Information processing device, information processing method, and program for supporting measures against polypharmacy

Abstract

Appropriately select targets for polypharmacy countermeasures. [Solution] The program disclosed herein is a program that causes a computer to function as each means of an information processing device that supports polypharmacy countermeasures, and the information processing device includes an estimation means that estimates the health risk of each of multiple subjects based on information on the subjects' prescription medications, and a selection means that selects targets for polypharmacy countermeasures based on the health risks of the multiple subjects.

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and a program for supporting measures against polypharmacy. [Background technology]

[0002] In recent years, there has been a demand for measures to combat polypharmacy. Polypharmacy, which stands for "Poly (many) Pharmacy (medicines)," refers to the taking of multiple medications. However, it does not simply mean taking a large number of medications; it refers to the occurrence of adverse events such as side effects due to taking many medications. These adverse events include any undesirable or unintended signs, symptoms, or illnesses that occur in patients who have been administered a drug, including those with no clear causal relationship to the drug.

[0003] There is no strict definition of the number of medications required to constitute polypharmacy, and appropriate prescriptions vary depending on the patient's condition, lifestyle, and environment. Adverse drug events increase roughly in proportion to the number of medications, with some data suggesting that six or more medications are particularly associated with an increased incidence of adverse drug events. However, while six or more medications may be necessary for treatment, problems may arise with only three medications; the nature of the medication is essentially important. Therefore, when addressing polypharmacy, rather than focusing solely on a uniform number of medications / types, it is necessary to optimize prescription content from the perspective of ensuring safety, etc. (Non-Patent Document 1).

[0004] Patent Document 1 proposes a database server that extracts users from a database whose prescribed medication types and numbers meet specific extraction conditions. The proposed database server extracts, for example, users aged 75 or older who have been prescribed six or more medications for three months or more as users at risk of polypharmacy due to multiple drug administration. The proposed database server notifies each extracted user that they are at risk of polypharmacy and should consult a pharmacist about their medication prescriptions. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2019-175341 [Non-patent literature]

[0006] [Non-Patent Document 1] Ministry of Health, Labor and Welfare, "Guidelines for Appropriate Use of Medicines for the Elderly (Compilation)" [online], May 2018, retrieved September 2, 2024, Internet<URL=https: / / www.mhlw.go.jp / content / 11121000 / kourei-tekisei_web.pdf> Summary of the Invention [Problem to be solved by the invention]

[0007] The system proposed in Patent Document 1 can extract subjects for whom polypharmacy countermeasure guidance should be provided by extracting subjects whose types and numbers of prescribed medications meet extraction conditions. However, the extraction conditions are uniformly determined based on the types and numbers of medications, and it is not possible to extract subjects for whom polypharmacy countermeasure guidance should be provided in relation to conditions where the causal relationship between medication and the medication is unclear, such as dementia, a state requiring nursing care, or lifestyle-related diseases.

[0008] The present invention has been made in view of the above problems, and its purpose is to realize a technology that makes it possible to appropriately select subjects for whom polypharmacy countermeasures should be taken. [Means for solving the problem]

[0009] In order to solve this problem, for example, a program according to the present disclosure is a program that causes a computer to function as each means of an information processing device that supports polypharmacy countermeasures, and the information processing device includes: an estimation means that estimates health risks of each of a plurality of subjects based on information on prescription drugs of the plurality of subjects; a behavior change probability estimation means for estimating a behavior change probability due to an intervention for each of the plurality of subjects based on the health information of the plurality of subjects;the health risks of the plurality of subjects and the behavior change probability and a selection means for selecting targets for polypharmacy countermeasures based on the above. [Effects of the Invention]

[0010] According to the present invention, it is possible to appropriately select targets for polypharmacy countermeasures. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a diagram showing an example of a polypharmacy countermeasure support system according to an embodiment of the present invention. [Figure 2] FIG. 1 is a block diagram showing an example of the functional arrangement of an information processing apparatus according to an embodiment. [Figure 3] FIG. 2 is a block diagram showing an example of the functional configuration of a terminal device according to the embodiment. [Figure 4] FIG. 2 is a diagram illustrating an example of the functional configuration of a health information database according to the embodiment. [Figure 5] 1 is a flowchart showing a series of processes related to polypharmacy countermeasure support according to an embodiment. [Figure 6] 10 is a flowchart showing a process for generating an estimation model according to the embodiment. [Figure 7] FIG. 4 is a diagram illustrating a process for generating an estimation model according to the embodiment. [Figure 8] FIG. 4 is a diagram illustrating the operation of an estimation model according to the embodiment. [Figure 9] 1 is a flowchart showing a process for selecting a subject of a polypharmacy countermeasure according to an embodiment. [Figure 10] FIG. 2 is a diagram illustrating a behavior change probability estimation model according to the embodiment. [Figure 11] 10 is a flowchart showing a process for generating a behavior change probability estimation model according to an embodiment. [Figure 12] FIG. 10 is a diagram illustrating polypharmacy countermeasure support using behavioral change probability according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, the embodiments will be described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the scope of the invention as claimed, and not all combinations of features described in the embodiments are necessarily essential to the invention. Two or more of the features described in the embodiments may be arbitrarily combined. Furthermore, the same reference numerals are used for the same or similar components, and redundant explanations will be omitted.

[0013] First Embodiment <Configuration of the polypharmacy countermeasure support system> The configuration of a polypharmacy countermeasure support system according to an embodiment of the present invention will be described with reference to Figure 1. The polypharmacy countermeasure support system 100 includes, for example, an information processing device 101, a health information database 102, and a terminal device 103 used by a user. Here, the user is assumed to be an employee of an organization or institution that implements polypharmacy countermeasures, but is not limited to this.

[0014] Possible organizations and institutions that take measures against polypharmacy include local governments, insurers (such as national health insurance associations, health insurance societies, and insurers under the Medical Care System for the Elderly, hereinafter referred to as "insurers"), pharmacies, and medical institutions. In this specification, local governments, pharmacies, and medical institutions are collectively referred to as insurers, etc. Insurers, etc. include local governments, health insurance associations, pharmacies, and medical institutions, but may also include other organizations and institutions.

[0015] The information processing device 101, the health information database 102, and the terminal device 103 are communicably connected via a network 104. The network may be any network such as a LAN, a WAN, or the Internet. In addition, all or part of the network may be connected via wireless communication.

[0016] <Configuration of information processing device> The information processing device 101 is, for example, a server managed by an insurer, etc., or a server managed by a company that provides services related to polypharmacy countermeasures to insurers, etc. In the following explanation, no particular distinction is made between insurers, etc. and companies that provide services, and they will be simply referred to as insurers, etc.

[0017] The terminal device 103 is, for example, a user terminal that allows an insurer or the like to access the information processing device 101 when obtaining information related to polypharmacy countermeasures. In this embodiment, an example will be described in which the terminal device 103 is, for example, a desktop computer. However, the terminal device 103 is not limited to a desktop computer, and may be any other device that can access the information processing device 101, such as a laptop computer, a tablet terminal, or a smartphone.

[0018] The health information database 102 is a database that stores various types of medical information and health information data. There may be multiple health information databases 102 for each type of data, or there may be only one. The health information database 102 may be owned by an insurer or other organization. It may be the health insurance claim information / specific health checkup information database (NDB) provided by the Ministry of Health, Labor and Welfare, a database (such as KDB) owned by an examination and payment organization (such as the Social Insurance Medical Fee Payment Fund, the National Health Insurance Association, and the National Health Insurance Federation), or other databases owned by private companies. The information processing device 101 may also have a database.

[0019] <Functional configuration of information processing device> Next, an example of the functional configuration of the information processing device 101 will be described with reference to Fig. 2. Note that each of the functional blocks described may be integrated or separated, and the functions described may be realized by different blocks. Also, what is described as hardware may be realized by software, and vice versa.

[0020] The communication unit 201 includes a communication circuit that communicates with various devices via a network. The communication unit 201 receives information processed by the control unit 204 from a communication partner device (e.g., the terminal device 103, etc.), and transmits information processed by the control unit 204 to the communication partner device (e.g., the terminal device 103). The power supply unit 202 is a power supply that provides power required for the operation of the information processing device 101.

[0021] The storage unit 203 includes a non-volatile storage medium such as a hard disk or semiconductor memory, and stores various programs executed by the control unit 204 of the information processing device 101, various data used by the control unit 204, and a DB 230. The various programs include an operating system, a framework, a library, and the like, as well as a program for executing processing to support polypharmacy countermeasures according to this embodiment. The various data include, for example, setting values ​​of the information processing device 101, data acquired from the health information database 102, data related to estimation models, training data, and estimation result data. The storage unit 203 also stores parameters for each estimation model. Data acquired from the health information database 102 may be stored in the DB 230.

[0022] The control unit 204 includes a CPU 210, which is a central processing unit, and a RAM 211. The control unit 204 controls the operation of each unit within the control unit 204 and the operation of each unit of the information processing device 101 by loading a program stored in the storage unit 203 into the RAM 211 and executing it. The control unit 204 also performs estimation of polypharmacy and selection of subjects for polypharmacy countermeasures, which will be described later.

[0023] The RAM 211 includes a volatile storage medium such as a DRAM, and temporarily stores parameters and processing results for the control unit 204 to execute programs.

[0024] The control unit 204 has functional blocks that are implemented by the CPU 210 reading out a program stored in the storage unit 203 into the RAM 211. The control unit 204 has functional blocks of an estimation model unit 220, a segment selection unit 221, a learning unit 222, a data acquisition unit 223, a user interface (IF) unit 224, and the segment selection unit 221.

[0025] The estimation model unit 220 uses a polypharmacy estimation model to estimate the risks associated with an individual's medication intake. The polypharmacy estimation model has a function of outputting a future fracture risk (probability) when, for example, data on medications taken by an individual is input. In addition to fracture risk, the estimation model unit 220 can estimate one or more predetermined risks, such as nursing care risk, dementia risk, and lifestyle-related disease risk. Furthermore, the estimation model unit 220 uses a behavior change probability estimation model to estimate the probability of behavior change due to intervention.

[0026] The segment selection unit 221 selects a segment of subjects for guidance and intervention by the insurer, etc., based on the individual's risk of polypharmacy estimated by the estimation model unit 220. For example, subjects who have been prescribed three or more benzodiazepine derivatives are considered to be a segment with a high fracture risk, and notifications and medication guidance are provided to subjects belonging to that segment.

[0027] The learning unit 222 learns the polypharmacy estimation model and generates a learned polypharmacy estimation model. The learning unit 222 learns the polypharmacy estimation model using data acquired from the health information database 102. In learning the polypharmacy estimation model, for example, the prescription data 450 can be used as data on medications taken by an individual. Furthermore, data on fracture hospitalization or consultation due to fracture in the medical data 440 can be used as correct answer data regarding the probability of fracture risk. Furthermore, data collected in advance by a follow-up survey or the like can also be used to assess fracture risk.

[0028] The data acquisition unit 223 acquires data necessary for estimating polypharmacy and selecting recipients of polypharmacy guidance from the health information database 102 or other data sources. The data acquisition unit 223 may store the acquired data in a database (DB) 230 of the storage unit 203.

[0029] The user IF unit 224 functions as an interface with the terminal device 103 of the insurer, etc. The user IF unit 224 displays an input screen and a setting screen on the terminal device 103, and accepts data input from the terminal device 103.

[0030] The segment selection unit 221 selects segments or subjects to be targeted for polypharmacy countermeasures based on the results of estimation using the polypharmacy estimation model.

[0031] The notification unit 225 notifies the selected segment or subject about polypharmacy. The notification unit 225 may notify the subject directly, or may send information about the subject to the terminal device 103. The insurer or the like can notify the subject of polypharmacy countermeasures by mail or email after receiving the notification on the terminal device 103.

[0032] <Configuration of terminal device> An example of the functional configuration of the terminal device 103 of the insurer or the like will be described with reference to Fig. 3. Note that each of the functional blocks described may be integrated or separated, and the described functions may be realized by different blocks. Also, what is described as hardware may be realized by software, and vice versa.

[0033] The communication unit 301 includes, for example, a communication circuit, and communicates with the information processing device 101 via mobile communication such as wired LAN or LTE, or via wireless communication such as WiFi, to send and receive the necessary data.

[0034] The operation unit 303 includes buttons and a touch panel provided on the terminal device 103, and accepts operations by a user such as an insurer to display information related to polypharmacy countermeasures. The display unit 304 includes a display panel such as an LCD or OLED, and displays GUIs for various operations. For example, the display unit 304 can display the number and information of subjects generated by the segment selection unit 221 of the information processing device 101.

[0035] The storage unit 305 includes, for example, a nonvolatile memory such as an HDD or semiconductor memory, and stores programs executed by the control unit 502, etc.

[0036] The control unit 302 includes a CPU 310 and a RAM 311, and controls the operation of each functional block in the control unit 302 and each unit in the terminal device 103 by the CPU 310 executing a program recorded in the storage unit 305, for example.

[0037] <Health information database configuration> Next, an example of the functional configuration of the health information database 102 will be described with reference to FIG. 4. Each of the functional blocks described may be integrated or separated, and the described functions may be implemented by different blocks. Furthermore, what is described as hardware may be implemented by software, and vice versa. The health information database 102 is a database server that stores various health information data according to this embodiment. The health information database 102 includes a communication unit 401, a power supply unit 402, a storage unit 403, and a control unit 404. The communication unit 401 includes, for example, a communication circuit, and communicates with the information processing device 101 via mobile communication such as a wired LAN or LTE, or via wireless communication such as WiFi, to transmit and receive necessary data. The power supply unit 402 is a power source that provides the power required for the operation of the health information database 102.

[0038] The control unit 404 includes a CPU 410 and a RAM 411, and controls the operation of each functional block in the control unit 404 and each unit in the health information database 102, for example, by the CPU 410 executing a program recorded in the storage unit 403.

[0039] The storage unit 403 includes nonvolatile memory such as an HDD or semiconductor memory, and stores various types of health information data and programs executed by the control unit 502. The storage unit 403 stores health checkup data 420, dental examination (dental checkup) data 430, medical treatment data 440, prescription receipt data 450, nursing care certification data 460, and intervention data 470. Some or all of these may be included in the health information database 102. The health information database 102 may also include other data. Each data is recorded for each patient (insured person) in association with the date of diagnosis, etc. The medical treatment data 440 and prescription receipt data 450 can be acquired from medical fee receipt data. The medical treatment data 440 can be acquired from medical receipt, dental receipt, and nursing care receipt data. Medical receipts, dental receipts, and nursing care receipts are created for each patient, each treatment month, and by type of inpatient / outpatient, etc. The health information data 102 may store health checkup data 420, dental checkup (dental examination) data 430, medical treatment data 440, prescription data 450, nursing care certification data 460, and intervention data 470, linked to a ledger. In recent years, with the use of a My Number card as a health insurance card, prescription data has been recorded in association with the My Number (individual number). Therefore, using the My Number, it is possible to obtain comprehensive data related to personal medical information for each individual, such as records of the health checkup data 420, dental checkup data 430, medical treatment data 440, prescription data 450, and nursing care certification data 460, as well as vital data obtained from a wearable device or the like, and electronic medical record data.

[0040] The health information database 102 may also acquire and use data from the Ministry of Health, Labor and Welfare's Health Insurance Claim Information and Specific Health Checkup Information Database (NDB), databases (such as the KDB) owned by screening and payment agencies (such as the Social Insurance Medical Fee Payment Fund, the National Health Insurance Association, and the National Health Insurance Federation), and other databases owned by private companies. The health information database 102 may also be the NDB, the KDB, or other databases owned by private companies. In the NDB, patient (insured) information is anonymized, but different types of patients are linked by the same hash ID. The data acquisition unit 223 of the information processing device 101 may also be configured to access the NDB to acquire data. The learning unit 222 and segment selection unit 221 of the information processing device 101 can use data from the NDB, the KDB, and other databases.

[0041] Each data item will be explained below. The health checkup data 420 is data on health checkups for a certain population. Municipalities and corporate health insurance associations conduct health checkups for their residents and members. The population of the health checkup data 420 may be a single municipality or a single company. The population of the health checkup data 420 may also be multiple municipalities or multiple companies. In the following, an example will be described in which the insurer is a company that provides services related to polypharmacy countermeasures, and uses the health checkup data 420 from multiple municipalities, multiple health insurance associations, etc. after obtaining permission to use the data. The same applies to the populations of the dental checkup data 430, medical treatment data 440, prescription data 450, nursing care certification data 460, and intervention data 470 described below.

[0042] The health checkup data 420 is data on health checkups that a subject has undergone, associated with the subject and accumulated by year. The health checkup data 420 stores information on whether the subject has undergone a health checkup, as well as measurement values ​​for each health checkup item. The health checkup data 420 includes interview (questionnaire) data that the subject answers in advance before the health checkup. The health checkup data includes specific health checkup data, which is data on the results of health checkups conducted for subjects aged 40 to 74 with a focus on metabolic syndrome, and late-stage elderly health checkup data conducted for subjects aged 75 or older. The health checkup data 420 may also include health checkup data conducted by insurers and other organizations for their employees in accordance with the Industrial Safety and Health Act. The health checkup data 420 includes data such as examinee information, specific health checkup result information, interview (questionnaire) information (medication, smoking history, etc.), whether the subject meets the metabolic syndrome criteria, and whether the subject meets the eligibility criteria for specific health guidance. Of the 420 health checkup data of the subjects, specific health checkup test items include health questions (medication history, smoking history), questionnaires, anthropometric measurements (height, weight, BMI, waist circumference), physical examination (physical examination), urinalysis (urine glucose, urinary protein), blood tests (lipids (triglycerides, HDL cholesterol, LDL cholesterol), glucose metabolism (fasting blood glucose or hemoglobin A1C), and liver function (GOT, GPT, γ-GTP), and additional items include anemia tests (red blood cell count, hemoglobin content, hematocrit value), electrocardiogram, fundus examination, and renal and urinary tract tests.

[0043] The dental examination data 430 is data on dental examinations that is associated with the subject and accumulated by year. The dental examination data 430 may also include data on dental examinations that are regular health checkups and specific health checkups. The dental examination data 430 accumulates whether the subject has undergone a dental examination or not, as well as the results of each dental examination item. The dental examination data 430 includes questionnaire data that the subject answers when undergoing a dental examination.

[0044] The medical data 440 is data that records the subject's visits to medical institutions and their admissions and discharges. It can be obtained from medical and dental receipt data created for each inpatient and outpatient visit. The medical data 440 includes the name of the medical institution, medical history, date of treatment, and name of medical procedure.

[0045] The dispensing receipt data 450 includes information on drugs dispensed at a dispensing pharmacy, information on the prescribing source, and information on the patient. The dispensing receipt data 450 is associated with each patient. The dispensing receipt data 450 includes the name of the medical institution or pharmacy, the dispensing date, the drug name, the ingredient name, the usage, the dosage, etc.

[0046] The nursing care certification data 460 is data of information related to nursing care certification. Nursing care certification is determined by a nursing care certification review board attached to a local government, based on an application for review, by calculating the standard time for nursing care certification etc. for five areas (direct assistance for daily living, indirect assistance for daily living, BPSD-related activities, functional training-related activities, and medical-related activities) and determining whether a person needs support level 1 to level 5 based on the total of the calculated time and the dementia surcharge. The nursing care certification data 460 may also include "independent," which is a state in which support such as nursing care services is not required for daily living. The nursing care certification data 460 may also include support level data.

[0047] The intervention data 470 includes data recording measures implemented by insurers and other organizations, correlating the implementer, implementation date, and details, personal medical data obtainable from electronic medical records, and vital data obtainable from wearable devices and other devices. Interventions are divided into individual and group interventions. An example of individual intervention is intervention that provides individual notification and guidance to subjects at high risk of adverse events due to polypharmacy (hereinafter referred to as "polypharmacy risk"). An example of group intervention is intervention that provides notification and guidance to a segment (group) at high risk of polypharmacy. Possible notifications and guidance include encouraging individuals to consult a pharmacist, providing nutritional and exercise guidance for those at high risk of fracture, supporting dementia prevention measures for those at high risk of dementia, and supporting independence measures for those at high risk of needing care. This polypharmacy countermeasure support system handles both group and individual interventions.

[0048] <Machine learning model> The polypharmacy estimation model for estimating polypharmacy risk according to this embodiment may use a machine learning model. Dispensing receipt data 450 from the health information database 102 is used as an explanatory variable. The machine learning model is trained using information on prescribed drugs in the dispensing receipt data 450. Data from the health information database 102 other than the dispensing receipt data 450 may be used. Using data other than the dispensing receipt data 450 can improve estimation accuracy. For example, if health checkup data 420, medical treatment data 440, etc. are used in addition to the dispensing receipt data 450, data on current illnesses, past illnesses, etc. can be used along with drug data. Polypharmacy can be estimated taking into account injury, disease, and contraindications, etc., thereby improving estimation accuracy.

[0049] The objective variables of the machine learning model include fracture risk, dementia risk, nursing care risk, and lifestyle-related disease risk. For training the machine learning model, for example, in the case of fracture risk, the dispensing receipt data 450 and medical data on fracture hospitalization or medical treatment due to fracture are required. For example, in the case of dementia risk, the dispensing receipt data 450 and dementia diagnosis data from the medical data 440 are required. Furthermore, for example, in the case of nursing care risk, the dispensing receipt data 450 and nursing care certification data from the nursing care certification data 460 are required. Furthermore, for example, in the case of lifestyle-related disease risk, the dispensing receipt data 450 and data on lifestyle-related disease hospitalization and diagnosis from the medical data 440 are required.

[0050] <Overall flow of this system> The overall flow of this embodiment will be described with reference to Fig. 5. The flowchart in Fig. 5 is implemented by the CPU 210 of the information processing device 101 loading a program stored in the storage unit 203 into the RAM 211 and executing it. In the following, the step numbers of each process included in the flowchart are indicated by numbers beginning with "S". This also applies to the subsequent flowcharts.

[0051] In S501, the learning unit 222 of the information processing device 101 generates a polypharmacy estimation model. The polypharmacy estimation model is a model that inputs prescription data 450 and the like for an individual subject and outputs the polypharmacy risk as a probability. Details of S501 will be described later.

[0052] Next, in S502, the estimation model unit 220 of the information processing device 101 performs a process of using a polypharmacy estimation model to obtain the probability of polypharmacy risk for each individual in a group of subjects. Based on the individual polypharmacy risk estimated by the estimation model unit 220, the segment selection unit 221 determines segments of subjects for which polypharmacy countermeasures are to be implemented. For example, segments with a high dementia risk are determined in order of drug prescription patterns. Details of S502 will be described later.

[0053] The polypharmacy estimation model of this embodiment can obtain the polypharmacy risk of an individual as a probability. It is also possible to determine who will be the target of polypharmacy risk countermeasures in order of individuals with the highest polypharmacy risk. One reason for calculating risk by segment is that if polypharmacy countermeasures are implemented in order of individuals with the highest polypharmacy risk, it would be difficult for insurers and others to understand what countermeasures should be implemented, since the reasons for the high risk vary from individual to individual. By calculating risk by segment and determining who will be the target of countermeasures, the same countermeasures and guidance can be implemented for all individuals belonging to the segment, improving the efficiency of implementing countermeasures and guidance.

[0054] Finally, in S503, the notification unit 225 of the information processing device 101 notifies the high-risk segments or high-risk subjects generated by the segment selection unit 221. Then, the user IF unit 224 transmits information about the selected segments or subjects to the terminal device 103. The terminal device 103 can display information about the segments or subjects with a high risk of polypharmacy on the display unit 304. By examining the information about subjects who are subject to polypharmacy countermeasures, insurers and the like can determine which segments of subjects require polypharmacy countermeasures. The content notified by the notification unit 225 of the information processing device 101 may include not only information about the segments or subjects with a high risk of polypharmacy, but also effective countermeasures for eliminating polypharmacy for the segments or subjects. For example, in addition to eliminating multiple medications, if the risk of fracture is high, measures such as recommending medical examinations such as bone density tests, starting osteoporosis medication and providing medication guidance, recommending exercise and improving nutrition, if the risk of dementia is high, measures to prevent dementia, if the risk of nursing care is high, measures to support independence, if the risk of lifestyle-related diseases is high, etc. are possible. Details of S503 will be described later.

[0055] <Polypharmacy> Before explaining the polypharmacy estimation model, we will explain polypharmacy. As mentioned above, polypharmacy refers to adverse events such as side effects caused by taking many medications. The main causes of polypharmacy are (1) drug combinations and (2) prescriptions from multiple medical institutions.

[0056] There are three cases of drug combinations in polypharmacy. The first is when drugs with the same ingredients and the same effects are prescribed, resulting in overlapping ingredients or effects. The second is when drugs are contraindicated for certain illnesses, meaning that patients with certain illnesses should be careful when taking them. The third is when drugs are contraindicated for co-administration, meaning that there is a possibility of drug interactions, and a combination of drugs is prescribed.

[0057] One cause of polypharmacy due to multiple medical institutions is when a patient suffers from multiple illnesses and receives prescriptions for each illness from multiple medical departments at different medical institutions. Another cause is when a patient visits a medical institution with symptom A, receives a prescription for drug X, experiences an adverse event from drug X, visits another medical institution, receives a prescription for drug Y, and then visits another medical institution with an adverse event from drug Y, repeating this process, creating a condition known as a prescription cascade, which leads to polypharmacy.

[0058] Another point to consider is the long-term prescription of medicines containing ingredients that can lead to tolerance and dependency with long-term use. Long-term prescriptions are also called careless prescriptions, and if these ingredients are prescribed continuously for a long period of time and adverse events with the same ingredients or similar effects occur at the same time, the risk of overprescription increases. Taking medication for a long period of time can also lead to tolerance and dependency. Taking these things into consideration, if there are any unused doses or if the medication is not suited to the patient's current symptoms, the prescription may be reviewed.

[0059] This polypharmacy countermeasure support system prioritizes polypharmacy cases and selects those for whom countermeasures should be prioritized. Pharmacists can detect identical ingredients, identical drugs, and contraindications for combined use if the medication notebook is properly managed. On the other hand, contraindications for injuries and illnesses require information on the patient's diagnosis, making it difficult to identify them from the medication notebook if the patient does not self-report them. This polypharmacy countermeasure support system prioritizes polypharmacy based on the severity of adverse event risk and whether it can be detected by other means, such as the medication notebook. For example, polypharmacy can be detected by prioritizing (1) identical ingredients, identical drugs, and contraindications across medical institutions, (2) contraindications for injuries and illnesses across medical institutions, (3) contraindications for combined use across medical institutions, (4) contraindications for injuries and illnesses at the same medical institution, (5) contraindications for combined use at the same medical institution, (6) multiple drugs at the same medical institution, and (7) long-term prescriptions at the same medical institution.

[0060] <Learning data used in the polypharmacy estimation model> The "Guidelines for Safe Drug Therapy for the Elderly" (Japan Geriatrics Society 2015) recommends that both benzodiazepine and non-benzodiazepine hypnotics not be administered long-term due to the risk of falls and fractures. However, many elderly people continue to take hypnotics long-term or multiple medications. In some cases, prescriptions continue to be given to patients who want medication because they cannot sleep.

[0061] As an example, this polypharmacy countermeasure support system first focuses on subjects who have been prescribed hypnotics. Furthermore, according to Non-Patent Document 1, benzodiazepine hypnotic analgesics carry risks such as over-sedation, deterioration of cognitive function, decreased motor function, falls, fractures, and delirium, and are therefore considered to be drugs that require particularly careful administration to elderly people.

[0062] In addition, the Japanese Anticholinergic Risk Scale, published by the Japanese Society of Geriatrics Medicine on May 17, 2024, lists drugs requiring particularly careful administration and drugs that should be avoided in the elderly for drug-induced geriatric syndromes (potentially inappropriate medications, PIMs). Typical PIMs include benzodiazepines and anticholinergics, both of which can cause or exacerbate various geriatric syndromes. Benzodiazepines, on the other hand, are often limited in their use to induce sleep or reduce anxiety, making them easy to identify from medication information. In contrast, anticholinergics are widely used in multiple organ and disease settings, ranging from those with primary anticholinergic activity to those with anticholinergic effects as a side effect, making them easily overlooked. Furthermore, in polypharmacy situations, multiple anticholinergics, varying in potency, are often included, making it difficult to assess their overall anticholinergic activity at a glance.

[0063] The Japanese Anticholinergic Risk Scale recommends assessing two aspects of risk. The first is a risk assessment of individual drugs. The strength of the risk due to the anticholinergic effects of each drug is rated on a scale of 1 to 3, with risk scale 3 being the highest risk category. If a drug with a high risk scale is being used, consideration should be given to switching to a drug with a lower risk scale. The second is a comprehensive risk assessment. Elderly people often suffer from multiple diseases and are prescribed multiple drugs. The risk scale scores for each drug are added together to calculate the patient's total anticholinergic burden. This makes it possible to understand the risk due to the anticholinergic effects of drug therapy as a whole. It has been shown that the total anticholinergic burden increases with the duration of use, so it is advisable to check the duration of use as well and intervene to reduce the total anticholinergic burden.

[0064] From these perspectives, the polypharmacy countermeasure support system of this embodiment first focuses on sleeping pills that are known to pose a risk of falls and fractures, and then, among sleeping pills, focuses on anti-anxiety drugs, benzodiazepine derivatives used as sleeping pills, and SSRIs and SNRIs used as antidepressants, which are likely to lead to dizziness, falls, etc. in the elderly.

[0065] GABA receptors, which are activated by GABA (gamma-aminobutyric acid), an inhibitory neurotransmitter in the brain, are deeply involved in anxiolysis, hypnosis, and sedation, and form a complex with benzodiazepine (BZD) receptors. When BZD receptors are stimulated, chloride ions (Cl) are released into this complex. - ) flows into the brain, suppressing brain excitement and producing anti-anxiety, hypnotic and sedative effects. Benzodiazepine derivatives act on the BZD receptor and by binding to this receptor increase the activity of GABA, producing anti-anxiety, hypnotic and sedative effects, improving symptoms such as anxiety disorders and psychosomatic disorders.

[0066] SSRIs are an abbreviation for "selective serotonin reuptake inhibitors." SSRIs are thought to increase serotonin levels in the brain by inhibiting the reuptake of released serotonin into cells, smoothing neurotransmission and exerting antidepressant and anxiolytic effects. They are typically used to treat depression and depressive states in adults, and are also indicated for obsessive-compulsive disorder, social anxiety disorder, panic disorder, and post-traumatic stress disorder. SNRIs are an abbreviation for "serotonin-noradrenaline reuptake inhibitors." SNRIs are thought to increase serotonin and noradrenaline levels in the brain by inhibiting the reuptake of released serotonin and noradrenaline into cells, smoothing neurotransmission and exerting antidepressant and anxiolytic effects. Among antidepressants, SSRIs primarily enhance serotonin, while SNRIs increase both serotonin and noradrenaline.

[0067] As an example, this polypharmacy countermeasure support system focuses on benzodiazepine derivatives and SSRIs / SNRIs, which are likely to lead to factors such as falls, and in addition to these, it uses a risk scale score of 1 to 3 on anticholinergic drugs.

[0068] First, regarding the sleeping pills that are the focus of this polypharmacy countermeasure support system, the definition of sleeping pills in the National Health Insurance Database (KDB) and the anticholinergic drug risk scale are shown in Table 1 below. [Table 1]

[0069] Next, Table 2 below shows the definition of benzodiazepine-type hypnotics, which are recommended to be avoided especially for elderly people, and the anticholinergic risk scale. [Table 2]

[0070] Furthermore, the definitions of antidepressants SSRIs and SNRIs and the anticholinergic risk scale are shown in Table 3 below. [Table 3]

[0071] <Generation of polypharmacy estimation model> Details of S501 in Fig. 5 will be described with reference to Fig. 6 to Fig. 8. First, using estimation of fracture risk as an example, the generation of a polypharmacy estimation model will be described with reference to the flowchart in Fig. 6 and the relationship between data and models in Fig. 7. The flowchart in Fig. 6 is implemented by the CPU 210 of the information processing device 101 reading a program stored in the storage unit 203 into the RAM 211 and executing it.

[0072] Here, the medical care data 440 and the dispensing receipt data 450 are used as learning data. In S601, the data acquisition unit 223 of the information processing device 101 acquires the dispensing receipt data 450 for fiscal year 2021 from the health information database 102 and stores it in the storage unit 203.

[0073] Next, in S602, the data acquisition unit 223 of the information processing device 101 acquires the medical data 440 for fiscal years 2021 and 2022 from the health information database 102 and stores it in the memory unit 203.

[0074] Next, in S603, the learning unit 222 of the information processing device 101 identifies subjects who were hospitalized for a fracture in fiscal year 2021 from the medical data 440. Then, the learning unit 222 excludes data of subjects who were hospitalized for a fracture in fiscal year 2021 from the dispensing receipt data 450 for fiscal year 2021 to obtain the population. Subjects who were hospitalized for a fracture can be obtained from the medical data 440. Here, subjects who were hospitalized for a fracture in fiscal year 2021 were excluded, but subjects who visited the hospital for a fracture in fiscal year 2021 may also be excluded.

[0075] Next, in S604, the learning unit 222 identifies subjects who were newly hospitalized for a fracture in fiscal year 2022 from the population based on the medical data 440 for fiscal year 2022. Here, subjects who were newly hospitalized for a fracture were identified, but it is also possible to identify subjects who were newly hospitalized for a fracture in fiscal year 2022.

[0076] Next, in S605, the learning unit 222 trains the machine learning model to learn prescription drug patterns from the 2021 prescription receipts of subjects who will be newly hospitalized for a fracture in 2022 for the aforementioned population. This allows the machine learning model to learn which prescription drug patterns subjects have a high risk of fracture. Training the machine learning model using data on fracture hospitalizations makes it possible to estimate the risk of being hospitalized for a fracture. Since the risk of being hospitalized for a fracture can also be said to be the risk of having a fracture, training the machine learning model using data on fracture hospitalizations makes it possible to estimate the risk of fracture. The well-known gradient boosting is used as the learning algorithm, but the learning algorithm is not limited to this.

[0077] In the above learning, 450 prescription data from fiscal year 2021 and 440 medical data from fiscal years 2021 and 2022 were used, but any fiscal year can be used for the health data. Furthermore, the period is not limited to fiscal year, as long as it is data from one period to the next. It can be a calendar year, or a period of two years or more. It can also be in monthly or monthly units. A longer period allows for prescription drug patterns to be obtained based on prescription drug data over a longer period. In the above explanation, information on hospitalization due to fracture was extracted from medical data 440. Information on hospitalization due to fracture can also be obtained from hospitalization prescription data. Information on hospitalization due to fracture can also be obtained from other prescriptions, such as follow-up surveys.

[0078] The machine learning model is trained on the characteristics of prescription medication patterns of subjects who are newly hospitalized for a fracture or who have received medical treatment for a fracture, and an estimation model for estimating fracture risk due to polypharmacy is generated (S606).

[0079] Furthermore, if patterns of health checkup data 420, dental checkup data 430, medical treatment data 440, and nursing care certification data 460 are used as training data for the machine learning model in addition to the dispensing receipt data 450, an estimation model that enables more accurate estimation of fracture risk can be generated. The prescription drug data in the dispensing receipt data 450 is essential, and the other data to be used is optional.

[0080] As other data, by using questionnaire data from the medical care data 440 and health checkup data 420, data on current illnesses and past medical conditions can be obtained. This makes it possible to estimate polypharmacy, which is a contraindication to certain illnesses. In this way, by using data other than prescription drug data, the accuracy of estimating the risk of polypharmacy can be improved.

[0081] Other specific data used include BMI from health checkup data, items from the health checkup questionnaire, one of the following illness names from the prescription data: cerebrovascular disease, ischemic heart disease, dementia, and fractures, whether or not a patient has visited a dentist from dental prescription data, and a questionnaire from the frailty health checkup.

[0082] The medical interview (questionnaire) data included in the health checkup data 420 also includes information on the subject's responses to questions such as whether or not they use medications, including blood pressure-lowering medications, insulin injections or blood sugar-lowering medications, cholesterol-lowering medications, medical history such as stroke, heart disease, and kidney failure, smoking habits, weight gain since age 20, exercise habits, eating habits, drinking habits, and sleep status. Because this medical interview (questionnaire) data includes information on past medications, lifestyle habits, and medical history, it can improve the accuracy of polypharmacy estimation models. For example, people who do not exercise regularly may have low bone density and tend to have a higher fracture risk. Furthermore, drinking habits increase the risk of falls due to alcohol consumption, which tends to increase the fracture risk. Combining exercise habits and drinking habits with the prescription data 450 can improve the accuracy of fracture risk estimation.

[0083] The polypharmacy estimation model described above uses benzodiazepine derivatives, SSRIs / SNRIs, and anticholinergics as examples of prescription medications. The estimation model may also be trained using one or more of the following prescription medication examples: sedative hypnotics / anti-anxiety medications, antidepressants (including sulpiride), medications for BPSD, medications for hypertension, medications for diabetes, medications for dyslipidemia, anticoagulants, and medications for peptic ulcers, to estimate the risk of polypharmacy.

[0084] Some BPSD medications are directly related to the state of dementia and are linked to the risk of dementia and fractures. Medications for hypertension, diabetes, and dyslipidemia are directly related to lifestyle-related diseases and are linked to the risk of lifestyle-related diseases. Medications for hypertension, diabetes, and dyslipidemia, which are related to lifestyle-related diseases, are likely to be taken over a long period of time, and in combination with other types of medication, often cause polypharmacy. Anticoagulants are directly related to cerebral infarction and are linked to the risk of lifestyle-related diseases. Anticoagulants are contraindicated in subjects at risk of bleeding, and are therefore contraindicated in cases of injury or illness, and can cause polypharmacy. Medications for peptic ulcers may contain anticholinergics, which can cause polypharmacy.

[0085] FIG. 8 shows the operation of the polypharmacy estimation model generated by the learning unit 222. The polypharmacy estimation model generated by the flow of FIG. 6 receives input of information on prescribed medications from prescription data for a certain subject in a certain year, and outputs the subject's fracture risk (probability) for the following year (within one year). The input and output of the prediction model do not necessarily have to match the training data used to train the prediction model. For example, the prediction model may be trained using five years' worth of data, and medication information for a certain subject at a certain time may be input to output the fracture risk for the following year (within one year). Furthermore, the output of the prediction model is not limited to the fracture risk within one year, and may be for a period shorter or longer than one year.

[0086] <Polypharmacy estimation model for estimating dementia risk> For the polypharmacy estimation model that estimates dementia risk, the learning unit 222 trains the polypharmacy estimation model by treating subjects in the medical data 440 who were not diagnosed with dementia in fiscal year 2021 but were diagnosed with dementia in fiscal year 2022 as subjects who newly developed dementia, instead of using fracture hospitalization data. The rest of the model is the same as the fracture risk estimation model. There is no known clear causal relationship between taking sleeping pills and the onset of dementia. A machine learning model can estimate risks such as these, where the causal relationship is unknown.

[0087] <Polypharmacy estimation model for estimating nursing care risk> For the polypharmacy estimation model that estimates nursing care risk, the learning unit 222 trains the estimation model by using, instead of fracture hospitalization data, individuals who were not certified as requiring nursing care in FY2021 but were certified as requiring nursing care level 2 or higher in FY2022 in the nursing care certification data 460 as subjects who have newly become required to care. Newly certified individuals requiring nursing care may be individuals certified as requiring nursing care level 1 or higher, or individuals certified as requiring nursing care level 3 or higher. They may also be individuals with a level of support required. Physical frailty data may also be included. The rest is the same as the fracture risk estimation model. The accuracy of the estimation by the polypharmacy estimation model can be improved due to the number of subjects and the greater change in condition for individuals requiring nursing care level 2 compared to individuals requiring nursing care level 1.

[0088] <Polypharmacy estimation model for estimating lifestyle-related disease risk> For the polypharmacy estimation model estimating lifestyle-related disease risk, the learning unit 222 trains the estimation model by considering subjects in the medical data 440 who were not hospitalized for lifestyle-related diseases in fiscal year 2021 but were hospitalized for lifestyle-related diseases in fiscal year 2022 as subjects newly hospitalized for lifestyle-related diseases, instead of fracture hospitalization data. Alternatively, a prediction model may be trained for subjects in the medical data who were not treated for lifestyle-related diseases in fiscal year 2021 but were treated for lifestyle-related diseases in fiscal year 2022 as subjects newly diagnosed with lifestyle-related diseases. Furthermore, a prediction model may be trained by comparing medical expenses related to lifestyle-related diseases between fiscal year 2021 and fiscal year 2022, and considering subjects whose medical expenses increased by more than a certain amount or percentage as subjects newly diagnosed with lifestyle-related diseases or whose lifestyle-related diseases became more severe. Based on a comparison of medical expenses, data on subjects whose lifestyle-related diseases became more severe can be used as training data. Furthermore, based on a comparison of medical expenses, data on all subjects can be used as training data without excluding those previously diagnosed with lifestyle-related diseases from the population. Other aspects are similar to those of the fracture risk estimation model.

[0089] <Target for polypharmacy countermeasures> The detailed processing of S502 in Fig. 5 will be described with reference to Fig. 9. The flowchart in Fig. 9 is implemented by the CPU 210 of the information processing device 101 reading a program stored in the storage unit 203 into the RAM 211 and executing it.

[0090] This section explains how the polypharmacy estimation model estimates fracture risk. Insurers and other entities can implement polypharmacy countermeasures for segments or subjects with a high fracture risk. Because fracture risk can easily lead to a bedridden state and a need for nursing care, as an example, subjects with a high fracture risk are targeted for polypharmacy countermeasures.

[0091] By using a polypharmacy estimation model that estimates fracture risk, it is possible to estimate the fracture risk (probability) for the following year (within one year) by inputting the prescribed medications from the prescription data of a subject for a certain year. In addition to the prescribed medications, disease data from the medical care data 440 and medical history and lifestyle habits from the interview (questionnaire) data of the health checkup data 420 may also be input to the polypharmacy estimation model. Furthermore, the estimation period is not limited to within one year, and may be shorter or longer than one year.

[0092] Here, the population for which polypharmacy is estimated does not necessarily have to be the same as the population of subjects used to generate the polypharmacy estimation model. Since accuracy is generally required for generating an estimation model, the more data there is, the better. On the other hand, risk estimation using an estimation model requires the selection of subjects taking into account factors such as the size of the insurer and the composition of subjects, and therefore this is the population against which insurers, etc., are taking measures against polypharmacy.

[0093] In S901, the estimation model unit 220 of the information processing device 101 first estimates the fracture risk (probability) for each individual in the next year for all subjects of the polypharmacy countermeasure project using the polypharmacy fracture risk estimation model. Here, persons prescribed sleeping pills are the population of subjects of the polypharmacy countermeasure project.

[0094] Next, in S902, the segment selection unit 221 of the information processing device 101 calculates the fracture risk for each segment by averaging the fracture risks of the subjects belonging to each segment. Here, the average value of the subjects' probabilities within the group is calculated, but other values ​​such as the median or mode may also be used. As a result of calculating the average value, the fracture risk for each segment is obtained.

[0095] Here, a segment is a group of subjects divided according to certain conditions. In the polypharmacy estimation model, prescription drugs from the dispensing receipt data 450 are used as input data. Each of the prescription drug patterns from the dispensing receipt data 450 is divided into multiple groups according to conditions, and each segment is a pattern. A prescription drug pattern is a pattern of the type and number of drugs prescribed.

[0096] As an example, Table 4 shows the relationship between the number of types of benzodiazepine derivatives prescribed and fracture risk for patients prescribed sleeping pills. Table 4 shows that the more types of benzodiazepine derivatives prescribed, the higher the fracture risk. Although the number of people prescribed three or more types of benzodiazepine derivatives is small in proportion, they are at high risk of fracture. [Table 4]

[0097] Next, as an example, Table 5 shows the relationship between the number of anticholinergic drugs (risk scale 3) prescribed and fracture risk, with subjects prescribed sleeping pills. Even if only one type of anticholinergic drug is prescribed in the same month as a sleeping pill, it can be said that the fracture risk is high. For subjects prescribed two or more types, although the percentage is small, the fracture risk is even higher. [Table 5]

[0098] Next, as an example, Table 6 shows the relationship between the number of SSRIs and SNRIs prescribed and fracture risk for subjects prescribed sleeping pills. Although there are few subjects who are prescribed sleeping pills and SSRIs or SNRIs at the same time, it can be said that their fracture risk is high. [Table 6]

[0099] Next, in S903, the segment selection unit 221 of the information processing device 101 prioritizes the implementation of polypharmacy countermeasures for each segment. Furthermore, if the risk of a segment is low, if the number of subjects belonging to the segment is small, or if the proportion of subjects to the total number of subjects is small, multiple segments may be integrated.

[0100] For example, segments in Table 4 with one or more, two or more, or three or more prescribed types of benzodiazepine derivatives may be combined into one segment. The segments can be combined taking into account the difference in average risk for the number of prescribed types of benzodiazepine derivatives and the proportion of the number to all subjects.

[0101] For example, segments may be grouped by the number of prescription types of anticholinergic drugs in Table 5 being 1 or more, 2 or more, or 3 or more. The method of grouping segments may be determined according to the average risk of the number of prescription types of anticholinergic drugs and the ratio to all subjects.

[0102] It is also possible to combine SSRIs and SNRIs in Table 6. Both SSRIs and SNRIs are antidepressants, and because they both have the effect of enhancing serotonin, they tend to have similar fracture risks. It is not limited to SSRIs and SNRIs, but it is also possible to combine segments of multiple drugs that have similar drug actions and similar average risks.

[0103] Among the subjects prescribed the sleeping pills shown in Tables 4 to 6, those prescribed SSRIs / SNRIs, anticholinergics, and benzodiazepine derivatives were found to have a higher risk of fracture.

[0104] Next, in S904, the segment selection unit 221 of the information processing device 101 selects a priority for each segment and selects a segment for which polypharmacy countermeasures will be implemented. The segments for which countermeasures will be implemented are selected based on Table 10, in descending order of fracture risk, from segments based on the number of SSRI / SNRI prescriptions and segments based on the number of anticholinergic drug prescriptions.

[0105] From Tables 4 to 6, all subjects (those prescribed sleeping pills) may be classified by the number of prescriptions for benzodiazepine derivatives, the number of prescriptions for anticholinergic drugs, and the number of prescriptions for SSRIs and SNRIs. For example, segments may be created by first dividing them by the number of prescriptions for benzodiazepine derivatives, then by the number of prescriptions for anticholinergic drugs, and then by the number of prescriptions for SSRIs and SNRIs. The order of dividing them into stages is not limited to this, and segments may be created in any order, such as by the number of prescriptions for benzodiazepine derivatives, the number of prescriptions for SSRIs and SNRIs, and the number of prescriptions for anticholinergic drugs, or by the number of prescriptions for SSRIs and SNRIs, the number of prescriptions for anticholinergic drugs, and the number of prescriptions for benzodiazepine derivatives.

[0106] In Tables 4 to 6, the proportion of each segment to all subjects is added up, and segments that exceed a predetermined percentage are selected as segments for which measures should be taken. The predetermined percentage may be 20%. Alternatively, the number of subjects for which measures should be taken may be determined in advance, and segments may be selected in descending order of risk until the number of segments exceeds the predetermined number. The predetermined number may be 200.

[0107] <Notice on measures to combat polypharmacy> The detailed process of S503 in FIG. 5 will be described.

[0108] As a countermeasure, the notification unit 225 notifies the subject selected by the segment selection unit 221 by email of medication information urging the subject to consult a pharmacist about polypharmacy so that the subject can take medication appropriately. The notification unit 225 may notify the terminal device 103 of the segment or the subject's information. When transmitting the subject's information to the terminal device 103, the notification unit 225 may also transmit information related to the medication information to be notified.

[0109] Insurers, etc., can receive the notification from notification unit 225 and notify the subject by mail or email of a medication information notice encouraging the subject to consult with a pharmacist. By referring to the medication information notice, the subject who receives the notice can consult with a pharmacist, take appropriate medication, and reduce the risk of fractures due to polypharmacy. Insurers, etc. can also reduce medical expenses by eliminating polypharmacy and by reducing the risk of fractures.

[0110] <Dementia risk> In the above-described embodiment, attention is focused on the risk of fracture due to polypharmacy. Next, a polypharmacy countermeasure focusing on the risk of dementia will be described.

[0111] Regarding dementia, if a subject's medical data does not show a definitive diagnosis of dementia in one year, but shows a definitive diagnosis in the following year's data, it can be said that the subject has newly developed dementia. It is also possible to evaluate the severity of dementia and memory loss. The learning unit 222 can generate a dementia risk prediction model in the prediction model flow based on the data of these subjects instead of the fracture hospitalization data.

[0112] Once the polypharmacy estimation model for dementia risk has been generated, the estimation model unit 220 can predict the dementia risk of each individual subject as a probability. In S901, the estimation model unit 220 of the information processing device 101 first uses the polypharmacy estimation model for dementia risk to estimate the dementia risk (probability) for each individual subject of the project for the following year. The project targets individuals who prescribe sleeping pills, but other drugs may also be used.

[0113] Next, in S902, the segment selection unit 221 of the information processing device 101 calculates the dementia risk of the segment by taking the average value of the dementia risk of the subjects belonging to each segment. Here, the average value of the probability of the subjects within the group is taken, but other values ​​such as the median or mode may also be used. As a result of taking the average value, the dementia risk of each segment is obtained.

[0114] Next, in S903, the segment selection unit 221 of the information processing device 101 prioritizes the implementation of polypharmacy countermeasures for each segment. Furthermore, if the risk of a segment is low, if there are few subjects belonging to the segment, or if the proportion of subjects is low, multiple segments may be integrated.

[0115] Next, in S904, the segment selection unit 221 of the information processing device 101 selects a priority for each segment and selects a target person for which measures will be taken. The target person for which measures will be taken is selected in order of the segment with the highest dementia risk, i.e., the target person with the highest dementia risk.

[0116] <Nursing care risk> Next, we will explain polypharmacy countermeasures that focus on nursing care risks.

[0117] For the estimation model that estimates nursing care risk, the learning unit 222 trains the estimation model by treating, instead of fracture hospitalization data, subjects in the nursing care certification data 460 who were not certified as needing nursing care in a certain year but who were newly certified as needing nursing care level 2 or higher in the following year as subjects who have newly received nursing care certification. The learning period is not limited to one year, and may be shorter or longer than one year. Newly certified individuals needing nursing care may be those certified as needing nursing care level 1 or higher, or those certified as needing nursing care level 3 or higher. They may also be those with a level of support required. Physical frailty data may also be included. The accuracy of estimations by the polypharmacy estimation model can be improved due to the number of subjects and the greater change in condition for those needing nursing care level 2 compared to those needing nursing care level 1.

[0118] Once the polypharmacy estimation model for nursing care risk is generated, the estimation model unit 220 can predict the nursing care risk of each individual subject as a probability. In S901, the estimation model unit 220 of the information processing device 101 first uses the polypharmacy estimation model for nursing care risk to estimate the nursing care risk (probability) for each individual subject of the project for the next year (within one year). The estimation period is not limited to within one year, and may be shorter or longer than one year. While those who prescribe sleeping pills are the subject of the project, other drugs may also be used.

[0119] Next, in S902, the segment selection unit 221 of the information processing device 101 calculates the nursing care risk of the segment by taking the average value of the nursing care risk of the subjects belonging to each segment. Here, the average value of the subjects' probabilities within the group is taken, but other values ​​such as the median or mode may also be used. As a result of taking the average value, the nursing care risk of each segment is obtained.

[0120] Next, in S903, the segment selection unit 221 of the information processing device 101 prioritizes the implementation of polypharmacy countermeasures for each segment. Furthermore, if the risk of a segment is low, or if the number of subjects or the proportion of subjects belonging to a segment is small, multiple segments may be integrated.

[0121] Next, in S904, the segment selection unit 221 of the information processing device 101 selects a priority order for each segment and selects a target person for which measures will be taken. The target person for which measures will be taken is selected in order of segments with higher nursing care risks.

[0122] <Risk of lifestyle-related diseases> Next, we will explain measures to combat polypharmacy that focus on the risk of lifestyle-related diseases.

[0123] For the estimation model estimating lifestyle-related disease risk, the learning unit 222 trains the prediction model by using, instead of fracture hospitalization data, subjects in the medical data 440 who were not hospitalized for lifestyle-related diseases in a given year but who were hospitalized for lifestyle-related diseases in the following year as subjects hospitalized for lifestyle-related diseases. This does not limit the subjects to those hospitalized for lifestyle-related diseases, but may also include subjects newly diagnosed with lifestyle-related diseases. Furthermore, by comparing medical expenses related to lifestyle-related diseases between one year and the next, subjects whose expenses increase by more than a certain amount or a certain percentage may be considered as subjects who have newly developed lifestyle-related diseases or whose lifestyle-related diseases have become more severe. The learning period is not limited to one year, and may be shorter or longer than one year. This allows data on subjects whose lifestyle-related diseases have become more severe to be used as training data. In this case, it is possible to use, as training data, subjects who have been diagnosed with lifestyle-related diseases in advance from the population without excluding them in advance.

[0124] Once the polypharmacy estimation model for lifestyle-related disease risk is generated, the estimation model unit 220 can predict the lifestyle-related disease risk of each individual subject as a probability. In S901, the estimation model unit 220 of the information processing device 101 first uses the polypharmacy estimation model for lifestyle-related disease risk to estimate the lifestyle-related disease risk (probability) for each individual subject of the project for the following year. The estimation period is not limited to one year, and may be shorter or longer than one year. While the project targets individuals prescribed sleeping pills, other drugs may also be used.

[0125] Next, in S902, the segment selection unit 221 of the information processing device 101 calculates the lifestyle-related disease risk of each segment by averaging the lifestyle-related disease risk of the subjects belonging to each segment. Here, the average probability of the subjects within the group is calculated, but other values ​​such as the median or mode may also be used. As a result of calculating the average, the lifestyle-related disease risk of each segment is obtained.

[0126] Next, in S903, the segment selection unit 221 of the information processing device 101 prioritizes the implementation of polypharmacy countermeasures for each segment. Furthermore, if the risk of a segment is low, if there are few subjects belonging to the segment, or if the proportion of subjects is low, multiple segments may be integrated.

[0127] Next, in S904, the segment selection unit 221 of the information processing device 101 selects a priority order for each segment and selects targets for measures to be taken. Targets for measures to be taken are selected in order of segments with a high risk of lifestyle-related diseases.

[0128] <Second embodiment> In the first embodiment, segments with a high risk of polypharmacy are selected as targets for implementing polypharmacy countermeasures. In the second embodiment, segments with a high risk of polypharmacy are selected as targets for implementing countermeasures by using the probability of behavioral change along with the risk of polypharmacy.

[0129] Behavioral change refers to a change in a subject's behavior as a result of intervention by an insurer, etc. An example would be a subject who has not visited a medical institution, etc., receiving a recommendation from an insurer, etc. to do so and beginning to do so. Another example would be a subject receiving guidance from an insurer, etc. to improve their lifestyle habits and working to improve their lifestyle habits. The probability of behavioral change is the probability that a subject's behavior will change as a result of intervention by an insurer, etc. An example would be the probability that a subject will visit a medical institution, etc., as a result of an insurer, etc. recommending the subject to visit a medical institution, etc.

[0130] When an insurer etc. encourages a target individual to visit a medical institution etc., the target individual may respond differently, either completely ignoring the recommendation or not seeing treatment or improvement even after the recommendation. It also depends on the content of the recommendation made by the insurer etc. The way the target individual perceives the recommendation and their subsequent behavior will be different depending on whether the insurer etc. sends direct mail to the target individual or whether the insurer etc. calls or visits the target individual to encourage them to visit a medical institution. Furthermore, people who have never had a health checkup and people who receive a health checkup every year have different awareness of health and will behave differently in response to the recommendation.

[0131] <Generation of a behavioral change estimation model> In this embodiment, in response to intervention by an insurer or the like, health checkup data 420 and medical care data 440 of a subject who has visited a medical institution or the like are used as training data to train a behavior change probability estimation model.

[0132] The estimation models for the behavioral change probability in this embodiment include a consultation probability model, a response model, and an uplift model. Each of these models will be described with reference to FIG.

[0133] 10(a) is a conceptual diagram illustrating the generation of a medical examination probability model. In the medical examination probability model, health checkup data 420, medical treatment data 440, and dispensing receipt data 450 are used as learning data. The data acquisition unit 223 of the information processing device 101 acquires the health checkup data 420, medical treatment data 440, and dispensing receipt data 450 from the health information database 102 and stores them in the storage unit 203. The period of the health checkup data 420, medical treatment data 440, and dispensing receipt data 450 acquired by the data acquisition unit 223 is arbitrary.

[0134] Next, the learning unit 222 identifies subjects who have actually taken action. For example, subjects who have undergone a health checkup are identified. The learning unit 222 identifies attributes that tend to lead to undergoing a health checkup from the health checkup data 420, medical care data 440, and prescription data 450 of the subjects who have undergone the health checkup, and generates a checkup probability model.

[0135] As a result, when the health checkup data 420, medical treatment data 440, and prescription data 450 of a certain subject are input into the medical examination probability model, a probability indicating a tendency to undergo a health examination is output. Here, it is assumed that subjects who undergo a health examination also have a high tendency to visit a medical institution, etc., and the output of the medical examination probability model is the probability that a certain subject will visit a medical institution, etc. The medical examination probability model is an estimation model that outputs the probability that a subject will visit a medical institution, etc., even if an insurer, etc., does not take any intervention to recommend the subject.

[0136] Next, the response model will be described. FIG. 10(b) is a conceptual diagram illustrating the generation of a response model. In the response model, health checkup data 420, medical treatment data 440, dispensing receipt data 450, and intervention data 470 are used as learning data. The data acquisition unit 223 of the information processing device 101 acquires the health checkup data 420, medical treatment data 440, dispensing receipt data 450, and intervention data 470 from the health information database 102 and stores them in the storage unit 203. The health checkup data 420, medical treatment data 440, dispensing receipt data 450, and intervention data 470 acquired by the data acquisition unit 223 may be from any period. Furthermore, only a portion of these data may be acquired, or other data may be included.

[0137] The intervention data 470 includes, for example, data in which an insurer or the like notifies a subject of a recommendation to visit a medical institution or the like based on the subject's health checkup data 420 or the like. The learning unit 222 identifies a subject whose behavior has changed due to an intervention based on the intervention data 470. Here, as an example, an insurer or the like recommends a subject to visit a medical institution, and identifies a subject who visited the medical institution or the like in accordance with the recommendation. The learning unit 222 identifies attributes that tend to make the subject more likely to visit a medical institution or the like from the health checkup data 420, medical data 440, and dispensing receipt data 450 of a subject who visited a medical institution or the like in accordance with the recommendation, and generates a response model. In this case, the learning unit 222 may identify attributes that make the subject less likely to visit a medical institution or the like using the health checkup data 420, medical data 440, and dispensing receipt data 450 of a subject who did not visit a medical institution or the like despite being recommended to do so by an insurer or the like, in addition to the attributes that tend to make the subject more likely to visit a medical institution or the like, and use these as learning data to generate a response model.

[0138] When the medical checkup data 420, medical treatment data 440, and prescription data 450 of a certain subject are input into the response model, the output is the probability that the subject will actually visit a medical institution, etc. when the subject is encouraged to visit a medical institution, etc. The response model has the advantage of being able to generate an estimation model even if the insurer, etc. has a low track record of encouraging patients to visit a medical institution, etc.

[0139] Next, the uplift model will be described with reference to Fig. 10(c) and Fig. 11. Fig. 10(c) is a conceptual diagram for explaining generation of the uplift model. The flowchart in Fig. 11 is realized by the CPU 210 of the information processing device 101 reading a program stored in the storage unit 203 into the RAM 211 and executing it.

[0140] In the uplift model, the health check data 420, medical treatment data 440, dispensing receipt data 450, and intervention data 470 are used as learning data. In S1101, the data acquisition unit 223 of the information processing device 101 acquires the health check data 420, medical treatment data 440, and dispensing receipt data 450 from the health information database 102 and stores them in the storage unit 203. The period of the health check data 420, medical treatment data 440, and dispensing receipt data 450 acquired by the data acquisition unit 223 is arbitrary.

[0141] In S1102, the data acquisition unit 223 of the information processing device 101 acquires intervention data 470 and stores it in the storage unit 203. The intervention data 470 includes, for example, data of a notification sent by an insurer or the like to a subject, encouraging the subject to visit a medical institution or the like, based on the subject's health checkup data 420 or the like. The intervention data 470 here includes, for example, data of subjects who were not encouraged to visit a medical institution or the like, data of subjects who were encouraged to visit a medical institution or the like by a regular paper notification, and data of subjects who were encouraged to visit a medical institution or the like by a detailed paper notification. The recommendation notification is not limited to these, and may include, for example, a telephone call, an email, or a visit, or may be a paper notification with further modified content. The intervention data 470 prepared as learning data may be a notification of a recommendation to visit a medical institution or the like at a plurality of different levels.

[0142] Next, in S1103, the learning unit 222 of the information processing device 101 causes the uplift model to learn the attributes of subjects who actually visited a medical institution, etc., for subjects who were not notified (non-intervention) of a recommendation to visit a medical institution, etc. This enables the uplift model to estimate the probability of visiting a medical institution, etc., according to the attributes of the subject in the non-intervention case.

[0143] Next, in S1104, the learning unit 222 of the information processing device 101 causes the uplift model to learn the attributes of subjects who have actually visited a medical institution, etc., and who have been encouraged to visit a medical institution, etc. through a written notification (intervention). This allows the uplift model to estimate the probability of visiting a medical institution, etc., according to the attributes of the subject in the case of intervention.

[0144] Next, in S1105, the learning unit 222 of the information processing device 101 compares the consultation probability in the case of no intervention in S1103 with the consultation probability in the case of a paper notification (intervention) in S1104. For example, if the consultation probability estimated in S1104 is higher than the consultation probability estimated in S1103, the learning unit 222 can determine that the paper notification (intervention) is effective, excluding the influence of non-intervention. Since highly health-conscious subjects will naturally visit a medical institution or the like even without intervention from an insurer or the like, in S1105, the effect of the intervention, excluding the influence of no intervention, and the increase in the probability of visiting a medical institution or the like due to the intervention can be calculated.

[0145] In S1106, the learning unit 222 trains the uplift model to learn the difference between the visit probability in the case of no intervention in S1105 and the visit probability in the case of intervention, according to the attributes of the subject. In the example described above, the learning unit 222 compared the visit probability in the case of no intervention with the visit probability in the case of intervention, but it may also compare the visit probability in the case of a normal notification with the visit probability in the case of a detailed notification. In this case, it is possible to determine the extent to which the probability of not visiting a medical institution, etc. with a normal notification increases, but the probability of visiting a medical institution, etc. increases with a detailed notification.

[0146] When the health checkup data 420, medical treatment data 440, and prescription data 450 of a certain subject are input into the generated uplift model, the increase in the probability that the subject will actually visit a medical institution when the subject is encouraged to visit a medical institution or the like is output.

[0147] The learning unit 222 may further verify and re-learn the generated uplift model. In S1107, the learning unit 222 inputs the health check data 420, the medical care data 440, and the prescription data 450, and outputs the increase in the probability that the subject will visit a medical institution or the like. The learning unit 222 then re-learns the uplift model depending on whether the subject who has received the intervention actually visits a medical institution or the like. If the subject actually visits a medical institution or the like, the uplift model's output probability is corrected to increase, and if the subject does not visit a medical institution or the like for a certain period of time, the uplift model's output probability is corrected to decrease.

[0148] According to the behavior change probability estimation model, by inputting health information, it is possible to output the probability of a subject's behavior change. For example, the behavior change probability estimation model can estimate the following: Older subjects who underwent a health checkup in the previous year are more likely to undergo a health checkup in response to a recommendation. Subjects who have not undergone a health checkup in the past 1-2 years, or subjects who only undergo a health checkup once every few years, are more likely to undergo a health checkup when recommended to do so. Furthermore, subjects who have never undergone a health checkup in the past are less likely to undergo a health checkup even when recommended to do so. Subjects who visit medical institutions with a certain frequency are more likely to respond to recommendations. Elderly people are more likely to respond to recommendations to undergo a health checkup via telephone. In this way, the behavior change probability estimation model outputs the probability of behavior change by inputting health information data.

[0149] <Selection of target segments using a behavior change probability estimation model> Next, selection of a subject segment for which measures are to be taken using a polypharmacy estimation model and a behavioral change probability estimation model will be described with reference to Figure 12. The behavioral change probability estimation model may be any of a consultation probability model, a response model, and an uplift model. In the second embodiment, the behavioral change probability estimation model is combined with the polypharmacy estimation model of the first embodiment to select a subject segment.

[0150] First, the estimation model unit 220 of the information processing device 101 inputs prescription data 1201 of an individual subject into a polypharmacy model 1202. Then, a fracture risk (probability) 1203 for the next fiscal year (within one year) for a certain subject is output. Here, fracture risk is taken as an example, but any risk caused by polypharmacy may be used. It may also be the nursing care risk, dementia risk, or lifestyle-related disease risk mentioned above.

[0151] Furthermore, the estimation model unit 220 of the information processing device 101 inputs the subject's individual health checkup data, medical treatment data, and prescription data 1204 into a behavior change probability estimation model 1205. This then obtains, for a certain subject, a probability 1206 of visiting a medical institution or the like when an insurer or the like intervenes.

[0152] Next, the estimation model unit 220 calculates the probability 1207 of a subject visiting a medical institution or the like as a result of an intervention, by multiplying the subject's fracture risk (probability) 1203 for the next fiscal year (within one year) by the probability 1206 that the subject will visit a medical institution or the like as a result of the intervention. The calculation performed by the estimation model unit 220 may also be to weight either the fracture risk 1203 or the behavior change probability 1206 and then take the product.

[0153] By reflecting the probability of behavioral change due to intervention in the fracture risk using the polypharmacy estimation model, it is possible to identify subjects who are at high risk due to polypharmacy and require countermeasures, and who are likely to visit a medical institution or the like if the insurer intervenes. After determining the probability 1207 of a subject's polypharmacy risk being visited a medical institution as a result of intervention, a segment for which countermeasures will be taken is selected according to the flow in FIG. 9. According to the second embodiment, it is possible to select a segment for which countermeasures are more effective. As with the first embodiment, high-risk subjects may be selected.

[0154] The invention is not limited to the above-described embodiment, and various modifications and variations are possible within the scope of the gist of the invention. [Explanation of symbols]

[0155] 101: Information processing device, 102: Health information database, 103: Terminal device

Claims

1. A program that causes a computer to function as each means of an information processing device that supports measures against polypharmacy, the information processing device comprising: an estimation means for estimating a health risk of each of a plurality of subjects based on information on the prescription drugs of the plurality of subjects; a behavior change probability estimation means for estimating a behavior change probability due to an intervention for each of the plurality of subjects based on the health information of the plurality of subjects; a selection means for selecting targets for polypharmacy countermeasures based on the health risks and the behavioral change probabilities of the plurality of subjects; A program that includes:

2. The selection means selects a target segment for the polypharmacy countermeasure based on the health risks of multiple subjects for each segment. The program according to claim 1.

3. The segments are defined by the type and number of prescription drugs. The program according to claim 2.

4. The plurality of subjects have been prescribed a sleeping pill. The program according to claim 1.

5. The prescribed medication includes at least one of a benzodiazepine derivative, an SSRI / SNRI, and an anticholinergic drug. The program according to claim 1.

6. The health risk includes at least one of a fracture risk, a nursing care risk, a dementia risk, and a lifestyle-related disease risk. The program according to claim 1.

7. The estimation means includes an estimation model that learns based on information on prescription drugs of subjects whose health status has changed from a certain period to a subsequent period. The program according to claim 1.

8. The selection means selects segments of the polypharmacy countermeasure until a predetermined number or a predetermined ratio is exceeded. The program according to claim 2.

9. An information processing device that supports measures against polypharmacy, an estimation means for estimating a health risk of each of a plurality of subjects based on information on the prescription drugs of the plurality of subjects; a behavior change probability estimation means for estimating a behavior change probability due to an intervention for each of the plurality of subjects based on the health information of the plurality of subjects; and a selection means for selecting targets for polypharmacy countermeasures based on the health risks and the behavioral change probabilities of the plurality of subjects. Information processing device.

10. An information processing method executed by an information processing device that supports measures against polypharmacy, an estimation step of estimating a health risk of each of a plurality of subjects based on information on the prescription drugs of the plurality of subjects; a behavior change probability estimation step of estimating a behavior change probability due to an intervention for each of the plurality of subjects based on health information of the plurality of subjects; and a selection step of selecting targets for polypharmacy countermeasures based on the health risks of the plurality of subjects and the probability of behavioral change. Information processing methods.

Images