Medical staff evaluation dynamic generation method and system
By constructing dynamic correction coefficients and comprehensive evaluation degree values, the problems of one-sided evaluation dimensions and insufficient dynamic response in the evaluation of medical staff have been solved, realizing accurate quantification and cross-cycle trend tracking of medical staff evaluation, and improving the fairness and accuracy of the evaluation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-02-10
- Publication Date
- 2026-03-27
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing evaluation methods for medical staff are too one-sided in their assessment dimensions, making it difficult to fully reflect professional quality control factors such as compliance of treatment pathways, stability of medical record quality, and peer review. Furthermore, they lack the ability to dynamically respond to sudden emergencies and performance leaps across cycles, which affects the fairness and incentive effect of the evaluation.
By acquiring data on the workload, quality control indicators, multidimensional feedback, and emergency event records of medical staff, a dynamic correction coefficient is constructed to generate a comprehensive evaluation value and map it to a dimensionless scoring space, thereby achieving precise quantification of the evaluation and cross-cycle trend tracking.
It significantly improves the fairness and accuracy of the evaluation, can reflect the actual performance of medical staff in handling abnormal events in a timely manner, and can identify continuous changes in performance trends.
Smart Images
Figure CN121745774A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of personnel evaluation, in particular to a medical staff evaluation dynamic generation method and system. BACKGROUND
[0002] Medical staff evaluation refers to the technology of comprehensively evaluating the work performance and professional ability level of medical staff by collecting and analyzing multi-dimensional data such as business load, medical quality, service feedback and emergency performance of medical staff within the evaluation period. It is an important means for medical institutions to implement fine management and optimize human resources.
[0003] The prior art often uses a method based on a single business indicator (such as the number of outpatient visits or the number of surgical cases) or a simple satisfaction score, combined with subjective experience evaluation, to divide performance levels and achieve evaluation. However, the evaluation dimensions in the prior art are one-sided and cannot fully reflect professional quality control elements such as compliance with diagnosis and treatment paths, stability of medical record quality, and peer review. In addition, the static weight configuration lacks dynamic response capability to sudden emergency events and cross-period performance transitions, resulting in evaluation results that cannot timely reflect the actual performance of medical staff in handling abnormal events, and difficulty in identifying the continuity of performance trends, affecting the fairness and incentive effect of evaluation. SUMMARY
[0004] To solve the above technical problems, the purpose of the present application is to provide a medical staff evaluation dynamic generation method and system, which realizes the accurate quantification of evaluation comprehensive index and the intelligent tracking of cross-period trend, significantly improving the fairness and accuracy of evaluation generation.
[0005] To achieve the above purpose, the present application provides a medical staff evaluation dynamic generation method, comprising:
[0006] Obtaining business load data, quality control indicator data, multi-dimensional feedback data and emergency event record data of medical staff within the evaluation period, the business load data including outpatient visit sequence, inpatient management bed day sequence, and surgical operation classification count sequence, the quality control indicator data including diagnosis and treatment path compliance rate sequence, medical record A-level rate sequence, and rational drug use score sequence, the multi-dimensional feedback data including patient satisfaction score sequence, peer review score sequence, and complaint event marker sequence; Determining the fluctuation synchronization of the outpatient visit sequence and the inpatient management bed day sequence, determining the cross-class operation proportion in the surgical operation classification count sequence, and comprehensively determining the business load balance characteristic value; Determining a quality control stability characteristic value according to the quality control indicator data, and determining a service response characteristic value according to the multi-dimensional feedback data; generate a comprehensive evaluation degree value based on a coupling relationship between the service load balancing feature value and the quality control stability feature value, and a deviation amplitude of the service response feature value from a preset reference threshold value; construct a dynamic correction coefficient based on an identification result of an abnormal response event in the emergency event record data and a transition amplitude of the comprehensive evaluation degree value between adjacent evaluation periods; weight and fuse the comprehensive evaluation degree value and the dynamic correction coefficient and map them to a dimensionless score space to obtain a medical staff evaluation comprehensive index.
[0007] Further, determine the fluctuation synchronization of the outpatient reception person sequence and the inpatient management bed day sequence, determine the cross-level operation proportion in the operation grading count sequence, and comprehensively determine the service load balancing feature value, including: extract the phase difference of the periodic fluctuations of the outpatient reception person sequence and the inpatient management bed day sequence, respectively, and if the phase difference is less than a preset synchronization tolerance, generate a load coordination marker; statistically determine the ratio of the number of operations exceeding the practice range level in the operation grading count sequence to the total number of operations as the cross-level operation proportion; perform negative correlation normalization processing on the difference between the sequence coincidence degree corresponding to the load coordination marker and the cross-level operation proportion to obtain the service load balancing feature value.
[0008] Further, determine a quality control stability feature value according to the quality control indicator data and a service response feature value according to the multi-dimensional feedback data, including: determine the quality control stability feature value according to the stability of the diagnosis and treatment path compliance rate sequence, the dispersion coefficient of the medical record A-level rate sequence, and the range of the reasonable drug use score sequence; determine the service response feature value according to the distribution skewness of the patient satisfaction score sequence, the concentration trend deviation of the peer review score sequence, and the density of the complaint event marker sequence.
[0009] Further, determine a quality control stability feature value according to the stability of the diagnosis and treatment path compliance rate sequence, the dispersion coefficient of the medical record A-level rate sequence, and the range of the reasonable drug use score sequence, including: calculate the moving standard deviation of the diagnosis and treatment path compliance rate sequence, and if the moving standard deviation is less than a preset fluctuation threshold, assign a stability weighting factor; extract the quartile dispersion coefficient of the medical record A-level rate sequence as the dispersion coefficient; obtain the range of the maximum value and the minimum value in the reasonable drug use score sequence and perform standard deviation normalization; multiply the stability weighting factor, the reciprocal of the quartile dispersion coefficient, and the normalized range to obtain the quality control stability feature value.
[0010] Further, according to the distribution skewness of the patient satisfaction score sequence, the central tendency deviation of the peer review score sequence, and the density of the complaint event marker sequence, a service response feature value is determined, including: If the skewness coefficient of the patient satisfaction score sequence is greater than zero, a negative skewness marker is generated, and the absolute value of the skewness corresponding to the negative skewness marker is taken as a first input parameter; The proportion of scores below a preset pass line in the peer review score sequence is calculated as a central tendency deviation; The number of markers per unit time in the complaint event marker sequence is counted as a complaint density; The first input parameter, the central tendency deviation, and the complaint density are positively correlated and normalized to obtain a service response feature value.
[0011] Further, based on the coupling relationship between the business load balancing feature value and the quality control stability feature value, and the deviation amplitude of the service response feature value from a preset reference threshold, a comprehensive evaluation degree value is generated, including: A two-dimensional coupling matrix of the business load balancing feature value and the quality control stability feature value is constructed, and the eigenvalue of the two-dimensional coupling matrix is extracted as a coupling strength parameter; The relative distance between the service response feature value and the preset reference threshold is calculated, and if the relative distance is greater than a preset warning distance, a deviation warning marker is generated; The coupling strength parameter and the correction weight corresponding to the deviation warning marker are weighted and summed to obtain a comprehensive evaluation degree value.
[0012] Further, according to the identification result of the abnormal response event in the emergency event record data and the transition amplitude of the comprehensive evaluation degree value between adjacent evaluation periods, a dynamic correction coefficient is constructed, including: Response timeout events, improper handling events, and cover-up and omission events are filtered from the emergency event record data, are given differentiated abnormal weights and are accumulated to obtain an emergency abnormality cumulative value; The absolute value of the difference between the comprehensive evaluation degree values of the current evaluation period and the previous evaluation period is calculated, and if the absolute value of the difference is greater than a preset transition threshold, a transition anomaly is identified; The emergency abnormality cumulative value and the determination result of the transition anomaly are positively correlated and mapped to generate a dynamic correction coefficient.
[0013] Further, the comprehensive evaluation degree value and the dynamic correction coefficient are weighted and fused and mapped to a dimensionless score space to obtain a medical staff evaluation comprehensive index, including: If the dynamic correction coefficient is greater than a preset correction threshold, an inhibitory correction is performed on the comprehensive evaluation degree value by using a decreasing type weighting function; If the dynamic correction coefficient is less than or equal to the preset correction threshold, a gain correction is performed on the comprehensive evaluation degree value by using an increasing type weighting function. The corrected comprehensive evaluation degree value is mapped to a preset dimensionless score space through standard score conversion to obtain a medical staff evaluation comprehensive index.
[0014] Further, after the comprehensive evaluation degree value and the dynamic correction coefficient are weighted and fused and mapped to the dimensionless score space to obtain the medical staff evaluation comprehensive index, the method further includes: According to the ranking quantile value of the medical staff evaluation comprehensive index in all medical staff evaluation objects, a grade evaluation mark is generated.
[0015] In order to achieve the above-mentioned purpose, the present application further provides a medical staff evaluation dynamic generation system, comprising: A data acquisition module is configured to acquire service load data, quality control index data, multi-dimensional feedback data and emergency event record data of medical staff in an evaluation period, wherein the service load data includes a clinic visit sequence, a hospital management bed day sequence and a surgical operation grading count sequence, the quality control index data includes a diagnosis and treatment path compliance rate sequence, a medical record A-level rate sequence and a rational drug use score sequence, and the multi-dimensional feedback data includes a patient satisfaction score sequence, a peer review score sequence and a complaint event mark sequence. A first determination module is configured to determine the fluctuation synchronization of the clinic visit sequence and the hospital management bed day sequence, determine the cross-grade operation proportion in the surgical operation grading count sequence, and comprehensively determine a service load balance characteristic value. A second determination module is configured to determine a quality control stability characteristic value according to the quality control index data, and determine a service response characteristic value according to the multi-dimensional feedback data. A numerical generation module is configured to generate a comprehensive evaluation degree value based on the coupling relationship between the service load balance characteristic value and the quality control stability characteristic value and the deviation amplitude of the service response characteristic value from a preset reference threshold. A dynamic correction module is configured to construct a dynamic correction coefficient according to the identification result of an abnormal response event in the emergency event record data and the transition amplitude of the comprehensive evaluation degree value between adjacent evaluation periods. A comprehensive evaluation module is configured to weight and fuse the comprehensive evaluation degree value and the dynamic correction coefficient and map them to a dimensionless score space to obtain a medical staff evaluation comprehensive index.
[0016] Compared with the prior art, the present application has the following advantages: The application discloses a medical staff evaluation dynamic generation method and system, acquires business load data, quality control index data and multi-dimensional feedback data, determines fluctuation synchronization of outpatient service and inpatient management bed day sequences, determines cross-level operation proportion of operation grading count sequences, and determines business load balance characteristic values; determines quality control stability characteristic values according to the quality control index data, and determines service response characteristic values according to the multi-dimensional feedback data; generates a comprehensive evaluation degree value based on a coupling relationship between the business load balance characteristic values and the quality control stability characteristic values and a deviation amplitude between the service response characteristic values and a preset reference threshold; constructs a dynamic correction coefficient according to the recognition result and the comprehensive evaluation degree value; and obtains an evaluation comprehensive index according to the comprehensive evaluation degree value and the dynamic correction coefficient, so that the accurate quantification of the evaluation comprehensive index and the intelligent tracking of cross-cycle trends are realized, and the fairness and accuracy of evaluation generation are significantly improved. BRIEF DESCRIPTION OF DRAWINGS
[0017] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of preferred embodiments and are not meant to limit the present application. Furthermore, the same reference numerals in different drawings are intended to represent the same components throughout the several drawings. In the drawings: Figure 1 A flowchart of a medical staff evaluation dynamic generation method in an embodiment of the application is shown; Figure 2 A structure diagram of a medical staff evaluation dynamic generation system in an embodiment of the application is shown. DETAILED DESCRIPTION
[0018] The specific embodiments of the application will be further described in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the application, but are not used to limit the scope of the application.
[0019] In the description of the present application, it should be understood that the terms "center", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.
[0020] The terms "first", "second", "third", etc. are only used for description purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second", etc. can explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise stated, the meaning of "multiple" is two or more.
[0021] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connecting" should be understood in a broad sense, for example, can be fixed connection, can also be detachable connection, or integral connection; can be mechanical connection, can also be electrical connection; can be direct connection, can also be indirect connection through an intermediate medium, can be internal communication of two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.
[0022] The following is a description of the preferred embodiments of the present application in conjunction with the accompanying drawings.
[0023] As Figure 1 shown, the embodiments of the present application disclose a medical staff evaluation dynamic generation method, characterized in that, comprising: S110: acquiring service load data, quality control index data, multi-dimensional feedback data and emergency event record data of medical staff in an evaluation period, the service load data comprising outpatient visit sequence, inpatient management bed day sequence, operation classification count sequence, the quality control index data comprising diagnosis and treatment path compliance rate sequence, medical record first-class rate sequence, rational drug use score sequence, the multi-dimensional feedback data comprising patient satisfaction score sequence, peer review score sequence, complaint event marker sequence; S120: determining the fluctuation synchronization of the outpatient visit sequence and the inpatient management bed day sequence, determining the cross-class operation proportion in the operation classification count sequence, and comprehensively determining the service load balance characteristic value; S130: determining a quality control stability characteristic value according to the quality control index data, and determining a service response characteristic value according to the multi-dimensional feedback data; S140: generating a comprehensive evaluation degree value based on the coupling relationship between the service load balance characteristic value and the quality control stability characteristic value, and the deviation amplitude of the service response characteristic value and the preset reference threshold value; S150: constructing a dynamic correction coefficient according to the identification result of abnormal response event in the emergency event record data and the transition amplitude of the comprehensive evaluation degree value in adjacent evaluation periods; S160: weighting and fusing the comprehensive evaluation degree value and the dynamic correction coefficient and mapping to a dimensionless score space to obtain a medical staff evaluation comprehensive index.
[0024] In this embodiment, the evaluation period is set to a quarter, for example, the first quarter of 2024 (January 1 to March 31). The business load data is extracted from the hospital HIS system: the outpatient service sequence is the number of patients treated each day, for example, the sequence value is [45, 52, 48, 50...n1], a total of 90 data points. The inpatient management bed sequence is the number of inpatient beds managed each day multiplied by the number of days, for example, [320, 335, 310...n2]. The operation classification count sequence is counted according to the hospital operation classification standard, for example, 5 cases of first-class operation, 12 cases of second-class operation, 8 cases of third-class operation, and 2 cases of fourth-class operation. The quality control indicator data comes from the electronic medical record quality control system: the diagnosis and treatment path compliance rate sequence is the matching percentage of the actual diagnosis and treatment of each discharged patient and the clinical path, for example, the sequence value is [95%, 92%, 96%...n3]. The medical record first-class rate sequence is the proportion of medical records rated as first-class each month, for example, [0.88, 0.85, 0.90]. The rational drug use score sequence is automatically scored by the clinical pharmacist system according to the rules of antibacterial drug use, compatibility contraindications, etc., for example, [92 points, 88 points, 95 points]. Multidimensional feedback data comes from the patient satisfaction survey system, the peer review platform of the department, and the complaint management system: the patient satisfaction score sequence is the satisfaction score of each outpatient patient (1-10 points), for example, [9, 8, 9, 7...n4]. The peer review score sequence is the score of the professional ability and cooperative spirit of the department colleagues within the quarter (1-100 points), for example, [85, 78, 82]. The complaint event marker sequence is a 0 / 1 marker of whether it is complained about each day, for example, [0, 0, 1, 0, 0, 0, 2...n5] (2 means 2 complaints on the same day). Emergency event record data comes from the hospital emergency command center, recording participation in rescue, public health emergencies, etc.
[0025] In some embodiments of the present application, the fluctuation synchronization of the outpatient service sequence and the inpatient management bed sequence is determined, the proportion of cross-level operations in the operation classification count sequence is determined, and the business load balance feature value is comprehensively determined, including: Respectively extracting the phase difference of the periodic fluctuation of the outpatient service sequence and the inpatient management bed sequence, if the phase difference is less than a preset synchronization tolerance, a load coordination marker is generated; Statistically determining the ratio of the number of operations exceeding the level of the practice range in the operation classification count sequence to the total number of operations as the proportion of cross-level operations; The difference between the sequence coincidence degree corresponding to the load coordination marker and the proportion of cross-level operations is negatively correlated and normalized to obtain the business load balance feature value.
[0026] In this embodiment, the phase difference of periodic fluctuations is calculated by extracting the main periodic components through Fourier transform. For example, the main period of the outpatient visit sequence is 7 days, and the main period of the inpatient management bed day sequence is 7 days. The phase difference is 0 days, which is less than the preset synchronization tolerance of 1 day, and the load coordination label is 1. The sequence overlap is calculated as the percentage of days when the two sequences are simultaneously at their peak / trough within the cycle. For example, if there are 60 days of synchronization within 90 days, the overlap is 0.67. In the surgical operation grade counting sequence, Zhang San is registered as a level 2 surgical qualification. He performed 8 level 3 surgeries and 2 level 4 surgeries. His scope of practice is level 2 and below. The number of cross-level operations is 8+2=10, and the total number of surgeries is 27, with a cross-level operation ratio of 0.37. The difference is 0.3. The negative correlation normalization process uses (1-0.30) / (1+0.30)=0.54 to obtain a business load balance characteristic value of 0.54. The closer it is to 1, the more balanced the load and the fewer cross-level operations. The preset synchronization tolerance is set to 1 or 2 days based on the hospital's scheduling cycle. Cross-level operations are determined according to the physician's practice registration scope, and operations exceeding the highest registration level are included.
[0027] The beneficial effects of the above technical solution are: phase difference extraction quantitatively assesses the synergy between outpatient and inpatient services; load synergy markers identify efficient working modes; cross-level operation ratios directly quantify the compliance risks of practice; difference calculation balances synergy and compliance; negative correlation normalization standardizes feature values; accurately reflects the degree of business load balance; and provides key inputs for comprehensive evaluation.
[0028] In some embodiments of this application, determining quality control stability characteristic values based on the quality control index data and determining service response characteristic values based on the multidimensional feedback data includes: Based on the stability of the treatment pathway compliance rate sequence, the coefficient of variation of the grade A medical record rate sequence, and the range of the rational drug use score sequence, the quality control stability characteristic value is determined. Service response characteristic values are determined based on the distribution skewness of the patient satisfaction score sequence, the central tendency deviation of the peer review score sequence, and the density of the complaint event marker sequence.
[0029] In some embodiments of this application, quality control stability characteristic values are determined based on the stability of the treatment pathway compliance rate sequence, the coefficient of variation of the grade A medical record rate sequence, and the range of the rational drug use score sequence, including: Calculate the moving standard deviation of the medical pathway compliance rate sequence; if the moving standard deviation is less than a preset fluctuation threshold, assign a stability weighting factor. Extract the quartile coefficients of the grade A rate sequence of the medical records as the coefficients of variation; Obtain the range between the maximum and minimum values in the rational drug use score sequence, and normalize the standard deviation. The product of the stability weighting factor, the reciprocal of the quartile dispersion coefficient and the normalized range is obtained.
[0030] In this embodiment, the moving standard deviation calculation window of the diagnosis and treatment path compliance rate sequence is set to 30 days, and the compliance rate of a medical staff for 90 days is [95%, 92%, 96%, 94%...n6], and the moving standard deviation calculation result is 2.1%. The preset fluctuation threshold is set to 3.0%, and the stability weighting factor is 1.1. The setting of the stability weighting factor follows the principle of reward and punishment incentive and the requirement of differentiated evaluation, avoids excessive punishment due to normal fluctuation, and embodies the fault tolerance of evaluation. The medical record A-level rate sequence is monthly data, for example [0.88, 0.85, 0.90, 0.87], the quartile dispersion coefficient is calculated as the difference between the upper quartile 0.89 and the lower quartile 0.86 divided by the median 0.875, which is 0.034. The rational drug use score sequence is [92, 88, 95, 90, 87], the maximum value is 95, the minimum value is 87, and the range is 8. The sequence standard deviation is calculated to be 3.2, and the normalized range is 8÷3.2=2.5. The product operation result is 80.9. The product operation result is standardized by dividing by 100 to obtain the quality control stability characteristic value 0.81. The moving standard deviation threshold is set according to the hospital quality control standard, the quartile dispersion coefficient reflects the fluctuation degree of the A-level rate, the reciprocal thereof embodies that the smaller the fluctuation is, the larger the characteristic value is, the normalized range eliminates the influence of the score value range, the product operation comprehensively evaluates the stability of the three quality control indicators, and the higher the value is, the more stable the quality control is.
[0031] The beneficial effects of the above technical solution are: the moving standard deviation tracks the compliance rate fluctuation in real time, the weighting factor encourages stable diagnosis and treatment behavior, the quartile dispersion coefficient quantifies the consistency of medical record quality, the reciprocal design strengthens low fluctuation and high score, the range normalization eliminates the influence of absolute score, and the product comprehensively evaluates the stability of quality control, providing reliable basis for quality dimension in evaluation.
[0032] In some embodiments of the present application, according to the distribution skewness of the patient satisfaction score sequence, the central tendency deviation of the peer review score sequence and the density of the complaint event marker sequence, a service response characteristic value is determined, comprising: If the skewness coefficient of the patient satisfaction score sequence is greater than zero, a negative skewness marker is generated, and the absolute value of the skewness corresponding to the negative skewness marker is taken as a first input parameter; The proportion of scores lower than a preset qualified line in the peer review score sequence is calculated as a central tendency deviation; The number of markers in a unit time in the complaint event marker sequence is counted as a complaint density; The first input parameter, the central tendency deviation and the complaint density are positively correlated and normalized to obtain a service response characteristic value.
[0033] In this embodiment, the patient satisfaction score sequence is [9, 8, 9, 7, 6, 5, 8, 9... n7], the statistical skewness coefficient is 0.85>0, the negative skewness label is 1, indicating that there are more low-score patients, and the absolute value of skewness 0.85 is used as the first input parameter. The skewness coefficient calculation method is more complex here, and will not be introduced here. The peer review score sequence is [85, 78, 82, 75, 70, 80], the preset pass line is 75 points, and there are 2 (70, 75) below 75 points, accounting for 2 / 6≈0.333, as the concentration tendency deviation. There are 5 complaint labels in the complaint event label sequence within 90 days, and the unit time density is 5÷90≈0.056 times / day. The positive correlation fusion adopts weighted summation, the skewness weight is 0.4, the deviation weight is 0.3, and the complaint density weight is 0.3. The weighted sum is 0.457. Normalize to the [0, 1] interval, and divide by the historical maximum service response characteristic value.
[0034] The beneficial effects of the above technical solutions are: the skewness quantifies the patient evaluation distribution form, the negative skewness label identifies the satisfaction risk, the deviation reflects the lack of peer recognition, the complaint density directly captures the service defects, the positive correlation fusion strengthens the cumulative effect of negative information, the normalization processing realizes the score comparability, and the service response characteristic value accurately depicts the service quality level. Provide key feedback dimensions for comprehensive evaluation.
[0035] In some embodiments of the application, based on the coupling relationship between the service load balancing characteristic value and the quality control stability characteristic value, and the deviation amplitude of the service response characteristic value and the preset reference threshold, a comprehensive evaluation degree value is generated, including: A two-dimensional coupling matrix of the service load balancing characteristic value and the quality control stability characteristic value is constructed, and the eigenvalue of the two-dimensional coupling matrix is extracted as a coupling strength parameter; Calculate the relative distance between the service response characteristic value and the preset reference threshold. If the relative distance is greater than the preset warning distance, generate a deviation warning label; The coupling strength parameter and the correction weight corresponding to the deviation warning label are weighted and summed to obtain a comprehensive evaluation degree value.
[0036] In this embodiment, the service load balancing characteristic value u = 0.54, the quality control stability characteristic value v = 0.81, and a two-dimensional coupling matrix [[u, 0], [0, v]] is constructed, that is, [[0.54, 0], [0, 0.81]]. The matrix characteristic values are extracted as λ1=0.54, λ2=0.81, the coupling strength parameter is 0.81, which is the larger characteristic value, indicating that quality control stability is dominant. The preset reference threshold is set to 0.8, the service response characteristic value w = 0.762, and the relative distance |0.762-0.8| / 0.8 = 0.0475. The preset alert distance is set to 0.05, 0.0475 < 0.05, no deviation warning mark is generated, and the correction weight is 0. The coupling strength parameter 0.81 and the correction weight 0 are weighted and summed to obtain the comprehensive evaluation degree value 0.81. If the service response characteristic value is 0.70, the relative distance 0.125 > 0.05, a deviation warning mark is generated, the correction weight is set to -0.15 (negative penalty), and the comprehensive evaluation degree value = 0.81 + (-0.15) = 0.66.
[0037] The beneficial effects of the above technical solutions are: the two-dimensional coupling matrix quantifies the internal correlation between service and quality control, the characteristic value extraction identifies the dominant factor, the relative distance calculation evaluates the service deviation, the alert distance setting provides the early warning boundary, the correction weight dynamically adjusts the comprehensive evaluation degree value, and the comprehensive evaluation degree value realizes multi-dimensional performance fusion to provide the core evaluation quantity for evaluation.
[0038] In some embodiments of the present application, a dynamic correction coefficient is constructed according to the identification result of the abnormal response event in the emergency event record data and the transition amplitude of the comprehensive evaluation degree value between adjacent evaluation periods, including: In the emergency event record data, the response timeout event, the improper handling event, and the concealment and omission event are screened, and are respectively given a differentiated abnormal weight and accumulated to obtain an emergency abnormal cumulative value; The absolute value of the difference between the comprehensive evaluation degree values of the current evaluation period and the previous evaluation period is calculated, and if the absolute value of the difference is greater than a preset transition threshold, it is identified as a transition anomaly; The emergency abnormal cumulative value and the determination result of the transition anomaly are positively correlated, and a dynamic correction coefficient is generated.
[0039] In this embodiment, the emergency event record data contains 3 emergency events participated by a medical staff in a quarter. It contains fields such as timestamp, event type, response duration, treatment process record, reporting record, etc. The first emergency response duration is 190 seconds, the standard response duration is 120 seconds, and the first abnormal weight is 0.4. The second emergency treatment did not perform the compression frequency of cardiopulmonary resuscitation according to the standard, there was 1 key node violation, and the second abnormal weight was 0.3. The patient's family complained after the third emergency, and the record showed that it was not reported within 2 hours, which was marked as concealing and missing, and the third abnormal weight was 0.3. The setting of the first abnormal weight, the second abnormal weight and the third abnormal weight follows the medical quality and safety priority principle and the hospital quality management practice. The first abnormal weight (response timeout, weight 0.4) is the highest weight, the emergency response time is the core indicator of the golden life window, and directly determines the prognosis of the patient. Response timeout has the highest weight on patient safety, improper treatment has the second highest weight on medical quality, and concealing has the third highest weight on management transparency. This mechanism accurately quantifies the negative performance in emergency events. Emergency abnormal cumulative value = 0.4 + 0.3 + 0.3 = 1.0. The comprehensive evaluation degree value of the previous evaluation period is 0.85, and the comprehensive evaluation degree value of the current evaluation period is 0.81, and the absolute difference is 0.04. The preset transition threshold is set to 0.05, and 0.04<0.05 is not recognized as a transition anomaly, and the transition anomaly is marked as 0. The positive correlation mapping weights the emergency abnormal cumulative value 1.0 and the transition anomaly 0, and the emergency weight is 0.8, the transition weight is 0.2, and 0.80 is obtained. If the absolute difference is 0.06>0.05, the transition anomaly is marked as 1, and 1.0 is obtained. The dynamic correction coefficient reflects the comprehensive influence of abnormal events and performance fluctuations, and the larger the value, the stronger the correction required.
[0040] The beneficial effects of the above technical solutions are: three types of emergency events accurately capture abnormal performance, differentiated weights reflect the difference in severity, emergency abnormal cumulative value quantifies the short board of emergency capability, transition threshold identifies dramatic changes in performance, positive correlation mapping integrates emergency and transition effects, and dynamic correction coefficient realizes dynamic adjustment of evaluation, providing correction basis for final score.
[0041] In some embodiments of the present application, the comprehensive evaluation degree value and the dynamic correction coefficient are weighted and fused and mapped to a dimensionless score space to obtain a medical staff evaluation comprehensive index, including: If the dynamic correction coefficient is greater than the preset correction threshold, the comprehensive evaluation degree value is inhibited and corrected by using a decreasing type weighting function; If the dynamic correction coefficient is less than or equal to the preset correction threshold, the comprehensive evaluation degree value is gain-corrected by using an increasing type weighting function; The corrected comprehensive evaluation degree value is converted and mapped to a preset dimensionless score space through a standard score to obtain a medical staff evaluation comprehensive index.
[0042] In this embodiment, the preset correction threshold is set to 0.75. The decreasing type weighting function adopts an exponential decay form, the comprehensive evaluation degree value is 0.81, the dynamic correction coefficient is 0.80, and the corrected comprehensive evaluation degree value = 0.81 x e^(-0.3 x 0.80) = 0.638, e^ is a natural exponential function, which is used to realize nonlinear penalty decay. The increasing type weighting function adopts a linear growth form, and the corrected comprehensive evaluation degree value = comprehensive evaluation degree value x (1 + 0.2 x dynamic correction coefficient), which embodies the gain effect. The standard score conversion takes the mean value μ and the standard deviation σ of the comprehensive evaluation degree value of the medical staff in the hospital as the benchmark to calculate the standard score Z = (0.638 - μ) / σ. Assuming that μ = 0.60 and σ = 0.08, then Z = (0.638 - 0.60) / 0.08 = 0.475. The preset dimensionless interval is [0, 100], and the T-score conversion is used to map to [0, 100]: the medical staff evaluation comprehensive index = 50 + 10 x Z, that is, 50 + 10 x 0.475 = 54.75 points. The index 54.75 points indicates that a certain medical staff is in the middle slightly above the level of the whole hospital in the evaluation of this quarter. The decreasing type function punishes objects with poor emergency performance or large performance fluctuations, the increasing type function rewards individuals with stable performance, the standard score conversion realizes the comparability of the whole hospital, and the difference in the score scale is eliminated.
[0043] The above technical scheme has the beneficial effects that: the correction threshold intelligently judges the correction direction, the decreasing type and the increasing type function realize differentiated rewards and punishments, the exponential decay strengthens the punishment effect of the high correction coefficient, the linear gain moderately rewards stable performance, the standard score conversion provides a statistically comparable score, the dimensionless score space unifies the evaluation standard of the whole hospital, the comprehensive index is objectively comparable, and the evaluation fairness is significantly improved.
[0044] In some embodiments of the present application, after the comprehensive evaluation degree value and the dynamic correction coefficient are weighted and fused and mapped to a dimensionless score space to obtain a medical staff evaluation comprehensive index, the following steps are further included: According to the ranking quantile value of the medical staff evaluation comprehensive index in all medical staff evaluation objects, a grade evaluation mark is generated.
[0045] In this embodiment, all evaluation objects are 200 clinical doctors in the hospital. The ranking quantile value of a certain medical staff comprehensive index 54.75 points is calculated as the ranking position divided by the total number of people, for example, ranking 120th, and the quantile value is 120 ÷ 200 = 0.60. The grade evaluation mark is divided: the top 10% (0-0.10) is A level (excellent), the top 10%-30% (0.10-0.30) is B level (good), the top 30%-60% (0.30-0.60) is C level (qualified), and the last 40% (0.60-1.00) is D level (to be improved). Therefore, the certain medical staff is evaluated as C level.
[0046] The beneficial effects of the above technical solutions are: the ranking quantile value realizes the relative ranking of the whole hospital, the level mark objectively positions the performance level, the closed-loop management promotes the ability improvement of medical staff, and the efficiency of hospital human resource management is significantly improved.
[0047] In order to further illustrate the technical idea of the application, the technical solutions of the application will be described in combination with specific application scenarios.
[0048] Correspondingly, as shown in Figure 2 The application also provides a medical staff evaluation dynamic generation system, which comprises: A data acquisition module is configured to acquire service load data, quality control index data, multi-dimensional feedback data and emergency event record data of medical staff in an evaluation period, wherein the service load data comprises a clinic visit sequence, an inpatient management bed day sequence and a surgical operation classification count sequence, the quality control index data comprises a diagnosis and treatment path compliance rate sequence, a medical record A-level rate sequence and a rational drug use score sequence, and the multi-dimensional feedback data comprises a patient satisfaction score sequence, a peer review score sequence and a complaint event marker sequence. A first determination module is configured to determine the fluctuation synchronization of the clinic visit sequence and the inpatient management bed day sequence, determine the cross-class operation proportion in the surgical operation classification count sequence, and comprehensively determine a service load balance characteristic value. A second determination module is configured to determine a quality control stability characteristic value according to the quality control index data, and determine a service response characteristic value according to the multi-dimensional feedback data. A numerical generation module is configured to generate a comprehensive evaluation degree value based on the coupling relationship between the service load balance characteristic value and the quality control stability characteristic value, and the deviation amplitude of the service response characteristic value and a preset reference threshold. A dynamic correction module is configured to construct a dynamic correction coefficient according to the identification result of an abnormal response event in the emergency event record data and the transition amplitude of the comprehensive evaluation degree value between adjacent evaluation periods. A comprehensive evaluation module is configured to perform weighted fusion and mapping of the comprehensive evaluation degree value and the dynamic correction coefficient to a dimensionless score space to obtain a medical staff evaluation comprehensive index.
[0049] In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.
[0050] Although the present application has been described with reference to the above embodiments, various modifications can be made to the application and equivalents thereof without departing from the scope of the application. In particular, features of the disclosed embodiments can be used in any combination without departing from the scope of the application, and the description of the various embodiments does not imply that the combinations of features are not combinable unless the description states that a combination is not possible. The description of the various embodiments is not meant to limit the application but merely to provide examples of the application.
[0051] It is to be understood that the above description is merely a preferred example of the application and is not intended to limit the application, and that modifications can be made by those ordinarily skilled in the art with the content of the above without departing from the scope of the application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application shall be included in the scope of the protection of the application.
Claims
1. A method for dynamically generating performance evaluations for medical personnel, characterized in that, include: The system acquires workload data, quality control indicator data, multidimensional feedback data, and emergency event record data of medical staff during the evaluation period. The workload data includes outpatient visit sequence, inpatient bed-day sequence, and surgical operation grade count sequence. The quality control indicator data includes treatment pathway compliance rate sequence, medical record grade A rate sequence, and rational drug use score sequence. The multidimensional feedback data includes patient satisfaction score sequence, peer review score sequence, and complaint event marking sequence. Determine the synchronicity of fluctuations between the outpatient visitor count sequence and the inpatient bed day sequence, determine the proportion of cross-level operations in the surgical operation grade count sequence, and comprehensively determine the business load balancing characteristic value; The quality control stability characteristic value is determined based on the quality control index data, and the service response characteristic value is determined based on the multidimensional feedback data. Based on the coupling relationship between the business load balancing characteristic value and the quality control stability characteristic value, and the deviation of the service response characteristic value from the preset benchmark threshold, a comprehensive evaluation degree value is generated. Based on the identification results of abnormal response events in the emergency event record data and the jump amplitude of the comprehensive evaluation degree value between adjacent evaluation cycles, a dynamic correction coefficient is constructed. The comprehensive evaluation score is weighted and fused with the dynamic correction coefficient and mapped to a dimensionless scoring space to obtain the comprehensive evaluation index for medical staff.
2. The method for dynamically generating medical personnel performance evaluations according to claim 1, characterized in that, Determine the synchronicity of fluctuations between the outpatient visit frequency sequence and the inpatient bed-day sequence, determine the proportion of cross-level operations in the surgical procedure grading count sequence, and comprehensively determine the workload balancing characteristic value, including: Extract the periodic fluctuation phase difference between the outpatient visit sequence and the inpatient bed day sequence respectively. If the phase difference is less than the preset synchronization tolerance, generate a load coordination marker. The percentage of surgeries exceeding the scope of practice in the surgical operation grading sequence is calculated as the percentage of cross-level surgeries. The difference between the sequence overlap degree corresponding to the load coordination label and the cross-level operation ratio is subjected to negative correlation normalization to obtain the service load balancing feature value.
3. The method for dynamically generating medical personnel performance evaluations according to claim 1, characterized in that, Based on the quality control index data, quality control stability characteristic values are determined, and based on the multidimensional feedback data, service response characteristic values are determined, including: Based on the stability of the treatment pathway compliance rate sequence, the coefficient of variation of the grade A medical record rate sequence, and the range of the rational drug use score sequence, the quality control stability characteristic value is determined. Service response characteristic values are determined based on the distribution skewness of the patient satisfaction score sequence, the central tendency deviation of the peer review score sequence, and the density of the complaint event marker sequence.
4. The method for dynamically generating medical personnel performance evaluations according to claim 3, characterized in that, Based on the stability of the treatment pathway compliance rate sequence, the coefficient of variation of the grade A medical record rate sequence, and the range of the rational drug use score sequence, quality control stability characteristic values are determined, including: Calculate the moving standard deviation of the medical pathway compliance rate sequence; if the moving standard deviation is less than a preset fluctuation threshold, assign a stability weighting factor. Extract the quartile coefficients of the grade A rate sequence of the medical records as the coefficients of variation; Obtain the range between the maximum and minimum values in the rational drug use score sequence, and normalize the standard deviation. The stability weighting factor, the reciprocal of the quartile dispersion coefficient, and the normalized range are multiplied to obtain the quality control stability characteristic value.
5. The method for dynamically generating medical personnel performance evaluations according to claim 3, characterized in that, Based on the distribution skewness of the patient satisfaction rating sequence, the central tendency deviation of the peer review score sequence, and the density of the complaint event marker sequence, service response characteristic values are determined, including: If the skewness coefficient of the patient satisfaction score sequence is greater than zero, a negative skewness label is generated, and the absolute value of the skewness corresponding to the negative skewness label is used as the first input parameter. The percentage of scores below a preset passing line in the peer review score sequence is calculated as the central tendency deviation. The number of markers per unit time in the complaint event marker sequence is counted as the complaint density; The first input parameter, the central tendency deviation, and the complaint density are positively correlated and fused and normalized to obtain the service response feature value.
6. The method for dynamically generating medical personnel performance evaluations according to claim 1, characterized in that, Based on the coupling relationship between the load balancing characteristic value and the quality control stability characteristic value, and the deviation of the service response characteristic value from the preset benchmark threshold, a comprehensive evaluation level value is generated, including: Construct a two-dimensional coupling matrix between the business load balancing characteristic value and the quality control stability characteristic value, and extract the characteristic value of the two-dimensional coupling matrix as the coupling strength parameter; Calculate the relative distance between the service response feature value and a preset benchmark threshold. If the relative distance is greater than a preset warning distance, generate a deviation warning flag. The coupling strength parameter and the correction weight corresponding to the deviation warning mark are weighted and summed to obtain the comprehensive evaluation degree value.
7. The method for dynamically generating medical personnel performance evaluations according to claim 1, characterized in that, Based on the identification results of abnormal response events in the emergency event record data and the jump amplitude of the comprehensive evaluation level value between adjacent evaluation cycles, a dynamic correction coefficient is constructed, including: Filter out response timeout events, improper handling events, and concealment / failure events from the emergency event record data, assign differentiated anomaly weights to each, and sum them up to obtain the cumulative value of emergency anomalies; Calculate the absolute value of the difference between the comprehensive evaluation level value of the current evaluation cycle and the previous evaluation cycle. If the absolute value of the difference is greater than the preset transition threshold, it is identified as a transition anomaly. The cumulative value of the emergency anomaly is positively correlated with the judgment result of the transition anomaly to generate a dynamic correction coefficient.
8. The method for dynamically generating medical personnel performance evaluations according to claim 1, characterized in that, The comprehensive evaluation score is weighted and fused with the dynamic correction coefficient and mapped to a dimensionless scoring space to obtain the comprehensive evaluation index for medical staff, including: If the dynamic correction coefficient is greater than the preset correction threshold, a decreasing weighting function is used to suppress the correction of the comprehensive evaluation value. If the dynamic correction coefficient is less than or equal to the preset correction threshold, an incremental weighting function is used to perform a gain correction on the comprehensive evaluation value. The revised comprehensive assessment score is mapped to a pre-defined dimensionless scoring space through standard score conversion to obtain the comprehensive evaluation index for medical staff.
9. The method for dynamically generating medical personnel performance evaluations according to claim 1, characterized in that, After weighting and fusing the comprehensive assessment score with the dynamic correction coefficient and mapping it to a dimensionless scoring space to obtain the comprehensive index for medical staff evaluation, the following steps are also included: A grade rating label is generated based on the ranking percentile of the comprehensive evaluation index of medical personnel among all medical personnel evaluation subjects.
10. A dynamic evaluation generation system for medical personnel, applied to the dynamic evaluation generation method for medical personnel as described in any one of claims 1-9, characterized in that, include: The data acquisition module is used to acquire the workload data, quality control indicator data, multidimensional feedback data, and emergency event record data of medical staff during the evaluation period. The workload data includes outpatient visit sequence, inpatient bed-day sequence, and surgical operation grade count sequence. The quality control indicator data includes treatment pathway compliance rate sequence, medical record grade A rate sequence, and rational drug use score sequence. The multidimensional feedback data includes patient satisfaction score sequence, peer review score sequence, and complaint event marking sequence. The first determining module is used to determine the fluctuation synchronicity between the outpatient visitor count sequence and the inpatient bed day sequence, determine the proportion of cross-level operations in the surgical operation grade counting sequence, and comprehensively determine the business load balancing characteristic value. The second determining module is used to determine the quality control stability characteristic value based on the quality control index data and to determine the service response characteristic value based on the multidimensional feedback data. The numerical generation module is used to generate a comprehensive evaluation value based on the coupling relationship between the business load balancing characteristic value and the quality control stability characteristic value, and the deviation of the service response characteristic value from the preset benchmark threshold. The dynamic correction module is used to construct a dynamic correction coefficient based on the identification results of abnormal response events in the emergency event record data and the jump amplitude of the comprehensive evaluation degree value between adjacent evaluation cycles. The comprehensive evaluation module is used to weight and fuse the comprehensive evaluation level value with the dynamic correction coefficient and map it to the dimensionless scoring space to obtain the comprehensive evaluation index of medical staff.