Administration management device, administration management method, and program

Abstract

A medication management device that acquires input information including at least one of: a hemoglobin level at the time of examination; a characteristic decrease amount and a characteristic increase amount of an erythropoiesis-stimulating agent; correction moving-average values of these; a dosage amount of the erythropoiesis-stimulating agent, the dosage amount being for setting the hemoglobin level to a target hemoglobin level after the elapse of a unit period; and an item value at the time of examination, the item value pertaining to a plurality of inhibitory items that inhibit increases in the hemoglobin value. The medication management device also acquires output information indicated by the post-administration hemoglobin level at a point in time when the unit period has elapsed from when a dosage amount was administered to a patient. The medication management device estimates the post-administration hemoglobin level, which is output information on the basis of the input of the input information, or estimates the difference between the post-administration hemoglobin level and the target hemoglobin level, on the basis of a plurality of relationships that respectively indicate the relationships between the input information and the output information

Description

[Technical field]

[0001] The present invention relates to an administration management device, an administration management method, and a program for an erythropoietin stimulating factor preparation. This application claims priority based on Japanese Patent Application No. 2020-005654, filed on January 17, 2020, the disclosure of which is incorporated herein in its entirety. [Background technology]

[0002] Many hemodialysis patients suffer from renal anemia, which is one of the major factors that worsen the prognosis. In clinical practice, when a patient's hemoglobin level falls outside the target range, the attending physician determines the dosage of erythropoietin stimulating agent (ESA), a renal anemia treatment drug, to control the patient's hemoglobin level. Patent Document 1 discloses a technology that determines the hemoglobin production rate that will result in the patient's target hemoglobin level from the red blood cell life span, and determines the required ESA dosage. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6190632 Summary of the Invention [Problem to be solved by the invention]

[0004] Here, the technology of the above-mentioned Patent Document 1 determines the hemoglobin production rate assuming that the amount of decrease in the patient's hemoglobin value (hemoglobin concentration) per unit period per day is constant, and determines the required ESA dosage. However, the amount of decrease in hemoglobin value per unit period differs depending on the patient and the characteristics of the patient. In addition, the amount of increase in hemoglobin value per unit period according to the dosage of an erythropoietin stimulating factor preparation also differs depending on the patient and the characteristics of the patient, and is affected by the amount of decrease in hemoglobin value per unit period. Therefore, a technology for managing hemoglobin values ​​more appropriately according to the patient is required.

[0005] Therefore, an object of the present invention is to develop a technique for predicting the amount of change in hemoglobin level, and to provide an administration management device, an administration management method, and a program that solve the above-mentioned problems. [Means for solving the problem]

[0006] According to a first aspect of the present invention, an administration management device includes: a hemoglobin level at examination indicating a hemoglobin level at the time of examination of a patient; an intrinsic decrease amount per unit period indicating an amount by which the hemoglobin level of the patient will decrease if an erythropoietin stimulating factor preparation is not administered in a unit period; a statistical value of the intrinsic decrease amount; an intrinsic increase amount per unit period indicating an amount by which the hemoglobin level of the patient will increase if an erythropoietin stimulating factor preparation is administered in the unit period at a maximum allowable unit amount per unit period, the intrinsic increase amount per unit period indicating an increase amount relative to the hemoglobin level in the case where the erythropoietin stimulating factor preparation is not administered in the unit period; a statistical value of the intrinsic increase amount; the dosage of the erythropoietin stimulating factor preparation for setting the patient's hemoglobin value to a target hemoglobin value after a unit period of time has elapsed, input information including at least one of item values ​​of the patient at the time of the examination related to a plurality of inhibitory items that inhibit an increase in the hemoglobin value, and output information indicated by a post-administration hemoglobin value that indicates the hemoglobin value at a point in time when the unit period has elapsed after the dosage of the erythropoietin stimulating factor preparation is administered to the patient; and a learning unit that generates a hemoglobin value estimation model by machine learning based on the plurality of relationships, which estimates the post-administration hemoglobin value, which is the output information, based on the input of the input information.

[0007] According to a second aspect of the present invention, an administration management device includes: a hemoglobin level at examination indicating a hemoglobin level at the time of examination of a patient; an intrinsic decrease amount per unit period indicating an amount by which the hemoglobin level of the patient will decrease if an erythropoietin stimulating factor preparation is not administered during a unit period; a statistical value of the intrinsic decrease amount; an intrinsic increase amount per unit period indicating an amount by which the hemoglobin level of the patient will increase if an erythropoietin stimulating factor preparation is administered during the unit period at a maximum allowable unit amount per unit period, the intrinsic increase amount per unit period indicating an increase amount relative to the hemoglobin level in the case where the erythropoietin stimulating factor preparation is not administered during the unit period; and a statistical value of the intrinsic increase amount. The device comprises an acquisition unit that acquires a plurality of relationships indicating the relationship between input information including at least one of the dosage of the erythropoietin stimulating factor preparation for setting the hemoglobin value to a target hemoglobin value, the patient's item values ​​at the time of the examination related to a plurality of inhibitory items that inhibit an increase in the hemoglobin value, and output information indicating the difference between a post-administration hemoglobin value, which indicates the hemoglobin value at the time when the unit period has elapsed after the dosage of the erythropoietin stimulating factor preparation is administered to the patient, and the target hemoglobin value, and a learning unit that generates a hemoglobin value estimation model by machine learning based on the plurality of relationships, which estimates the difference between the post-administration hemoglobin value and the target hemoglobin value, which is the output information, based on the input information.

[0008] The above-mentioned administration management device may further include an estimation unit that estimates the post-administration hemoglobin value, which will become the future output information at the time the unit period has elapsed, based on the hemoglobin value estimation model and the input information at the time of a new examination.

[0009] The above-mentioned administration management device may further include an estimation unit that estimates the difference between the post-administration hemoglobin value, which will become the future output information at the time the unit period has elapsed, and the target hemoglobin value based on the hemoglobin value estimation model and the input information at the time of a new examination.

[0010] The above-mentioned administration management device may be provided with an alert output unit that outputs alert information based on the comparison result between the difference between the post-administration hemoglobin value calculated by the estimation unit and the target hemoglobin value and a predetermined standard value.

[0011] The above-mentioned administration management device may be provided with an intrinsic quantity calculation unit that recalculates the intrinsic decrease amount and the intrinsic increase amount of the patient when the difference between the post-administration hemoglobin value calculated by the estimation unit and the target hemoglobin value exceeds the reference value.

[0012] The above-mentioned administration management device may further include a dosage calculation unit that calculates a new dosage of the erythropoietin stimulating factor preparation per unit period to be administered to the patient at the new examination based on the target hemoglobin value, the hemoglobin value at the time of examination, the recalculated intrinsic decrease amount, and the recalculated intrinsic increase amount.

[0013] According to a third aspect of the present invention, the administration management method includes a hemoglobin level at examination, which indicates the hemoglobin level of the patient at the time of examination, an intrinsic decrease amount per unit period, which indicates the amount by which the hemoglobin level of the patient will decrease if an erythropoietin stimulating factor preparation is not administered in a unit period, and a statistical value of the intrinsic decrease amount, an intrinsic increase amount per unit period, which indicates the amount by which the hemoglobin level of the patient will increase if an erythropoietin stimulating factor preparation is administered in the unit period at a maximum allowable unit amount per unit period, and indicates a relative increase amount based on the hemoglobin level in the case where the erythropoietin stimulating factor preparation is not administered in the unit period, and a statistical value of the intrinsic increase amount, The method includes obtaining a plurality of relationships indicating the relationship between input information including at least one of the dosage of the erythropoietin stimulating factor preparation for bringing the patient's hemoglobin value to a target hemoglobin value after the unit period has elapsed and the patient's item values ​​at the time of the examination related to a plurality of inhibitory items that inhibit an increase in the hemoglobin value, and output information indicated by a post-administration hemoglobin value that indicates the hemoglobin value at the time when the unit period has elapsed after administering the dosage of the erythropoietin stimulating factor preparation to the patient, and generating a hemoglobin value estimation model by machine learning based on the plurality of relationships that estimates the post-administration hemoglobin value, which is the output information, based on the input of the input information.

[0014] According to a fourth aspect of the present invention, the administration management method includes: a hemoglobin level at examination indicating a hemoglobin level at the time of examination of a patient; an intrinsic decrease amount per unit period indicating an amount by which the hemoglobin level of the patient will decrease if an erythropoietin stimulating factor preparation is not administered during a unit period; a statistical value of the intrinsic decrease amount; an intrinsic increase amount per unit period indicating an amount by which the hemoglobin level of the patient will increase if an erythropoietin stimulating factor preparation is administered during the unit period at a maximum allowable unit amount per unit period, the intrinsic increase amount per unit period indicating an increase amount relative to the hemoglobin level when the erythropoietin stimulating factor preparation is not administered during the unit period; and a statistical value of the intrinsic increase amount. The method includes acquiring a plurality of relationships indicating the relationship between input information including at least one of the dosage of the erythropoietin stimulating factor preparation for setting the patient's hemoglobin value to a target hemoglobin value and the patient's item values ​​at the time of the examination related to a plurality of inhibitory items that inhibit an increase in the hemoglobin value, and output information indicating the difference between a post-administration hemoglobin value, which indicates the hemoglobin value at the time when the unit period has elapsed after administration of the dosage of the erythropoietin stimulating factor preparation to the patient, and the target hemoglobin value, and generating a hemoglobin value estimation model by machine learning based on the plurality of relationships, which estimates the difference between the post-administration hemoglobin value and the target hemoglobin value, which is the output information, based on the input information.

[0015] The above-mentioned administration management method may further include estimating the post-administration hemoglobin value, which will become the future output information at the time the unit period has elapsed, based on the hemoglobin value estimation model and the input information at the time of a new examination.

[0016] According to a fifth aspect of the present invention, the program causes a computer of an administration management device to receive a hemoglobin level at examination, which indicates the hemoglobin level of the patient at the time of examination, a specific decrease amount per unit period, which indicates the amount by which the hemoglobin level of the patient will decrease if an erythropoietin stimulating factor preparation is not administered in a unit period, and a statistical value of the specific decrease amount, a specific increase amount per unit period, which is the amount by which the hemoglobin level of the patient will increase if an erythropoietin stimulating factor preparation is administered in the unit period at a maximum allowable unit amount per unit period, and which indicates a relative increase amount based on the hemoglobin level in the case where the erythropoietin stimulating factor preparation is not administered in the unit period, and a statistical value of the specific increase amount, The device functions as an acquisition means for acquiring a plurality of relationships indicating the relationships between input information including at least one of the dosage of the erythropoietin stimulating factor preparation for bringing the patient's hemoglobin value to a target hemoglobin value after the unit period has elapsed and the patient's item values ​​at the time of the examination related to a plurality of inhibitory items that inhibit an increase in the hemoglobin value, and output information indicated by a post-administration hemoglobin value that indicates the hemoglobin value at the time the unit period has elapsed after the dosage of the erythropoietin stimulating factor preparation is administered to the patient, and a learning means for generating a hemoglobin value estimation model by machine learning based on the plurality of relationships, which estimates the post-administration hemoglobin value, which is the output information, based on the input of the input information.

[0017] According to a sixth aspect of the present invention, the program causes a computer of an administration management device to receive a hemoglobin value at examination indicating a hemoglobin value at the time of examination of a patient, an intrinsic decrease amount per unit period indicating an amount by which the hemoglobin value of the patient will decrease if an erythropoietin stimulating factor preparation is not administered in a unit period, a statistical value of the intrinsic decrease amount, an intrinsic increase amount per unit period indicating an amount by which the hemoglobin value of the patient will increase if an erythropoietin stimulating factor preparation is administered in the unit period at a maximum allowable unit amount per unit period, the intrinsic increase amount per unit period indicating an increase amount relative to the hemoglobin value in the case where the erythropoietin stimulating factor preparation is not administered in the unit period, and a statistical value of the intrinsic increase amount, and The device functions as an acquisition means for acquiring a plurality of relationships indicating the relationships between input information including at least one of the dosage of the erythropoietin stimulating factor preparation for setting the patient's hemoglobin value to a target hemoglobin value and the patient's item values ​​at the time of the examination related to a plurality of inhibitory items that inhibit an increase in the hemoglobin value, and output information indicating the difference between a post-administration hemoglobin value, which indicates the hemoglobin value at the time when the unit period has elapsed after administration of the dosage of the erythropoietin stimulating factor preparation to the patient, and the target hemoglobin value, and a learning means for generating a hemoglobin value estimation model by machine learning based on the plurality of relationships, which estimates the difference between the post-administration hemoglobin value and the target hemoglobin value, which is the output information, based on the input information.

[0018] The above-mentioned program may further cause the computer to function as an estimation means for estimating the post-administration hemoglobin value, which will become the future output information at the time the unit period has elapsed, based on the hemoglobin value estimation model and the input information at the time of a new examination. Effect of the Invention

[0019] According to an embodiment of the present invention, it is possible to appropriately manage hemoglobin levels according to the patient. [Brief description of the drawings]

[0020] [Figure 1] 1 is a schematic diagram of a dosage management system including a dosage management device according to an embodiment of the present invention; [Diagram 2] FIG. 2 is a diagram illustrating a hardware configuration of an administration management device according to an embodiment of the present invention. [Diagram 3] FIG. 1 is a functional block diagram of an administration management device according to a first embodiment of the present invention. [Figure 4] FIG. 1 is a first diagram showing a processing flow of an administration management device according to a first embodiment of the present invention. [Diagram 5] FIG. 2 is a second diagram showing the processing flow of the administration management device according to the first embodiment of the present invention. [Figure 6] FIG. 13 is a functional block diagram of a modified example of the administration management device. [Figure 7] 1 is a diagram illustrating an inherent decrease amount and an inherent increase amount according to an embodiment of the present invention. FIG. [Figure 8] FIG. 4 is a functional block diagram of an administration management device 1 according to a second embodiment of the present invention. [Figure 9] FIG. 11 is a first diagram showing a processing flow of an administration management device according to a second embodiment of the present invention. [Figure 10] FIG. 11 is a second diagram showing the processing flow of the administration management device according to the second embodiment of the present invention. [Figure 11] FIG. 11 is a diagram showing an overview of the processing of an administration management device according to a second embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] Hereinafter, an administration management device according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a schematic diagram of an administration management system 100 including an administration management device 1 according to the present embodiment. As shown in Fig. 1, the administration management device 1 is a computer. The administration management device 1 may be communicatively connected to an electronic medical record server 2 and may acquire electronic medical record information of patients. The administration management system 100 is configured by the administration management device 1 and the electronic medical record server 2 being communicatively connected to each other.

[0022] The administration management device 1 according to the present embodiment acquires the patient's target hemoglobin value, the patient's current hemoglobin value, and the specific decrease amount per unit period, which indicates the amount by which the patient's hemoglobin value decreases when the erythropoiesis stimulating agent preparation (Erythropoiesis Stimulating Agents) is not administered in the unit period. The administration management device 1 also acquires the specific increase amount per unit period, which is the sum of the amount by which the patient's hemoglobin value increases based on the patient's current hemoglobin value when the maximum allowable unit amount of erythropoiesis stimulating agent preparation per unit period is administered in the unit period and the specific decrease amount per unit period. The administration management device 1 then calculates the dosage of the erythropoiesis stimulating agent preparation per unit period for the patient using the patient's target hemoglobin value, the patient's current hemoglobin value, the patient's specific decrease amount per unit period, and the patient's specific increase amount per unit period. The word "specific" in the terms "specific decrease amount" and "specific increase amount" means that the decrease amount and increase amount are specific to each patient (or the group to which the patient belongs).

[0023] FIG. 2 is a diagram showing a hardware configuration of the administration management device 1 according to this embodiment. As shown in FIG. 2, the administration management device 1 is a computer equipped with various hardware components such as a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, a database 104, a communication module 105, and the like.

[0024] First Embodiment FIG. 3 is a functional block diagram of the administration management device 1 according to the first embodiment. The administration management device 1 starts up when the power is turned on and executes a pre-stored administration management program. As a result, the administration management device 1 is provided with the functions of a control unit 11, a past dose specification unit 12, a hemoglobin value change amount calculation unit 13, an intrinsic increase amount calculation unit 14, an intrinsic decrease amount calculation unit 15, a required dose calculation unit 16, and an administration schedule management unit 17.

[0025] The control unit 11 controls each functional unit of the administration management device 1. The past dose specifying unit 12 calculates the past dose of an erythropoietin stimulating factor preparation administered to a patient during a unit period in order to bring the patient's hemoglobin level to a target hemoglobin level. The hemoglobin value change amount calculation unit 13 calculates the amount of change in hemoglobin value per unit time in the past. The hemoglobin value change amount calculation unit 13 also calculates a predicted value of the amount of change in the hemoglobin value of the patient per unit time based on the amount of erythropoietin stimulating factor preparation administered to the patient per unit time in the past, the specific amount of decrease per unit time in the past, and the specific amount of increase per unit time in the past. The inherent increase amount calculation section 14 calculates an inherent increase amount per unit period for each patient. The intrinsic decrease amount calculation unit 15 calculates the intrinsic decrease amount per unit period for each patient. The required dosage calculation unit 16 calculates the dosage of erythropoietin stimulating factor preparation to be administered to the patient per unit period using the patient's target hemoglobin level, the patient's current hemoglobin level, the patient's specific decrease per unit period, and the patient's specific increase per unit period. The administration schedule management unit 17 generates administration schedule information of an erythropoietin stimulating factor preparation to a patient based on the amount of the erythropoietin stimulating factor preparation to be administered to the patient per unit period.

[0026] FIG. 4 is a first diagram showing the process flow of the administration management device 1. Next, the process flow of the administration management device 1 will be explained step by step. For example, a doctor collects blood from a dialysis patient when performing dialysis on the patient. The doctor passes the blood of the dialysis patient to an examiner. The examiner then measures the patient's current hemoglobin level from the components of the patient's blood. The examiner communicates the hemoglobin level to the doctor. The doctor inputs the patient's current hemoglobin level into the administration management device 1. The administration management device 1 acquires the patient's current hemoglobin level (step S101).

[0027] The control unit 11 of the administration management device 1 transmits information such as the patient's ID (identification information), current hemoglobin value, date and time, etc. to the electronic medical record server 2. The electronic medical record server 2 stores information such as the current hemoglobin value, hemoglobin values ​​for each past blood collection date, each dose of erythropoietin stimulating agent (hereinafter referred to as ESA) administered to the patient once or multiple times per unit period, each unique decrease amount of the patient per unit period calculated once or multiple times in the past and its corrected moving average value, each unique increase amount of the patient per unit period calculated once or multiple times in the past and its corrected moving average value, age, sex, weight, and identification information (group ID) of the group to which the patient belongs based on the patient's characteristics. The group ID may be, for example, identification information of a group set based on the region, medical institution, degree of renal anemia, age, sex, weight, etc., among the patient's characteristics.

[0028] An example of ESA is erythropoietin (EPO), and in this case, the unit period is set to one week. The unit period may vary depending on the type of ESA. For example, for an ESA that has a long-lasting erythropoiesis stimulating effect, a longer unit period such as 2 to 4 weeks is set. The specific decrease per unit period indicates the amount by which the patient's hemoglobin level decreases when the ESA is not administered during the unit period. The specific increase per unit period indicates the increase (specific decrease + increase from the start of measurement of hemoglobin level) assuming that there is no specific decrease in the patient's hemoglobin level when the maximum allowable unit amount of ESA per unit period is administered during the unit period. In other words, the specific increase per unit period is the amount by which the patient's hemoglobin level increases when the maximum allowable unit amount of ESA per unit period is administered during the unit period, and indicates the relative increase based on the hemoglobin level when the erythropoiesis stimulating factor preparation is not administered during the unit period. These specific decrease and increase amounts vary from patient to patient. For example, the intrinsic decrease and increase are indicators that vary depending on the age, sex, weight, blood volume, red blood cell life span, hematopoietic function, residual blood in the dialysis circuit, blood sampling volume and frequency, etc. of each individual patient. When the ESA is EPO, the maximum allowable unit amount that can be administered to a patient with renal anemia undergoing dialysis per unit period in Japan is 9000 IU (IU = international unit). In other words, in Japan, doctors are permitted to administer 0 to 9000 IU of EPO to patients per unit period, i.e., one week.

[0029] The administration management device 1 receives the input of the patient's current hemoglobin value, and when the doctor gives an instruction to start calculating the dosage, it detects the input of the instruction (step S102). The control unit 11 of the administration management device 1 obtains information required for the calculation process of the dosage from the electronic medical record server 2 (step S103). For example, the administration management device 1 obtains from the electronic medical record server 2, based on the patient ID, the hemoglobin value for each past blood collection date, each dose of ESA administered to the patient once or multiple times per unit period in the past, each specific decrease amount of the patient per unit period calculated once or multiple times in the past, each specific increase amount of the patient per unit period calculated once or multiple times in the past, age, sex, weight, and group ID.

[0030] Then, the administration management device 1 starts calculating a new dose of ESA per unit period for the patient (step S104). Specifically, first, the target hemoglobin value of the patient is identified (step S105). For example, the doctor inputs the target hemoglobin value into the administration management device 1, and the control unit 11 acquires and identifies the target hemoglobin value. The target hemoglobin value may be stored in the administration management device 1 in advance. As an example, the target hemoglobin value is 10.5 g / dl. The control unit 11 also identifies the current hemoglobin value input by the doctor (step S106). The control unit 11 also instructs the intrinsic increase amount calculation unit 14 to calculate the intrinsic increase amount per unit period for the patient. The control unit 11 instructs the intrinsic decrease amount calculation unit 15 to calculate the intrinsic decrease amount per unit period for the patient.

[0031] The following describes the case where the intrinsic decrease and increase are calculated from the average dose of ESA (e.g., EPO) or the dose of ESA (e.g., EPO) and the change in hemoglobin level. Compared to calculating the intrinsic decrease and increase from the average dose of ESA, calculating the intrinsic decrease and increase from the dose of ESA and the change in hemoglobin level allows the intrinsic decrease and increase for each individual to be calculated with high accuracy.

[0032] The inherent increase amount calculation unit 14 acquires an inherent increase amount calculation formula stored in association with the group ID of the patient (step S107). The inherent increase amount calculation unit 14 calculates the inherent increase amount of the patient using the inherent increase amount calculation formula (step S108). The inherent increase amount calculation formula is a calculation formula identified by using statistical processing from information (intrinsic increase amount, average EPO dose per unit period, etc.) of each patient in the patient characteristic group indicated by the group ID. As an example, the inherent increase amount calculation formula is shown by the following formula (1). This calculation formula changes depending on the group. For example, this inherent increase amount calculation formula may be a calculation formula other than formula (1) as long as it is a calculation formula obtained by using a statistical method for each patient group, for example, for each data of patients visiting a clinic, for each data of patients of a medical institution belonging to a medical corporation group, or for each data of patients of a medical institution belonging to a region.

[0033]

number

[0034] The intrinsic decline amount calculation unit 15 acquires the intrinsic decline amount calculation formula stored in association with the group ID of the patient (step S109). The intrinsic decline amount calculation unit 15 calculates the intrinsic decline amount of the patient using the intrinsic decline amount calculation formula (step S110). The intrinsic decline amount calculation formula is a calculation formula specified by statistical processing from information (intrinsic decline amount, average EPO dose per unit period, etc.) of each patient in the patient characteristic group indicated by the group ID. As an example, the intrinsic decline amount calculation formula is shown by the following formula (2). The calculation formula changes depending on the group. This intrinsic decline amount calculation formula may be a calculation formula other than formula (2) as long as it is a calculation formula obtained by using a statistical method for each patient group, for example, for each data of patients visiting a clinic, for each data of patients of a medical institution belonging to a medical corporation group, or for each data of patients of a medical institution belonging to a region.

[0035]

number

[0036] The administration management device 1 calculates the intrinsic increase amount using the above formula (1) and the intrinsic decrease amount using formula (2) only when the patient is a new patient. When the patient is not a new patient, the administration management device 1 calculates the intrinsic increase amount and intrinsic decrease amount of the patient in the past with higher accuracy by correcting the intrinsic increase amount and intrinsic decrease amount at the current examination (for example, using the following formula (6)) and the intrinsic decrease amount at the current examination (for example, using the following formula (7)) as shown in FIG. 5 described later. The calculation of the intrinsic decrease amount is performed when no ESA (for example, EPO) has been administered between the previous examination and the current examination.

[0037] The required dosage calculation unit 16 then multiplies the difference between the target hemoglobin level and the patient's current hemoglobin level by the patient's intrinsic decrease per unit time period, and the ESA unit amount per increase in the patient's hemoglobin level, obtained by dividing the patient's maximum allowable unit amount by the patient's intrinsic increase (see formula (3)). In this way, the required dosage calculation unit 16 calculates the patient's dosage of ESA (EPO) per unit time period (dosage index value) (step S111). The ESA unit amount per increase in the patient's hemoglobin level may be a linear approximation of multiple relationships between the patient's intrinsic increase in the past and the ESA dosage.

[0038]

number

[0039] By the above processing, the ESA dosage per unit period for the patient can be calculated. This dosage is the dosage per unit period calculated based on the specific decrease and increase in hemoglobin value per unit period specified for a group specified based on patient characteristics. Therefore, the administration management device 1 can calculate a more appropriate ESA dosage per unit period for the patient. Furthermore, by displaying the dosage on the monitor of the administration management device 1, the doctor can easily grasp the more appropriate ESA dosage per unit period for the patient.

[0040] FIG. 5 is a second diagram showing the process flow of the administration management device 1. In the above process, the intrinsic increase amount and the intrinsic decrease amount are calculated using a calculation formula according to the group to which the patient belongs based on the patient characteristics, but the intrinsic increase amount and the intrinsic decrease amount may be specified or calculated by other methods. The process flow in this case will be described with reference to FIG.

[0041] The processes from step S101 to step S106 are the same as the process flow shown in Fig. 4. Next, the past dose specifying unit 12 acquires the dose of ESA administered to the patient in the past during a unit period in order to bring the patient's hemoglobin level into the target hemoglobin level (step S207). The past dose specifying unit 12 may calculate this statistical value. Alternatively, the past dose specifying unit 12 may acquire this statistical value, which has been calculated in advance by the electronic medical record server 2, from the electronic medical record server 2. The calculation of the statistical value of the dose may be, for example, the average value of the doses per one or more past unit periods.

[0042] The hemoglobin value change amount calculation unit 13 then acquires the amount of change in hemoglobin value per unit time in the past. The hemoglobin value change amount calculation unit 13 also acquires the statistical value of the amount of ESA (EPO) administered to the patient per unit time in the past, the patient's specific decrease per unit time in the past, and the patient's specific increase per unit time in the past. These pieces of information are included in the information acquired from the electronic medical record server 2.

[0043] The hemoglobin level change calculation unit 13 may multiply the ratio of the statistical value of the ESA dose per unit period to the maximum allowable unit amount of ESA per unit period by the intrinsic increase per unit period, and subtract the intrinsic decrease per unit period from the multiplied value to calculate the change in hemoglobin level of the patient per unit period. In other words, the hemoglobin level change calculation unit 13 can calculate the change in hemoglobin level per unit period after administering ESA to the patient to reach the target hemoglobin level using the statistical value of the dose, the intrinsic decrease, and the intrinsic increase according to formula (4) (step S208).

[0044]

number

[0045] More specifically, when the ESA is EPO, formula (4) can be expressed as formula (5).

[0046]

number

[0047] Then, the intrinsic increase amount calculation unit 14 calculates a new intrinsic increase amount per unit period based on the amount of change in hemoglobin value actually measured by the patient per unit period (amount of change in hemoglobin value after administration) and the amount of change in hemoglobin value calculated by the above formula (5) (step S209). In many cases, the intrinsic increase amount calculation unit 14 calculates the intrinsic increase amount based on the amount of change in hemoglobin value actually measured (amount of change in hemoglobin value after administration). Specifically, since the ESA (EPO) dose and the amount of change in hemoglobin value per unit period at the time of administration of that dose are known, the new intrinsic increase amount per unit period can be calculated by formula (6). This intrinsic increase amount is a patient-specific intrinsic increase amount.

[0048]

number

[0049] Reticulocyte hemoglobin measurement can measure the amount of newly produced hemoglobin. Basal hemoglobin production is the amount when there is no hemopoietic stimulation, and reactive hemoglobin production is the relative increase from basal hemoglobin production when there is hemopoietic stimulation. Reactive hemoglobin production is highly correlated with ESA concentration. The intrinsic decrease determined by a normal hemoglobin test is inversely correlated with basal hemoglobin production. The relative increase in hemoglobin due to ESA administration compared to the intrinsic decrease in a normal hemoglobin test corresponds to reactive hemoglobin production and is highly correlated with the ESA dose. This relative increase in hemoglobin is equal to the intrinsic increase when the maximum allowable unit amount of erythropoietin stimulating factor preparation is administered in a unit period. Intrinsic increase and decrease will cause errors in a single measurement. For this reason, the effect of errors can be reduced by repeatedly measuring and averaging the intrinsic increase and decrease.

[0050] The intrinsic decline amount calculation unit 15 calculates a new intrinsic decline amount per unit period using formula (7) (step S210). Since the ESA (EPO) dose and the change in hemoglobin value per unit period when that dose is administered are known, the intrinsic decline amount calculation unit 15 can calculate a new intrinsic decline amount per unit period using formula (7). This intrinsic decline amount is a patient-specific intrinsic decline amount.

[0051]

number

[0052] The intrinsic decrease amount calculation unit 15 may receive an input of an intrinsic decrease amount observed in actual clinical practice, and use this received intrinsic decrease amount. For example, when a patient's condition is good, it is expected that the increase in hemoglobin value per unit period during ESA administration is relatively large, so a doctor may decide to temporarily suspend ESA administration to the patient. At this time, the decrease in hemoglobin value per unit period may be observed without administering ESA, and this value may be estimated as the intrinsic decrease amount. This intrinsic decrease amount based on the observation results of actual clinical practice is an intrinsic decrease amount specific to the patient. Alternatively, the intrinsic decrease amount calculation unit 15 may obtain clinical data including daily hemoglobin values ​​of a patient in good condition for whom ESA administration has been discontinued, and calculate the intrinsic decrease amount per unit period based on this clinical data.

[0053] The intrinsic increase amount calculation unit 14 and the intrinsic decrease amount calculation unit 15 may use the intrinsic increase amount and intrinsic decrease amount per unit period of the past of the patient and the intrinsic increase amount and intrinsic decrease amount per new unit period to calculate a weighted average in which the intrinsic increase amount and intrinsic decrease amount per unit period of the past are weighted more than the intrinsic increase amount and intrinsic decrease amount per new unit period, and may set these values ​​as the intrinsic increase amount (an example of a statistical value of the intrinsic increase amount) and the intrinsic decrease amount (an example of a statistical value of the intrinsic decrease amount) per unit period of the new patient. In this case, for example, the intrinsic increase amount calculation unit 14 calculates the intrinsic increase amount per new unit period by the formula (8).

[0054]

number

[0055] Alternatively, for example, the intrinsic degradation amount calculation unit 15 calculates a new intrinsic degradation amount per unit period by using equation (9).

[0056]

number

[0057] The inherent increase calculation unit 14 may calculate a new inherent increase per unit period according to equation (10) using a modified moving average (MMA) method.

[0058]

number

[0059] The inherent degradation amount calculation unit 15 may calculate a new inherent degradation amount per unit period according to equation (11) using a modified moving average (MMA) method.

[0060]

number

[0061] As shown in FIG. 6, the administration management device 1 may include an acquisition unit 18 instead of the intrinsic increase amount calculation unit 14 and the intrinsic decrease amount calculation unit 15, and the acquisition unit 18 may acquire the intrinsic increase amount and the intrinsic decrease amount recorded in association with the patient ID from the electronic medical record server 2. For example, the intrinsic increase amount and the intrinsic decrease amount may be values ​​determined based on the measurement results of the reticulocyte hemoglobin measurement and recorded in the electronic medical record server 2. In this embodiment, it is assumed that the above-mentioned statistical values ​​of the intrinsic increase amount and the intrinsic decrease amount calculated by the administration management device 1 based on the intrinsic increase amount and the intrinsic decrease amount of the patient in the past are registered in the electronic medical record server 2. The above-mentioned statistical values ​​of the intrinsic increase amount and the intrinsic decrease amount may be a moving average value or a modified moving average value of the intrinsic increase amount and the intrinsic decrease amount. The moving average value is the average value of the past n values ​​(n represents an integer of 2 or more). The modified moving average value is the average value calculated by weighting the past n values ​​in a predetermined period in ascending order of time.

[0062] FIG. 7 is a diagram for explaining the inherent decrease amount and the inherent increase amount. If the patient's current hemoglobin level at time A falls below the target hemoglobin level, the doctor will administer ESA so that the hemoglobin level will reach the target hemoglobin level at time B, one week after time A as a unit period. It is desirable that this administration will improve the patient's hemoglobin level over the unit period, so that it reaches the target hemoglobin level, as shown by arrow L1 in Figure 7. Arrow L1 in Figure 7 is an example, and if the patient is in poor physical condition or if there is a large discrepancy between the hemoglobin level and the target hemoglobin level, it is desirable for the patient's hemoglobin level to reach near the target hemoglobin level more quickly. Although not shown in Figure 7, if the patient's hemoglobin level significantly exceeds the target hemoglobin level, measures such as interrupting administration of ESA will be taken.

[0063] An example of the change in hemoglobin level of a patient when ESA is not administered during a unit period is shown in L2. The amount of decrease in hemoglobin level from time point A, which is the start point of the unit period, to time point B after the unit period when ESA is not administered is defined as the intrinsic decrease. An example of the change in hemoglobin level of a patient when the maximum allowable unit amount of ESA per unit period is administered is shown in L3. When the maximum allowable unit amount of ESA per unit period is administered, the intrinsic increase is the sum of the amount of increase in hemoglobin level from time point A, which is the start point of the unit period, to time point B after the unit period, and the intrinsic decrease. The intrinsic increase amount calculation unit 14 and the intrinsic decrease amount calculation unit 15 may calculate the intrinsic increase amount and the intrinsic decrease amount by any of the methods described above, and may use past intrinsic increase amounts and intrinsic decrease amounts and the latest intrinsic increase amounts and intrinsic decrease amounts to reflect inter-individual and intra-individual differences, and calculate the intrinsic increase amount and the intrinsic decrease amount for each patient by machine learning.

[0064] The required dosage calculation unit 16 multiplies the difference between the target hemoglobin level and the patient's current hemoglobin level by the patient's specific decrease per unit time, and the ESA unit amount per increase in the patient's hemoglobin level, which is obtained by dividing the patient's maximum allowable unit amount by the patient's specific increase (see formula (3)). In this way, the required dosage calculation unit 16 calculates the patient's new future ESA (EPO) dosage per unit time (dosage index value) (step S111). The administration management device 1 then repeats the above-mentioned process when calculating the ESA dosage, based on the data newly stored in the electronic medical record server 2 or the like.

[0065] If the calculated dosage exceeds the maximum allowable unit amount per unit period (for example, 9000 IU for EPO), the required dosage calculation unit 16 identifies the maximum allowable unit amount (for example, 9000 IU for EPO) as the ESA (EPO) per unit period for the patient in the future.

[0066] When the calculated dosage is less than the minimum dosage set for an individual patient, the required dosage calculation section 16 may specify the minimum dosage as the future dosage of ESA (EPO) per unit period for the patient.

[0067] When the input current hemoglobin value is close to the target hemoglobin value that is included in a predetermined range based on the target hemoglobin value, the control unit 11 of the administration management device 1 does not need to calculate a new ESA (EPO) dosage per unit period for the patient in the future. In this case, the control unit 11 of the administration management device 1 may specify the dosage previously calculated for the patient as the ESA (EPO) dosage per unit period for the patient in the future.

[0068] Then, the administration schedule management unit 17 generates administration schedule information indicating the calculated future ESA dosage per unit period for the patient, and outputs it to the monitor of the administration management device 1. This allows the doctor to confirm the future ESA dosage per unit period for the patient, and administers ESA based on that ESA dosage per unit period at the doctor's discretion.

[0069] According to the above process, it is possible to calculate and output an appropriate ESA dosage per unit period based on the intrinsic increase or decrease of each patient. Furthermore, a doctor can check this administration schedule information and easily administer ESA based on the ESA dosage per unit period appropriate for the patient.

[0070] Furthermore, according to the processing of the administration management device described above, the ESA dosage can be calculated and presented without the need for any special additional examination. By administering ESAs in the right dosage, hemoglobin levels can be stabilized, reducing the frequency of ESA dosage adjustments, improving patient prognosis, and reducing the clinical workload of doctors and nursing staff. Furthermore, the above-described process can reduce errors in calculating dosages and recording errors in electronic medical charts, etc. In addition, by properly calculating the dosage for each patient, overdosing can be prevented, and a reduction in the amount of ESA administered is expected to reduce medical costs.

[0071] The administration schedule management unit 17 determines the allocation amount based on the new ESA (EPO) dosage amount per unit period for the patient calculated by the above-mentioned process. For example, containers containing ESA drug solutions are managed in different sizes, and the volumes of each drug solution are different. The doctor must determine which size of container to use and when to administer to the patient based on the dosage amount per unit period for the patient, the planned number of visits per unit period, and the difference between the patient's current hemoglobin value and the target hemoglobin value. Usually, the drug solution contained in the container is administered during one patient visit to the hospital.

[0072] In such a case, the administration schedule management unit 17 judges whether the patient's condition is good or bad based on, for example, the difference between the patient's current hemoglobin value and its target hemoglobin value. The administration schedule management unit 17 may judge the patient's condition to be bad if the difference between the patient's current hemoglobin value and its target hemoglobin value is equal to or greater than a predetermined threshold. Similarly, the administration schedule management unit 17 may judge the patient's condition to be bad if the current hemoglobin value and the predicted hemoglobin value are equal to or greater than a predetermined threshold. When the administration schedule management unit 17 judges the patient's condition to be bad, it may issue an alert and allocate the dosage in a unit period to the number of planned visits to the hospital, and allocate a larger amount of ESA to be administered early.

[0073] The administration schedule management unit 17 uses a predetermined dosage distribution management calculation formula to determine the size of the liquid medicine container to be used for each scheduled number of visits per unit period. In the case of receiving hemodialysis treatment three times a week, the administration schedule management unit 17 generates administration schedule information so that ESA is administered every other day or every other day. However, in generating this administration schedule information, if the administration interval becomes long, the administration schedule management unit 17 may determine the size of the liquid medicine container so that the administration amount is larger on the scheduled visit day before the interval so that the hemoglobin value does not decrease during that time.

[0074] When the administration schedule management unit 17 determines that the patient's condition is not bad based on the patient's current hemoglobin value, the administration schedule management unit 17 may determine the size of the liquid medicine container so that the same amount of ESA can be administered for each scheduled visit date. The administration schedule management unit 17 may also set a gradient distribution of the dosage so that a larger amount of ESA can be administered first to reach the target hemoglobin value early. The size of the liquid medicine container determined for each scheduled visit date varies depending on the type of ESA, the dosage per unit period, the capacity of the liquid medicine container, the patient's condition at the time of calculation, the number of scheduled visits per unit period, and other factors, but the size of the liquid medicine container for each visit is determined using a predetermined algorithm.

[0075] Second Embodiment Next, an administration management device according to a second embodiment will be described. FIG. 8 is a functional block diagram of the administration management device 1 according to the second embodiment. The administration management device 1 according to the second embodiment further includes an acquisition unit 18, a learning unit 19, an estimation unit 20, and an alert output unit 21 in addition to the configuration of the administration management device 1 according to the first embodiment. The acquisition unit 18 acquires information input by a doctor or nurse or information stored in a storage device such as a database 104. The learning unit 19 calculates a hemoglobin value estimation model using a machine learning technique. The estimation unit 20 uses the hemoglobin value estimation model to estimate a future post-administration hemoglobin value when a dose of ESA (EPO) per unit time is administered to a patient, and a difference between the post-administration hemoglobin value and a target hemoglobin value. The alert output unit 21 outputs alert information indicating an abnormality (abnormality in the dose of ESA (EPO)) when the difference between the post-administration hemoglobin value calculated by the estimation unit 20 and the target hemoglobin value is equal to or greater than a predetermined threshold value.

[0076] (Generation of hemoglobin value estimation model) FIG. 9 is a first diagram showing the process flow of the administration management device 1 according to the second embodiment. A doctor inputs the hemoglobin value measured at the previous examination of a dialysis patient into the administration management device 1. The acquisition unit 18 of the administration management device 1 acquires the hemoglobin value measured at the previous examination of the patient (hereinafter referred to as the hemoglobin value at examination) as input information (step S301). The doctor also inputs item values ​​at the previous examination related to one or more inhibitory items that inhibit an increase in the hemoglobin value of the dialysis patient. The acquisition unit 18 of the administration management device 1 acquires item values ​​related to those inhibitory items as input information (step S302). An example of an inhibitory item is one or more item values ​​related to iron (FER (ferritin), TSAT (transferrin saturation)), CRP (C-reactive protein), body temperature, BUN (urea nitrogen), nutrition, fatigue, carnitine, hyperthyroidism (osteitis fibrosa), bleeding tendency, etc. The frequency with which the doctor inputs item values ​​into the administration management device 1 varies depending on the inhibitory item. Furthermore, the acquisition unit 18 of the administration management device 1 may acquire an index value of the dose of ESA (EPO) per unit time period of the patient as input information. For example, this index value may be the dose of ESA (EPO) calculated by the required dose calculation unit 16 (see step S111). In this case, the acquisition unit 18 of the administration management device 1 acquires the calculated dose of ESA (EPO) from the required dose calculation unit 16.

[0077] The acquisition unit 18 of the administration management device 1 acquires as input information the intrinsic increase amount and its corrected moving average value calculated at the previous examination in step S108 or step S209 of the first embodiment, the intrinsic decrease amount and its corrected moving average value calculated at the previous examination in step S110 or step S210, and the actual dose administered based on the dose of ESA (EPO) calculated at the previous examination in step S111 (step S303). The administration management device 1 transmits these input information to the electronic medical record server 2 together with the patient's ID. As a result, the electronic medical record server 2 stores the input information in association with the patient's ID. If these input information have already been registered in the electronic medical record server 2, instead of the doctor inputting the input information to the administration management device 1, the administration management device 1 may acquire the input information from the electronic medical record server 2.

[0078] The doctor also collects blood from the dialysis patient when performing dialysis on the patient during this consultation. The doctor hands the dialysis patient's blood to an examiner. The examiner then measures the patient's current hemoglobin level from the components of the patient's blood. The examiner communicates the hemoglobin level to the doctor. The doctor inputs the patient's current hemoglobin level (hereinafter, post-administration hemoglobin level) to the administration management device 1. The acquisition unit 18 of the administration management device 1 acquires the dialysis patient's post-administration hemoglobin level as output information (step S304). Alternatively, the acquisition unit 18 of the administration management device 1 acquires the difference between the dialysis patient's post-administration hemoglobin level and the target hemoglobin level as output information.

[0079] The acquisition unit 18 records the input information and output information in the result accumulation table of the database 104 in association with each other. Based on the doctor's operation, the administration management device 1 repeatedly records the input information and output information in the result accumulation table of the database 104 and the electronic medical record server 2 in association with each other every time a large number of dialysis patients are examined. As described above, the input information includes the previous dose of ESA (EPO), the hemoglobin value measured in the previous examination, the intrinsic increase in hemoglobin value, the corrected moving average of the intrinsic increase, the intrinsic decrease in hemoglobin value, the corrected moving average of the intrinsic decrease, and one or more inhibiting items that inhibit the increase in the hemoglobin value of the dialysis patient. However, the input information may be information that includes any one or more of the dose of ESA (EPO) in the previous examination, the hemoglobin value measured in the previous examination, the intrinsic increase in hemoglobin value calculated in the previous examination, the corrected moving average of the intrinsic increase calculated in the previous examination, the intrinsic decrease in hemoglobin value calculated in the previous examination, the corrected moving average of the intrinsic decrease calculated in the previous examination, and one or more inhibiting items identified by the previous examination, etc. The new input information may also include input information specified by examination, calculation, or acquisition during a specified period from the previous time, the time before last, or even more than that.

[0080] Then, based on the doctor's instruction or a predetermined timing, the learning unit 19 of the administration management device 1 acquires from the database 104 a large number of relationships each indicating a relationship between the input information and the output information. The learning unit 19 uses the relationships between the input information and the output information to calculate parameters for specifying a hemoglobin level estimation model that estimates the post-administration hemoglobin level, which is the output information, or the difference between the post-administration hemoglobin level, which is the output information, and the target hemoglobin level, when the input information is input, by machine learning (step S305). The calculation of parameters for specifying a hemoglobin level estimation model by machine learning is also called generation of a hemoglobin level estimation model. The hemoglobin level estimation model is specified by parameters such as weight coefficients of a neural network that outputs output information corresponding to the input information when the input information is input. The calculation of parameters for specifying a hemoglobin level estimation model may be performed by using a known machine learning method. The learning unit 19 records the parameters for specifying a hemoglobin level estimation model in the database 104 or the like (step S306).

[0081] Through the above processing, the administration management device 1 can calculate parameters for generating a hemoglobin value estimation model that calculates the patient's post-administration hemoglobin value after a unit period, or the difference between the post-administration hemoglobin value and the target hemoglobin value, based on input information consisting of the patient's hemoglobin value at the time of examination, the item values ​​of multiple inhibitory items that inhibit an increase in hemoglobin value, the patient's intrinsic increase amount and its corrected moving average value, the intrinsic decrease amount and its corrected moving average value, the ESA (EPO) dose calculated at the previous examination, and the hemoglobin value measured at the previous examination, and output information indicating the patient's hemoglobin value after a unit period has elapsed.

[0082] The learning unit 19 may identify which of a plurality of groups a dialysis patient belongs to based on the difference in the patient's symptoms, and calculate parameters for generating a hemoglobin level estimation model specific to each group based on the relationship between the input information and the output information for each group. In this case, the learning unit 19 records in the database 104 a group ID for identifying a group, in association with parameters for generating a hemoglobin level estimation model for the group indicated by the ID.

[0083] (Estimation of post-administration hemoglobin levels using a hemoglobin level estimation model) FIG. 10 is a second diagram showing the process flow of the administration management device 1 according to the second embodiment. FIG. 11 is a diagram showing an outline of the process of the administration management device 1 according to the second embodiment. After the hemoglobin level estimation model is generated, the doctor can instruct the administration management device 1 to calculate the dialysis patient's post-administration hemoglobin level after a unit period of time using the hemoglobin level estimation model, or the difference between the post-administration hemoglobin level and the target hemoglobin level, from the time of the next new examination.

[0084] A doctor collects blood from a dialysis patient when performing dialysis on the patient in a new consultation. The doctor hands the blood of the dialysis patient to an examiner. The examiner then measures the patient's current hemoglobin value from the components of the patient's blood. The examiner communicates the hemoglobin value to the doctor. The doctor inputs the patient's current hemoglobin value into the administration management device 1. The acquisition unit 18 of the administration management device 1 acquires the patient's current hemoglobin value (hereinafter referred to as the hemoglobin value at the time of examination) as input information (step S401). The doctor also identifies item values ​​related to multiple inhibiting items that inhibit an increase in the hemoglobin value of the dialysis patient through a new consultation or an examination at that time, and inputs the item values ​​related to the inhibiting items into the administration management device 1. The acquisition unit 18 of the administration management device 1 acquires item values ​​related to the inhibiting items identified through the patient's new consultation or examination as input information (step S402).

[0085] The acquisition unit 18 of the administration management device 1 acquires as input information (step S403) the intrinsic increase amount and its corrected moving average value calculated in a new examination or stored in association with the patient's ID in step S108 or step S209 of the first embodiment, the intrinsic decrease amount and its corrected moving average value calculated in a new examination or stored in association with the patient's ID in step S110 or step S210, the dose of ESA (EPO) calculated in the previous examination in step S111, and the hemoglobin value measured in the previous examination in step S101. As described above, the input information may include any one or more of the dose of ESA (EPO) in the previous examination, the hemoglobin value measured in the previous examination, the intrinsic increase amount of hemoglobin value calculated last time, the corrected moving average value of the intrinsic increase amount calculated last time, the intrinsic decrease amount of hemoglobin value calculated last time, the corrected moving average value of the intrinsic decrease amount calculated last time, and one or more inhibition items identified by the previous examination or the like.

[0086] Then, the doctor instructs the administration management device 1 to estimate the post-administration hemoglobin level of the patient at a future time point after the unit period has passed. The administration management device 1 accepts the input of the estimation instruction (step S404). The estimation unit 20 of the administration management device 1 acquires parameters for generating a hemoglobin level estimation model from the database 104 based on the instruction to estimate the post-administration hemoglobin level, and inputs input information to the hemoglobin level estimation model specified based on the parameters. As a result, the estimation unit 20 calculates output information corresponding to the input information using the hemoglobin level estimation model (step S405). The estimation unit 20 outputs the output information to the alert output unit 21. If the hemoglobin level estimation model is a model for calculating the post-administration hemoglobin level, the estimation unit 20 calculates and outputs the post-administration hemoglobin level of the patient after the unit period. For example, the estimation unit 20 calculates and outputs the post-administration hemoglobin level of the patient after the unit period when the dose of ESA (EPO) calculated by the required dose calculation unit 16 is administered. Alternatively, if the hemoglobin level model is a model that calculates the difference between a post-administration hemoglobin level and a target hemoglobin level, the estimation unit 20 outputs the difference between the patient's post-administration hemoglobin level after a unit period and the patient's target hemoglobin level as output information to the alert output unit 21. For example, the estimation unit 20 outputs the difference between the patient's post-administration hemoglobin level after a unit period and the patient's target hemoglobin level when the dose of ESA (EPO) calculated by the required dose calculation unit 16 is administered to the alert output unit 21 as output information.

[0087] In the above estimation process, the estimation unit 20 may estimate output information by further using the hemoglobin values ​​at the time of examination of the patient at multiple past examinations as input information. In this case, the estimation unit 20 acquires the hemoglobin values ​​at the time of examination measured at multiple past examinations that are linked to the patient's ID and recorded in the electronic medical record server 2, adds the acquired values ​​as further input information, and inputs the input information into the hemoglobin value estimation model. Then, the estimation unit 20 calculates output information of the hemoglobin value estimation model in the same manner as the above process.

[0088] Alternatively, in the above-mentioned estimation process, the estimation unit 20 may estimate the output information using, as input information, moving average values ​​of the hemoglobin value at examination and the dosage of the patient at multiple past examinations, which are included in the input information. In this case, the estimation unit 20 acquires the hemoglobin value at examination and the dosage measured at multiple past examinations that are linked to the patient's ID and recorded in the electronic medical record server 2, and calculates the moving average values. The estimation unit 20 then further adds the moving average values ​​of the hemoglobin value at examination and the dosage as input information, and inputs the input information into the hemoglobin value estimation model. The estimation unit 20 then calculates the output information of the hemoglobin value estimation model in the same manner as in the above-mentioned process.

[0089] The estimation unit 20 records the statistical values ​​and category reference values ​​of the output information estimated in the past for each patient. Alternatively, the estimation unit 20 may record the statistical values ​​and category reference values ​​of the output information estimated in the past for each group including the patient. The category reference value is, for example, a threshold value that determines whether the post-administration hemoglobin value, which is the output information, or the difference between the post-administration hemoglobin value and the target hemoglobin value is in the range of an excessive value, the range of an expected value, or the range of an insufficient value. This category reference value may be a reference value that is set in advance for the group to which the patient belongs.

[0090] The alert output unit 21 receives an input of an estimated post-administration hemoglobin level, or a difference between the estimated post-administration hemoglobin level and the target hemoglobin level, or a statistical value thereof, or a category reference value corresponding to such information. When the received information is an estimated post-administration hemoglobin level, the alert output unit 21 calculates the difference between the post-administration hemoglobin level and the target hemoglobin level. The alert output unit 21 compares the value of the difference between the estimated post-administration hemoglobin level and the target hemoglobin level with the category reference value to determine whether to output alert information (step S406).

[0091] When the difference value (absolute value) between the post-administration hemoglobin level and the target hemoglobin level exceeds the division reference value, the difference value is within the range of the excessive value or the undervalue, so the alert output unit 21 displays the difference value. When the difference value (absolute value) between the estimated post-administration hemoglobin level and the target hemoglobin level exceeds the division reference value (when it is excessive or undervalue), the alert output unit 21 outputs alert information indicating an abnormality on the screen (step S407).

[0092] When the difference (absolute value) between the post-administration hemoglobin level and the target hemoglobin level does not exceed the classification reference value (when it is within the range of the expected value), the alert output unit 21 notifies the estimation unit 20 of information of no alert (no abnormality). When the estimation unit 20 receives the notification of the no alert information, if the output information is the difference between the post-administration hemoglobin level and the target hemoglobin level, the estimation unit 20 adds the difference to the target hemoglobin level to calculate the estimated post-administration hemoglobin level. The estimation unit 20 displays the calculated estimated post-administration hemoglobin level as an estimation result on the monitor of the administration management device 1 (step S408). The estimation unit 20 also displays the dose of ESA (EPO) obtained in step S403 and used to calculate the estimated post-administration hemoglobin level as an index value on the monitor of the administration management device 1 (step S409). The doctor checks the target hemoglobin level, the estimated post-administration hemoglobin level, the dose of ESA (EPO) used to calculate the post-administration hemoglobin level, the difference between the estimated post-administration hemoglobin level and the target hemoglobin level, alert information, etc., and then uses the confirmed information as an indicator to determine the dose he or she has determined himself or herself, and administers this dose to the patient as the dose of ESA (EPO) for this examination.

[0093] The alert output unit 21 may indicate which of multiple ranges separated by more threshold values, for example multiple ranges separated into five or ten levels, to which the difference between the post-administration hemoglobin level and the target hemoglobin level falls, and output alert information based on the difference between the post-administration hemoglobin level and the target hemoglobin level.

[0094] When the difference between the post-administration hemoglobin level and the target hemoglobin level is a large value that falls within the range of excessive values ​​or insufficient values ​​that generate an alert, the alert output unit 21 instructs the intrinsic increase amount calculation unit 14, the intrinsic decrease amount calculation unit 15, and the required dosage calculation unit 16 to perform recalculation. In response to this instruction, the alert output unit 21 outputs the estimated post-administration hemoglobin level or the difference between the estimated post-administration hemoglobin level and the target hemoglobin level to the intrinsic increase amount calculation unit 14 and the intrinsic decrease amount calculation unit 15. The intrinsic decrease amount calculation unit 15 uses the difference between the estimated post-administration hemoglobin level and the target hemoglobin level to calculate a new intrinsic decrease amount by equation (12), which is similar to equation (7) (step S410).

[0095]

number

[0096] The intrinsic increase amount calculation unit 14 calculates a new intrinsic increase amount using the new intrinsic decrease amount and the difference between the estimated post-administration hemoglobin level and the target hemoglobin level (post-administration hemoglobin level change amount) according to equation (13), which is similar to equation (6) (step S411).

[0097]

number

[0098] Then, the required dosage calculation unit 16 calculates the dosage of ESA (EPO) in the same manner as in step S111, using the patient's intrinsic increase amount newly calculated by the intrinsic increase amount calculation unit 14 and the patient's intrinsic decrease amount newly calculated by the intrinsic decrease amount calculation unit 15, based on the instruction of the alert output unit 21 (step S412). This processing by the required dosage calculation unit 16 is one aspect of the processing for calculating a new dosage of an erythropoietin stimulating factor preparation per unit period to be administered to the patient at a new examination, based on the target hemoglobin value, the examination hemoglobin value, the recalculated intrinsic decrease amount, and the recalculated intrinsic increase amount. The maximum value of the value calculated in step S412 is the maximum dosage set by law (for example, 9000 IU for EPO).

[0099] The required dosage calculation unit 16 displays the new calculated dosage of ESA (EPO) on the monitor (step S413). The doctor checks the estimated post-administration hemoglobin value and the ESA (EPO) dosage index used to calculate that value, and administers the dosage he / she determined based on that dosage to the patient as the dosage of ESA (EPO) for the current examination.

[0100] Through the above process, the administration management device 1 can estimate the post-administration hemoglobin value after a unit period of time by administration of the calculated dose of ESA (EPO). If the difference (absolute value) between the estimated post-administration hemoglobin value after a unit period of time by administration of the calculated dose of ESA (EPO) and the target hemoglobin value does not exceed a predetermined classification reference value, the administration management device 1 can notify the doctor of administration using the dose of ESA (EPO) to achieve the estimated post-administration hemoglobin value after the unit period as an index. If the difference (absolute value) between the estimated post-administration hemoglobin value after a unit period of time by administration of the calculated dose of ESA (EPO) and the target hemoglobin value exceeds a predetermined threshold value, the administration management device 1 can calculate a new dose to achieve the patient's hemoglobin value after the unit period of time to the target hemoglobin value.

[0101] The administration management device 1 described above has a computer system therein. The programs for causing the administration management device 1 to perform the above-mentioned processes are stored in a computer-readable recording medium of the administration management device 1, and the computer of the administration management device 1 reads and executes the programs to perform the above-mentioned processes. The computer-readable recording medium here refers to a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, etc. Also, the computer program may be distributed to a computer via a communication line, and the computer that receives the program may execute the program.

[0102] The above program may be for implementing some of the functions of each of the above-mentioned processing units. Furthermore, the above program may be a so-called differential file (differential program) that can implement the above-mentioned functions in combination with a program already recorded in the computer system. [Industrial Applicability]

[0103] The present invention may be applied to an administration management device, an administration management method, and a program for an erythropoietin stimulating factor preparation. [Explanation of symbols]

[0104] 1. Administration management device 2. Electronic medical record server 11. Control section 12. Past Dosage Identification Section 13...Hemoglobin value change calculation section 14... Intrinsic rise amount calculation unit (intrinsic amount calculation unit) 15... Intrinsic degradation amount calculation unit (intrinsic amount calculation unit) 16. Required dosage calculation section 17. Administration Schedule Management Section 18...Acquisition part 19. Learning Department 20...Estimation part 21 Alert output section

Claims

1. a hemoglobin level at examination, which indicates the hemoglobin level at the time of examination of the patient; a specific decrease amount per unit time, which indicates the amount by which the hemoglobin level of the patient decreases when an erythropoietin stimulating factor preparation is not administered during the unit time; and A statistic value of the inherent drop amount; an intrinsic increase per unit time, which is an increase in the hemoglobin level of the patient when the erythropoietin stimulating factor preparation is administered in the unit time period at a maximum allowable unit amount per unit time, and which indicates a relative increase in the hemoglobin level based on a case in which the erythropoietin stimulating factor preparation is not administered in the unit time period; A statistic value of the inherent increase amount; an amount of the erythropoietin stimulating factor preparation to be administered to bring the patient's hemoglobin level to a target hemoglobin level after the unit period has elapsed; and The item values ​​of the patient at the time of the medical examination are inhibitory items that inhibit an increase in the hemoglobin value, and include at least one or more of the following item values: iron (FER (ferritin), TSAT (transferrin saturation)), CRP (C-reactive protein), body temperature, BUN (urea nitrogen), nutrition, fatigue, carnitine, hyperthyroidism (osteitis fibrosa), and bleeding diathesis; Input information including at least one of the following: output information indicating a post-administration hemoglobin value indicating a hemoglobin value at a time point when the unit period has elapsed after the administration of the dose of the erythropoietin stimulating factor preparation to the patient; An acquisition unit that acquires a plurality of relationships each indicating a relationship between the a learning unit that generates a hemoglobin level estimation model by machine learning based on the plurality of relationships, the hemoglobin level estimation model estimating the post-administration hemoglobin level, which is the output information, based on the input of the input information; An administration management device comprising:

2. a hemoglobin level at examination, which indicates the hemoglobin level at the time of examination of the patient; a specific decrease amount per unit time, which indicates the amount by which the hemoglobin level of the patient decreases when an erythropoietin stimulating factor preparation is not administered during the unit time; and A statistic value of the inherent drop amount; an intrinsic increase per unit time, which is an increase in the hemoglobin level of the patient when the erythropoietin stimulating factor preparation is administered in the unit time period at a maximum allowable unit amount per unit time, and which indicates a relative increase in the hemoglobin level based on a case in which the erythropoietin stimulating factor preparation is not administered in the unit time period; A statistic value of the inherent increase amount; an amount of the erythropoietin stimulating factor preparation to be administered to bring the patient's hemoglobin level to a target hemoglobin level after the unit period has elapsed; and The item values ​​of the patient at the time of the medical examination are inhibitory items that inhibit an increase in the hemoglobin value, and include at least one or more of the following item values: iron (FER (ferritin), TSAT (transferrin saturation)), CRP (C-reactive protein), body temperature, BUN (urea nitrogen), nutrition, fatigue, carnitine, hyperthyroidism (osteitis fibrosa), and bleeding diathesis; Input information including at least one of the following: output information showing a difference between a post-administration hemoglobin value, which indicates a hemoglobin value at a time point when the unit period has elapsed after administration of the dose of the erythropoietin stimulating factor preparation to the patient, and the target hemoglobin value; An acquisition unit that acquires a plurality of relationships each indicating a relationship between the a learning unit that generates a hemoglobin level estimation model by machine learning based on the plurality of relationships, the hemoglobin level estimation model estimating a difference between the post-administration hemoglobin level and the target hemoglobin level, which is the output information, based on the input information; An administration management device comprising:

3. an estimation unit that estimates the post-administration hemoglobin value, which will be the future output information at the time when the unit period has elapsed, based on the hemoglobin value estimation model and the input information at the time of a new examination; The administration management device of claim 1 or claim 2, further comprising:

4. Based on the hemoglobin value estimation model and the input information at the time of a new examination, The post-administration hemoglobin value and the target hemoglobin value, which will be the future output information at a time point when a certain period of time has passed, are an estimation unit for estimating a difference from a raw value; The administration management device of claim 1 or claim 2, further comprising:

5. an alert output unit that outputs alert information based on a comparison result between the difference between the post-administration hemoglobin level and the target hemoglobin level calculated by the estimation unit and a predetermined reference value; The administration management device of claim 3 or claim 4, further comprising:

6. an intrinsic amount calculation unit that recalculates the intrinsic amount of decrease and the intrinsic amount of increase of the patient when the difference between the post-administration hemoglobin value and the target hemoglobin value calculated by the estimation unit exceeds the reference value; The administration management device of claim 5 further comprising:

7. a dosage calculation unit that calculates an index of a new dosage of the erythropoietin stimulating factor preparation per unit period to be administered to the patient at the new examination based on the target hemoglobin value, the hemoglobin value at the time of examination, the recalculated intrinsic decrease amount, and the recalculated intrinsic increase amount; The administration management device of claim 6 further comprising:

8. a hemoglobin level at examination, which indicates the hemoglobin level at the time of examination of the patient; a specific decrease amount per unit time, which indicates the amount by which the hemoglobin level of the patient decreases when an erythropoietin stimulating factor preparation is not administered during the unit time; and A statistic value of the inherent drop amount; an intrinsic increase per unit time, which is an increase in the hemoglobin level of the patient when the erythropoietin stimulating factor preparation is administered in the unit time period at a maximum allowable unit amount per unit time, and which indicates a relative increase in the hemoglobin level based on a case in which the erythropoietin stimulating factor preparation is not administered in the unit time period; A statistic value of the inherent increase amount; an amount of the erythropoietin stimulating factor preparation to be administered to bring the patient's hemoglobin level to a target hemoglobin level after the unit period has elapsed; and The item values ​​of the patient at the time of the medical examination are inhibitory items that inhibit an increase in the hemoglobin value, and include at least one or more of the following item values: iron (FER (ferritin), TSAT (transferrin saturation)), CRP (C-reactive protein), body temperature, BUN (urea nitrogen), nutrition, fatigue, carnitine, hyperthyroidism (osteitis fibrosa), and bleeding diathesis; Input information including at least one of the following: output information indicating a post-administration hemoglobin value indicating a hemoglobin value at a time point when the unit period has elapsed after the administration of the dose of the erythropoietin stimulating factor preparation to the patient; Obtain multiple relationships, each of which indicates a relationship between A hemoglobin level estimation model for estimating the post-administration hemoglobin level, which is the output information, based on the input of the input information is generated by machine learning based on the multiple relationships. The administration method includes:

9. a hemoglobin level at examination, which indicates the hemoglobin level at the time of examination of the patient; a specific decrease amount per unit time, which indicates the amount by which the hemoglobin level of the patient decreases when an erythropoietin stimulating factor preparation is not administered during the unit time; and A statistic value of the inherent drop amount; an intrinsic increase per unit time, which is an increase in the hemoglobin level of the patient when the erythropoietin stimulating factor preparation is administered in the unit time period at a maximum allowable unit amount per unit time, and which indicates a relative increase in the hemoglobin level based on a case in which the erythropoietin stimulating factor preparation is not administered in the unit time period; A statistic value of the inherent increase amount; an amount of the erythropoietin stimulating factor preparation to be administered to bring the patient's hemoglobin level to a target hemoglobin level after the unit period has elapsed; and The item values ​​of the patient at the time of the medical examination are inhibitory items that inhibit an increase in the hemoglobin value, and include at least one or more of the following item values: iron (FER (ferritin), TSAT (transferrin saturation)), CRP (C-reactive protein), body temperature, BUN (urea nitrogen), nutrition, fatigue, carnitine, hyperthyroidism (osteitis fibrosa), and bleeding diathesis; Input information including at least one of the following: output information showing a difference between a post-administration hemoglobin value, which indicates a hemoglobin value at a time point when the unit period has elapsed after administration of the dose of the erythropoietin stimulating factor preparation to the patient, and the target hemoglobin value; Obtain multiple relationships, each of which indicates a relationship between A hemoglobin level estimation model that estimates a difference between the post-administration hemoglobin level, which is the output information, and the target hemoglobin level based on the input of the input information is generated by machine learning based on the multiple relationships. The administration method includes:

10. The post-administration hemoglobin value, which will be the future output information at the time when the unit period has elapsed, is estimated based on the hemoglobin value estimation model and the input information at the time of a new examination. The administration management method according to claim 8 or claim 9, further comprising:

11. The computer of the administration management device, a hemoglobin level at examination, which indicates the hemoglobin level at the time of examination of the patient; a specific decrease amount per unit time, which indicates the amount by which the hemoglobin level of the patient decreases when an erythropoietin stimulating factor preparation is not administered during the unit time; and A statistic value of the inherent drop amount; an intrinsic increase per unit time, which is an increase in the hemoglobin level of the patient when the erythropoietin stimulating factor preparation is administered in the unit time period at a maximum allowable unit amount per unit time, and which indicates a relative increase in the hemoglobin level based on a case in which the erythropoietin stimulating factor preparation is not administered in the unit time period; A statistic value of the inherent increase amount; an amount of the erythropoietin stimulating factor preparation to be administered to bring the patient's hemoglobin level to a target hemoglobin level after the unit period has elapsed; and The item values ​​of the patient at the time of the medical examination are inhibitory items that inhibit an increase in the hemoglobin value, and include at least one or more of the following item values: iron (FER (ferritin), TSAT (transferrin saturation)), CRP (C-reactive protein), body temperature, BUN (urea nitrogen), nutrition, fatigue, carnitine, hyperthyroidism (osteitis fibrosa), and bleeding diathesis; Input information including at least one of the following: output information indicating a post-administration hemoglobin value indicating a hemoglobin value at a time point when the unit period has elapsed after the administration of the dose of the erythropoietin stimulating factor preparation to the patient; An acquisition means for acquiring a plurality of relationships each indicating the relationship between the a learning means for generating a hemoglobin level estimation model by machine learning based on the plurality of relationships, the model estimating the post-administration hemoglobin level, which is the output information, based on the input of the input information; A program that functions as a

12. The computer of the administration management device, a hemoglobin level at examination, which indicates the hemoglobin level at the time of examination of the patient; a specific decrease amount per unit time, which indicates the amount by which the hemoglobin level of the patient decreases when an erythropoietin stimulating factor preparation is not administered during the unit time; and A statistic value of the inherent drop amount; an intrinsic increase per unit time, which is an increase in the hemoglobin level of the patient when the erythropoietin stimulating factor preparation is administered in the unit time period at a maximum allowable unit amount per unit time, and which indicates a relative increase in the hemoglobin level based on a case in which the erythropoietin stimulating factor preparation is not administered in the unit time period; A statistic value of the inherent increase amount; an amount of the erythropoietin stimulating factor preparation to be administered to bring the patient's hemoglobin level to a target hemoglobin level after the unit period has elapsed; and The item values ​​of the patient at the time of the medical examination are inhibitory items that inhibit an increase in the hemoglobin value, and include at least one or more of the following item values: iron (FER (ferritin), TSAT (transferrin saturation)), CRP (C-reactive protein), body temperature, BUN (urea nitrogen), nutrition, fatigue, carnitine, hyperthyroidism (osteitis fibrosa), and bleeding diathesis; Input information including at least one of the following: output information showing a difference between a post-administration hemoglobin value, which indicates a hemoglobin value at a time point when the unit period has elapsed after administration of the dose of the erythropoietin stimulating factor preparation to the patient, and the target hemoglobin value; An acquisition means for acquiring a plurality of relationships each indicating the relationship between the a learning means for generating a hemoglobin level estimation model by machine learning based on the plurality of relationships, the model estimating a difference between the post-administration hemoglobin level and the target hemoglobin level, which is the output information, based on the input of the input information; A program that functions as a

13. The computer further comprises: an estimation means for estimating the post-administration hemoglobin value, which will be the future output information at the time point when the unit period has elapsed, based on the hemoglobin value estimation model and the input information at the time of a new examination; The program according to claim 11 or 12, which causes the program to function as follows:

Images