Intervention effect analysis system and intervention effect analysis method

Abstract

To properly analyzes an effect of intervention of healthcare business or preventive care business, from care, medical, healthcare business, or preventive care data.SOLUTION: An intervention effect analysis system is provided for analyzing an effect of intervention measures. An input unit receives an input of intervention data of an intervention group to which the intervention measures have been implemented and non-intervention data of a non-intervention group to which the intervention measures are not implemented. A calculation unit calculates, using multiple machine learning methods, trend scores in each of the intervention data and the non-intervention data from background information included in each of the intervention data and the non-intervention data, generates multiple pairs of data with intervention data and the non-intervention data having a small difference of the calculated trend scores, calculates standardized effect amount indicating a difference of the background information between the multiple pairs of data, and calculates an intervention effect from the multiple pairs of data on the basis of the calculated standardized effect amount.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an intervention effect analysis system for analyzing the effect of an intervention measure. [Background technology]

[0002] Local governments are required to analyze the effectiveness of intervention measures such as health checkups, health programs, and nursing care prevention programs in order to extend the healthy life expectancy of residents and optimize social security costs. In order to implement effective and efficient intervention measures, it is important to utilize data accumulated by local governments to determine whether intervention measures are effective or ineffective, prioritize budget allocation to effective projects, and realize sustainable social security services.

[0003] The following prior art is included as background art in this technical field. Patent Document 1 (JP Patent Publication No. 2019-192065 A) describes a device for verifying the effect of care intervention on subjects, which is a device that verifies a portion of similar subjects based on past data on the subjects. A care intervention effect verification device is described, which comprises a means for generating a group consisting of subjects, a means for dividing the some of the subjects constituting the group into a first subgroup to which care intervention will be performed and a second subgroup to which care intervention will not be performed, a means for accepting verification data including values ​​of predetermined evaluation indexes for the first subgroup and the second subgroup after a specific period of time has elapsed, and a means for determining the effectiveness of the care intervention based on a comparison result of the values ​​of the predetermined evaluation indexes for the first subgroup and the second subgroup (see claim 1 of the patent document).

[0004] Patent Document 2 (JP Patent Publication 2014-225177A) describes an analysis system that has a processor that executes a program and a memory that stores the program, and analyzes the effectiveness of a health project by executing the program. The analysis system includes an input unit into which a subscriber's medical expenses, intervention information indicating whether an intervention service has been provided, and a start date of the intervention service are input, and the system analyzes the relationship between the medical expenses before the intervention service is provided and the intervention information, and calculates a propensity score for an intervention group that indicates the probability that the intervention service has been provided and a propensity score for an intervention group that indicates the probability that the intervention service has been provided from the medical expenses before the intervention service is provided, and a propensity score for an intervention group that indicates the probability that the intervention service has been provided and a propensity score for an intervention group that indicates the probability that the intervention service has not been provided. An analysis system is described that has: a propensity score calculation unit that calculates a propensity score of a non-intervention group, which indicates a probability; and an adjusted medical expense calculation unit that calculates adjusted medical expenses of the intervention group before and after the provision of the intervention service by multiplying the inverse of the propensity score of the intervention group by the medical expenses of the intervention group before and after the provision of the intervention service, and calculates adjusted medical expenses of the non-intervention group before and after the provision of the intervention service by multiplying the medical expenses of the non-intervention group before and after the provision of the intervention service by the inverse of the propensity score of the non-intervention group (see claim 1 of Patent Document 2). [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2019-192065 A [Patent Document 2] JP 2014-225177 A Summary of the Invention [Problem to be solved by the invention]

[0006] Even when comparing groups that differ in background information such as age and medical visit status in addition to whether or not they received an intervention, it is difficult to properly evaluate the effectiveness of an intervention measure. Therefore, it is necessary to use propensity scores to make the distribution of the groups equal, create groups that differ only in whether or not they received the intervention, and track these groups to properly evaluate the effectiveness of the measure.

[0007] In the technology described in Patent Document 1, subjects are divided into a care intervention group and a non-care intervention group so that the variation in background information of subjects before the care intervention is equalized, and the effects on the level of care and care costs after the intervention are examined. However, the technology described in Patent Document 1 divides the intervention group and the non-intervention group in the future, and it is difficult to apply it to past data after intervention has been performed. In addition, Patent Document 1 does not specifically describe a method for equalizing the variation in background information.

[0008] Furthermore, in the technology described in Patent Document 2, as a method for analyzing the effect of an intervention, background information (covariates) of an intervention group and a non-intervention group is adjusted using a propensity score, and then compared and analyzed. Here, the propensity score is a score that consolidates background information into one, and is the probability of intervention calculated from the background information. However, in the technology described in Patent Document 2, a single propensity score is used to adjust background information and analyze the effect of an intervention, but stable adjustment of the background information is not taken into consideration.

[0009] The present invention aims to accurately analyze the effects of interventions such as health care projects and nursing care prevention projects based on nursing care, medical care, health care projects, and nursing care prevention data. [Means for solving the problem]

[0010] A representative example of the invention disclosed in the present application is as follows: That is, an intervention effect analysis system for analyzing the effect of an intervention measure, comprising a calculation device for executing a predetermined process, an input unit to which data is input, and a storage device accessible to the calculation device, the input unit receives input of intervention data of an intervention group in which the intervention measure is implemented and non-intervention data of a non-intervention group in which the intervention measure is not implemented, and the calculation device: A plurality of machine learning models trained using background information contained in each of the intervention data and the non-intervention data and each of the propensity scores of the intervention data and the non-intervention data. Using the above, a propensity score is calculated for each of the intervention data and the non-intervention data from background information included in each of the intervention data and the non-intervention data, and a plurality of paired data are generated using the intervention data and the non-intervention data having a small difference between the calculated propensity scores, and the plurality of paired data are Intervention and non-intervention data inand calculating a standardized effect size indicating a difference in background information between the plurality of paired data based on the calculated standardized effect size. Showing the difference between a predetermined index value of the intervention data and a predetermined index value of the non-intervention data It is characterized by calculating the intervention effect. Effect of the Invention

[0011] According to one aspect of the present invention, the intervention effect of health care projects and nursing care prevention projects can be properly analyzed from nursing care, medical care, health care projects, and nursing care prevention data. Problems, configurations, and effects other than those described above will be made clear by the following description of the embodiments. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a block diagram showing the configuration of an intervention effect analysis system according to a first embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of the configuration of an intervention data storage unit according to the first embodiment. [Diagram 3] FIG. 2 is a diagram illustrating an example of the configuration of a non-intervention data storage unit according to the first embodiment. [Figure 4] FIG. 2 is a diagram illustrating an example of the configuration of a propensity score storage unit according to the first embodiment. [Diagram 5] FIG. 2 is a diagram illustrating an example of the configuration of a matching data storage unit according to the first embodiment. [Figure 6] FIG. 2 is a diagram illustrating an example of the configuration of a matching data storage unit according to the first embodiment. [Figure 7] FIG. 1 is a diagram showing an example of the format of the standardized effect amount in the first embodiment. [Figure 8] FIG. 11 is a diagram illustrating a process performed by a matching data extraction unit in the first embodiment. [Figure 9] 1 is a flowchart of an intervention effect analysis process in Example 1. [Figure 10] FIG. 13 is a diagram showing an example of a machine learning method setting screen in the first embodiment. [Figure 11] FIG. 13 is a diagram showing an example of a background information setting screen in the first embodiment. [Figure 12] FIG. 13 is a diagram showing an example of an intervention effect index setting screen in the first embodiment. [Figure 13] FIG. 13 is a diagram showing an example of an intervention effect display screen in Example 1. [Figure 14] FIG. 11 is a block diagram showing the configuration of an intervention effect analysis system according to a second embodiment. [Figure 15] 13 is a flowchart of an intervention effect analysis process in Example 2. [Figure 16] FIG. 11 is a diagram showing a process performed by a composite matching data generating unit in the second embodiment. [Figure 17] FIG. 11 is a block diagram showing the configuration of an intervention effect analysis system according to a third embodiment. [Figure 18] FIG. 13 is a diagram showing an example of a background information group setting screen according to the third embodiment. [Figure 19] FIG. 13 is a diagram showing an example of an intervention effect display screen in Example 3. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] <Example 1> FIG. 1 is a block diagram showing the configuration of an intervention effect analysis system according to the first embodiment.

[0014] The intervention effect analysis system of the first embodiment includes an analysis terminal 100 and a database 130.

[0015] The analysis terminal 100 is a computer having an input unit 102, an output unit 103, a processor (CPU) 104, a memory 105, and a storage medium 106.

[0016] The input unit 102 is a human interface such as a mouse or keyboard, and accepts input to the analysis terminal 100. The output unit 103 is a display or printer that outputs the results of calculations performed by the analysis terminal 100. The input unit 102 and the output unit 103 may include an interface that inputs and outputs data via a storage medium or a network. The storage medium 106 is a storage device that stores various programs that execute the intervention effect analysis process by the analysis terminal 100, and the execution results of the intervention effect analysis process, and is composed of, for example, a non-volatile storage medium (magnetic disk drive, non-volatile memory, etc.).

[0017] The memory 105 includes a ROM, which is a non-volatile storage element, and a RAM, which is a volatile storage element. The ROM stores unchanging programs (e.g., BIOS) and the like. The RAM is a high-speed, volatile storage element such as a DRAM (Dynamic Random Access Memory), and temporarily stores programs executed by the processor 104 and data used when the programs are executed. That is, the programs stored in the storage medium 106 are deployed in the memory 105.

[0018] The processor 104 is a calculation device that executes a program loaded into the memory 105, and is, for example, a CPU or a GPU. The processes and calculations described below are executed by the processor 104. Note that a part of the processes executed by the processor 104 by executing a program may be executed by another calculation device (for example, hardware such as a Field Programmable Gate Array (FPGA) or an Application Specific Integrated Circuit (ASIC)).

[0019] The programs executed by the processor 104 are provided to each server via a removable medium (such as a CD-ROM or a flash memory) or a network, and are stored in a non-volatile storage device, which is a non-transitory storage medium. For this reason, the computer system may have an interface for reading the removable medium.

[0020] The intervention effect analysis system of this embodiment is a computer system configured on one computer or on multiple logically or physically configured computers, and may operate in separate threads on the same computer, or may operate on a virtual computer constructed on multiple physical computer resources.

[0021] The storage medium 106 stores programs that realize a machine learning method setting unit 111, a background information setting unit 112, a propensity score calculation unit 113, a matching data creation unit 114, a standardized effect amount calculation unit 115, a matching data extraction unit 116, an intervention effect index setting unit 117, and an intervention effect calculation unit 118.

[0022] The machine learning method setting unit 111 sets the machine learning method to be used in the intervention effect analysis, and stores the selection result in the machine learning method storage unit 133.

[0023] The background information setting unit 112 sets background information to be used in the intervention effect analysis, and stores the selected background information in the background information storage unit 134. The background information is information that can be a covariate for the presence or absence of an intervention, and is, as shown as options on the background information setting screen (FIG. 11), the subject's basic information (gender, age, address, etc. from the resident basic register), nursing care information (level of required nursing care, nursing care costs, etc.), medical information (presence or absence of disease, medical costs by disease, etc.), and health check information (test values, interview results, etc.).

[0024] The propensity score calculation unit 113 calculates the propensity score using a machine learning model based on the machine learning method selected by the machine learning method setting unit 111, and stores the calculation result in the propensity score storage unit 135. The propensity score is the probability of intervention (probability of intervention) obtained from background information, and is calculated using a machine learning model that uses background information as an explanatory variable and the presence or absence of intervention as a response variable.

[0025] The matching data creation unit 114 creates matching data in which data with similar background information is combined using the propensity scores, and stores the matching data in the matching data storage unit 136.

[0026] The standardized effect size calculation unit 115 evaluates the matching result and calculates the standardized effect size. The standardized effect size is a standardized value indicating the difference in background information between the intervention group and the non-intervention group in each matching data.

[0027] The matching data extraction unit 116 selects the best machine learning method based on the evaluation result of the matching data using the standardized effect size.

[0028] The intervention effect index setting unit 117 sets the intervention effect index selected by the user as the intervention effect index to be displayed on the intervention effect display screen (FIG. 13). The intervention effect index is a numerical value for quantitatively evaluating the effect of an intervention, and may be, for example, the presence or absence of a disease, the disease incidence rate, medical expenses by disease, nursing care expenses, the level of nursing care, test values, and interview results.

[0029] The intervention effect calculation unit 118 calculates an intervention effect index from the matching data in the machine learning method selected based on the evaluation result of the matching data, and generates display data for the intervention effect display screen (FIG. 13).

[0030] The database 130 is composed of an intervention data storage unit 131 , a non-intervention data storage unit 132 , a machine learning method storage unit 133 , a background information storage unit 134 , a propensity score storage unit 135 , and a matching data storage unit 136 .

[0031] The intervention data storage unit 131 stores intervention data for multiple years for people in the intervention group. As shown in FIG. 2, the intervention data storage unit 131 includes an individual ID 1311, basic information 1312 of the subject, nursing care information 1313, medical information 1314, health check information 1315, and intervention information 1316. The individual ID 1311 is a unique identifier indicating a single resident. The basic information 1312 of the subject is the gender, age, address, etc. of the subject, and can be acquired from the resident basic register. The nursing care information 1313 is information related to nursing care, such as the degree of nursing care required and nursing care costs. The medical information 1314 is information related to medical care, such as the presence or absence of a disease and medical costs by disease. The health check information 1315 is information on the results of a health check, such as test values ​​and interview results. The intervention information 1316 is information on the presence or absence of an intervention, that is, information on whether an intervention measure for improving health or quality of life (QOL) related to medical care, health check, or nursing care, such as nursing care prevention activities or medical procedures, has been received. Among the intervention data, nursing care information 1313, medical information 1314, and health check information 1315 change over time, so they are recorded by year (year T, year T+1, etc.). In addition, data for the year in which the intervention measure was implemented is stored as background information. In the example of Fig. 2, data for year T indicates the background information.

[0032] The non-intervention data storage unit 132 stores non-intervention data for multiple years for people in the non-intervention group. As shown in FIG. 3, the non-intervention data storage unit 132 includes the same items as the intervention data storage unit 131, such as an individual ID 1321, basic information 1322 of the subject (gender, age, address, etc., in the resident basic register), nursing care information 1323 (level of nursing care required, nursing care costs, etc.), medical information 1324 (presence or absence of disease, medical costs by disease, etc.), health check information 1325 (test values, interview results, etc.), and intervention information 1326 (presence or absence of intervention). Among the non-intervention data, the nursing care information 1323, medical information 1324, and health check information 1325 change over time, so they are recorded by year (year T, year T+1, etc.). In addition, data for the year in which the intervention measure was implemented is stored as background information. In the example of FIG. 3, data for year T indicates background information.

[0033] The propensity score storage unit 135 stores the propensity scores of the intervention group and the non-intervention group calculated from the background information by a plurality of machine learning methods. As shown in FIG. 4, the propensity score storage unit 135 includes the same items as the intervention data storage unit 131 and the non-intervention data storage unit 132, such as an individual ID 1351, basic information 1352 of the subject (gender, age, address, etc. in the resident basic register), nursing care information 1353 (level of nursing care required, nursing care costs, etc.), medical information 1354 (presence or absence of disease, medical costs by disease, etc.), medical checkup information 1355 (test values, interview results, etc.), and intervention information 1356 (presence or absence of intervention). Furthermore, the propensity score storage unit 135 includes propensity scores 1357 of the intervention group and the non-intervention group calculated from the background information using machine learning models by a plurality of machine learning methods. In this embodiment, as shown in FIG. 4, the propensity score is calculated using methods such as logistic regression analysis and neural network.

[0034] The matching data storage unit 136 stores the result of matching based on the propensity score calculated by a certain machine learning method. As shown in FIG. 5 and FIG. 6, the matching data storage unit 136 includes the same items as the propensity score storage unit 135, such as an individual ID 1361, basic information 1362 of the subject (gender, age, address, etc. in the resident basic register), nursing care information 1363 (level of nursing care required, nursing care costs, etc.), medical information 1364 (presence or absence of disease, medical costs by disease, etc.), medical checkup information 1365 (test values, interview results, etc.), intervention information 1366 (presence or absence of intervention), and propensity score 1367. Furthermore, the matching data storage unit 136 includes matching information 1368 between data of an intervention group and data of a non-intervention group having similar propensity scores. FIG. 5 shows the result of matching when the propensity score is calculated by logistic regression analysis, and FIG. 6 shows the result of matching when the propensity score is calculated by a neural network. In the figure, the matching information 1368 is represented by lines linking rows in a data table, but a column for recording the other party's personal ID may be provided, or it may be recorded in a table linking data from the intervention group with data from the non-intervention group.

[0035] Even if the background information is the same, the propensity scores differ depending on the machine learning method, and the matching pairs differ. For this reason, it is preferable to create the matching data storage unit 136 for each propensity score calculation method (FIGS. 5 and 6).

[0036] FIG. 7 is a diagram showing an example of the format of the standardized effect amount calculated by the standardized effect amount calculation unit 115. The standardized effect amount is stored in the memory 105 or the storage medium 106 as an internal variable of the intervention effect analysis system in the illustrated table format. The standardized effect amount includes the same items as those in the matching data storage unit 136, such as basic information 1402 of the subject (gender, age, address, etc. in the resident basic register), nursing care information 1403 (level of nursing care required, nursing care costs, etc.), medical information 1404 (presence or absence of disease, medical costs by disease, etc.), and health check information 1405 (test values, interview results, etc.). Furthermore, the standardized effect amount includes a propensity score calculation method 1401 and a representative value 1406 of the standardized effect amount. The representative value 1406 of the standardized effect amount is a value obtained by statistically processing the standardized effect amount of the data pair in each propensity score calculation method. Various statistical values ​​such as the average value, maximum value, minimum value, and median value can be used for the statistical processing of the standardized effect amount.

[0037] FIG. 9 is a flowchart of the intervention effect analysis process of the first embodiment.

[0038] First, the input unit 102 receives input of intervention data and non-intervention data, and stores them in the intervention data storage unit 131 and the non-intervention data storage unit 132 (S101).

[0039] Next, the machine learning method setting unit 111 displays a machine learning method setting screen (Figure 10), sets the machine learning method selected on the machine learning method setting screen as the machine learning method to be used for the intervention effect analysis, and stores the selection result in the machine learning method memory unit 133 (S102).

[0040] Next, the background information setting unit 112 displays a background information setting screen (FIG. 11), sets the background information selected on the background information setting screen as the background information to be used in the intervention effect analysis, and stores the selection result in the background information storage unit 134 (S103).

[0041] Next, the propensity score calculation unit 113 calculates a propensity score using a machine learning model based on the machine learning method selected by the machine learning method setting unit 111, and stores the calculation result in the propensity score storage unit 135 (S104). If multiple machine learning methods are selected in step S102, multiple propensity scores are calculated for each individual's data for each selected machine learning method.

[0042] Next, the matching data creation unit 114 creates matching data for the propensity scores and stores it in the matching data storage unit 136 (S105). For example, data with similar propensity scores between the intervention group and the non-intervention group are matched to create a data pair. When multiple propensity scores are calculated, matching data including different data pairs is created for each propensity score.

[0043] Next, the standardized effect amount calculation unit 115 evaluates the matching result, calculates the standardized effect amount, and calculates a representative value of the calculated standardized effect amount (S106). For example, the standardized effect amount calculation unit 115 calculates the age difference between the matched data pairs of the intervention group and the non-intervention group, and calculates the standardized effect amount by dividing the average value of the calculated age difference by the standard deviation of the age difference. The standardized effect amount makes it possible to know the magnitude of the difference in specific background information (e.g., age), and to compare multiple pieces of background information without considering the unit or the magnitude of the value. The standardized effect amount calculated by the standardized effect amount calculation unit 115 is stored in the memory 105 or the storage medium 106 as an internal variable of the intervention effect analysis system in the table format shown in FIG. 7. Then, the standardized effect amount calculation unit 115 statistically processes the standardized effect amount of the data pair in each propensity score calculation method to calculate a representative value of the standardized effect amount. Various statistical values ​​such as the average value, maximum value, minimum value, and median value can be adopted for the statistical processing of the standardized effect amount. When multiple propensity scores are calculated, a different standardized effect amount and a representative value of the standardized effect amount are calculated for each propensity score.

[0044] Next, the matching data extraction unit 116 selects the best machine learning method (propensity score calculation method) based on the evaluation result of the matching data by the standardized effect size. For example, as shown in FIG. 8, the calculation method of the propensity score with the smallest representative value is selected, and matching data based on the propensity score calculated by the calculation method is extracted (S108).

[0045] Next, the intervention effect index setting unit 117 displays the intervention effect index setting screen (FIG. 12), and sets the intervention effect index selected on the intervention effect index setting screen as the intervention effect index to be displayed on the intervention effect display screen (FIG. 13) (S109).

[0046] Next, the intervention effect calculation unit 118 calculates an intervention effect index from the matching data in the machine learning method selected based on the evaluation result of the matching data, and generates display data for the intervention effect display screen (FIG. 13) (S110).

[0047] Next, the intervention effect calculation unit 118 calculates the significance probability of the intervention effect calculated in step S109, and tests the calculated value of the intervention effect (S111).

[0048] FIG. 10 is a diagram showing an example of a machine learning method setting screen.

[0049] The machine learning method setting screen includes a list of machine learning methods prepared in advance and a selection field for selecting each machine learning method. The machine learning method setting screen is displayed by the machine learning method setting unit 111 in step S102, and prompts the user to select a machine learning method to be used in the intervention effect analysis. After selecting one or more machine learning methods in the selection field, the user operates the "Done" button to store the selection result in the machine learning method storage unit 133.

[0050] FIG. 11 is a diagram showing an example of a background information setting screen.

[0051] The background information setting screen includes a list of background information, which is an item of intervention data and non-intervention data recorded in the intervention data storage unit 131 and the non-intervention data storage unit 132, and a selection field for selecting each piece of background information. The background information setting screen is displayed by the background information setting unit 112 in step S103, and prompts the user to select background information to be used in the intervention effect analysis. After selecting one or more pieces of background information in the selection field, the user operates the "Done" button to store the selection result in the background information storage unit 134.

[0052] FIG. 12 is a diagram showing an example of an intervention effect index setting screen.

[0053] The intervention effect index setting screen includes a list of intervention effect indexes calculated as a result of the intervention effect analysis, a selection field for selecting each intervention effect index, and an input field for the period for calculating the intervention effect. The intervention effect index setting screen is displayed by the intervention effect index setting unit 117 in step S109, and prompts the user to select an intervention effect index to be displayed as a result of the intervention effect analysis. The user selects one or more intervention effect indexes in the selection field, inputs the effect calculation period, and then operates the "Complete" button to transmit the selection result to the intervention effect index setting unit 117. In addition, the intervention effect calculation unit 118 calculates the intervention effect according to the effect calculation period inputted in the intervention effect index setting screen.

[0054] FIG. 13 is a diagram showing an example of an intervention effect display screen.

[0055] The intervention effect display screen includes an intervention effect indicator selection area in which the user selects the intervention effect indicator they wish to display, a number display area that displays the difference in numbers between the intervention group and the non-intervention group in a simple comparison and a comparison using matching data, a graph display area that displays the intervention effect in a graph in a simple comparison and a comparison using matching data, and a numerical display area that displays the intervention effect as a numerical value.

[0056] The user selects the intervention effect index selected on the intervention effect index setting screen (Figure 12) in the intervention effect index selection area to switch between the intervention effect index displayed in the graph display area and the numerical display area. The graph display area displays the intervention effect for the period entered in the effect calculation period on the intervention effect index setting screen so that the intervention group and non-intervention group can be compared. The numerical display area displays the difference in the intervention effect between the intervention group and non-intervention group for the period entered in the effect calculation period on the intervention effect index setting screen, and the significance probability of the numerical value of the intervention effect. The significance probability allows the user to check the statistical certainty of the intervention effect value.

[0057] In addition, when comparing the intervention group and the non-intervention group, the intervention effect index before the intervention is different in a simple comparison, but in the matching data, the intervention effect index before the intervention is the same. In addition, background information other than the intervention effect index before the intervention, which was set in the background information setting screen in Figure 11, is also the same for the intervention group and the non-intervention group, and the difference between the intervention group and the non-intervention group is whether or not they received the intervention. Therefore, the difference in the intervention effect index between the intervention group and the non-intervention group after the intervention can be calculated as the intervention effect, and the correct intervention effect can be analyzed.

[0058] As described above, in the first embodiment, a plurality of matching data are created and matching data with similar background information is selected, so that the correct effect of an intervention can be known.

[0059] <Example 2> In the second embodiment of the present invention, appropriate data is selected from other matching data, and composite matching data is created. In the second embodiment, differences from the first embodiment are mainly described, and the description of the same configuration and effects as the first embodiment is omitted.

[0060] FIG. 14 is a block diagram showing the configuration of an intervention effect analysis system according to the second embodiment.

[0061] The intervention effect analysis system of the second embodiment includes an analysis terminal 100 and a database 130.

[0062] The analysis terminal 100 is a computer having an input unit 102, an output unit 103, a processor (CPU) 104, a memory 105, and a storage medium 106.

[0063] The storage medium 106 stores programs that realize a machine learning method setting unit 111, a background information setting unit 112, a propensity score calculation unit 113, a matching data creation unit 114, a standardized effect amount calculation unit 115, a matching data extraction unit 116, an intervention effect index setting unit 117, an intervention effect calculation unit 118, a background information extraction unit 119, a paired data extraction unit 120, a composite matching data creation unit 121, and a standardized effect amount improvement judgment unit 122.

[0064] The background information extraction unit 119 compares the standardized effect amount of each piece of background information and extracts background information with a large standardized effect amount.

[0065] The paired data extraction section 120 extracts paired data between the intervention group and the non-intervention group having a large difference in the background information extracted by the background information extraction section 119.

[0066] The composite matching data creation unit 121 extracts other paired data from the matching data in which the propensity score has been calculated by another machine learning method. Then, the data of the non-intervention group of the paired data extracted by the paired data extraction unit 120 is replaced with the data of the non-intervention group of the extracted paired data to create new paired data, thereby creating composite matching data.

[0067] The standardized effect amount improvement determination unit 122 calculates the standardized effect amount of the composite matching data created by the composite matching data creation unit 121, and evaluates the matching result. If the standardized effect amount is reduced and the evaluation result is better, the new paired data is better, so the new paired data is adopted.

[0068] FIG. 15 is a flowchart of the intervention effect analysis process of the second embodiment.

[0069] Steps S101 to S108 are the same as the intervention effect analysis process (FIG. 9) of the first embodiment described above.

[0070] After extracting the matching data (S108), it is determined whether the search for background information is complete (S118). For example, if the replacement of pair data has been performed once or more for all background information, it may be determined that the search is complete. Also, if the standardized effect amount is smaller than a predetermined threshold, it may be determined that the search is complete.

[0071] If the search for background information is not complete, the background information extraction unit 119 compares the standardized effect amount of each piece of background information and extracts background information with a larger standardized effect amount (S112).

[0072] Next, the paired data extraction unit 120 extracts paired data having a large difference in background information extracted in step S112 (S113).

[0073] Next, the synthetic matching data creation unit 121 extracts another pair data from the matching data in which the propensity score has been calculated by another machine learning method, and replaces the data of the non-intervention group of the pair data extracted in step S113 with the data of the non-intervention group of the extracted pair data to create new pair data (S114). That is, the data of the intervention group is replaced with data of another non-intervention group whose background information is similar to that of the intervention group, and new pair data is created to create synthetic matching data. The replacement of data constituting this pair data may be performed automatically by the intervention effect analysis system, or the pair data in which replacement occurs may be displayed on the screen so that the user can select whether to replace it.

[0074] For example, as shown in Figure 16 , when the background information with the largest standardized effect size is age, paired data with a large age difference between the intervention and non-intervention groups is extracted, and if the age difference of the paired data including the intervention data among other matching data with propensity score 1 is small, the non-intervention data of the original paired data (with propensity score 2) is replaced with the non-intervention data of the paired data (with propensity score 1) to create new paired data, and composite matching data is created.

[0075] Next, the standardized effect amount improvement judgment unit 122 calculates the standardized effect amount of the composite matching data created with the new paired data, and evaluates the matching result based on whether the standardized effect amount has improved (S115). If the standardized effect amount has decreased, the new paired data is better, so the new paired data is adopted (S116). On the other hand, if the standardized effect amount has not decreased, there is no improvement due to the new paired data, so the new paired data is not adopted (S117).

[0076] Thereafter, it is determined whether the search for background information is complete (S118). If the search for background information is complete, the intervention effect index setting unit 117 displays an intervention effect index setting screen (FIG. 12), and sets the intervention effect index selected on the intervention effect index setting screen as the intervention effect index to be displayed on the intervention effect display screen (FIG. 13) (S119).

[0077] Next, the intervention effect calculation unit 118 calculates an intervention effect index from the composite matching data created in the processes of steps S112 to S117, and generates display data for the intervention effect display screen (FIG. 13) (S120).

[0078] Next, the intervention effect calculation unit 118 calculates the significance probability of the intervention effect calculated in step S120, and tests the calculated value of the intervention effect (S121).

[0079] As described above, in Example 2, appropriate pair data is selected from other matching data to create synthesized matching data, thereby improving the similarity between the intervention data and the non-intervention data and enabling the true effect of the intervention measure to be known.

[0080] <Example 3> In the third embodiment of the present invention, a group of background information is set. Note that in the third embodiment, differences from the first and second embodiments will be mainly described, and descriptions of the same configurations and effects as the first and second embodiments will be omitted.

[0081] FIG. 17 is a block diagram showing the configuration of an intervention effect analysis system according to the third embodiment.

[0082] The intervention effect analysis system of the third embodiment includes an analysis terminal 100 and a database 130.

[0083] The analysis terminal 100 is a computer having an input unit 102, an output unit 103, a processor (CPU) 104, a memory 105, and a storage medium 106.

[0084] The storage medium 106 stores programs that realize a machine learning method setting unit 111, a background information setting unit 112, a propensity score calculation unit 113, a matching data creation unit 114, a standardized effect amount calculation unit 115, a matching data extraction unit 116, an intervention effect index setting unit 117, an intervention effect calculation unit 118, a background information extraction unit 119, a paired data extraction unit 120, a composite matching data creation unit 121, a standardized effect amount improvement judgment unit 122, and a background information group setting unit 123.

[0085] The background information group setting unit 123 displays a background information group setting screen (FIG. 18), and sets a group of background information selected on the background information group setting screen. The intervention effect is calculated for each group of background information that has been set.

[0086] FIG. 18 is a diagram showing an example of a background information group setting screen according to the third embodiment.

[0087] The background information group setting screen includes a list of background information, which is an item of intervention data and non-intervention data recorded in the intervention data storage unit 131 and the non-intervention data storage unit 132, and a selection field for selecting each background information. The background information group setting screen is displayed by the background information group setting unit 123, and prompts the user to select background information for analyzing the intervention effect for each group. After selecting one or more pieces of background information in the selection field, the user operates the "Complete" button to transmit the selection result to the background information group setting unit 123. The background information group may be set in advance for each piece of background information (for example, age is divided into three groups of 60 to 79 years old for each decade, and the ages are set to the 60s, 70s, and 80s), or after operating the "Complete" button, a sub-screen for setting the details of the group may be displayed to prompt the user to input the range of the group (60 to 79 years old) and the interval (every 10 years).

[0088] FIG. 19 is a diagram showing an example of an intervention effect display screen in the third embodiment.

[0089] The intervention effect display screen of Example 3 includes an intervention effect index selection area in which the user selects the intervention effect index they wish to display, an effect calculation method selection area in which they select between displaying a simple comparison and displaying matching data, a number display area in which the difference in number of people for each group of background information between the intervention group and the non-intervention group in the simple comparison and the comparison using matching data, a graph display area in which the intervention effect is displayed in a graph for each group of background information in the simple comparison and the comparison using matching data, and a numerical display area in which the intervention effect for each group of background information is displayed as a numerical value.

[0090] The user selects the intervention effect index selected on the intervention effect index setting screen (Figure 12) in the intervention effect index selection area to switch the intervention effect index displayed in the graph display area and numerical value display area. In addition, the user selects either simple comparison or matching data in the effect calculation method selection area to switch the data displayed in the graph display area and numerical value display area. The graph display area displays the intervention effect for the period entered in the effect calculation period on the intervention effect index setting screen for each background information group so that the intervention group and the non-intervention group can be compared. The numerical value display area displays the difference between the intervention group and the non-intervention group in the intervention effect for the period entered in the effect calculation period on the intervention effect index setting screen and the significance probability of the intervention effect value for each background information group. The significance probability allows the user to check the statistical certainty of the intervention effect value.

[0091] As described above, in the third embodiment, it is possible to know the effect of an intervention measure for each group of other background information (for example, age).

[0092] The present invention is not limited to the above-described embodiments, and includes various modified examples and equivalent configurations within the spirit of the appended claims. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the configurations described. Furthermore, a part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Furthermore, the configuration of another embodiment may be added to the configuration of one embodiment. Furthermore, a part of the configuration of each embodiment may be added, deleted, or replaced with another configuration.

[0093] In addition, each of the above-mentioned configurations, functions, processing units, processing means, etc. may be realized in hardware, for example by designing some or all of them as an integrated circuit, or may be realized in software by a processor interpreting and executing a program that realizes each function.

[0094] Information such as programs, tables, and files that realize each function can be stored in a storage device such as a memory, a hard disk, or an SSD (Solid State Drive), or in a recording medium such as an IC card, an SD card, or a DVD.

[0095] In addition, the control lines and information lines shown are those considered necessary for the explanation, and do not necessarily show all the control lines and information lines necessary for implementation. In reality, it can be considered that almost all components are connected to each other. [Explanation of symbols]

[0096] 100 Analysis Terminal 102 Input section 103 Output section 104 processors 105 Memory 106 Storage medium 111 Machine Learning Method Setting Section 112 Background information setting section 113 Propensity Score Calculation Unit 114 Matching Data Creation Department 115 Standardized Effect Size Calculation Section 116 Matching Data Extraction Unit 117 Intervention Effect Indicator Setting Division 118 Intervention Effect Calculation Department 119 Background information extraction part 120 Pair Data Extraction Unit 121 Synthetic Matching Data Creation Department 122 Standardized Effect Size Improvement Assessment Section 123 Background Information Group Setting Section 130 Databases 131 Intervention data storage unit 132 Non-intrusive data storage unit 133 Machine Learning Memory Unit 134 Background information storage unit 135 Propensity score memory 136 Matching data storage unit

Claims

1. An intervention effect analysis system for analyzing the effect of an intervention measure, The present invention comprises a computing device that executes a predetermined process, an input unit to which data is input, and a storage device that can be accessed by the computing device, The input unit receives input of intervention data of an intervention group in which the intervention measure is implemented and non-intervention data of a non-intervention group in which the intervention measure is not implemented, The computing device includes: Using background information included in each of the intervention data and the non-intervention data and a plurality of machine learning models trained with the propensity scores of each of the intervention data and the non-intervention data, a propensity score is calculated for each of the intervention data and the non-intervention data from the background information included in each of the intervention data and the non-intervention data; A plurality of paired data are generated using the intervention data and the non-intervention data having a small difference in the calculated propensity score, Calculate a standardized effect size indicating the difference in background information between the intervention data and the non-intervention data in the plurality of paired data; An intervention effect analysis system characterized by calculating an intervention effect indicating the difference between a specified index value of intervention data and a specified index value of non-intervention data from the multiple paired data based on the calculated standardized effect amount.

2. The intervention effect analysis system according to claim 1, The computing device calculates an intervention effect from paired data having a minimum representative value of the standardized effect size, The intervention effect analysis system, wherein the representative value is one of an average value, a maximum value, a minimum value, and a median value.

3. The intervention effect analysis system according to claim 1, The computing device includes: Extracting the background information with the largest standardized effect size; An intervention effect analysis system characterized by replacing at least a portion of the non-intervention data among the extracted paired data with other non-intervention data based on the difference in the standardized effect amount in the background information of the paired data, thereby creating new paired data.

4. The intervention effect analysis system according to claim 1, The intervention effect analysis system is characterized in that the calculation device divides the background information into groups and calculates the intervention effect for each group.

5. The intervention effect analysis system according to claim 1, The calculation device generates data for displaying an intervention effect display screen including an intervention effect index selection area for selecting the intervention effect index to be displayed, a number display area for displaying the difference between the number of people in the intervention group and the number of people in the non-intervention group, a graph display area for displaying the intervention effect in a graph, and a numerical display area for displaying the intervention effect as a numerical value.

6. The intervention effect analysis system according to claim 1, The intervention effect analysis system is characterized in that the intervention measure is an action for improving health or quality of life related to medical care, health checkups, or nursing care.

7. An intervention effect analysis method in which a computer analyzes the effect of an intervention measure, comprising: The computer includes an arithmetic unit that executes a predetermined process, an input unit to which data is input, and a storage device that is accessible by the arithmetic unit; The method comprises: The input unit receives input of intervention data of an intervention group in which the intervention measure is implemented and non-intervention data of a non-intervention group in which the intervention measure is not implemented, The computing device calculates a propensity score for each of the intervention data and the non-intervention data from background information included in each of the intervention data and the non-intervention data, using a plurality of machine learning models trained with background information included in each of the intervention data and the non-intervention data and the propensity scores of each of the intervention data and the non-intervention data; The computing device generates a plurality of paired data using intervention data and non-intervention data having a small difference between the calculated propensity scores, The arithmetic device calculates a standardized effect size indicating a difference in background information between the intervention data and the non-intervention data in the plurality of paired data, The intervention effect analysis method is characterized in that the calculation device calculates an intervention effect indicating a difference between a predetermined index value of intervention data and a predetermined index value of non-intervention data from the multiple paired data based on the calculated standardized effect amount.

8. The intervention effect analysis method according to claim 7, The calculation device calculates the intervention effect from the paired data having the smallest representative value of the standardized effect size, The method for analyzing intervention effect, wherein the representative value is one of an average value, a maximum value, a minimum value, and a median value.

9. The intervention effect analysis method according to claim 7, The computing device extracts the background information having the largest standardized effect amount, The method for analyzing intervention effects is characterized in that the calculation device replaces at least a portion of the non-intervention data among the extracted paired data with other non-intervention data based on the difference in the standardized effect amount in the background information of the paired data, thereby creating new paired data.

10. The intervention effect analysis method according to claim 7, The method for analyzing an intervention effect, wherein the calculation device divides the background information into groups and calculates the intervention effect for each group.

11. The intervention effect analysis method according to claim 7, The intervention effect analysis method, characterized in that the calculation device generates data for displaying an intervention effect display screen including an intervention effect index selection area for selecting the intervention effect index to be displayed, a number display area for displaying the difference between the number of people in the intervention group and the number of people in the non-intervention group, a graph display area for displaying the intervention effect in a graph, and a numerical display area for displaying the intervention effect as a numerical value.

12. The intervention effect analysis method according to claim 7, The intervention effect analysis method, wherein the intervention measure is an action for improving health or quality of life related to medical care, health checkups, or nursing care.

Images