JP2016103092A - System for evaluating effect of preventive care service and program for evaluating effect of preventive care service - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a system for evaluating effect of preventive care service and a program for evaluating effect of preventive care service capable of quantitatively evaluating the effect of preventive care service.SOLUTION: The preventive care service effect evaluation system includes: a data acquisition section; and a preventive care service effect calculation section. The data acquisition section acquires a data which represents a situation of utilization of a medical institution and a data relevant to the implementation of the preventive care service. The preventive care service effect calculation section calculates the effect of the preventive care service by calculating changes in utilization of hospital by implementing the preventive care service based on the data acquired by the data acquisition section.SELECTED DRAWING: Figure 2

Description

  Embodiments described herein relate generally to a preventive service effect evaluation apparatus and a preventive service effect evaluation program.

  It is expected that the prevention of illness will be prevented and the medical expenses will be reduced by implementing preventive services for preventing illness. Since implementation of preventive services is costly, there is a high need to evaluate its effectiveness. However, with the conventional technology, it has been difficult to quantitatively evaluate the effect of the preventive service.

JP 2012-128670 A

  The problem to be solved by the present invention is to provide a preventive service effect evaluation apparatus and a preventive service effect evaluation program capable of quantitatively evaluating the effect of the preventive service.

  The preventive service effect evaluation apparatus of the embodiment has a data acquisition unit and a preventive service effect calculation unit. The data acquisition unit acquires data indicating the usage status of the medical institution and data related to the implementation of the prevention service. The preventive service effect calculating unit calculates the effect of the preventive service by deriving a change in the hospitalization situation due to the execution of the preventive service based on the data acquired by the data acquiring unit.

The figure which shows the hardware constitutions and communication environment of the prevention service effect evaluation apparatus 1 which concern on 1st Embodiment. The figure which shows the function structure of the prevention service effect evaluation apparatus 1. FIG. The figure which shows an example of the subscriber data stored in the subscriber data storage part 22. FIG. The figure which shows an example of the medical examination data stored in the medical examination data storage part 24. FIG. The flowchart which shows the flow of the whole process performed by the preventive service effect evaluation apparatus 1 which concerns on 1st Embodiment. The flowchart which shows the flow of the subscriber state determination process performed by the state determination part 12. FIG. The figure which shows notionally the event which each element of the state transition probability matrix P shows. The flowchart which shows the flow of the prevention service effect calculation process performed by the prevention service effect calculation part 14. FIG. It is a figure which shows the function structure of the preventive service effect evaluation apparatus 1A which concerns on 2nd Embodiment. The flowchart which shows the flow of the whole process performed by the preventive service effect evaluation apparatus 1A which concerns on 2nd Embodiment.

  Hereinafter, a preventive service effect evaluation apparatus and a preventive service effect evaluation program according to embodiments will be described with reference to the drawings.

(First embodiment)
FIG. 1 is a diagram illustrating a hardware configuration and a communication environment of the preventive service effect evaluation apparatus 1 according to the first embodiment. The preventive service effect evaluation apparatus 1 includes, for example, a processor 10 such as a CPU (Central Processing Unit), a storage device 20, a drive device 30, an input / output device 40, and a communication interface 50.

  The storage device 20 includes, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive), a flash memory, and the like. In addition to the program executed by the processor 10, the storage device 20 stores various information used by the preventive service effect evaluation device 1 for processing. The drive device 30 is loaded with a portable storage medium that stores programs and various types of information. The input / output device 40 includes an input device such as a keyboard, a mouse, and a touch panel, and a display device such as an LCD (Liquid Crystal Display) or an organic EL (Electroluminescence) display device.

  The communication interface 50 includes a network card for connecting to the network NW. The network NW is a communication network such as a LAN (Local Area Network) or a WAN (Wide Area Network). Connected to the network NW are an information providing source device 70 that provides information to the preventive service effect evaluation device 1, a terminal device 80 that receives service provision from the preventive service effect evaluation device 1, and the like.

  The program executed by the processor 10 may be stored in the storage device 20 in advance when the preventive service effect evaluation apparatus 1 is shipped, or may be installed by a portable storage medium mounted on the drive device 30. Good. Further, the program executed by the processor 10 may be downloaded from another device via the network NW.

  In this embodiment, the preventive service effect evaluation apparatus 1 is described as an apparatus that provides a cloud service. However, the preventive service effect evaluation apparatus 1 is an apparatus that outputs an evaluation result when a user directly operates. It may be.

  FIG. 2 is a diagram illustrating a functional configuration of the preventive service effect evaluation apparatus 1. The preventive service effect evaluation apparatus 1 includes a data acquisition unit 11, a state determination unit 12, a state transition probability calculation unit 13, and a preventive service effect calculation unit 14 as software function units that function when the processor 10 executes a program. With. In addition, the preventive service effect evaluation apparatus 1 includes, in the storage device 20, a parameter storage unit 21, a subscriber data storage unit 22, a receipt data storage unit 23, a medical examination data storage unit 24, and a preventive service related data store. Unit 25 and a state transition model storage unit 26.

  For example, the data acquisition unit 11 receives information on the target year and target disease input by the user from the terminal device 80 from the terminal device 80 via the communication unit interface 50 and the network NW, and stores the received information as parameters. Stored in the unit 21. In addition, the data acquisition unit 11 receives, for example, data stored in the subscriber data storage unit 22, the receipt data storage unit 23, the medical examination data storage unit 24, and the preventive service related data storage unit 25 from the information providing source device 70. Receive via the communication unit interface 50 and the network NW. The data acquisition unit 11 arranges the target person in a format suitable for processing, such as associating the target person with a common ID (identification information), and stores the data in each storage unit. The information providing source device 70 is a database server managed by, for example, a health insurance association, a medical institution, or a public institution. Instead of (or in addition to) these, the data acquisition unit 11 may store data input to the input / output device 40 in each storage unit.

  The state determination unit 12 determines the state of the subject based on data indicating the usage status of the medical institution, such as receipt data or medical examination data. The state of the subject includes a state indicating the out-of-hospital status such as a non-examination or a medical examination. The state transition probability calculation unit 13 compares the state of the subject determined by the state determination unit 12 between, for example, years, and calculates the probability of transition from one state to another state. The preventive service effect calculation unit 14 calculates the effect of the preventive service. The functions of these functional units will be described in detail later.

  The parameter storage unit 21 stores information on a target year that is a reference for generating a state transition model and a target disease that is a processing target. The subscriber data storage unit 22 stores data on subscribers to the health insurance association for each target year. FIG. 3 is a diagram illustrating an example of subscriber data stored in the subscriber data storage unit 22. In the figure, “individual ID” is identification information of the subject. Further, the “withdrawal flag” is a flag that is set to 1 when, for example, withdrawing from the health insurance association in the target year, and 0 when the participation in the health insurance association is maintained.

  The receipt data storage unit 23 stores receipt data. Receipt data is a description of medical remuneration charged to an insurer by a medical institution, and describes the medical treatment received by the examinee. The receipt data includes, for example, the name of the examinee, the date of medical treatment, the medical facility, the name of the diagnosed disease, the actual number of medical treatments, the type of medical treatment, the number of claims, and the like. By confirming the receipt data, it is possible to confirm the type of medical treatment and the number of times each subject has received in the designated month in units of months.

  The medical examination data storage unit 24 stores medical examination data related to the results of the medical examination for each target year. FIG. 4 is a diagram illustrating an example of medical examination data stored in the medical examination data storage unit 24.

  The preventive service related data storage unit 25 stores preventive service related data. Preventive service-related data is data related to the implementation of various disease severity prevention services. For example, (1) whether or not a notification to encourage attendance as a preventive service has been given, and (2) attendance as a result of notification It is the data which stored whether it was made for every subject. In the state transition model storage unit 26, for example, a state transition model as a processing result of the preventive service effect evaluation apparatus 1 is stored.

  Hereinafter, the flow of processing executed by the preventive service effect evaluation apparatus 1 will be described. FIG. 5 is a flowchart showing the flow of overall processing executed by the preventive service effect evaluation apparatus 1 according to the first embodiment. First, the state determination unit 12 reads information on a target year as a reference for generating a state transition model and a target disease to be processed from parameters stored in the parameter storage unit 21 (step S100). In the following description, it is assumed that the target disease is diabetes.

  The state determination unit 12 reads the data of the subscribers to the health insurance association in the target year from the subscriber data storage unit 22 (step S102). Next, the state determination unit 12 reads the receipt data in the target year from the receipt data storage unit 23 for each subscriber read in step S102 (step S104). Next, for each of the subscribers read in step S102, the state determination unit 12 uses the data related to the target disease read in step S100 among the data related to the health check performed in the target year. 24 is read (step S106). The test value is, for example, HbA1c (hemoglobin A1 sea) related to diabetes. These data are read from an auxiliary storage device such as an HDD or a flash memory, and expanded in a work area such as a RAM and used for processing described later.

  Next, the state determination unit 12 extracts one subscriber according to a predetermined order from the subscribers in the target year read in Step S102 (Step S108). At this time, if there is no extractable subscriber, the process proceeds to step S112. When there is an extractable subscriber, the state determination unit 12 performs a subscriber state determination process (step S110; FIG. 6).

  Prior to the description of the processing after step S112 in FIG. 5, the subscriber status determination processing will be described. The state determination unit 12 selects the state of the subject from one of seven states, for example, “not yet visited”, “received”, “mild”, “moderate”, “severe”, “healthy”, and “withdrawal”. Classify into: “Non-visit” means a state in which the patient is admitted to the hospital despite having the target disease. “Visit” refers to a condition in which there is a possibility of suffering from the target disease and he / she has been admitted to the hospital for an examination or the like but has not been treated. “Mild” refers to a condition in which the subject disease is affected but the symptoms are mild and a mild treatment (for example, medication) is performed. “Medium” refers to a state of suffering from the target disease but moderate in severity and undergoing moderate treatment (for example, insulin injection). “Severe” refers to a condition in which the subject is afflicted with a target disease, the symptoms are severe, and severe treatment (for example, artificial dialysis treatment) is performed. “Healthy” refers to a state in which the subject is not suffering from the disease. “Leaving” means leaving the health insurance association. These states are determined, for example, by referring to subscriber data, receipt data, and medical examination data in the target year of the subscriber extracted in step S108.

  FIG. 6 is a flowchart showing the flow of the subscriber state determination process executed by the state determination unit 12. First, the state determination unit 12 refers to the subscriber data read in step S102, and determines whether or not the subject has joined the health insurance association (step S200). If the target person has not joined the health insurance association, the state determination unit 12 determines the state of the target person as “withdrawal” (step S202), and ends the process of this flowchart.

  When the target person has joined the health insurance association, the state determination unit 12 refers to the receipt data read in step S104 and performs the following processing. The state determination unit 12 determines whether the number of receipts (the number of receipt data) in the target year of the subject is equal to or greater than the threshold value N1 (step S204). The threshold value N1 is an arbitrary positive integer.

  When the number of receipts is equal to or greater than the threshold value N1, the state determination unit 12 determines whether the subject has received severe treatment in the target year (step S206). When severe treatment is received, the state determination unit 12 determines that the state of the subject is “severe” (step S208), and ends the process of this flowchart.

  When severe treatment has not been received, the state determination unit 12 determines whether or not the subject has received neutral treatment in the target year (step S210). When receiving the treatment for the symptom, the state determination unit 12 determines that the state of the subject is “middle illness” (step S212), and ends the process of this flowchart.

  When not receiving the treatment for the symptom, the state determination unit 12 determines whether or not the subject has received the mild treatment in the target year (step S214). When receiving mild treatment, the state determination unit 12 determines that the subject's state is “mild” (step S216), and ends the process of this flowchart.

  When the mild treatment is not received, the state determination unit 12 refers to the data read in step S106 regarding the treatment of the target disease, and whether the test value related to the target disease in the target year of the subject is equal to or greater than the threshold A1. It is determined whether or not (step S218). If the test value is equal to or greater than the threshold value A1, the state determination unit 12 determines that the state of the subject is “not yet visited” (step S222), and ends the process of this flowchart. When the test value is less than the threshold value A1, the state determination unit 12 determines that the state of the subject is “consultation” (step S220), and ends the process of this flowchart.

  On the other hand, when it is determined in step S204 that the number of receipts is less than the threshold value N1, the state determination unit 12 determines whether the test value related to the target disease in the target year of the target person is equal to or greater than the threshold value A2. (Step S224). The threshold value A2 may be the same value as the threshold value A1, or may be a different value. When the test value is equal to or greater than the threshold value A2, the state determination unit 12 determines that the state of the subject is “not yet visited” (step S222), and ends the process of this flowchart. When the inspection value is less than the threshold value A2, the state determination unit 12 determines that the state of the subject is “healthy” (step S226), and ends the process of this flowchart.

  Returning to FIG. 5, the processing after step S112 will be described. The process in step S110 (the process in the flowchart in FIG. 6) is performed for all subscribers read in step S102. When the process for all subscribers is completed, the process proceeds to step S112. The state determination unit 12 determines whether or not the target year has been updated while the flowchart is executed (step S112). If the target year has not been updated, the state determination unit 12 updates the target year by one (step S114). And the state determination part 12 performs the process from step S102 to S110 about the updated object year. As a result, the state of the target person is determined for each of two years, the initially set target year and the next target year.

When the states of subscribers for two years (including those who are subscribers in the target year and withdrawn in the next year) are thus determined, the state transition probability calculation unit 13 is integrated by the following processing. The number of transitions is initialized (step S116). For example, the number of state transitions is integrated as an element of the state transition number matrix O represented by Expression (1). The state transition number matrix O is a matrix in which the state of the target year is a first argument and the state of the next year after the target year is a second argument. “Initialization” in step S116 means, for example, that all values of the memory area corresponding to the elements of this matrix are set to zero. In the formula, n is “visited”, h is “unvisited”, m is “mild”, i is “moderate”, d is “severe”, f is “healthy”, and e is “withdrawal”. ing. For example, n _hm indicates the number of subscribers who have changed from the state h “not visited” to the state n “received” from the target year to the next year.

  Next, the state transition probability calculating unit 13 extracts one subscriber according to a predetermined order from the target person in the target year read in Step S102 (Step S118). At this time, if there is no extractable subscriber, the process proceeds to step S124. When there is an extractable subscriber, the state transition probability calculating unit 13 determines whether the state in the next year of the target year has been determined for the subscriber extracted in step S118 (step S120; next) Fiscal year judgment). When the state in the year following the target year has been determined, the state transition probability calculation unit 13 adds 1 to the corresponding element in the state transition number matrix O described above (step S122), and returns to step S118. Moreover, also when the state in the following fiscal year is not determined, the process returns to step S118. Such processing is executed for all subscribers in the target year.

When there are no more subscribers that can be extracted, the state transition probability calculating unit 13 calculates a state transition probability (step S124). The state transition probability is represented by, for example, a state transition probability matrix P expressed by Equation (2). In the equation, p _xy indicates the percentage of subscribers who have transitioned to the state y in the next year among the subscribers determined to be the state x in the target year. The state transition probability calculating unit 13 obtains a total value in the row direction in the state transition number matrix O shown in the equation (1), and divides each element by the total value in the row direction for each row, so that the state transition probability matrix P Is derived. The state transition probability matrix P has a total value of 1 when added in the row direction. FIG. 7 is a diagram conceptually showing events represented by the elements of the state transition probability matrix P.

  When the state transition probability matrix P is obtained, the preventive service effect calculating unit 14 reads the preventive service related data from the preventive service related data storage unit 25 (step S126). Then, the preventive service effect calculation unit 14 performs an operation for calculating the effect of the preventive service based on the preventive service related data read in step S126 (step S128; FIG. 8).

  FIG. 8 is a flowchart showing the flow of the preventive service effect calculating process executed by the preventive service effect calculating unit 14. First, the preventive service effect calculation unit 14 extracts one preventive service notifier who notified the preventive service from the subscribers extracted in step S118 based on the preventive service related data read in step S126 (step S300). When one preventive service notifier is extracted, the preventive service effect calculating unit 14 determines whether or not the extracted preventive service notifier has received the preventive service (step S302). The preventive service effect calculation unit 14 determines whether or not the preventive service notifier has changed in behavior when the preventive service notifier has taken the preventive service or not (steps S304 and S310).

  For example, the preventive service effect calculating unit 14 receives a preventive service (may or may not include a notification month) and the number of receipts within a predetermined period (for example, several months) is greater than or equal to a threshold value. In some cases, it is determined that behavioral change has occurred. As a result, the preventive service effect calculation unit 14 divides the preventive service notifier into four classes: “taken / changed”, “taken / not changed”, “not-taken / changed”, and “not-taken / changed”. Classification is performed (steps S306, S308, S312, and S314).

When all preventive service notifiers are extracted, the preventive service effect calculating unit 14 calculates the state transition rate α (step S316). The state transition rate α is the sum of the number n _{cr of} preventive service notifiers classified as “taken / changed” and the number n _cn of preventive service notifiers classified as “not attended / changed”. It is obtained by dividing by the total number of preventive service notifiers n _all (see equation (3)). The state transition rate α obtained in this way is a value indicating the proportion of the preventive service notifiers whose behavior has changed due to the notification of the preventive service.

The state transition rate α may be obtained by dividing the number n _cr of preventive service notifiers classified as “taken / changed” by the total number n _all of preventive service notifiers (formula (4)). reference). The state transition rate α obtained in this way is a value indicating the proportion of the preventive service notifiers whose behavior has changed due to the notification and attendance of the preventive service.

Next, the preventive service effect calculation unit 14 calculates the state number ratio r in the target year (step S318). The state / person ratio r is expressed by a vector represented by Expression (5). Each element in the vector indicates the ratio of the number of persons in each state to all the target persons in the target year. In the formula, r _n is the percentage of people who are "consultation", r _h is the percentage of people who are "not consulted", r _m is the percentage of people who are "mild", r _i is " is the proportion of people is a medium disease ", r _d is the percentage of people who are" severe ", r _f is the percentage of people who are" healthy ", the percentage of people r _e is a" leave " Respectively.

Here, prevention services notifier (when adopting the formula (4) prevention service students) consider a hypothetical state number ratio r _a determined for only. Effect of prevention services notification, and percentage r _n people is "consultation", assuming appear in proportion r _h of human "not visited", the state number ratio r _a is represented by the formula (6) Will be.

Here, the preventive service effect matrix Q calculated by the preventive service effect calculating unit will be described. Preventive Services effect calculating unit 14 calculates a preventive service impact matrix Q state number ratio r _a by multiplying the state number ratio r is obtained (step S322). The preventive service effect matrix Q is a matrix that satisfies the relational expression Q · r = r _a . Impact of prevention services notification, and the ratio r _n of the people is a "consultation", that appeared in the ratio r _h of the people is "not consulted", and the total value of the row direction of the prevention service effect matrix Q is 1 In the case where the normalization condition is satisfied, the preventive service effect matrix Q is expressed by Expression (7).

Therefore, by solving this equation (7) for q _nn and q _hh , the relationship of equation (8) can be obtained.

Therefore, by newly setting q _nn and q _hh as the preventive service effect rate q, the effect matrix can be set as described in Equation (9).

  As described above, the preventive service effect calculating unit 14 can calculate the preventive service effect matrix Q after calculating the preventive service effect rate q (step S320).

  Returning to the description of FIG. When the preventive service effect matrix Q is calculated, the preventive service effect calculating unit 14 multiplies the state transition probability matrix P by the preventive service effect matrix Q to obtain a state transition probability matrix Pa reflecting the effect of the preventive service, for example, It is calculated as a state transition model (step S130) and stored in the state transition model storage unit 26. The preventive service effect calculation unit 14 stores a state transition probability matrix P that does not reflect the effect of the preventive service together with the state transition probability matrix Pa that reflects the effect of the preventive service in the state transition model storage unit 26 as a state transition model. May be. And the prevention service effect evaluation apparatus 1 outputs a state transition model to the terminal device 80, for example (step S132).

This state transition probability matrix Pa is information representing the effect of the preventive service. For example, by subtracting the state transition probability matrix P from the state transition probability matrix Pa, it is possible to quantitatively evaluate in which part of the state transition probability the effect of the prevention service appears. Further, by multiplying the state transition probability matrix Pa over the years, a future state transition probability matrix Pa such as two years later, three years later, and so on can be calculated. For example, the state transition probability matrix Pa after 5 years is obtained by Pa ^5.

In addition, the preventive service effect calculation unit 14 may output another index value as the effect of the preventive service instead of (or in addition to) the state transition probability matrix Pa. For example, the preventive service effect calculation unit 14, as shown in the expressions (10) to (12), the service effect rate β representing the quality of the service and the service target person representing the scale of the service. It is possible to perform evaluation in consideration of the effect of the service and the scale of the service by decomposing it into the rate γ and outputting these index values.

  According to the preventive service effect evaluation apparatus 1 according to the first embodiment described above, the subscriber data, the receipt data, and the medical examination data indicating the use status of the medical institution, and the data related to the execution of the preventive service. Based on the acquired data, the values (state transition rate α, preventive service effect rate q) indicating the change in the hospital visit status due to the implementation of the preventive service are derived, and the preventive service effect matrix Q based on these is used to prevent Since the effect of the service is calculated, the effect of the preventive service can be quantitatively evaluated.

  Moreover, since the state transition probability matrix Pa is obtained by multiplying the state transition probability matrix P by the preventive service effect matrix Q, it is useful for quantifying the effect of the preventive service, which is a change in the state transition probability caused by the preventive service. Information can be obtained.

(Second Embodiment)
Hereinafter, the second embodiment will be described. FIG. 9 is a diagram illustrating a functional configuration of the preventive service effect evaluation apparatus 1A according to the second embodiment. The preventive service effect evaluation apparatus 1A according to the second embodiment includes, in addition to the components included in the preventive service effect evaluation apparatus 1 according to the first embodiment, a state prediction unit 15, a state prediction result comparison unit 16, and A state prediction result storage unit 27 and a state comparison result storage unit 28 are provided. In the second embodiment, the parameter storage unit 21 stores information on a period for performing prediction (prediction period).

  The state prediction unit 15 calculates the number of states that reflect the effect of the prevention service and the number of states that do not reflect the effect of the prevention service in the year in which the prediction period has elapsed from the target year. The state prediction result comparison unit 16 calculates a result of comparing the number of persons that reflects the effect of the preventive service with the number of persons that does not reflect the effect of the preventive service. The state prediction result storage unit 27 stores the processing result by the state prediction unit 15, and the state comparison result storage unit 28 stores the processing result by the state prediction result comparison unit 16.

  FIG. 10 is a flowchart showing the flow of overall processing executed by the preventive service effect evaluation apparatus 1A according to the second embodiment. First, the state determination unit 12 reads information on a target year as a reference for generating a state transition model, a target disease to be processed, and a prediction period from parameters stored in the parameter storage unit 21 ( Step S400). Next, the state determination part 12 reads the data of the subscriber to the health insurance association in the target year stored in the subscriber data storage part 22 (step S402). Next, the state determination unit 12 reads the receipt data for the target year from the receipt data storage unit 23 for each subscriber read in step S102 (step S404). Next, for each of the subscribers read in step S102, the state determination unit 12 uses the data related to the target disease read in step S100 among the data related to the health check performed in the target year. 24 is read (step S406). These data are read from an auxiliary storage device such as an HDD or a flash memory, and expanded in a work area such as a RAM and used for processing described later.

Next, the state determination unit 12 extracts one subscriber according to a predetermined order from the subscribers in the target year read in Step S402 (Step S408). At this time, if there is no extractable subscriber, the process proceeds to step S414. If there is an extractable subscriber, the state determination unit 12 performs a subscriber state determination process (step S410). As the subscriber status determination process, the process shown in the flowchart of FIG. 6 is performed as in the first embodiment. The result of the subscriber determination process is added by the state prediction unit 15 as the number of subscribers belonging to each state in the target year (step S412). By executing the processes of steps S408 to S412 for all subscribers in the target year, the number of subscribers belonging to each state in the target year is calculated. The state number of subscribers belonging to each state is represented as a state number vector N (see equation (13)). In the formula, n _x represents the number of subscribers that belong to the state x.

  Next, the state prediction unit 15 obtains a state transition probability matrix P that does not reflect the effect of the prevention service and a state transition probability matrix Pa that reflects the effect of the prevention service from the state transition model storage unit 26 as the state transition model. Read (step S414). Next, the state prediction unit 15 sets the state number calculated in steps S408 to S412 as an initial state of the state number (step S416).

Next, the state predicting unit 15 calculates the state number reflecting the effect of the preventive service in the next year of the set year by obtaining the product N ^T Pa of the state number vector N and the state transition probability matrix Pa (step S418). ). Here, the superscript T represents transposition. The set year is a year in which the target year is set as an initial state, and is incremented by one year each time the processes of step S418 and step S420 are executed.

Next, the state predictor 15, a state number which does not reflect the effects of prevention services in the next year setting year, is calculated by determining the product N ^{T P} state number vector N and the state transition probability matrix P (step S420 ).

Next, the state prediction unit 15 determines whether the calculation for the prediction period has been completed (step S422). When the prediction period is, for example, 5 years, the state prediction unit 15 sets the state number obtained by the processing in step S418 and step S420 as a new state number and multiplies the state transition probability matrix Pa or P by 5 Run repeatedly. Thus, the state predictor 15, the forecast period y elapsed year from the target year, the state number N ^T Pa ^y reflecting the effect of prevention services, and a state number N ^T P ^y that does not reflect the effects of prevention services Can be calculated. These values are stored in the state prediction result storage unit 27.

Next, the state prediction result comparison unit 16 compares the state number ^N T Pa ^y reflecting the effect of prevention services, and a state number ^N T ^{P y} that does not reflect the effects of prevention services (step S424), for example those and stores the difference ^{{N T Pa y -N T P} y} to calculate the state comparison result storage unit 28. Then, the preventive service effect evaluation apparatus 1A outputs the state comparison result to, for example, the terminal device 80 (step S426).

The state prediction result comparison unit 16 according to the second embodiment, the difference between the state number ^{{N T Pa y -N T P} y}, by multiplying the cost vector represented by the formula (14), calculates the amount of variation in the cost due to the effect of prevention services ^{{N T Pa y -N T P} y} C, may output the variation amount of the costs as a preventative service impact. Where c _x is the annual cost for subscribers in state x.

  According to the preventive service effect evaluation apparatus 1A according to the second embodiment described above, in addition to the same effect as the first embodiment, the difference in the number of state persons reflecting the preventive service effect, or the cost based on this It is possible to calculate an index value that is easy to understand directly, such as a fluctuation amount.

  According to at least one embodiment described above, a data acquisition unit 11 that acquires data indicating the use status of a medical institution such as receipt data and medical examination data, and preventive service related data related to the implementation of the preventive service; Based on the data acquired by the data acquisition unit 11, the prevention service effect calculation unit 14 that calculates the effect of the prevention service by deriving the change in the hospitalization status due to the implementation of the prevention service, and the prevention service effect calculation unit 14 The effect can be evaluated quantitatively.

  Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and the equivalents thereof.

  For example, in the above embodiment, the severity of the disease is set to three levels of “mild”, “moderate”, and “severe”, but a state in which different numbers of stages are provided according to the disease and its treatment characteristics is set. May be.

  Further, in the calculation of the service effect matrix Q, it is assumed that the effect of the service has no effect other than “consultation” and “non-consultation”, but the service effect matrix Q may be calculated as extending to other states. . At this time, since the value of each element of the service effect matrix Q is not necessarily uniquely determined, the state transition probability value of a specific state in the state transition probability when the service effect matrix Q is applied to the state transition probability matrix The service effect matrix Q may be derived by setting a target condition such as maximal or minimal and applying a method such as an approach method.

  In the second embodiment, the difference in the number of persons in the state when the preventive service is implemented and the case where the preventive service is not implemented, or the cost fluctuation is individually compared. However, the weight corresponding to each state is multiplied. By integrating, as one evaluation value, the effect when the service is not performed with respect to the case where the preventive service is performed may be evaluated.

  In addition, as data for judging the effectiveness of preventive services, we use data based on whether or not you have attended a service and whether or not you have changed your behavior. Data that considers the degree of behavior modification, or the case where there are a plurality of different types of behavior modification may be used.

DESCRIPTION OF SYMBOLS 1, 1A ... Preventive service effect evaluation apparatus, 10 ... Processor, 11 ... Data acquisition part, 12 ... State determination part, 13 ... State transition probability calculation part, 14 ... Preventive service effect calculation part, 15 ... State prediction part, 16 ... State prediction result comparison unit, 20 ... storage device, 21 ... parameter storage unit, 22 ... subscriber data storage unit, 23 ... receipt data storage unit, 24 ... medical examination data storage unit, 25 ... preventive service related data storage unit, 26 ... state transition model storage unit, 27 ... state prediction result storage unit, 28 ... state comparison result storage unit, 50 ... communication interface

Claims (6)

A data acquisition unit that acquires data indicating the use status of medical institutions and data related to the implementation of preventive services;
Based on the data acquired by the data acquisition unit, by deriving the change of the hospitalization situation due to the implementation of the preventive service, the preventive service effect calculating unit for calculating the effect of the preventive service;
A preventive service effect evaluation apparatus. A state determination unit that determines the state of the subject including the visit status of the subject based on the data indicating the use status of the medical institution obtained by the data acquisition unit;
A state transition probability calculating unit that calculates the state transition probability of the subject based on the state of the subject determined by the state determining unit;
The preventive service effect calculating unit calculates a value indicating a change in hospitalization status due to the implementation of the preventive service, and the calculated value acts on the state transition probability of the subject calculated by the state transition probability calculating unit. To calculate the effectiveness of preventive services,
The preventive service effect evaluation apparatus according to claim 1. The state transition probability calculating unit derives a state transition matrix having the state of the subject at a reference time as a first argument and the state of the subject at a time later than the reference time as a second argument,
The preventive service effect calculating unit calculates the effect of the preventive service by multiplying the state transition matrix by a preventive service effect matrix including a value indicating a change in hospital visit status due to the execution of the preventive service.
The preventive service effect evaluation apparatus according to claim 2. Based on the effect calculated by the preventive service effect calculating unit, a predicting unit that predicts a difference in the number of states of the target person due to the implementation of the preventive service,
The preventive service effect evaluation apparatus according to claim 2 or 3. Based on the effect calculated by the preventive service effect calculating unit, a prediction unit that predicts cost fluctuations due to the implementation of the preventive service,
The preventive service effect evaluation apparatus according to any one of claims 2 to 4. On the computer,
Get data indicating the usage status of medical institutions and data related to the implementation of preventive services,
Based on the acquired data, the effect of the preventive service is calculated by deriving the change in the hospitalization status due to the implementation of the preventive service.
Prevention service effectiveness assessment program.

Images