RISK ASSESSMENT SYSTEM, RISK ASSESSMENT METHOD, AND COMPUTER PROGRAM

The risk evaluation system analyzes incident reports to derive a comprehensive risk score, addressing the lack of effective risk comparison in medical institutions, thereby enhancing risk assessment and management.

JP7595882B2Active Publication Date: 2024-12-09NAT UNIV CORP TOKAI NAT HIGHER EDUCATION & RES SYST +1
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
JP2021553591
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2019-10-30
Filing Date
2020-10-26
Publication Date
2024-12-09
Estimated Expiration
2040-10-26

AI Technical Summary

Technical Problem

Existing systems fail to accurately compare and evaluate the risk magnitude faced by multiple medical institutions and their departments, limiting the ability to assess and prevent medical accidents and errors effectively.

Method used

A risk evaluation system and method that analyzes incident reports using fault, severity, and impact term scores to derive a comprehensive risk score, enabling quantitative assessment and comparison of risks across different organizational units.

Benefits of technology

Enables accurate evaluation and visualization of risk magnitude, facilitating comparisons between medical institutions and departments, and supporting proactive risk management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007595882000001
    Figure 0007595882000001
  • Figure 0007595882000002
    Figure 0007595882000002
  • Figure 0007595882000003
    Figure 0007595882000003
Patent Text Reader

Abstract

A negligence term score storage unit 22 stores, for each of a plurality of words, a negligence term score, which is an index representing a tendency for each word to appear in a report indicating that an event has occurred as a result of negligence in an organization. A severe term score storage unit 24 stores, for each of the plurality of words, a severe term score, which is an index representing a tendency for each word to appear in a report indicating that a serious event has occurred. A score deriving unit 42 extracts a plurality of words appearing in a report being analyzed, and derives a risk score, which is an index representing the magnitude of risk for the organization, on the basis of the negligence term score for each extracted word, the serious term score for each extracted word, and a predetermined risk evaluation weighting for each of the negligence of the organization and the seriousness of the event.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to data processing technology, and in particular to a risk assessment system, a risk assessment method, and a computer program. [Background technology]

[0002] In medical institutions, reports called incident reports are prepared. An incident here refers to a medical accident, medical malpractice, or an event that may lead to them in the course of daily medical practice. Incident reports are used to analyze incidents and prevent the recurrence of similar incidents, as well as to prevent the occurrence of medical accidents and medical malpractice. The applicant of the present invention has proposed a technology for accurately classifying incident reports in the following Patent Document 1. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2018-081334 A Summary of the Invention [Problem to be solved by the invention]

[0004] Until now, the magnitude of risk held by multiple medical institutions has not been compared sufficiently among medical institutions. In addition, the magnitude of risk held by multiple departments within a medical institution has not been compared sufficiently among departments. The present inventor thought that it would be possible to accurately estimate the magnitude of risk held by an organization by analyzing incident reports.

[0005] The present disclosure has been made based on the above-mentioned idea of ​​the inventor, and one objective is to provide a technology for accurately evaluating the magnitude of risk to an organization based on reports of events that have occurred in the organization. [Means for solving the problem]

[0006] In order to solve the above problems, a risk assessment system of one embodiment of the present disclosure is a system that analyzes a report that reports on an event that has occurred in an organization and assesses the magnitude of risk to the organization, and is equipped with a first memory unit that stores a negligence term score, which is an index representing the likelihood of each of a plurality of words to appear in a report indicating that the event occurred due to the negligence of the organization, a second memory unit that stores a serious term score, which is an index representing the likelihood of each of the plurality of words to appear in a report indicating that a serious event has occurred, and a score derivation unit that extracts a plurality of words written in the report to be analyzed, and derives a risk score, which is an index representing the magnitude of risk to the organization, based on the negligence term score of each extracted word, the serious term score of each extracted word, and predetermined weights in risk assessment for each of the organization's negligence and the seriousness of the event.

[0007] Another aspect of the present disclosure is a risk assessment method, in which a computer that can access a first storage unit and a second storage unit analyzes a report reporting an event that has occurred in an organization to assess the magnitude of the organization's risk, the method comprising: a step of extracting a plurality of words described in the report to be analyzed, the step of storing a negligence term score for each of a plurality of words being an index representing the likelihood of the word appearing in a report indicating that an event has occurred due to the organization's negligence, the step of storing a serious term score for each of the plurality of words being an index representing the likelihood of the word appearing in a report indicating that a serious event has occurred, and a step of deriving a risk score, which is an index representing the magnitude of the organization's risk, based on the negligence term score for each of the extracted words, the serious term score for each of the extracted words, and a weight in risk assessment that is predetermined for each of the organization's negligence and the seriousness of the event.

[0008] Any combination of the above components, and conversion of the present disclosure into an apparatus, a computer program, a recording medium storing a computer program, and the like, are also valid aspects of the present disclosure. Effect of the Invention

[0009] According to the present disclosure, the magnitude of risk to an organization can be accurately evaluated based on reports that report events that have occurred in the organization. [Brief description of the drawings]

[0010] [Figure 1] FIG. 13 is a diagram illustrating a method for determining a negligence term score. [Diagram 2] FIG. 1 is a schematic diagram showing a method for determining a severe term score. [Diagram 3] FIG. 2 is a diagram illustrating a method for determining an impact term score. [Figure 4] FIG. 2 is a block diagram showing functional blocks of the risk assessment device according to the first embodiment. [Diagram 5] 13 is a flowchart showing the flow of a negligence score calculation process. [Figure 6] FIG. 13 is a diagram showing the weights of risk assessment factors in a hospital. [Figure 7] FIG. 13 is a diagram showing possible methods for calculating a risk score. [Figure 8] FIG. 13 is a diagram showing the correlation between ranking evaluation by GRM and ranking evaluation by each calculation method. [Figure 9] FIG. 11 is a diagram illustrating an example of a risk evaluation result in the first embodiment. [Figure 10] FIG. 11 is a diagram illustrating an example of a risk evaluation result in the first embodiment. [Figure 11] FIG. 1 is a scatter plot showing the relationship between reporting volume and risk score. [Figure 12] FIG. 13 is a diagram showing a specific example of risk deviation derivation. [Figure 13] FIG. 11 is a diagram illustrating an example of a risk evaluation result according to the second embodiment. [Figure 14] FIG. 11 is a diagram illustrating an example of a risk evaluation result according to the second embodiment. [Figure 15] FIG. 11 is a diagram illustrating an example of a risk evaluation result according to the second embodiment. [Figure 16]FIG. 11 is a diagram illustrating an example of a risk evaluation result according to the second embodiment. [Figure 17] FIG. 11 is a diagram illustrating an example of a risk evaluation result according to the second embodiment. [Figure 18] FIG. 11 is a diagram illustrating an example of a risk evaluation result according to the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] <First Example> In the first embodiment, a technology is proposed for accurately evaluating the magnitude of risk held by a medical institution based on incident reports created by the medical institution. This technology makes it possible to visualize the magnitude of risk held by organizations of various granularities, such as medical institutions and departments within medical institutions, and also makes it possible to compare risks between organizations. In the following description, the medical institution is assumed to be a hospital.

[0012] In the first embodiment, a negligence term score, a seriousness term score, and an impact term score are used as indices corresponding to the "characteristic degree" described in the above Patent Document 1. Fig. 1 shows a schematic diagram of a method for determining a negligence term score. The negligence term score is an index indicating the likelihood of each of a plurality of words appearing in an incident report indicating that an incident occurred due to negligence of a hospital (in other words, a medical worker).

[0013] Figure 2 shows a schematic diagram of a method for determining the severe term score. The severe term score is an index that indicates the likelihood of each of a plurality of words appearing in an incident report indicating that a severe situation, i.e., a serious event, has occurred to a patient, specifically, that the patient has become seriously ill (that a severe situation has occurred to the patient due to an incident).

[0014] Figure 3 shows a schematic diagram of how the impact term score is determined. The impact term score is an index that indicates the likelihood of a word appearing in an incident report that is judged by a human to contain important information, or in other words, that a word requires careful consideration or some kind of action.

[0015] Fig. 4 is a block diagram showing the functional blocks of the risk assessment device of the first embodiment. Each block shown in the block diagram in this specification can be realized in terms of hardware by elements such as a computer processor, CPU, and memory, electronic circuits, and mechanical devices, and in terms of software by a computer program, etc., but here, the functional blocks realized by the cooperation of these are depicted. Therefore, it will be understood by those skilled in the art that these functional blocks can be realized in various forms by combining hardware and software.

[0016] The risk assessment device 10 is an information processing device that analyzes incident reports and evaluates the magnitude of risk held by a hospital or each department within a hospital. The risk assessment device 10 includes a control unit 12 and a memory unit 14. The control unit 12 executes various data processing for risk assessment. The memory unit 14 stores data referenced or updated by the control unit 12. The memory unit 14 includes a classification information memory unit 20, a negligence term score memory unit 22, a seriousness term score memory unit 24, and an impact term score memory unit 26.

[0017] The classification information storage unit 20 stores classification information indicating whether each of a plurality of incident reports created by a hospital corresponds to an incident report indicating that an incident occurred due to negligence of the hospital (in other words, a medical professional). A set of corresponding incident reports is called a "negligence report group," and a set of non-corresponding incident reports is called a "non-negligence report group." As shown in FIG. 1, a hospital GRM (general risk manager, medical safety manager) checks the contents of a plurality of incident reports and judges whether each incident report corresponds to a negligence report group or a non-negligence report group. Classification information indicating this judgment result is stored in the classification information storage unit 20. For example, the classification information storage unit 20 may store identification information (report number, etc.) of one or more incident reports corresponding to a negligence report group and identification information of one or more incident reports corresponding to a non-negligence report group.

[0018] The classification information storage unit 20 also stores classification information indicating whether each of a plurality of incident reports created in the hospital corresponds to an incident report indicating that a patient has become seriously ill. A set of corresponding incident reports is called a "severe report group," and a set of non-corresponding incident reports is called a "non-severe report group." As shown in FIG. 2, the GRM of the hospital checks the contents of the plurality of incident reports and determines whether each incident report corresponds to the severe report group or the non-severe report group. Classification information indicating this determination result is stored in the classification information storage unit 20. For example, the classification information storage unit 20 may store identification information of one or more incident reports corresponding to the severe report group and identification information of one or more incident reports corresponding to the non-severe report group.

[0019] The classification information storage unit 20 also stores classification information indicating whether each of the multiple incident reports created in the hospital corresponds to an incident report that a human has determined to contain important content. A set of corresponding incident reports is called an "impact report group," and a set of non-corresponding incident reports is called a "non-impact report group." As shown in FIG. 3, the GRM of the hospital checks the contents of the multiple incident reports and determines whether each incident report corresponds to the impact report group or the non-impact report group. Classification information indicating this determination result is stored in the classification information storage unit 20. For example, the classification information storage unit 20 may store identification information of one or more incident reports that correspond to the impact report group and identification information of one or more incident reports that correspond to the non-impact report group.

[0020] In the embodiment, eight GRMs were asked to vote on whether each incident report belonged to the impact report group. Then, incident reports that received three or more votes in favor of belonging to the impact report group were classified into the impact report group, and incident reports that received less than three votes were classified into the non-impact report group.

[0021] Returning to Fig. 4, the negligence term score storage unit 22 stores the negligence term score of each of the multiple words extracted from the incident report. The serious term score storage unit 24 stores the serious term score of each of the multiple words extracted from the incident report. The impact term score storage unit 26 stores the impact term score of each of the multiple words extracted from the incident report.

[0022] The control unit 12 includes a report reading unit 30, a tissue information reading unit 32, a term score determination unit 34, a score derivation unit 42, an output unit 52, and an image generation unit 54. A computer program including a plurality of modules corresponding to these plurality of functional blocks may be stored in a recording medium and installed in the storage of the risk assessment device 10 via the recording medium. Alternatively, this computer program may be installed in the storage of the risk assessment device 10 via a network. The CPU of the risk assessment device 10 may fulfill the functions of the above-mentioned plurality of functional blocks by reading this computer program into the main memory and executing it.

[0023] The report reading unit 30 reads incident report data (e.g., text data) from an electronic file in which the incident report is recorded. Data items of the incident report include the report number, creator information (including occupation such as doctor, nurse, etc.), occurrence date (date on which the incident occurred), department name (name of the department where the incident occurred), and free text. The risk assessment device 10 scores the magnitude of risk based on the free text description of the incident report.

[0024] The organizational information reading unit 32 reads organizational information from an electronic file in which information about each department of a hospital (also referred to as "organizational information") is recorded. The organizational information includes the name (division name) of each of multiple departments in a hospital and the number of employees. The organizational information may include, for each of multiple hospitals, the names of multiple departments established in each hospital and the number of employees.

[0025] The term score determination unit 34 includes an erroneous term score determination unit 36 ​​that determines an erroneous term score for each of the plurality of words, a severe term score determination unit 38 that determines a severe term score for each of the plurality of words, and an impact term score determination unit 40 that determines an impact term score for each of the plurality of words.

[0026] A method for determining a negligence term score will be described with reference to Fig. 1. The negligence term score determination unit 36 ​​extracts a plurality of words (15966 words in the example of Fig. 1) contained in a plurality of incident reports read by the report reading unit 30. For example, the negligence term score determination unit 36 ​​may extract nouns (or nouns and verbs) from sentences written in free text fields of a plurality of incident reports using a publicly known technique such as morphological analysis. In addition, the negligence term score determination unit 36 ​​refers to the classification information stored in the classification information storage unit 20 to identify whether each of the plurality of incident reports read by the report reading unit 30 corresponds to a negligence report group or a non-negligence report group.

[0027] Here, the word for which the negligence term score is calculated is called the target word. The negligence term score determination unit 36 ​​specifies the number A of reports in the negligence report group that include the target word in the free text field, the number B of reports in the negligence report group that do not include the target word in the free text field, the number C of reports in the non-negligence report group that include the target word in the free text field, and the number D of reports in the non-negligence report group that do not include the target word in the free text field. The negligence term score determination unit 36 ​​calculates the ratio of the occurrence rate of the target word in the negligence report group and the non-negligence report group as the negligence term score of the target word according to formula 1 or formula 2 in Fig. 1.

[0028] The negligence term score determination unit 36 ​​executes the above process for each of the extracted words as a target word in sequence, and calculates the negligence term score of each word. The negligence term score determination unit 36 ​​stores the words and the negligence term scores of each word in the negligence term score storage unit 22 in association with each other.

[0029] As shown in Equation 1 and Equation 2, the negligence term score is a value less than -1 or greater than or equal to 1. The larger the negligence term score of a word, the more likely the word is to appear in the negligence report group than in the non-negligence report group. In the example of Figure 1, the word "same surname" appears about 45 times more frequently in the negligence report group than in the non-negligence report group. Also, the smaller the negligence term score of a word, the more likely the word is to appear in the non-negligence report group than in the negligence report group.

[0030] A method for determining the severe term score will be described with reference to Fig. 2. Similar to the negligence term score determination unit 36, the severe term score determination unit 38 extracts a plurality of words (15966 words in the example of Fig. 2) contained in the plurality of incident reports read by the report reading unit 30. Furthermore, the severe term score determination unit 38 refers to the classification information stored in the classification information storage unit 20 and identifies whether each of the plurality of incident reports read by the report reading unit 30 corresponds to a severe report group or a non-severe report group.

[0031] Here, a word for which a severe term score is calculated is called a target word. The severe term score determination unit 38 specifies the number A of reports in the severe report group that include the target word in the free text field, the number B of reports in the severe report group that do not include the target word in the free text field, the number C of reports in the non-severe report group that include the target word in the free text field, and the number D of reports in the non-severe report group that do not include the target word in the free text field. The severe term score determination unit 38 calculates the ratio of the occurrence rate of the target word in the severe report group and the non-severe report group as the severe term score of the target word according to formula 3 or formula 4 in Fig. 2.

[0032] The severe term score determination unit 38 executes the above process for each of the extracted words as a target word in sequence, and calculates the severe term score for each word. The severe term score determination unit 38 stores the severe term scores of each word in the severe term score storage unit 24 in association with the words.

[0033] Similar to the negligence term score, the severe term score can be a number less than -1 or greater than or equal to 1. The higher the severe term score of a word, the more likely the word is to appear in the severe report group than in the non-severe report group. In the example of Figure 2, the word "cardiac arrest" appears about 74 times more frequently in the severe report group than in the non-severe report group. Also, the lower the severe term score of a word, the more likely the word is to appear in the non-severe report group than in the severe report group.

[0034] A method for determining an impact term score will be described with reference to Fig. 3. Similar to the negligence term score determination unit 36, the impact term score determination unit 40 extracts a plurality of words (14,000 words in the example of Fig. 3) contained in the plurality of incident reports read by the report reading unit 30. Furthermore, the impact term score determination unit 40 refers to the classification information stored in the classification information storage unit 20 and identifies whether each of the plurality of incident reports read by the report reading unit 30 corresponds to an impact report group or a non-impact report group.

[0035] Here, a word for which an impact term score is calculated is called a target word. The impact term score determination unit 40 specifies the number A of reports in the impact report group that include the target word in the free-text field, the number B of reports in the impact report group that do not include the target word in the free-text field, the number C of reports in the non-impact report group that include the target word in the free-text field, and the number D of reports in the non-impact report group that do not include the target word in the free-text field. The impact term score determination unit 40 calculates the ratio of the occurrence rate of the target word in the impact report group and the non-impact report group as the impact term score of the target word according to formula 5 or formula 6 in FIG. 3.

[0036] The impact term score determination unit 40 executes the above process for each of the extracted words as a target word in turn, and calculates the impact term score of each word. The impact term score determination unit 40 stores the multiple words and their impact term scores in the impact term score storage unit 26 in association with each other.

[0037] Like the negligence term score, the impact term score can be a number less than -1 or greater than or equal to 1. The higher the impact term score of a word, the more likely the word is to appear in impact reports than in non-impact reports. In the example of Figure 3, the word "furious" appears about 99 times more frequently in impact reports than in non-impact reports. Also, the lower the impact term score of a word, the more likely the word is to appear in non-impact reports than in impact reports.

[0038] The score derivation unit 42 derives various scores that quantitatively indicate the magnitude of risk held by the hospital and each department within the hospital. The score derivation unit 42 includes a negligence score derivation unit 44, a seriousness score derivation unit 46, an impact score derivation unit 48, and a risk score derivation unit 50.

[0039] The negligence score derivation unit 44 derives a negligence score, which is an index showing the magnitude of risk from the viewpoint of hospital negligence (in other words, medical staff negligence), based on the negligence term score stored in the negligence term score storage unit 22. FIG. 5 is a flowchart showing the flow of the negligence score calculation process. The negligence score derivation unit 44 extracts a plurality of words described in the incident report (hereinafter also referred to as the "analysis target report") of the analysis target read by the report reading unit 30 (S10). The negligence score derivation unit 44 may extract a plurality of words included in the free text field of the analysis target report using a publicly known technique such as morphological analysis. The negligence score derivation unit 44 counts the number of occurrences of each word in the analysis target report ("word count" in FIG. 5).

[0040] The negligence score derivation unit 44 calculates the product of the number of occurrences ("word count" in FIG. 5) and the negligence term score ("word-specific aggregate value" in FIG. 5) for each extracted word. The negligence score derivation unit 44 derives the negligence score as the result of division (quotient) using the sum of the word-specific aggregate values ​​of each extracted word ("ΣNR" in FIG. 5) as the dividend and the sum of the number of words of each extracted word ("ΣN" in FIG. 5) as the divisor (S12).

[0041] When multiple analysis target reports are read by the report reading unit 30, the negligence score derivation unit 44 repeats the processes of S10 and S12 for each analysis target report, and derives a negligence score for each analysis target report (report unit). The negligence score derivation unit 44 counts the negligence scores of the multiple analysis target reports at various granularities (S14).

[0042] For example, the negligence score derivation unit 44 derives the negligence score for each department (department unit) by obtaining statistics based on the negligence scores of the analysis target reports of each department based on the department names included in each of the multiple analysis target reports. In other words, the negligence score derivation unit 44 derives the negligence score for each of the multiple departments based on multiple incident reports reporting incidents that occurred in each of the multiple departments. In addition, the negligence score derivation unit 44 derives the negligence score for the entire hospital (hospital unit) by obtaining statistics based on the negligence scores of multiple analysis target reports (e.g., all analysis target reports) created in a certain hospital. In other words, the negligence score derivation unit 44 derives the negligence score for each of the multiple hospitals based on multiple incident reports reporting incidents that occurred in each of the multiple hospitals.

[0043] The negligence score derivation unit 44 also derives negligence scores for each period (e.g., by year and month) by calculating statistics based on the negligence scores of the analysis target reports corresponding to each of a plurality of periods (by year or month) based on the occurrence dates included in each of the plurality of analysis target reports. The negligence score derivation unit 44 also derives negligence scores by year and month for each of a plurality of departments based on the department names and occurrence dates included in each of the plurality of analysis target reports.

[0044] The above statistics may be any one of the average value (arithmetic mean value or moving average value), median value, and mode value of the negligence scores of the multiple analysis target reports, or any combination thereof. The above statistics may further include the minimum value, maximum value, first quartile point, and third quartile point of the negligence scores of the multiple analysis target reports. The same applies to the statistics of the severity score, impact score, and risk score described below.

[0045] The seriousness score derivation unit 46 derives a seriousness score, which is an index representing the magnitude of risk from the viewpoint of the severity of the patient (in other words, the seriousness of the incident), based on the seriousness term scores stored in the seriousness term score storage unit 24. The method of deriving the seriousness score is similar to the method of deriving the negligence score shown in FIG.

[0046] That is, the severity score derivation unit 46 extracts a plurality of words written in the analysis target report read by the report reading unit 30, and calculates a word-specific aggregate value based on the severe term score for each extracted word (replacing the negligence term score in FIG. 5 with the severe term score). The severity score derivation unit 46 derives the result of division (quotient) using the sum of the word-specific aggregate values ​​of each extracted word ("ΣNR" in FIG. 5) as the dividend and the sum of the number of words of each extracted word ("ΣN" in FIG. 5) as the divisor as the severity score.

[0047] The severity score derivation unit 46 derives a severity score for each analysis target report (report unit) when multiple analysis target reports are read by the report reading unit 30. The severity score derivation unit 46 tallies the severity scores of the multiple analysis target reports at various granularities.

[0048] For example, the severity score derivation unit 46 derives a severity score for each department (department unit) by calculating statistics based on the severity scores of the analysis target reports of each department based on the department names included in each of the multiple analysis target reports. In other words, the severity score derivation unit 46 derives a severity score for each of the multiple departments based on multiple incident reports reporting incidents that occurred in each of the multiple departments. In addition, the severity score derivation unit 46 derives a severity score for the entire hospital (hospital unit) by calculating statistics based on the severity scores of multiple analysis target reports (e.g., all analysis target reports) created in a certain hospital. In other words, the severity score derivation unit 46 derives a severity score for each of the multiple hospitals based on multiple incident reports reporting incidents that occurred in each of the multiple hospitals.

[0049] The severity score derivation unit 46 also derives a severity score for each period (e.g., by year and month) by calculating statistics based on the severity scores of the analysis target reports corresponding to each of a plurality of periods (by year or month) based on the occurrence date included in each of the plurality of analysis target reports. The severity score derivation unit 46 also derives a severity score for each of a plurality of departments based on the department name and occurrence date included in each of the plurality of analysis target reports.

[0050] The impact score derivation unit 48 derives an impact score, which is an index representing the magnitude of risk from the viewpoint of a person determining that an incident report contains important content, based on the impact term scores stored in the impact term score storage unit 26. The method of deriving the impact score is similar to the method of deriving the negligence score shown in FIG.

[0051] That is, the impact score derivation unit 48 extracts a plurality of words written in the analysis target report read by the report reading unit 30, and calculates a word-specific aggregate value based on the impact term score for each extracted word (the negligence term score in FIG. 5 is replaced with the impact term score). The impact score derivation unit 48 derives the impact score as the result of division (quotient) using the sum of the word-specific aggregate values ​​of each extracted word ("ΣNR" in FIG. 5) as the dividend and the sum of the number of words of each extracted word ("ΣN" in FIG. 5) as the divisor.

[0052] The impact score derivation unit 48 derives an impact score for each analysis target report (report unit) when multiple analysis target reports are read by the report reading unit 30. The impact score derivation unit 48 counts the impact scores of the multiple analysis target reports at various granularities.

[0053] For example, the impact score derivation unit 48 derives an impact score for each department (department unit) by calculating statistics based on the impact scores of the analysis target reports of each department based on the department names included in each of the multiple analysis target reports. In other words, the impact score derivation unit 48 derives an impact score for each of the multiple departments based on multiple incident reports reporting incidents that occurred in each of the multiple departments. In addition, the impact score derivation unit 48 derives an impact score for the entire hospital (hospital unit) by calculating statistics based on the impact scores of multiple analysis target reports (e.g., all analysis target reports) created in a certain hospital. In other words, the impact score derivation unit 48 derives an impact score for each of the multiple hospitals based on multiple incident reports reporting incidents that occurred in each of the multiple hospitals.

[0054] The impact score derivation unit 48 also derives an impact score for each period (e.g., by year and month) by calculating statistics based on the impact scores of the analysis target reports corresponding to each of multiple periods (by year or month) based on the occurrence date included in each of the multiple analysis target reports. The impact score derivation unit 48 also derives an impact score for each of multiple departments based on the department name and occurrence date included in each of the multiple analysis target reports.

[0055] In this embodiment, a "risk score" is introduced as an index that comprehensively represents the magnitude of risk held by a hospital or each department within the hospital. First, the risk score will be described.

[0056] Figure 6 shows the weights of risk assessment factors in a hospital. The figure shows the results (average values) of weighting by seven evaluators (doctors and nurses including GRM) for multiple risk assessment factors in a hospital, obtained by AHP (Analytic Hierarchy Process) analysis. Among multiple risk assessment factors, negligence and seriousness were weighted heavily, and the ratio of the weights of negligence and seriousness was approximately 4:3. Based on this result, the inventor selected the negligence score and seriousness score as parameter candidates for deriving the risk score.

[0057] On the other hand, there are various possible combinations of negligence scores and seriousness scores. Figure 7 shows possible methods for calculating risk scores. The figure shows calculation methods (1) to (10). The weighting by AHP analysis in calculation methods (9) and (10) involves adding up the negligence scores and seriousness scores in the ratio shown in Figure 6 (approximately 4:3).

[0058] FIG. 8 shows the correlation between the ranking evaluation by GRM and the ranking evaluation by each calculation method. The inventor had seven evaluators (doctors and nurses including GRM) evaluate the ranking of multiple types of incidents in the hospital. As shown in FIG. 8, there are 12 types of incidents: "unexpected death," "operating room related," ..., and "administrative / procedural." The GRM row in FIG. 8 shows the ranking results (average value) by the seven evaluators. Also, rows (1) to (10) in FIG. 8 show the ranking results for each risk score calculation method shown in FIG. 7. As a result of the correlation analysis, it was found that the calculation method (9) in FIG. 7 has the highest correlation with the judgment of GRM. Therefore, in the embodiment, the risk score is derived using the calculation method (9) in FIG. 7.

[0059] Returning to Fig. 4, the risk score derivation unit 50 derives a risk score based on the negligence term score of each word extracted from the analysis target report (in the embodiment, the negligence score which is a statistic thereof), the serious term score of each word extracted from the analysis target report (in the embodiment, the seriousness score which is a statistic thereof), and the weights in the risk assessment determined in advance for the negligence of the hospital and the severity of the patient (in the embodiment, the negligence score: seriousness score = 4.106: 3.245). Specifically, the risk score derivation unit 50 derives the total value of the product of the negligence score and the negligence weight (4.106) derived by the negligence score derivation unit 44 and the product of the seriousness score and the seriousness weight (3.245) derived by the seriousness score derivation unit 46 as the risk score.

[0060] The risk score derivation unit 50 derives a risk score for each analysis target report (report unit) when multiple analysis target reports are read by the report reading unit 30. The risk score derivation unit 50 tallies the risk scores of the multiple analysis target reports at various granularities.

[0061] For example, the risk score derivation unit 50 derives a risk score for each department (department unit) by calculating statistics based on the risk scores of the analysis target reports of each department based on the department names included in each of the multiple analysis target reports. In other words, the risk score derivation unit 50 derives a risk score for each of the multiple departments based on multiple incident reports reporting incidents that occurred in each of the multiple departments. In addition, the risk score derivation unit 50 derives a risk score for the entire hospital (hospital unit) by calculating statistics based on the risk scores of multiple analysis target reports (e.g., all analysis target reports) created in a certain hospital. In other words, the risk score derivation unit 50 derives a risk score for each of the multiple hospitals based on multiple incident reports reporting incidents that occurred in each of the multiple hospitals.

[0062] Furthermore, the risk score derivation unit 50 derives a risk score for each period (e.g., by year and month) by calculating statistics based on the risk scores of the analysis target reports corresponding to each of a plurality of periods (by year or month) based on the occurrence dates included in each of the plurality of analysis target reports. For example, the risk score derivation unit 50 may derive a risk score for each of a plurality of departments by year and month based on the department names and occurrence dates included in each of the plurality of analysis target reports.

[0063] The output unit 52 outputs the multiple types of scores derived by the score derivation unit 42 to a predetermined device or medium. For example, the output unit 52 may store in a predetermined storage device a CSV (comma-separated values) file in which the negligence score, seriousness score, risk score, and impact score of each of the multiple hospitals are recorded. The output unit 52 may also store in a predetermined storage device a CSV file in which the negligence score, seriousness score, risk score, and impact score of each hospital are recorded by year and month (for example, January 2012, February 2012, etc.).

[0064] The output unit 52 may store in a predetermined storage device a CSV file that records the negligence score, seriousness score, risk score, and impact score of each department for multiple departments in the hospital. The output unit 52 may store in a predetermined storage device a CSV file that indicates the negligence score, seriousness score, risk score, and impact score by year and month for each department. The output unit 52 may display the multiple types of scores derived by the score derivation unit 42 on a display device.

[0065] The image generation unit 54 generates an image including a statistical chart (statistical graph, etc.) based on the multiple types of scores derived by the score derivation unit 42. The output unit 52 outputs the image data generated by the image generation unit 54 to a predetermined device or medium. Examples of images generated by the image generation unit 54 will be described later.

[0066] For any of the negligence score, seriousness score, impact score, and risk score, the higher the value, the greater the risk that the hospital or each department within the hospital faces. According to the risk assessment device 10 of the first embodiment, it is possible to quantitatively grasp the magnitude of risk that various granularities of organizations, such as hospitals and departments within a hospital, face, etc., and also to compare risks between hospitals, departments, etc.

[0067] The operation of the risk assessment device 10 of the first embodiment having the above configuration will be described. The classification information storage unit 20 of the risk assessment device 10 prestores classification information by GRM for multiple incident reports, and the classification information is added and updated daily. When a term score setting instruction specifying multiple incident reports for setting a term score is input via a user interface or the like, the report reading unit 30 of the risk assessment device 10 reads data of the specified multiple incident reports.

[0068] The negligence term score determination unit 36 ​​determines a negligence term score for each of the words described in the read incident reports, and stores the negligence term score in the negligence term score storage unit 22. The serious term score determination unit 38 determines a serious term score for each of the words described in the read incident reports, and stores the serious term score in the serious term score storage unit 24. The impact term score determination unit 40 determines an impact term score for each of the words described in the read incident reports, and stores the impact term score in the impact term score storage unit 26. The negligence term score, serious term score, and impact term score may be updated by daily or monthly batch processing.

[0069] When a risk score calculation instruction is inputted via a user interface or the like, specifying a plurality of incident reports to be analyzed (analysis target reports) and organizational information, the report reading unit 30 of the risk assessment device 10 reads the plurality of analysis target reports. In addition, the organizational information reading unit 32 reads the organizational information.

[0070] The negligence score derivation unit 44 derives negligence scores for a plurality of analysis target reports based on the negligence term scores stored in the negligence term score storage unit 22. The negligence score derivation unit 44 also derives negligence scores for each hospital, department, and year and month by calculating statistics by aggregating the negligence scores for a plurality of analysis target reports by each hospital, department, and year and month.

[0071] The severe score derivation unit 46 derives severe scores for a plurality of analysis target reports based on the severe term scores stored in the severe term score storage unit 24. The severe score derivation unit 46 also derives severe scores for each hospital, each department, and each year and month by calculating statistics by aggregating the severe scores for the plurality of analysis target reports for each hospital, each department, and each year and month.

[0072] The impact score derivation unit 48 derives impact scores for a plurality of analysis target reports based on the impact term scores stored in the impact term score storage unit 26. The impact score derivation unit 48 also derives impact scores for each hospital, department, and year / month by calculating statistics for the impact scores for the plurality of analysis target reports.

[0073] The risk score derivation unit 50 derives risk scores for the multiple analysis target reports by adding up the negligence scores and severity scores of each of the multiple analysis target reports at a predetermined ratio. The risk score derivation unit 50 also derives risk scores for hospitals, departments, and months by aggregating the risk scores of the multiple analysis target reports by hospital, department, and month, respectively, to obtain statistics.

[0074] The output unit 52 records the negligence scores, severity scores, impact scores, and risk scores for each hospital, department, and month in a result file, and stores the result file in a specified storage area. The image generation unit 54 generates a specified statistical chart based on the negligence scores, severity scores, impact scores, and risk scores for each hospital, department, and month, and generates image data including the statistical chart. The output unit 52 stores the image data generated by the image generation unit 54 in a specified storage area.

[0075] An example of risk assessment by the risk assessment device 10 will now be described. The risk score deriving unit 50 derives a risk score for each of the multiple hospitals based on a group of incident reports created by each of the multiple hospitals. The image generating unit 54 generates an image showing the risk scores for each of the multiple hospitals side by side.

[0076] An example of this image is shown in Fig. 9. Fig. 9 includes a box plot showing the risk scores of each of six hospitals (Hospital A to Hospital F) as a risk assessment result. The box plot in Fig. 9 shows the minimum 60, maximum 62, first quartile 64, third quartile 66, median 68, and average 70 (arithmetic mean) of the risk scores for multiple incident reports for each hospital. In this way, the risk assessment device 10 of the first embodiment makes it possible to visualize the magnitude of risk held by multiple different organizations, making it easy to compare risks between organizations.

[0077] The risk score derivation unit 50 derives the risk scores of the hospital and each department for each of the multiple periods based on multiple incident reports that report incidents that occurred during each of the multiple periods. The image generation unit 54 generates an image showing the fluctuations in the risk scores of the hospital and each department over the multiple periods.

[0078] An example of this image is shown in FIG. 10. FIG. 10 includes a line graph showing the fluctuation of risk scores over time in a certain department in a hospital as a risk assessment result. The solid line among the elements of the line graph shows the risk score itself in the department. The dashed line shows the 5-month moving average value of the risk score, and the dashed line shows the 3-month moving average value of the risk score. The image generating unit 54 may generate an image including multiple line graphs showing the fluctuation of risk scores over time for multiple departments in a hospital. In this way, according to the risk assessment device 10 of the first embodiment, it is possible to visualize the fluctuation over time for the risks held by organizations of various granularities, and to assist in easily grasping the trend of the fluctuation.

[0079] <Second Example> The following description of the second embodiment will focus on the differences from the first embodiment, and will omit the description of the commonalities. It goes without saying that the features of the second embodiment can be arbitrarily combined with the features of the first embodiment and the modifications.

[0080] Figure 11 is a scatter plot showing the relationship between the amount of reports and the risk score. In this scatter plot, multiple points corresponding to multiple departments in the hospital are plotted. The amount of reports on the horizontal axis is the number of incident reports per person per year in each organization in the hospital. As this scatter plot shows, there was a tendency for departments with fewer reports to have higher risk scores, and departments with more reports to have lower risk scores. This is thought to be because a large number of reports does not necessarily mean that there are more high-risk incident reports, but rather that there are more low-risk incident reports, resulting in a lower overall score statistic. A similar trend was also seen in the negligence score, severity score, and impact score.

[0081] In this way, it is considered that the risk score, negligence score, severity score, and impact score are affected by the report volume (i.e., the number of incident reports). Therefore, in the second embodiment, a technology is proposed for more accurately evaluating the magnitude of the risk each organization faces by taking into account the report volume of each organization. In the second embodiment, risk deviation, negligence deviation, severity deviation, and impact deviation are introduced as indices showing the magnitude of the risk each organization faces by taking into account the report volume of each organization.

[0082] The functional blocks of the risk assessment device 10 of the second embodiment are similar to the functional blocks of the risk assessment device 10 of the first embodiment shown in Fig. 4. The risk score derivation unit 50 determines the correspondence between the number of incident reports created in each of a plurality of organizations, in other words, the number of incident reports reporting incidents that occurred in each organization, and the risk score of each organization (a value derived by the method of the first embodiment). The risk score derivation unit 50 derives the difference between a standard value of the risk score corresponding to the number of incident reports of a certain organization and the derived value of the risk score of the organization as the risk deviation of the organization.

[0083] 12 shows a specific example of risk deviation derivation. The risk score derivation unit 50 acquires the number of people in each organization according to the organization information read by the organization information reading unit 32. The risk score derivation unit 50 derives the number of incident reports in each organization per unit period based on the occurrence date and department name written in the incident report. The risk score derivation unit 50 derives the reporting volume of each organization based on the number of people in each organization and the number of incident reports in each organization per unit period. The reporting volume in the second embodiment is the number of incident reports per year and per person.

[0084] As a modified example, the reporting volume may be another numerical value based on the number of incident reports. For example, when the number of people in multiple organizations to be compared for risk is similar to each other, the number of incident reports per unit period may be the reporting volume. When the number of people in multiple organizations to be compared for risk and the incident report collection period are similar to each other, the number of incident reports itself may be the reporting volume. Furthermore, the average, median, or mode of the number of incident reports for each organization per unit period may be used as the reporting volume.

[0085] The risk score derivation unit 50 plots the positions of multiple organizations in a two-dimensional space with the reporting volume (i.e., the number of incident reports per year and per person) as the first axis and the risk score as the second axis, and obtains an approximation curve 80 (also called a spline curve or smoothing spline) for the position of each organization using a known method such as curve fitting. The risk score on the second axis may be a statistic (median value) of risk scores of multiple organizations with the same reporting volume. The approximation curve 80 indicates the standard value of the risk score corresponding to the reporting volume. The dashed lines above and below the approximation curve 80 in FIG. 12 indicate the 95% confidence interval of the approximation curve.

[0086] The risk score derivation unit 50 identifies a risk score on the approximation curve 80 corresponding to the reporting amount of a certain organization as a standard value of the risk score corresponding to that reporting amount. The risk score derivation unit 50 derives the difference between the standard value and the derived value of the risk score of the organization (a value obtained by the method of the first embodiment, for example, a median value) as the risk deviation of the organization. For example, when department A is plotted at point 82, the difference 84 between the risk score (standard value) on the approximation curve 80 corresponding to the reporting amount of point 82 and the risk score (derived value) of point 82 is derived as the risk deviation of department A. Also, when department B is plotted at point 86, the difference 88 between the risk score (standard value) on the approximation curve 80 corresponding to the reporting amount of point 86 and the risk score (derived value) of point 86 is derived as the risk deviation of department B.

[0087] The negligence score derivation unit 44 derives the negligence deviation in the same manner as the risk score derivation unit 50. That is, the negligence score derivation unit 44 obtains the correspondence between the number of incident reports created by each of a plurality of organizations, in other words, the number of incident reports reporting incidents that occurred in each organization, and the negligence score of each organization. The negligence score derivation unit 44 derives the difference between the standard value of the negligence score corresponding to the number of incident reports of a certain organization and the derived value of the negligence score of the organization as the negligence deviation of the organization.

[0088] Specifically, the negligence score derivation unit 44 plots the positions of multiple organizations in a two-dimensional space with the amount of incident reports (number of cases / number of people / year) as the first axis and the negligence score as the second axis, and obtains an approximation curve corresponding to the position of each organization. The negligence score on the second axis may be a statistic (median value) of the negligence scores of multiple organizations with the same amount of reports. It can be said that this approximation curve indicates the standard value of the negligence score corresponding to the amount of reports. The negligence score derivation unit 44 identifies the negligence score on the approximation curve corresponding to the amount of reports of a certain organization as the standard value of the negligence score corresponding to that amount of reports. The negligence score derivation unit 44 derives the difference between the standard value and the derived value of the negligence score of the organization (the value obtained by the method of the first embodiment, for example the median value) as the negligence deviation of the organization.

[0089] The severity score derivation unit 46 derives the severity deviation in the same manner as the risk score derivation unit 50. That is, the severity score derivation unit 46 obtains the correspondence between the number of incident reports created in each of a plurality of organizations, in other words, the number of incident reports reporting incidents that occurred in each organization, and the severity score of each organization. The severity score derivation unit 46 derives the difference between a standard value of the severity score corresponding to the number of incident reports of a certain organization and the derived value of the severity score of the organization as the severity deviation of the organization.

[0090] Specifically, the severity score derivation unit 46 plots the positions of multiple organizations in a two-dimensional space with the incident report volume (number of incidents / number of people / year) as the first axis and the severity score as the second axis, and obtains an approximation curve corresponding to the position of each organization. The severity score on the second axis may be a statistic (median) of the severity scores of multiple organizations with the same report volume. This approximation curve can be said to indicate the standard value of the severity score corresponding to the report volume. The severity score derivation unit 46 identifies the severity score on the approximation curve corresponding to the report volume of a certain organization as the standard value of the severity score corresponding to that report volume. The severity score derivation unit 46 derives the difference between the standard value and the derived value of the severity score of the organization (a value obtained by the method of the first embodiment, for example the median) as the severity deviation of the organization.

[0091] The impact score derivation unit 48 derives the impact deviation in the same manner as the risk score derivation unit 50. That is, the impact score derivation unit 48 obtains the correspondence between the number of incident reports created by each of a plurality of organizations, in other words, the number of incident reports reporting incidents that occurred in each organization, and the impact score of each organization. The impact score derivation unit 48 derives the difference between a standard value of the impact score corresponding to the number of incident reports of a certain organization and the derived value of the impact score of the organization as the impact deviation of the organization.

[0092] Specifically, the impact score derivation unit 48 plots the positions of multiple organizations in a two-dimensional space with the incident report volume (number of incidents / number of people / year) as the first axis and the impact score as the second axis, and obtains an approximation curve corresponding to the position of each organization. The impact score on the second axis may be a statistic (median) of the impact scores of multiple organizations with the same report volume. This approximation curve can be said to indicate a standard value of the impact score corresponding to the report volume. The impact score derivation unit 48 identifies the impact score on the approximation curve corresponding to the report volume of a certain organization as the standard value of the impact score corresponding to that report volume. The impact score derivation unit 48 derives the difference between the standard value and the derived value of the impact score of the organization (a value obtained by the method of the first embodiment, for example the median) as the impact deviation of the organization.

[0093] According to the risk assessment device 10 of the second embodiment, the difference between the derived value of the risk index based on the contents of the incident report and the standard value of the risk index corresponding to the report volume (the number of incident reports per unit period and unit number of people) is set as a new risk index value (risk deviation, negligence deviation, severity deviation, impact deviation). This makes it possible to eliminate the influence of the number of incident reports of each organization from the risk index value of each organization and to more accurately assess the magnitude of the risk held by each organization.

[0094] Furthermore, according to the risk assessment device 10 of the second embodiment, the accuracy of risk comparison between multiple organizations can be further improved. For example, the risk score of department A (point 82) in FIG. 12 is 1.0, and the risk score of department B (point 86) is 0.5. Therefore, at first glance, it appears that department B has a lower risk than department A. However, the risk deviation of department A is -0.4 as shown by difference 84, and the risk deviation of department B is 0.3 as shown by difference 88. Therefore, by taking into account the amount of reports, it can be determined that department A has a lower risk than department B.

[0095] The operation of the risk assessment device 10 of the second embodiment will be described. The negligence score derivation unit 44 derives the negligence score for each of the multiple hospitals and multiple departments through the same operation as in the first embodiment. The negligence score derivation unit 44 further derives the negligence deviation for each of the multiple hospitals and multiple departments based on the report volume and negligence score for each of the multiple hospitals and multiple departments. The severity score derivation unit 46 derives the severity score for each of the multiple hospitals and multiple departments through the same operation as in the first embodiment. The severity score derivation unit 46 further derives the severity deviation for each of the multiple hospitals and multiple departments based on the report volume and severity score for each of the multiple hospitals and multiple departments.

[0096] The impact score derivation unit 48 derives an impact score for each of the multiple hospitals and multiple departments through an operation similar to that of the first embodiment. The impact score derivation unit 48 further derives an impact deviation for each of the multiple hospitals and multiple departments based on the report volume and impact score for each of the multiple hospitals and multiple departments. The risk score derivation unit 50 derives a risk score for each of the multiple hospitals and multiple departments through an operation similar to that of the first embodiment. The risk score derivation unit 50 further derives a risk deviation for each of the multiple hospitals and multiple departments based on the report volume and risk score for each of the multiple hospitals and multiple departments.

[0097] The output unit 52 records the negligence deviation, severity deviation, impact deviation, and risk deviation for each hospital and department in a result file in addition to the output data in the first embodiment, and stores the result file in a specified storage area. The image generation unit 54 generates a specified statistical chart based on the negligence deviation, severity deviation, impact deviation, and risk deviation for each hospital and department, and generates image data including the statistical chart. The output unit 52 stores the image data generated by the image generation unit 54 in a specified storage area.

[0098] Fig. 13 shows an example of an image for risk assessment generated by the risk assessment device 10 of the second embodiment. The image generation unit 54 generates an image including a scatter diagram in which the positions of multiple organizations (hospitals, departments, etc.) are plotted according to the severity deviation and negligence deviation of each of the organizations in a two-dimensional space with the severity deviation as the first axis and the negligence deviation as the second axis. In the scatter diagram in the figure, the doctor's department is shown with a white circle, the nurse's department is shown with a black circle, and the medical staff's department (pharmacists, radiologists, laboratory technicians, nutritionists, etc.) are shown with a triangle.

[0099] 14 also shows an example of an image for risk assessment generated by the risk assessment device 10 of the second embodiment. The image generation unit 54 generates an image including a scatter diagram in which the position of each of a plurality of organizations (hospitals, departments, etc.) is plotted according to the risk deviation and impact deviation of each of the organizations in a two-dimensional space with the risk deviation as the first axis and the impact deviation as the second axis. In the scatter diagram in the figure, the name of each organization is placed at the position of each organization.

[0100] In the scatter plot of Figure 13, organizations placed in the first quadrant (upper right region) are organizations with both severity scores and negligence scores higher than the standard values, and can be determined to be organizations that require special attention and support in terms of safety management. Similarly, in the scatter plot of Figure 14, organizations placed in the first quadrant (upper right region) are organizations with both risk scores and impact scores higher than the standard values, and can be determined to be organizations that require special attention and support in terms of safety management.

[0101] FIG. 15 also shows an example of an image for risk assessment generated by the risk assessment device 10 of the second embodiment. FIG. 15 is an image corresponding to FIG. 13, and shows an image including a scatter diagram in which the positions of multiple organizations (hospitals, departments, etc.) are plotted according to the severity deviation and negligence deviation of each of the organizations in a two-dimensional space with the severity deviation as the first axis and the negligence deviation as the second axis. The scatter diagram in FIG. 13 plots the positions of each organization in a certain hospital (here, Hospital A), and the scatter diagram in FIG. 15 plots the positions of each organization in another hospital (here, Hospital B). As with the scatter diagram in FIG. 14, the scatter diagrams in FIG. 13 and FIG. 15 may also include organization names (e.g., gastrointestinal surgery, pediatrics, outpatient, etc.).

[0102] Comparing the scatter diagram of FIG. 13 for Hospital A with the scatter diagram of FIG. 15 for Hospital B, the severity deviation and negligence deviation are within a certain range in both scatter diagrams, and the usefulness of visualizing the risk of a hospital with the severity deviation and negligence deviation is confirmed. Furthermore, according to the risk assessment device 10 of the second embodiment, risk assessment between organizations (e.g., between hospitals) can be realized. For example, by comparing the scatter diagram of FIG. 13 for Hospital A with the scatter diagram of FIG. 15 for Hospital B, it is possible to easily determine which hospital has a higher risk in terms of the severity deviation and negligence deviation, and the factors (departments, etc.) that increase the risk.

[0103] FIG. 16 also shows an example of an image for risk assessment generated by the risk assessment device 10 of the second embodiment. FIG. 16 is an image corresponding to FIG. 14, and shows an image including a scatter diagram in which the position of each of a plurality of organizations (hospitals, departments, etc.) is plotted according to the risk deviation and impact deviation of each of the organizations in a two-dimensional space with the risk deviation as the first axis and the impact deviation as the second axis. The scatter diagram in FIG. 14 plots the position of each organization in a certain hospital (assumed to be Hospital A here), and the scatter diagram in FIG. 16 plots the position of each organization in another hospital (assumed to be Hospital B here). As with the scatter diagram in FIG. 14, the scatter diagram in FIG. 16 may also include organization names.

[0104] Comparing the scatter diagram of FIG. 14 for Hospital A with the scatter diagram of FIG. 16 for Hospital B, it is found that the risk deviation and impact deviation in both scatter diagrams are within a certain range, confirming the usefulness of visualizing the risks faced by hospitals using risk deviation and impact deviation. Furthermore, the risk assessment device 10 of the second embodiment can realize risk assessment between organizations (e.g., between hospitals). For example, by comparing the scatter diagram of FIG. 14 for Hospital A with the scatter diagram of FIG. 16 for Hospital B, it is possible to easily determine which hospital is at higher risk in terms of risk deviation and impact deviation, and the factors (departments, etc.) that increase the risk.

[0105] The image generating unit 54 of the risk assessment device 10 may generate an image in which an object indicating a reference value or range (called a "reference object") is added to the scatter diagrams of Figs. 13 to 16. The reference value or range may be the minimum value, the maximum value, or a range from the minimum value to the maximum value of the reference hospital or the comparison hospital. For example, the reference object may be a rectangular object indicating the range from the minimum value to the maximum value of the risk deviation (severity deviation) of the reference hospital and the range from the minimum value to the maximum value of the impact deviation (negligence deviation) of the reference hospital. The reference value or range may be the average value of the minimum value, the average value of the maximum value, or the range from the average value of the minimum value to the average value of the maximum value across multiple hospitals. This configuration makes it easier to evaluate the risks held by each organization and to compare the risks between organizations.

[0106] 17 and 18 also show examples of images for risk assessment generated by the risk assessment device 10 of the second embodiment. The image generation unit 54 of the risk assessment device 10 derives risk deviations of multiple hospitals to be compared based on the incident reports of the multiple hospitals, and generates an image including a graph (e.g., a bar graph) showing the risk deviations of the multiple hospitals side by side. FIG. 17 shows the risk deviations of cardiac surgery and vascular surgery of each hospital derived based on the incident reports of the cardiac surgery and vascular surgery of each hospital. Also, FIG. 18 shows the risk deviations of the cardiology departments of each hospital derived based on the incident reports of the cardiology departments of each hospital. This configuration makes it easier to compare risks between organizations (here, between hospitals).

[0107] The present disclosure has been described above based on the first and second embodiments. These embodiments are merely examples, and it will be understood by those skilled in the art that various modifications are possible in the combination of each component and each processing step, and that such modifications are also within the scope of the present disclosure.

[0108] Although not mentioned in the first and second embodiments, as a modified example, the risk assessment device 10 may further derive a core score and perform risk assessment based on the core score. Specifically, the classification information storage unit 20 of the risk assessment device 10 may store classification information indicating whether each of a plurality of incident reports created in a hospital corresponds to an incident report that has been judged by a meeting of a plurality of people (hereinafter also referred to as a "collegial body") to contain important content from the viewpoint of grasping and assessing risks. A set of corresponding incident reports is called a "core report group," and a set of non-corresponding incident reports is called a "non-core report group." The collegial body checks the contents of the plurality of incident reports and judges whether each incident report corresponds to the core report group or the non-core report group. Classification information indicating this judgment result may be stored in the classification information storage unit 20. The collegial body may be composed of members different from the GRM that judges the impact report, and the judgment of the core report may be different from the judgment of the impact report.

[0109] The term score determination unit 34 of the risk assessment device 10 may further include a core term score determination unit. The core term score is similar to the impact term score. The core term score is an index that indicates the likelihood of a term appearing in an incident report that the collegial body judges to contain important information from the viewpoint of risk understanding and assessment, in other words, an incident report that the collegial body judges to require careful consideration or some kind of action. The method of deriving the core term score may be similar to the method of deriving the impact term score shown in FIG. 3, and may be a method in which the "impact report group" in FIG. 3 is replaced with the "core report group" and the "non-impact report group" in FIG. 3 is replaced with the "non-core report group". The core term score of each word derived by the core term score determination unit may be stored in the storage unit 14 (core term score storage unit).

[0110] The score derivation unit 42 of the risk assessment device 10 may further include a core score derivation unit. The core score derivation unit derives a core score based on the core term score stored in the core term score storage unit. The core score is similar to the impact score, and is an index that indicates the magnitude of risk from the viewpoint of the panel judging that the incident report contains important content. The method of deriving the core score is the same as the method of deriving the negligence score shown in FIG. 5 (the negligence term score in FIG. 5 is replaced with the core term score). When multiple analysis target reports are read by the report reading unit 30, the core score derivation unit derives a core score for each analysis target report (report unit). The core score derivation unit aggregates the impact scores of the multiple analysis target reports at various granularities (hospital unit, department unit, etc.). The core score derivation unit derives core scores for organizations at various granularities and for various periods, similar to the impact score derivation unit.

[0111] The core score derivation unit derives the core deviation in the same manner as the risk score derivation unit 50. That is, the core score derivation unit obtains the number of incident reports created in each of the multiple organizations, in other words, the correspondence between the number of incident reports reporting incidents that occurred in each organization and the core score of each organization. The core score derivation unit derives the difference between a standard value of the core score corresponding to the number of incident reports of a certain organization and the derived value of the core score of the organization as the core deviation of the organization.

[0112] The output unit 52 may further record the core scores and core deviations for each hospital, department, and year and month in a result file. For example, the output unit 52 may generate and record a document file showing the negligence score, severity score, impact score, risk score, and core score of each of the multiple incident reports. The image generating unit 54 may generate a predetermined statistical chart including the core scores or core deviations for each hospital, department, and year and month, and generate image data including the statistical chart. For example, the image generating unit 54 may generate an image including a scatter diagram in which the impact deviations in FIG. 14 and FIG. 16 are replaced with the core deviations.

[0113] By using the core score and core deviation, it is possible to evaluate the risk of an organization from a more diverse perspective. Like the impact score, the core score is based on human judgment, but unlike the impact score, it is based on the results of a consensus among multiple people. Therefore, by introducing the core score and core deviation, it is possible to prevent oversights in risk assessment and support more appropriate risk assessment and appropriate judgment as an organization.

[0114] The risk assessment device 10 derives the seriousness score (severity deviation), the negligence score (negligence deviation), the risk score (risk deviation), and the core score (core deviation), but may be configured not to derive the impact score (impact deviation). This is because there may be organizations that cannot obtain a judgment based on GRM.

[0115] The risk assessment device 10 described in the first and second embodiments may include a communication unit that communicates with an external device according to a predetermined communication protocol, and may provide a risk assessment service as a cloud service or SaaS (Software as a Service). For example, a client device may upload an incident report and organizational information to the risk assessment device 10 and request a risk assessment from the risk assessment device 10. The control unit 12 of the risk assessment device 10 may obtain the incident report and organizational information via the communication unit and derive scores of various risk indexes. The output unit 52 of the risk assessment device 10 may transmit risk assessment results (risk scores, risk deviations, statistical charts, etc.) to the client device via the communication unit.

[0116] The multiple functional blocks of the risk assessment device 10 described in the first and second embodiments may be distributed among multiple devices. In this case, the multiple devices may communicate with each other and cooperate as a system to achieve the same functions as the risk assessment device 10 in the first and second embodiments.

[0117] The technical ideas (risk assessment technology) described in the first and second examples can be applied to other than analyzing incident reports created by medical institutions and assessing the risks faced by the medical institutions. For example, the reports to be analyzed may be incident reports in various industries such as the transportation industry or the construction industry. Furthermore, they are not limited to incident reports, and may be near-miss reports, papers, or news articles. The technical ideas described in the first and second examples can be widely applied to estimating and visualizing the risks faced by an organization where an event indicated in a report has occurred.

[0118] Any combination of the above-mentioned examples and modifications is also useful as an embodiment of the present invention. A new embodiment resulting from the combination has the effects of each of the combined examples and modifications. It will also be understood by those skilled in the art that the functions to be performed by each component recited in the claims can be realized by each component shown in the examples and modifications alone or in cooperation with each other. [Industrial Applicability]

[0119] The techniques of the present disclosure can be applied to systems or devices related to risk assessment. [Explanation of symbols]

[0120] 10 risk assessment device, 22 negligence term score memory unit, 24 seriousness term score memory unit, 26 impact term score memory unit, 30 report reading unit, 34 term score determination unit, 42 score derivation unit, 52 output unit, 54 image generation unit.

Claims

1. A system for analyzing a report that reports an event that has occurred in an organization and evaluating a level of risk to the organization, comprising: a first storage unit that stores, for each of a plurality of words, a negligence term score that is an index representing the likelihood of the word appearing in a report indicating that an event occurred due to the negligence of the organization; a second storage unit that stores a severe term score, which is an index representing the likelihood of each of a plurality of words appearing in a report indicating the occurrence of a severe event; a score derivation unit that derives a negligence score from the negligence term score of each word described in the report to be analyzed, derives a severity score from the severity term score of each word described in the report to be analyzed, and adds up the negligence score and the severity score in a ratio based on a weight in a predetermined risk assessment to derive a risk score that is an index representing the magnitude of risk to the organization; A risk assessment system comprising:

2. The risk assessment system described in claim 1, characterized in that the score derivation unit derives multiple risk scores based on multiple reports reporting events that have occurred in the organization, and derives a statistical quantity based on the multiple risk scores as the risk score for the organization.

3. Further comprising an image generating unit, The score derivation unit derives a risk score for each of the multiple organizations based on a plurality of reports reporting events that have occurred in each of the multiple organizations; The risk assessment system according to claim 1 or 2, wherein the image generating unit generates an image showing a risk score for each of the plurality of organizations.

4. Further comprising an image generating unit, The score derivation unit derives a risk score for the organization in each of a plurality of time periods based on a plurality of reports reporting events that occurred in each of the plurality of time periods; The risk assessment system according to claim 1 or 2, wherein the image generating unit generates an image showing a variation in the risk score of the tissue over the multiple periods.

5. The score derivation unit derives a risk score for each of the multiple organizations based on a plurality of reports reporting events that have occurred in each of the multiple organizations; The risk assessment system of claim 1 or 2, wherein the score derivation unit determines a correspondence between the number of reports and a standard risk score based on the number of reports and risk scores of each of the multiple organizations, and further derives the difference between the standard value of the risk score corresponding to the number of reports of a certain organization and the derived value of the risk score of the organization as the risk deviation of the organization.

6. a third storage unit that stores an impact term score, which is an index representing the likelihood of each of the plurality of words appearing in a report that a person judges to contain important content; An image generating unit, The score derivation unit further derives an impact score, which is an index representing the magnitude of risk to each of the multiple organizations, based on the impact term scores of each word described in a plurality of reports reporting events that occurred in each of the multiple organizations; the score derivation unit determines a correspondence relationship between the number of reports and a standard impact score based on the number of reports and the impact score of each of the plurality of organizations, and further derives an impact deviation for an organization as a difference between a standard value of the impact score corresponding to the number of reports of the organization and a derived value of the impact score of the organization; The risk assessment system described in claim 5, wherein the image generation unit generates an image in which the position of each organization is plotted according to the risk deviation and impact deviation of each organization in a two-dimensional space with risk deviation as the first axis and impact deviation as the second axis.

7. A method for evaluating a level of risk to an organization by a computer that can access a first storage unit and a second storage unit, the method comprising: analyzing a report that reports an event that has occurred in an organization; The first storage unit stores, for each of a plurality of words, a negligence term score which is an index representing the likelihood of the word appearing in a report indicating that an event occurred due to negligence of the organization; The second storage unit stores, for each of a plurality of words, a severe term score which is an index representing the likelihood of the word appearing in a report indicating the occurrence of a severe event; A step of extracting a plurality of words written in a report to be analyzed; deriving a negligence score from the negligence term score of each extracted word, deriving a severity score from the severity term score of each extracted word, and adding up the negligence score and the severity score in a ratio based on a weight in a predetermined risk assessment to derive a risk score that is an index representing the magnitude of risk to the organization; A risk assessment method comprising:

8. A computer program for causing a computer that can access a first storage unit and a second storage unit to analyze a report that reports an event that has occurred in an organization and evaluate a magnitude of a risk to the organization, the computer program comprising: The first storage unit stores, for each of a plurality of words, a negligence term score which is an index representing the likelihood of the word appearing in a report indicating that an event occurred due to negligence of the organization; The second storage unit stores, for each of a plurality of words, a severe term score which is an index representing the likelihood of the word appearing in a report indicating the occurrence of a severe event; The function to extract multiple words written in the report to be analyzed, a function of deriving a negligence score from the negligence term score of each extracted word, deriving a severity score from the severity term score of each extracted word, and adding up the negligence score and the severity score in a ratio based on a weight in a predetermined risk assessment to derive a risk score that is an index representing the magnitude of risk to the organization; A computer program for causing the computer to realize the above.

Citation Information

Patent Citations

  • Manufacture of roof tile

    JP1986072694A

  • Device, method and program for analyzing incident-accident report

    JP2008269354A

  • Classification system of report

    JP2018081334A

  • Risk scorecard

    US20110191138A1