Information processing device and information processing method

Abstract

We provide information to support health promotion programs and preventative care programs. [Solution] An information processing device comprising: a basic information acquisition unit that acquires basic information including the number of people certified as requiring long-term care, the number of people certified as requiring support, and the number of people diagnosed with frailty in a target area during a first period, information specific to the area during the first period, and information on health promotion and / or long-term care prevention projects implemented in the area during the first period; and a future prediction unit that estimates the incidence rate of people certified as requiring long-term care, the incidence rate of people certified as requiring support, and the incidence rate of people diagnosed with frailty in the area after a second period.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus and an information processing method for providing information to support a health promotion business and / or a long-term care insurance business.

Background Art

[0002] Conventionally, a technique for estimating future care recipients has been proposed (for example, Patent Document 1).

[0003] On the other hand, from the perspective of preventing care, the concept of frailty, which is a state in which a person can recover to a healthy state through appropriate measures and is a preventable state, has attracted attention.

[0004] However, it has not been realized to estimate the number of people diagnosed as frail (referred to as "frailty diagnosticians"), and appropriate information for supporting the planning and analysis of health promotion and care prevention projects in local governments and the like has not been provided.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] An object of the present invention is to provide information for supporting a health promotion business and / or a care prevention business.

Means for Solving the Problems

[0007] In order to solve the above problems, the present invention is A basic information acquisition unit acquires basic information including the number of people certified as requiring long-term care, the number of people certified as requiring support, and the number of people diagnosed with frailty in the target area during the first period, information specific to the area during the first period, and information on health promotion and / or long-term care prevention programs implemented in the area during the first period. A future prediction unit that estimates the incidence rate of those requiring long-term care, those requiring support, and those diagnosed with frailty in the aforementioned region after the second period, This is an information processing device characterized by having the following features.

[0008] According to this, it is possible to recognize how the incidence rate of those requiring long-term care, those requiring support, and those diagnosed with frailty will change in the target area after the second period. By providing useful information for the planning and evaluation of health promotion and / or long-term care prevention programs, it is possible to support health promotion and / or long-term care prevention programs.

[0009] Furthermore, in the present invention, The system further comprises an analysis information generation unit that generates analysis information based on the predicted incidence rate of persons requiring long-term care, the incidence rate of persons requiring support, and the incidence rate of persons diagnosed with frailty in the region after the second period, as predicted by the future prediction unit. The aforementioned analysis information is, The system may also include information that displays, in time series, the incidence rate of persons certified as requiring long-term care, the incidence rate of persons certified as requiring support, and the incidence rate of persons diagnosed with frailty in the region during the first period, as predicted by the future prediction unit, the incidence rate of persons certified as requiring long-term care, the incidence rate of persons certified as requiring support, and the incidence rate of persons diagnosed with frailty in the region after the second period.

[0010] According to this, it is possible to provide information that displays in time series the incidence rates of those certified as requiring long-term care, those certified as requiring support, and those diagnosed with frailty in the target area during the first period, as well as the incidence rates of those certified as requiring long-term care, those certified as requiring support, and those diagnosed with frailty in the target area after the second period, as predicted by the future prediction unit. This allows for the recognition of the current situation regarding the incidence of those certified as requiring long-term care, those certified as requiring support, and those diagnosed with frailty, along with the situation regarding those certified as requiring long-term care, those certified as requiring support, and those diagnosed with frailty after the second period, along a time axis. This provides useful information for planning and evaluating health promotion and / or long-term care prevention programs.

[0011] Furthermore, in the present invention, The aforementioned analysis information may include predictions of the effectiveness of the health promotion and / or long-term care prevention programs implemented in the region.

[0012] According to this, information on predicting the effectiveness of the aforementioned health promotion and / or long-term care prevention programs implemented in the target area is provided, thus offering useful information for planning health promotion and / or long-term care prevention programs in the target area.

[0013] Furthermore, in the present invention, The aforementioned analysis information may include proposals for health promotion and / or long-term care prevention programs suitable for the region.

[0014] According to this, it is possible to receive proposals for health promotion and / or long-term care prevention programs that are appropriate for the region, and thus provide useful information for planning health promotion and / or long-term care prevention programs in the target region.

[0015] Furthermore, the present invention is Steps include obtaining basic information including the number of people certified as requiring long-term care, the number of people certified as requiring support, and the number of people diagnosed with frailty in the target area during the first period, information specific to the area during the first period, and information on health promotion and / or long-term care prevention programs implemented in the area during the first period. A step of estimating the incidence rate of certified care-requiring persons, the incidence rate of certified support-requiring persons, and the incidence rate of frailty-diagnosed persons after the second period in the said region; An information processing method characterized by including this.

[0016] According to this, in the target region, since it is possible to recognize how the incidence rate of certified care-requiring persons, the incidence rate of certified support-requiring persons, and the incidence rate of frailty-diagnosed persons change after the second period, by providing information useful for the planning and evaluation of health promotion and / or care prevention business, the health promotion and / or care prevention business can be supported.

Effect of the Invention

[0017] According to the present invention, it is possible to provide information for supporting a health promotion business and / or a care prevention business.

Brief Description of the Drawings

[0018] [Figure 1] FIG. 1 is a conceptual diagram explaining the future prediction according to Example 1. [Figure 2] FIG. 2 is a diagram showing an example of prefectural data. [Figure 3] FIG. 3 is a diagram showing an example of municipal data. [Figure 4] FIG. 4 is a diagram showing an example of a questionnaire. [Figure 5] FIG. 5 is a diagram showing an example of analysis information displayed in three dimensions. [Figure 6] FIG. 6 is a diagram showing an example of analysis information displayed in two dimensions. [Figure 7] FIG. 7 is a schematic configuration diagram of a system including an information processing apparatus according to Example 1. [Figure 8] FIG. 8 is a diagram explaining a method for generating a future prediction model according to Example 1.

Mode for Carrying Out the Invention

[0019] 〔Example 1〕 The information processing device 100 according to Embodiment 1 of the present invention will be described in more detail below with reference to the drawings. However, the configuration of the device and system described in this embodiment should be appropriately modified depending on various conditions. In other words, the scope of this invention is not intended to be limited to the following embodiment.

[0020] Figure 1 is a diagram illustrating the information processing in the information processing device 100 according to Example 1. The information processing device 100 inputs data on the number of people certified as requiring long-term care, the number of people certified as requiring support, and the number of people diagnosed with frailty in the target prefecture or local government, as well as regional characteristics data and policy data (data related to projects) implemented by the local government regarding health promotion projects and long-term care insurance projects into the future forecasting unit 130. For example, data from fiscal year 2007 to fiscal year 2023 is input. In response to such input, the future forecasting unit 130 outputs the incidence rate of people diagnosed with frailty, the incidence rate of people certified as requiring support, and the incidence rate of people certified as requiring long-term care at least five years from now. Here, the input data is from fiscal year 2007 to fiscal year 2023, and the output data is five years from now, but it is possible to output data after an appropriate period has elapsed for input data of an appropriate period. As will be described later, the information output from the future forecasting unit 130 is not limited to the incidence rate of people diagnosed with frailty, the incidence rate of people certified as requiring support, and the incidence rate of people certified as requiring long-term care five years from now. The population of the target local government can be estimated from the regional characteristics data of the target prefecture or local government. Furthermore, while the incidence rates of frailty diagnosis, support needs certification, and long-term care certification are output here as indicators for evaluating the occurrence of frailty diagnosis, support needs certification, and long-term care certification, respectively, other evaluation indicators may also be used. Here, the period from 2007 to 2023 corresponds to the first period of the present invention, and the five years correspond to the second period of the present invention. Regional characteristic data corresponds to data specific to the region of the present invention.

[0021] Figure 2 shows an example of data related to prefectures among the regional characteristic data used as input data. Figure 3 shows an example of data related to municipalities among the regional characteristic data used as input data. This regional characteristic data is obtained from the Ministry of Internal Affairs and Communications' Housing and Land Survey and National Census, prefectural reports on regional health and health promotion projects, the Social Life Basic Survey and Public Health Administration Report, and the social and demographic statistics systems of prefectures and municipalities. Here, the data related to municipalities includes the number of people certified as requiring long-term care and the number of people certified as requiring support, but the region-specific data of the present invention corresponds to the information excluding these.

[0022] Figure 4 shows an example of a questionnaire for obtaining input data from the target municipalities. In the questionnaire shown in Figure 4, respondents answer by marking with a circle (○) the health promotion projects and long-term care prevention projects implemented by the target municipalities from fiscal year 2007 to 2023.

[0023] The content and sources of the data to be entered are not limited to those mentioned above; data from databases such as NDB (National Database: Receipt Information and Specific Health Checkup Information Database), KDB (National Health Insurance Database), and DPC (Diagnosis Procedure Combination) databases can also be used. You may do so.

[0024] Regional characteristic data includes the number of people certified as requiring long-term care and those certified as requiring support, as shown in the municipal data in Figure 3, as well as the number of people diagnosed with frailty. Frailty diagnosis includes frailty screening and basic checkups. Various diagnostic methods are used, such as checklists, the University of Tokyo Frailty Check, and the J-CHI criteria, and there is no standardized approach. Therefore, even if each local government conducts frailty assessments, the diagnostic methods may differ. However, the inventors have invented an AI that estimates the diagnostic results of a specific frailty assessment method, such as the J-CHI criteria, from the diagnostic results of frailty assessment methods such as frailty screening, basic checklists, and the University of Tokyo Frailty Check. By using this technology, even if the number of frailty assessment cases obtained from each local government using different frailty assessment methods, it is possible to convert this to the number of frailty assessment cases according to the J-CHI criteria before further processing.

[0025] Figure 5 shows an example of the display of data output from the future prediction unit 130. Here, the incidence rate of people certified as requiring support, the incidence rate of people certified as requiring long-term care, and the incidence rate of people diagnosed with frailty are plotted on three mutually orthogonal axes, and the graph is displayed in three dimensions. Here, the incidence rates of individuals certified as requiring support, individuals certified as requiring long-term care, and individuals diagnosed with frailty for a specific municipality from fiscal year 2007 to 2023 are plotted in a time series in a three-dimensional space. Then, the projected values ​​for the incidence rates of individuals certified as requiring support, individuals certified as requiring long-term care, and individuals diagnosed with frailty in fiscal year 2028 (five years later) are estimated and displayed by the future projection unit 130. The projected data for five years later can display projected data if the currently implemented health promotion programs and long-term care insurance programs are continued, but it can also display how the projected values ​​such as the incidence rate of frailty diagnosis in five years will change if the programs to be implemented in the future are entered. By obtaining information on the programs to be implemented in the future as basic information from the target municipality, the incidence rate of frailty diagnosis in five years will also change if those programs are implemented, thus providing information on predicting the effectiveness of the implemented programs. Furthermore, by obtaining information on the target municipality's target—how the projected values ​​such as the incidence rate of frailty diagnosis in five years should change—as basic information, it is possible to propose the optimal programs to achieve the target. Furthermore, since the incidence rates of individuals certified as requiring support and those certified as requiring long-term care are displayed in time series, this provides information to support the analysis and evaluation of health promotion and long-term care prevention programs that have already been implemented. For comparison, the graph may also display time-series data or single-year data for the incidence rates of individuals certified as requiring support, those certified as requiring long-term care, and those diagnosed with frailty in municipalities other than the municipality in question. Data from other municipalities may be displayed anonymized as needed. In addition, instead of displaying data for a single municipality, the data may be displayed averaged across higher-ranking municipalities or regions that include the municipality in question.

[0026] Figure 6 shows another example of how the data output from the future prediction unit 130 can be displayed. Here, the incidence rate of people requiring support and the incidence rate of people requiring long-term care are plotted on two mutually orthogonal axes, and the graph is displayed in two dimensions. Here, the incidence rates of individuals requiring support and those requiring long-term care for a specific municipality from fiscal year 2007 to 2023 are plotted in a two-dimensional space over time. Here, the predicted values ​​for the incidence rates of individuals requiring support and those requiring long-term care in fiscal year 2028 (five years later) are shown as white circles. Four regions, Region A, Region B, Region C, and Region D, are displayed, centered on the data for fiscal year 2023. Region A is the region where both the incidence rates of individuals requiring support and those requiring long-term care increase. Region B is the region where the incidence rate of individuals requiring support increases. Region C is the region where the incidence rate of individuals requiring long-term care increases. Region D is the region where both the incidence rates of individuals requiring support and those requiring long-term care decrease.

[0027] For example, if the program is not implemented from FY2023, the predicted value for FY2028 will be point P1, and both the incidence rate of those certified as requiring support and those certified as requiring long-term care will increase. If the home visit program is implemented from FY2023, the predicted value for FY2028 will be point P2, and the incidence rate of those certified as requiring support will increase. If the frailty check program is implemented from FY2023, the predicted value for FY2028 will be point P3, and the incidence rate of those certified as requiring long-term care will increase. From FY2023, education on nutrition and oral health will be implemented. Implementing the program will result in a projected value of point P4 for fiscal year 2028, with both the incidence rate of individuals requiring support and the incidence rate of individuals requiring long-term care decreasing. In this way, it is possible to predict in which direction the incidence rates of individuals requiring support and those requiring long-term care will change after 5 years, depending on which program is implemented. By predicting the effects of each implemented policy, it is possible to provide information that supports the planning of health promotion programs and long-term care prevention programs. Furthermore, since it is possible to predict in which direction the incidence rates of individuals requiring support and those requiring long-term care will change after 5 years, depending on which program is implemented, it is possible to propose the optimal policy according to the challenges and goals of the local government, such as reducing both the incidence rates of individuals requiring support and those requiring long-term care, thereby providing information that supports the planning of health promotion programs and long-term care prevention programs. In addition, since the incidence rates of individuals requiring support and those requiring long-term care are displayed in time series, it is possible to provide information that supports the analysis and evaluation of health promotion programs and long-term care prevention programs that have already been implemented.

[0028] Here, we have explained the graphs shown in Figure 6, which display the incidence rates of those certified as requiring support and those certified as requiring long-term care in two dimensions. However, the same principle applies to the three-dimensional graph shown in Figure 5, which includes the incidence rate of those diagnosed with frailty.

[0029] Figure 6 is a functional block diagram of a system including the information processing device 100 according to this embodiment. In this diagram, the information processing device 100, the user terminal 200, and the data source 300 are connected in a communication manner.

[0030] The user terminal 200 is a computer device managed and used by, for example, a local government or other organization that implements health promotion programs or elderly care prevention programs. The user terminal 200 is a general-purpose computer device that has a processing unit such as a CPU, main memory such as RAM and ROM, auxiliary storage such as EPROM, hard disk drive, and removable media, input units such as a keyboard, touch panel, and mouse, and output units such as a display and printer.

[0031] Data Source 300 is a database that records and provides open data such as the Ministry of Internal Affairs and Communications' Housing and Land Survey and National Census, prefectural reports on regional health and health promotion projects, social life basic surveys and public health administration reports, and prefectural and municipal social and demographic statistics systems. Data Source 300 may also include NDB, KDB, and DPC. Data Source 300 is a general computer device that has a processing unit such as a CPU, main memory such as RAM and ROM, and auxiliary storage devices for recording open data such as EPROM, hard disk drives, and removable media, and may have input units such as a keyboard, touch panel, and mouse, and output units such as a display and printer as needed.

[0032] The information processing device 100 is a computer device that has the function of providing information to support the planning or analysis of health promotion projects and long-term care prevention projects. The information processing device 100 is a general computer device that has an arithmetic unit such as a CPU, a main memory such as RAM and ROM and an auxiliary memory such as EPROM, a hard disk drive, and removable media, an input unit such as a keyboard, a touch panel, and a mouse, and an output unit such as a display. The information processing device 100 may consist of a single computer or may consist of multiple computers that cooperate with each other via a network.

[0033] The information processing device 100 includes a basic information acquisition unit 110, a preprocessing unit 120, a future prediction unit 130, and an analysis information generation unit 140. Each of these functional units is realized by loading a program stored in an auxiliary storage device into the main memory and executing it in the arithmetic unit.

[0034] The basic information acquisition unit 110 acquires policy data such as health promotion projects and long-term care prevention projects implemented by the user municipality via the network, and also acquires regional characteristic data from the data source 300. Here, the basic information is at least This includes, but is not limited to, policy data and regional characteristics data. The basic information acquisition unit 110 corresponds to the basic information acquisition unit of the present invention.

[0035] As described above, the preprocessing unit 120 has a function to preprocess data on the number of frailty diagnosed individuals so that it can be handled uniformly, for example, by estimating the number of frailty diagnosed individuals according to the J-CHS criteria, when the frailty diagnosis method differs from one municipality to another. Depending on the content of the information acquired by the basic information acquisition unit 110, preprocessing may be omitted as appropriate.

[0036] As shown in Figure 1, the future prediction unit 130 has the function of estimating the future incidence rates of people requiring support, people requiring long-term care, and people diagnosed with frailty in the target municipality, based on the basic information acquired by the basic information acquisition unit 110 or basic information obtained by preprocessing the basic information acquired by the basic information acquisition unit 110, using the future prediction model 130A. The future prediction unit 130 corresponds to the future prediction unit of the present invention.

[0037] The analysis information generation unit 140 generates the analysis information shown in Figures 4 and 5 based on the information predicted by the future prediction unit 130. The analysis information generation unit 140 corresponds to the analysis information generation unit of the present invention.

[0038] Figure 8 illustrates the method for generating the future prediction model 130A used in the future prediction unit 130 of the information processing device 100. The future prediction model 130A is trained using machine learning on healthcare-related data for each municipality, future population, future number of elderly people, and whether or not healthcare policies will be implemented each year. The healthcare-related data for each municipality includes data showing the annual trends in the number of people certified as requiring long-term care, the number of people certified as requiring support, the number of people diagnosed with frailty, and the population and number of elderly people by sex and age group. In addition, the healthcare-related data for each municipality includes data showing regional characteristics such as the annual trends in the number of medical facilities and nursing care facilities. By training a future prediction model that estimates and outputs the future incidence rates of people certified as requiring support, people certified as requiring long-term care, and people diagnosed with frailty using this data, a trained future prediction model 130A is generated. For the number of people diagnosed with frailty, it is also possible to use an AI that estimates the results of a specific diagnosis method, such as the J-CHS judgment criteria, from the frailty diagnosis results diagnosed by each municipality using an appropriate method. Alternatively, as a preprocessing step, the number of people diagnosed with frailty using the J-CHS judgment criteria may be estimated from the frailty diagnosis results diagnosed by each municipality using an appropriate method, and this estimation result may be used as training data to train the future prediction model.

[0039] The training data used to train the future prediction model 130A is not limited to that shown in Figure 7. It is preferable to train the model with the data shown in Figures 2 and 3 as local characteristic data for municipalities. The inventors' diligent research has revealed that not only medical expenses but also local characteristics (climate, industrial structure, disease incidence, local medical resources, local care resources, etc.) influence the occurrence of people certified as requiring long-term care or support. Therefore, by training the model with data that includes local characteristic data, it becomes possible to make more accurate future predictions that reflect local characteristics.

[0040] As described above, by using the future prediction model 130A, which was generated by training it with data on the annual trends in the number of people certified as requiring long-term care, it becomes possible to accurately estimate the future incidence rate of people certified as requiring long-term care based on regional characteristic data and implemented policy data over a predetermined period, as shown in Figure 1.

[0041] [Variation] In Figure 1, the future prediction unit 130 may estimate not only the incidence rate of frailty diagnosis, the incidence rate of people certified as requiring support, and the incidence rate of people certified as requiring long-term care, but also the healthy life expectancy, the number of deaths, and long-term care insurance premiums five years from now. In the above-described Example 1, we explained the case of providing information to support the planning of health promotion projects and long-term care prevention projects in local governments. However, by using machine learning on the utilization rate of local social resources and the participation rate of local events in the future prediction model 130A, this system can also be used to... It can also be used to propose appropriate lifestyle habits to individual residents, and to encourage participation in events such as community gatherings, health checkups for the elderly, and exercise classes. [Explanation of Symbols]

[0042] 100 Information Processing Devices 110 Basic information acquisition department 120 Pre-processing 130 Future Forecasting Department 140 Analysis information generation section 200 user terminals 300 data sources

Claims

1. A basic information acquisition unit acquires basic information including the number of people certified as requiring long-term care, the number of people certified as requiring support, and the number of people diagnosed with frailty in the target area during the first period, information specific to the area during the first period, and information on health promotion and / or long-term care prevention programs implemented in the area during the first period. A future prediction unit that estimates the incidence rate of those requiring long-term care, those requiring support, and those diagnosed with frailty in the aforementioned region after the second period, An information processing device characterized by having the following features.

2. The system further comprises an analysis information generation unit that generates analysis information based on the predicted incidence rate of persons requiring long-term care, the incidence rate of persons requiring support, and the incidence rate of persons diagnosed with frailty in the region after the second period, as predicted by the future prediction unit. The aforementioned analysis information is, The information processing device according to claim 1, characterized in that it includes information that displays in time series the incidence rate of persons certified as requiring long-term care, the incidence rate of persons certified as requiring support, and the incidence rate of persons diagnosed with frailty in the region during the first period, and the incidence rate of persons certified as requiring long-term care, the incidence rate of persons certified as requiring support, and the incidence rate of persons diagnosed with frailty in the region after the second period, as predicted by the future prediction unit.

3. The information processing device according to claim 2, characterized in that the analysis information includes predictions of the effects of the health promotion and / or long-term care prevention projects implemented in the region.

4. The information processing device according to claim 2, characterized in that the aforementioned analysis information includes proposals for health promotion and / or long-term care prevention programs suitable for the region.

5. Steps include obtaining basic information including the number of people certified as requiring long-term care, the number of people certified as requiring support, and the number of people diagnosed with frailty in the target area during the first period, information specific to the area during the first period, and information on health promotion and / or long-term care prevention programs implemented in the area during the first period. A step of estimating the incidence rate of those requiring long-term care, those requiring support, and those diagnosed with frailty in the aforementioned region after the second period, An information processing method characterized by including

Images