weighted combination triggering rule, and continuous exceeding threshold triggering rule. The single-indicator threshold triggering rule is as follows: for each of the multiple preset key quality indicators, a corresponding department-specific triggering threshold is set. When the actual measured value of any key quality indicator exceeds the department-specific triggering threshold corresponding to that indicator, the rule is triggered. The multi-indicator weighted combination triggering rule is as follows: For a set of preset key quality indicators, a corresponding weighted summation triggering threshold is set. When the result obtained by weighted summation of the actual measured values ​​of each indicator in the set exceeds the weighted summation triggering threshold, the rule is triggered. The continuous threshold triggering rule is as follows: For at least one preset key quality indicator, a duration threshold is set. When the actual measured value of the indicator continuously exceeds its corresponding department-specific triggering threshold for a duration that reaches the duration threshold, the rule is triggered.

[0009] Furthermore, in step S5, the final risk level is dynamically determined based on the comprehensive risk score and the multiple extreme risk triggering conditions, specifically including: S51: Compare the comprehensive risk score with a set of conventional risk thresholds to generate a preliminary risk level. The set of conventional risk thresholds includes a high-risk threshold and a medium-risk threshold, and the high-risk threshold is greater than the medium-risk threshold. S52: Statistically analyze the rules triggered in the extreme risk triggering rule set to obtain the total number of triggering rules and the distribution of triggering rule types. The distribution of triggering rule types includes the number of single-indicator over-threshold triggering rules, the number of multi-indicator weighted combination triggering rules, and the number of persistent over-threshold triggering rules. S53: Based on the total number of triggering rules and the distribution of triggering rule types, calculate the extreme risk penalty factor, as shown in the following formula: ; in, As an extreme risk penalty factor, This represents the number of rules triggered when a single metric exceeds its threshold. The number of multi-indicator weighted combination trigger rules that are triggered. The number of persistent over-threshold trigger rules that have been triggered; For single-indicator penalty weights; Penalty weights are applied to multiple indicators; For continuous penalty weights; S54: Based on the preliminary risk level and the extreme risk penalty factor, calculate the final risk score as shown in the following formula: in, For the final risk score, For comprehensive risk scoring; S55: The final risk score is compared again with the set of conventional risk thresholds to generate a final risk level. A high-risk level is generated when the final risk score is greater than or equal to the high-risk threshold, a medium-risk level is generated when the medium-risk threshold is less than or equal to the final risk score and the final risk score is less than the high-risk threshold, and a low-risk level is generated when the final risk score is less than the medium-risk threshold.

[0010] Furthermore, the preliminary risk level is generated in S51, specifically including: Obtain historical assessment data of the target medical unit, and extract the comprehensive risk score of the previous assessment from the historical assessment data; The trend value of the comprehensive risk score is calculated as shown in the following formula: ; in, This represents the change in the overall risk score. Based on the current comprehensive risk score, This is the previous comprehensive risk score; The trend correction factor is calculated based on the aforementioned trend value, as shown in the following formula: ; in, As a trend correction factor, The penalty coefficient is for upward trends; The reward coefficient represents a downward trend. The threshold is medium risk. The revised preliminary risk score is calculated as follows: The revised preliminary risk score is the sum of the comprehensive risk score and the trend correction factor. The revised preliminary risk score is compared with the set of conventional risk thresholds to generate a preliminary risk level, wherein: When the revised preliminary risk score is greater than or equal to the high risk threshold, the preliminary risk level is generated as a high risk level. When the medium risk threshold is less than or equal to the revised preliminary risk score and the revised preliminary risk score is less than the high risk threshold, the preliminary risk level is generated as medium risk level. When the revised preliminary risk score is less than the medium risk threshold, the preliminary risk level is generated as low risk.

[0011] Furthermore, in step S6, matching the corresponding closed-loop intervention instruction from a preset treatment strategy library specifically includes: Based on the final risk level, an intervention strategy template associated with the risk level is retrieved from the disposal strategy library. The intervention strategy template includes rectification action instructions, resource allocation instructions, and monitoring frequency instructions. Among them, the intervention strategy templates associated with high-risk levels include: the rectification action instruction is to suspend non-emergency elective surgeries, the resource allocation instruction is to increase the number of medical staff, and the monitoring frequency instruction is to monitor weekly; The intervention strategy template associated with medium-risk levels includes: rectification action instructions are to rectify within a time limit, resource allocation instructions are to optimize the shift schedule, and monitoring frequency instructions are to monitor monthly; The intervention strategy template for low-risk levels includes: the rectification action instruction is to maintain routine operations, the resource allocation instruction is to maintain existing resources, and the monitoring frequency instruction is to monitor every quarter.

[0012] Furthermore, it also includes: After the closed-loop intervention instructions are completed and a monitoring cycle has passed, S1 to S5 are executed again to obtain a new round of final risk level, a new round of comprehensive risk score and a new round of standardized scores for each dimension. Based on the standardized scores of the same dimension in the new round and the previous round, the difference between the standardized scores of each dimension is calculated as follows: ; in, Let be the standardized score change for the i-th dimension. For the new round of standardized scores of the i-th dimension, The previous standardized score for the i-th dimension; A comparative risk assessment report containing the difference is generated, and the effectiveness level of the implemented intervention strategy is automatically determined based on the difference. The effectiveness level includes effective, partially effective, and ineffective.

[0013] Furthermore, the automatic determination of the effectiveness level of the implemented intervention strategy based on the difference specifically involves: When the standardized score changes of all dimensions are positive, the effectiveness level is determined to be valid. When the standardized score change of some dimensions is positive and the standardized score change of other dimensions is negative or zero, the effectiveness level is determined to be partially effective. When the standardized score changes for all dimensions are non-positive, the effective level is deemed invalid.

[0014] In a second aspect of the present invention, a medical quality and safety risk assessment system based on a five-dimensional quantitative model is provided, comprising: The differentiated indicator library storage module is used to store the five-dimensional risk indicator system corresponding to different department categories, as well as the assignment standards and full scores of each indicator item; The department-specific threshold configuration module is used to store extreme risk triggering rule sets corresponding to different department categories. The extreme risk triggering rule sets include single indicator threshold exceeding triggering rules, multi-indicator weighted combination triggering rules, and continuous threshold exceeding triggering rules. The multi-source data acquisition interface module is used to connect with multiple data sources in the hospital information system to collect multi-dimensional evaluation data of the target medical unit; The dimensional calculation and mapping engine is used to call the corresponding five-dimensional risk indicator system from the differentiated indicator library storage module according to the department category of the target medical unit, map the collected multidimensional assessment data into the original actual scores of each indicator item, and calculate the standardized scores of each dimension. The weighted fusion scoring module is used to obtain a preset five-dimensional weight vector, and to sum the standardized scores of each dimension with the corresponding weight coefficients to generate a comprehensive risk score. The risk level dynamic determination module is used to read the extreme risk triggering rule set from the department-specific threshold configuration module, and dynamically output the final risk level by combining the comparison results of the comprehensive risk score and the conventional risk threshold and the satisfaction of the extreme risk triggering rule set. The disposal instruction matching and report generation module is used to match closed-loop intervention instructions according to the final risk level, generate a risk assessment report containing a comprehensive risk score, standardized scores of each dimension and closed-loop intervention instructions, and output the risk assessment report to the corresponding management terminal.

[0015] The embodiments of the present invention have at least the following beneficial effects: 1. This invention constructs a differentiated five-dimensional risk indicator system that uniquely corresponds to each department category. It deeply binds the personnel dimension, business quantity dimension, structural quality dimension, process quality dimension, and outcome quality dimension with departmental characteristics, and configures differentiated indicator assignment standards and full scores. This effectively solves the problem of the lack of departmental specificity and pertinence in the evaluation results caused by the use of a unified and fixed indicator system in the prior art. It enables the medical quality and safety risk assessment of different departments to accurately reflect their business characteristics and quality structure differences, and significantly improves the applicability of the evaluation system and the fairness of the evaluation results.

[0016] 2. This invention introduces an extreme risk triggering mechanism that includes single-indicator threshold exceeding trigger rules, multi-indicator weighted combination trigger rules, and persistent threshold exceeding trigger rules. It also combines extreme risk penalty factors to dynamically correct the comprehensive risk score. This effectively solves the problem that existing static scoring models are unable to capture complex extreme risk scenarios. It enables the timely identification and enhanced early warning of highly concealed and harmful risks such as single-indicator extreme anomalies, multi-indicator combination deterioration, and persistent indicator exceeding limits. This significantly improves the sensitivity of risk assessment and the ability to respond to extreme risk events.

[0017] 3. This invention incorporates a trend correction factor based on historical assessment data into risk level determination, dynamically compensating for changes in the overall risk score. Simultaneously, it establishes a closed-loop intervention strategy that matches the final risk level, including instructions for corrective actions, resource allocation, and monitoring frequency. After intervention, the effectiveness level of the intervention strategy is automatically determined by comparing dimensional scores from multiple rounds of assessment. This effectively solves the problems of static and rigid risk assessment, loose connection between assessment and intervention, and lack of quantitative feedback on effectiveness in existing technologies. It achieves complete closed-loop management from risk identification, graded early warning, strategy matching to effect verification, providing dynamic and traceable management support for continuous improvement of medical quality and safety. Attached Figure Description

[0018] The above and other objects, features, and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description taken in conjunction with the accompanying drawings. Several embodiments of the invention are illustrated in the drawings by way of example and not limitation, wherein: Figure 1 This is a flowchart illustrating a medical quality and safety risk assessment method based on a five-dimensional quantitative model provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of a medical quality and safety risk assessment system based on a five-dimensional quantitative model, provided in an embodiment of the present invention. Detailed Implementation

[0019] The principles and spirit of the invention will now be described with reference to several exemplary embodiments. It should be understood that these embodiments are provided merely to enable those skilled in the art to better understand and implement the invention, and are not intended to limit the scope of the invention in any way. Rather, these embodiments are provided to make the invention more thorough and complete, and to fully convey the scope of the invention to those skilled in the art.

[0020] Those skilled in the art will recognize that embodiments of the present invention can be implemented as a system, apparatus, device, method, or computer program product. Therefore, the present invention can be specifically implemented in the following forms: entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0021] It should be noted that the number of any elements in the accompanying drawings is for illustrative purposes only and not as a limitation, and any naming is for distinction only and has no limiting meaning.

[0022] The following is for reference. Figure 1 , Figure 1 This is a flowchart illustrating a medical quality and safety risk assessment method based on a five-dimensional quantitative model, provided in an embodiment of the present invention. Figure 1 As shown, a medical quality and safety risk assessment method based on a five-dimensional quantitative model includes: S1: Collect multidimensional assessment data of the target medical unit from multiple data sources. In S1, the system interfaces with multiple data sources within the hospital information system via a multi-source data acquisition interface module. This module supports medical data exchange protocols such as HL7, FHIR, or Web Service, and can interface with HIS (Hospital Information System), LIS (Laboratory Information System), PACS (Picture Archiving and Communication System), EMR (Electronic Medical Record System), operating room management system, infection control system, and human resources system, respectively. For example, it collects personnel configuration data from the human resources system, including the number of physicians on duty, the number of nurses, and the percentage of senior physicians; it collects workload data from the operating room management system, including the total number of surgeries, the percentage of level IV surgeries, the percentage of minimally invasive surgeries, and the average length of stay; it collects process quality data from the quality monitoring platform, including the surgical safety check execution rate and the timely handling rate of critical value reports; and it collects outcome quality data from the infection control system and quality assessment department, including the postoperative mortality rate, unplanned readmission rate, unplanned readmission rate, and surgical site infection rate. Data collection can be achieved through real-time extraction using ETL tools or batch import. The collected data is then aggregated and proceeds to the next step.

[0023] In this invention, the target medical unit refers to an organizational unit within a hospital that can independently conduct medical activities and undergo risk assessment, such as a surgical department, internal medicine department, emergency department, or intensive care unit. Multidimensional assessment data refers to raw data collected from different business systems, including five categories of data: staffing, workload, implementation of structural systems, process quality, and outcome quality.

[0024] S2: Based on the department category to which the target medical unit belongs, retrieve the five-dimensional risk indicator system that uniquely corresponds to that department category from the preset differentiated indicator library. In S2, the system first obtains the department category identifier of the target medical unit, such as the surgical series. Then, using this identifier as an index, it retrieves the corresponding indicator mapping table from the differentiated indicator database storage module. The differentiated indicator database storage module is a pre-configured database that stores the full score and scoring rules for each indicator item under different department categories. For example, the full score for the Level 4 surgery percentage indicator in the surgical department might be 10 points, while the same indicator in the internal medicine department might not be applicable or might have a full score of 0 points. This module is implemented using relational database tables, with one indicator mapping table corresponding to each department category. Each row in the table records one indicator item, and the fields include the indicator item name, data source field, scoring rule type and parameters, and full score.

[0025] The five-dimensional risk indicator system consists of four dimensions: personnel, business quantity, structural quality, process quality, and outcome quality. Each dimension contains multiple specific indicators; for example, the personnel dimension includes the doctor-to-nurse ratio, bed-to-nurse ratio, and the proportion of senior physicians. The full score or scoring rules for the same indicator item vary across different departmental categories, based on the business risk characteristics of that departmental category. For instance, for a surgical department, the full score for the postoperative mortality rate indicator might be set at 5 points, while for an internal medicine department, the full score might be set at 10 points, because postoperative mortality risk is more sensitive in surgery.

[0026] S3: Map the collected multidimensional assessment data to each indicator item in the five-dimensional risk indicator system to obtain the original actual score of each indicator item. Based on the sum of the original actual scores of all indicator items under each dimension and the sum of the full scores of that dimension, calculate the standardized score of each dimension. In S3, the following sub-operations are performed by the dimension calculation and mapping engine. This engine is a computational component implemented in Java or Python, which can be deployed independently and accessed via a REST API. Its internal processing flow is as follows: 1. Read the index mapping table obtained by S2.

[0027] 2. Iterate through each piece of raw data collected and calculate the original actual score for that indicator item according to the scoring rules in the mapping table. Example of scoring rules: For the nurse-bed ratio indicator, if the standard is 1:0.4 and the actual nurse-bed ratio is 1:0.33, then deduct 1 point for every 0.01 below the standard, with a maximum score of 10 points. The actual score = 10 - (0.4 - 0.33) / 0.01 = 3 points. This engine supports various rule types, including linear scoring, tiered scoring, and pass / fail scoring.

[0028] 3. For each dimension, sum the original actual scores of all indicators under that dimension to obtain the sum of the original actual scores. Simultaneously, the full scores of all indicators under this dimension are summed to obtain the total full score. .

[0029] 4. Calculate the standardized score according to the following formula.

[0030] The standardized score for the i-th dimension is calculated as follows: in, For the standardized score of the i-th dimension, This is the sum of the original actual scores for all indicators under the i-th dimension. This is the sum of the full scores of all indicators under the i-th dimension; i takes the values ​​1, 2, 3, 4, and 5, corresponding to the personnel dimension, business quantity dimension, structural quality dimension, process quality dimension, and result quality dimension, respectively.

[0031] Typically, for a surgical department, if the sum of the original actual scores for the personnel dimension is 39 points and the sum of the full scores is 75 points, then the standardized score for that dimension is calculated to be 52 points.

[0032] S4: Obtain a set of dimension weight coefficients, and sum the standardized scores of each dimension with the corresponding dimension weight coefficients to generate a comprehensive risk score.

[0033] In S4, the weighted fusion scoring module performs the calculation. This module is a lightweight mathematical function that receives a standardized score array from five dimensions, reads a preset weight coefficient vector, and performs a dot product operation. The weight coefficient vector satisfies the condition that the sum of all weight coefficients is 1. For example, a typical weight configuration is: personnel dimension weight coefficient 0.25, business quantity dimension weight coefficient 0.20, structural quality dimension weight coefficient 0.15, process quality dimension weight coefficient 0.20, and result quality dimension weight coefficient 0.20.

[0034] The comprehensive risk score is calculated as follows: in, For comprehensive risk scoring, For the standardized score of the i-th dimension, The preset weight coefficients are for the i-th dimension.

[0035] Using the previous example, if the standardized scores for the five dimensions are 52, 65, 75, 38, and 45 respectively, and the weighting coefficients are as above, then the calculated comprehensive risk score is 53.85.

[0036] S5: Obtain multiple extreme risk triggering conditions corresponding to the department category to which the target medical unit belongs, and dynamically determine the final risk level of the target medical unit based on the comprehensive risk score and the multiple extreme risk triggering conditions. S5 contains several sub-steps, as detailed below.

[0037] S51: Read the set of extreme risk triggering rules associated with the department category to which the target medical unit belongs from the department-specific threshold configuration library. The system reads the rule set from the department-specific threshold configuration module. This module is a database table that stores the thresholds and rules corresponding to different department categories. Extreme risk triggering conditions refer to the set of judgment rules used to identify extreme risk situations, including three types: single-indicator threshold exceeding triggering rules, multi-indicator weighted combination triggering rules, and persistent threshold exceeding triggering rules. For example, for surgical departments, a single-indicator threshold exceeding triggering rule can be set as: triggering when the postoperative mortality rate exceeds 5‰; a multi-indicator weighted combination triggering rule can be set as: triggering when the unplanned readmission rate multiplied by a weight of 0.6 plus the unplanned readmission rate multiplied by a weight of 0.4 results in a result exceeding 7%; a persistent threshold exceeding triggering rule can be set as: triggering when the standardized score for personnel dimensions exceeds 85 points for three consecutive days. Specific thresholds are set based on the department's historical data baseline.

[0038] S52: Compare the comprehensive risk score with a set of conventional risk thresholds to generate a preliminary risk level. This step first obtains historical assessment data for the target medical unit and extracts the comprehensive risk score from the previous assessment. For example, if the previous score was 50 points and the current score is 53.85 points, then the change in the comprehensive risk score is calculated.

[0039] The trend value of the comprehensive risk score is calculated as shown in the following formula: in, This represents the change in the overall risk score. Based on the current comprehensive risk score, This is the previous comprehensive risk score. In this example, the change is 3.85 points.

[0040] The trend correction factor is calculated based on the trend value, as shown in the following formula: in, As a trend correction factor, This is the penalty coefficient for an upward trend, typically set to 0.2. The reward coefficient for a downward trend is typically set to 0.1. This is the medium-risk threshold, typically set at 60 points. In this example... Calculated .

[0041] The revised preliminary risk score is calculated as follows: In this example, the corrected score is approximately 53.86.

[0042] The revised preliminary risk score is then compared with the standard risk threshold. (Standard high-risk threshold) A typical score is 70, which is the medium-risk threshold. The typical value is 50. Since 53.86 falls between 50 and 70, the initial risk level is determined to be medium risk.

[0043] S53: Statistically analyze the rules triggered in the extreme risk triggering rule set to obtain the total number of triggering rules and the distribution of triggering rule types. The system iterates through the rule set obtained by S51 and checks whether each rule has been triggered. For example, if the postoperative mortality rate of patients in this surgical department is 20‰, exceeding 5‰, then a single-indicator threshold exceeding rule is triggered. If no other rules are triggered, the total number of triggered rules is 1, the number of single-indicator triggering rules is 1, the number of multi-indicator weighted combination triggering rules is 0, and the number of persistent threshold exceeding triggering rules is 0.

[0044] S54: Calculate the extreme risk penalty factor based on the total number of triggering rules and the distribution of triggering rule types. The extreme risk penalty factor is calculated as follows: in, As an extreme risk penalty factor, This represents the number of rules triggered when a single metric exceeds its threshold. The number of multi-indicator weighted combination trigger rules that are triggered. The number of persistent over-threshold trigger rules that have been triggered; This is the penalty weight for a single indicator, typically set to 5. The penalty weight for multiple indicators is typically set to 10. This is a persistent penalty weight, typically set to 20. In this example... .

[0045] S55: Calculate the final risk score based on the preliminary risk level and the extreme risk penalty factor. The final risk score is calculated as follows: in, For the final risk score, This is a comprehensive risk score. In this example... point.

[0046] S56: Compare the final risk score with the set of conventional risk thresholds again to generate a final risk level. The score of 58.85 is compared with the high-risk threshold of 70 and the medium-risk threshold of 50. Since 58.85 is less than 70 and greater than or equal to 50, the final risk level is medium risk. However, it should be noted that in the embodiment given in this application, the postoperative mortality rate exceeded the threshold, triggering an extreme risk condition, and this condition itself was set to directly result in a high risk. In this example, only a penalty factor is added, reflecting hierarchical logic. In the actual system, the penalty factor can be adjusted according to the size and threshold setting. The system outputs the final risk level and generates a visual identifier: a red identifier for high risk, a yellow identifier for medium risk, and a green identifier for low risk.

[0047] S6: Based on the final risk level, match the corresponding closed-loop intervention instruction from the preset handling strategy library, generate a risk assessment report containing the comprehensive risk score, standardized scores of each dimension, and the closed-loop intervention instruction, and output the risk assessment report to the corresponding management terminal. In S6, the action instruction matching and report generation module performs the operation. This module retrieves the action strategy library based on the final risk level. The action strategy library is a relational table that stores intervention strategy templates corresponding to different risk levels. Each template includes rectification action instructions, resource allocation instructions, and monitoring frequency instructions. For example, for a medium-risk level, the matched intervention strategy template is: rectification action instruction is "rectification within a specified time limit," resource allocation instruction is "optimized scheduling," and monitoring frequency instruction is "monthly monitoring." The system generates a structured risk assessment report, which includes a comprehensive risk score of 58.85, standardized scores for each dimension, a medium-risk risk level with a yellow label, and the aforementioned closed-loop intervention instructions. The report is pushed to the hospital's quality management office terminal and the surgical department's management terminal via a message middleware (such as RabbitMQ).

[0048] S7: Closed-loop feedback and comparative evaluation steps The closed-loop management process includes four stages: assessment, rectification, monitoring, and reassessment. After the intervention instructions output by S6 are executed and a monitoring cycle (e.g., one month) has elapsed, the system automatically executes S1 to S5 again to obtain a new round of final risk level, a new round of comprehensive risk score, and a new round of standardized scores for each dimension. Then, the difference between the standardized scores for each dimension is calculated.

[0049] The standardized score difference for each dimension is calculated as follows: in, Let be the standardized score change for the i-th dimension. For the new round of standardized scores of the i-th dimension, is the standardized score of the i-th dimension in the previous round.

[0050] For example, if the standardized scores after intervention are 55 for personnel, 70 for business quantity, 80 for structural quality, 50 for process quality, and 60 for outcome quality, the changes would be +3, +5, +5, +12, and +15 respectively. A comparative risk assessment report is generated. The effectiveness level is then automatically determined based on the differences: if all dimensions show positive changes, the effectiveness level is considered effective. If some dimensions improve while others worsen, it is considered partially effective; if all dimensions show no improvement or worsen, it is considered ineffective. The system records the effectiveness level in the management log and can trigger higher-level interventions based on ineffective results, forming a continuous improvement cycle.

[0051] The above embodiments of the present invention have the following beneficial effects: like Figure 2 As shown in some embodiments, a medical quality and safety risk assessment system based on a five-dimensional quantitative model is provided. The system includes: The differentiated indicator library storage module 201 is used to store the five-dimensional risk indicator system corresponding to different department categories, as well as the assignment standards and full scores of each indicator item; The department-specific threshold configuration module 202 is used to store extreme risk triggering rule sets corresponding to different department categories. The extreme risk triggering rule sets include single indicator threshold exceeding triggering rules, multi-indicator weighted combination triggering rules, and continuous threshold exceeding triggering rules. The multi-source data acquisition interface module 203 is used to connect with multiple data sources in the hospital information system to collect multi-dimensional evaluation data of the target medical unit; The dimension calculation and mapping engine 204 is used to call the corresponding five-dimensional risk indicator system from the differentiated indicator library storage module according to the department category of the target medical unit, map the collected multidimensional assessment data into the original actual scores of each indicator item, and calculate the standardized scores of each dimension. The weighted fusion scoring module 205 is used to obtain a preset five-dimensional weight vector, and to sum the standardized scores of each dimension with the corresponding weight coefficients to generate a comprehensive risk score. The risk level dynamic determination module 206 is used to read the extreme risk triggering rule set from the department-specific threshold configuration module, and dynamically output the final risk level by combining the comparison results of the comprehensive risk score and the conventional risk threshold and the satisfaction of the extreme risk triggering rule set. The disposal instruction matching and report generation module 207 is used to match closed-loop intervention instructions according to the final risk level, generate a risk assessment report including a comprehensive risk score, standardized scores of each dimension and closed-loop intervention instructions, and output the risk assessment report to the corresponding management terminal.

[0052] It is understandable that the modules and references recorded in this medical quality and safety risk assessment system based on a five-dimensional quantitative model are... Figure 1 The steps described correspond to those in the medical quality and safety risk assessment method based on the five-dimensional quantitative model. Therefore, the operations, characteristics, and beneficial effects described above for the medical quality and safety risk assessment method based on the five-dimensional quantitative model are also applicable to the medical quality and safety risk assessment system based on the five-dimensional quantitative model and its constituent modules, and will not be repeated here.

[0053] Furthermore, the storage medium in the embodiments of this application stores program instructions capable of implementing all the above methods. These program instructions can be stored in the storage medium in the form of a software product, including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, or terminal devices such as computers, servers, mobile phones, and tablets.

[0054] The above description is merely an explanation of some preferred embodiments of the present invention and the technical principles employed. Those skilled in the art should understand that the scope of the invention as described in the embodiments of the present invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention.

Claims

1. A medical quality and safety risk assessment method based on a five-dimensional quantitative model, characterized in that, Includes the following steps: S1: Collect multidimensional evaluation data of the target medical unit from multiple data sources; S2: Based on the department category to which the target medical unit belongs, call the five-dimensional risk indicator system that uniquely corresponds to the department category from the preset differentiated indicator library. The five-dimensional risk indicator system includes personnel dimension, business quantity dimension, structural quality dimension, process quality dimension and result quality dimension. Each dimension is pre-bound with multiple indicator items as well as the assignment standard and full score associated with the department category. S3: Map the collected multidimensional assessment data to each indicator item in the five-dimensional risk indicator system to obtain the original actual score of each indicator item, and calculate the standardized score of each dimension based on the sum of the original actual scores of all indicator items under each dimension and the sum of the full score of that dimension. S4: Obtain a set of dimension weight coefficients, and perform a weighted sum of the standardized scores of each dimension and the corresponding dimension weight coefficients to generate a comprehensive risk score; S5: Obtain multiple extreme risk triggering conditions corresponding to the department category to which the target medical unit belongs, and dynamically determine the final risk level of the target medical unit based on the comprehensive risk score and the multiple extreme risk triggering conditions; S6: Based on the final risk level, match the corresponding closed-loop intervention instruction from the preset handling strategy library, generate a risk assessment report containing the comprehensive risk score, standardized scores of each dimension and the closed-loop intervention instruction, and output the risk assessment report to the corresponding management terminal.

2. The medical quality and safety risk assessment method based on a five-dimensional quantitative model according to claim 1, characterized in that, The standardized scores for each dimension calculated in S3 include: The standardized score for the i-th dimension is calculated as follows: ； in, For the standardized score of the i-th dimension, This is the sum of the original actual scores for all indicators under the i-th dimension. This is the sum of the full scores of all indicators under the i-th dimension; i takes the values ​​1, 2, 3, 4, and 5, corresponding to the personnel dimension, business quantity dimension, structural quality dimension, process quality dimension, and result quality dimension, respectively.

3. The medical quality and safety risk assessment method based on a five-dimensional quantitative model according to claim 1, characterized in that, The comprehensive risk score generated in S4 includes: Obtain a set of dimensional weight coefficients, which constitute a five-dimensional weight vector and satisfy the constraint that the sum of the weight coefficients of the personnel dimension, the business quantity dimension, the structural quality dimension, the process quality dimension, and the result quality dimension is equal to 1. The comprehensive risk score is calculated as follows: ； in, For comprehensive risk scoring, For the standardized score of the i-th dimension, The preset weight coefficients are for the i-th dimension.

4. The medical quality and safety risk assessment method based on a five-dimensional quantitative model according to claim 1, characterized in that, The S5 step involves obtaining multiple extreme risk triggering conditions corresponding to the department category of the target medical unit, including: Retrieve the extreme risk triggering rule set associated with the department category to which the target medical unit belongs from the department-specific threshold configuration library. The extreme risk triggering rule set includes three triggering types: single indicator exceeding threshold triggering rule, multi-indicator weighted combination triggering rule, and continuous exceeding threshold triggering rule. The single-indicator threshold triggering rule is as follows: for each of the multiple preset key quality indicators, a corresponding department-specific triggering threshold is set. When the actual measured value of any key quality indicator exceeds the department-specific triggering threshold corresponding to that indicator, the rule is triggered. The multi-indicator weighted combination triggering rule is as follows: For a set of preset key quality indicators, a corresponding weighted summation triggering threshold is set. When the result obtained by weighted summation of the actual measured values ​​of each indicator in the set exceeds the weighted summation triggering threshold, the rule is triggered. The continuous threshold triggering rule is as follows: For at least one preset key quality indicator, a duration threshold is set. When the actual measured value of the indicator continuously exceeds its corresponding department-specific triggering threshold for a duration that reaches the duration threshold, the rule is triggered.

5. The medical quality and safety risk assessment method based on a five-dimensional quantitative model according to claim 4, characterized in that, The final risk level is dynamically determined in S5 based on the comprehensive risk score and the multiple extreme risk triggering conditions, including: S51: Compare the comprehensive risk score with a set of conventional risk thresholds to generate a preliminary risk level. The set of conventional risk thresholds includes a high-risk threshold and a medium-risk threshold, and the high-risk threshold is greater than the medium-risk threshold. S52: Statistically analyze the rules triggered in the extreme risk triggering rule set to obtain the total number of triggering rules and the distribution of triggering rule types. The distribution of triggering rule types includes the number of single-indicator over-threshold triggering rules, the number of multi-indicator weighted combination triggering rules, and the number of persistent over-threshold triggering rules. S53: Based on the total number of triggering rules and the distribution of triggering rule types, calculate the extreme risk penalty factor, as shown in the following formula: ； in, As an extreme risk penalty factor, This represents the number of rules triggered when a single metric exceeds its threshold. The number of multi-indicator weighted combination trigger rules that are triggered. The number of persistent over-threshold trigger rules that have been triggered; For single-indicator penalty weights; Penalty weights are applied to multiple indicators; For continuous penalty weights; S54: Based on the preliminary risk level and the extreme risk penalty factor, calculate the final risk score as shown in the following formula: in, For the final risk score, For comprehensive risk scoring; S55: The final risk score is compared again with the set of conventional risk thresholds to generate a final risk level. A high-risk level is generated when the final risk score is greater than or equal to the high-risk threshold, a medium-risk level is generated when the medium-risk threshold is less than or equal to the final risk score and the final risk score is less than the high-risk threshold, and a low-risk level is generated when the final risk score is less than the medium-risk threshold.

6. The medical quality and safety risk assessment method based on a five-dimensional quantitative model according to claim 5, characterized in that, The preliminary risk level is generated in S51, including: Obtain historical assessment data of the target medical unit, and extract the comprehensive risk score of the previous assessment from the historical assessment data; The trend value of the comprehensive risk score is calculated as shown in the following formula: ； in, This represents the change in the overall risk score. Based on the current comprehensive risk score, This is the previous comprehensive risk score; The trend correction factor is calculated based on the aforementioned trend value, as shown in the following formula: ； in, As a trend correction factor, The penalty coefficient is for upward trends; The reward coefficient represents a downward trend. The threshold is medium risk. The revised preliminary risk score is the sum of the comprehensive risk score and the trend correction factor; The revised preliminary risk score is compared with the set of conventional risk thresholds to generate a preliminary risk level, wherein: When the revised preliminary risk score is greater than or equal to the high risk threshold, the preliminary risk level is generated as a high risk level. When the medium risk threshold is less than or equal to the revised preliminary risk score and the revised preliminary risk score is less than the high risk threshold, the preliminary risk level is generated as medium risk level. When the revised preliminary risk score is less than the medium risk threshold, the preliminary risk level is generated as low risk.

7. The medical quality and safety risk assessment method based on a five-dimensional quantitative model according to claim 1, characterized in that, The step S6 involves matching the corresponding closed-loop intervention command from a preset treatment strategy library, specifically including: Based on the final risk level, an intervention strategy template associated with the risk level is retrieved from the disposal strategy library. The intervention strategy template includes rectification action instructions, resource allocation instructions, and monitoring frequency instructions. Among them, the intervention strategy templates associated with high-risk levels include: the rectification action instruction is to suspend non-emergency elective surgeries, the resource allocation instruction is to increase the number of medical staff, and the monitoring frequency instruction is to monitor weekly; The intervention strategy template associated with medium-risk levels includes: rectification action instructions are to rectify within a time limit, resource allocation instructions are to optimize the shift schedule, and monitoring frequency instructions are to monitor monthly; The intervention strategy template for low-risk levels includes: the rectification action instruction is to maintain routine operations, the resource allocation instruction is to maintain existing resources, and the monitoring frequency instruction is to monitor every quarter.

8. The medical quality and safety risk assessment method based on a five-dimensional quantitative model according to claim 1, characterized in that, Also includes: After the closed-loop intervention instructions are completed and a monitoring cycle has passed, S1 to S5 are executed again to obtain a new round of final risk level, a new round of comprehensive risk score and a new round of standardized scores for each dimension. Based on the standardized scores of the same dimension in the new round and the previous round, the difference between the standardized scores of each dimension is calculated as follows: ； in, Let be the standardized score change for the i-th dimension. For the new round of standardized scores of the i-th dimension, The previous standardized score for the i-th dimension; A comparative risk assessment report containing the difference is generated, and the effectiveness level of the implemented intervention strategy is automatically determined based on the difference. The effectiveness level includes effective, partially effective, and ineffective.

9. The medical quality and safety risk assessment method based on a five-dimensional quantitative model according to claim 8, characterized in that, The automatic determination of the effectiveness level of the implemented intervention strategy based on the difference is as follows: When the standardized score changes of all dimensions are positive, the effectiveness level is determined to be valid. When the standardized score change of some dimensions is positive and the standardized score change of other dimensions is negative or zero, the effectiveness level is determined to be partially effective. When the standardized score changes for all dimensions are non-positive, the effective level is deemed invalid.

10. A medical quality and safety risk assessment system based on a five-dimensional quantitative model, used to execute the method of any one of claims 1 to 9, characterized in that, include: The differentiated indicator library storage module is used to store the five-dimensional risk indicator system corresponding to different department categories, as well as the assignment standards and full scores of each indicator item; The department-specific threshold configuration module is used to store extreme risk triggering rule sets corresponding to different department categories. The extreme risk triggering rule sets include single indicator threshold exceeding triggering rules, multi-indicator weighted combination triggering rules, and continuous threshold exceeding triggering rules. The multi-source data acquisition interface module is used to connect with multiple data sources in the hospital information system to collect multi-dimensional evaluation data of the target medical unit; The dimensional calculation and mapping engine is used to call the corresponding five-dimensional risk indicator system from the differentiated indicator library storage module according to the department category of the target medical unit, map the collected multidimensional assessment data into the original actual scores of each indicator item, and calculate the standardized scores of each dimension. The weighted fusion scoring module is used to obtain a preset five-dimensional weight vector, and to sum the standardized scores of each dimension with the corresponding weight coefficients to generate a comprehensive risk score. The risk level dynamic determination module is used to read the extreme risk triggering rule set from the department-specific threshold configuration module, and dynamically output the final risk level by combining the comparison results of the comprehensive risk score and the conventional risk threshold and the satisfaction of the extreme risk triggering rule set. The disposal instruction matching and report generation module is used to match closed-loop intervention instructions according to the final risk level, generate a risk assessment report containing a comprehensive risk score, standardized scores of each dimension and closed-loop intervention instructions, and output the risk assessment report to the corresponding management terminal.

Images