WO2015189958A1

WO2015189958A1 - Analysis system and analysis method

Info

Publication number: WO2015189958A1
Application number: PCT/JP2014/065619
Authority: WO
Inventors: 島田　和之; 淳平佐藤
Original assignee: 株式会社日立製作所
Priority date: 2014-06-12
Filing date: 2014-06-12
Publication date: 2015-12-17

Abstract

Provided is an analysis system comprising a processor and a storage device that is connected to the processor. The storage device stores case information which indicates the disease and length of hospital stay of each patient, prescription information which indicates the prescription period of each medicine for each patient, test information which indicates the test results for each patient, and knowledge information which indicates side-effect determination conditions based on the test results. The processor extrapolates, for each patient, the medicine that is the cause of a side effect, and does so on the basis of the relationship between a period in which the test results for each patient satisfy the side-effect determination conditions, and the prescription period of each medicine for each patient, and the processor outputs the results of the extrapolation.

Description

Analysis system and analysis method

The present invention relates to an information system that supports analysis of side effects of prescription drugs in hospitals and the like.

As a background art in this technical field, there is JP-A-2002-342484 (Patent Document 1). Patent Document 1 states that “the present invention includes a system for predicting possible drug side effects by using a search engine that compares integrated data from laboratory and pharmacy information systems. Is compared to a predetermined ADE that defines the normal range for a particular laboratory test, and if an abnormal test value is received and a drug in the patient's drug regimen satisfies a drug included in the ADE rules, To determine if the procedure has been triggered and the patient's examination and dispensing data are taking appropriate corrective action, no corrective action being taken during that period, or improper corrective action being taken Patient test data and dispensing data are monitored and health care providers are alerted of potential ADEs. " About reference).

Patent Document 1: Japanese Patent Application Laid-Open No. 2002-342484

In recent years, the expansion of medical expenses has become a global social problem as the population ages and the life span increases. Against this background, government agencies, insurers, and hospitals are strongly required to optimize medical processes that maintain medical quality while reducing medical costs. With regard to support for analysis of such a medical process, an information system that analyzes effects or side effects of pharmaceuticals using medical data is being studied. Said patent document 1 is the example.

However, with the conventional techniques as described above, it is difficult to comprehensively grasp the medical treatment process from diagnosis of disease name to treatment. In particular, in a wide variety of medications, there is a problem that the therapeutic effect and side effects are comprehensively grasped and it is not known whether an appropriate drug is selected or whether the hospitalization period has increased due to side effects or the like.

An object of the present invention is to provide an information system that visualizes the position (drug position) of effects, side effects, medical expenses, etc. of individual drugs in a wide variety of medications and supports appropriate medications. .

In order to solve the above problems, the present invention is an analysis system comprising a processor and a storage device connected to the processor, wherein the storage device indicates case information indicating the disease and hospitalization period of each patient. And prescription information indicating the prescription period of each drug to each patient, test information indicating the test result of each patient, and knowledge information indicating determination conditions for side effects based on the test results, For each disease, the processor causes a side effect based on the relationship between the time when the test result of each patient satisfies the determination condition for the side effect and the prescription period of each drug for each patient. The medicine is estimated and the result of the estimation is output.

According to one aspect of the present invention, since side effects for each drug are presented, it is possible to easily grasp whether or not an appropriate drug has been selected, and to support optimization of a medical process such as appropriate medication. .

Issues, configurations, and effects other than those described above will be clarified by the following description of the embodiments.

It is a block diagram which shows the structure of the drug positioning visualization system of the Example of this invention. It is explanatory drawing of the example of the case database of the Example of this invention. It is explanatory drawing of the example of the knowledge database of the Example of this invention. It is a flowchart which shows operation | movement of the data server of the Example of this invention. It is a flowchart which shows operation | movement when the side effect determination part of the Example of this invention performs the analysis process which calculates a side effect index. In the Example of this invention, when a user selects the disease made into analysis and visualization, it is explanatory drawing of the example of the disease selection screen which an input / output terminal displays. It is explanatory drawing of the example of a screen which the input / output terminal of the Example of this invention displays the result of an analysis process. It is explanatory drawing of the example of a screen which the input / output terminal of the Example of this invention displays case data.

FIG. 1 is a block diagram showing a configuration of a drug positioning visualization system according to an embodiment of the present invention.

The drug positioning visualization system of the present embodiment is a computer system that analyzes the position (drug position) of the effect, side effect, medical cost, etc. of each drug, visualizes the result, and outputs the result. The network 140 and the input / output terminal 120 used by the user are configured.

In this embodiment, the input / output terminal 120 includes an input unit (not shown) such as a keyboard, mouse, or touch panel, an output unit (not shown) such as a display, and a communication unit (not shown) that communicates with the data server 100 and the like. ) And one or more personal computers. In addition, a portable terminal such as a PDA, PHS, mobile phone, smartphone, or tablet terminal having an input unit such as a button or a touch panel, an output unit such as a display, and a communication unit that communicates with the data server 100 or the like is an input / output terminal. 120 can also be used.

In this system, the input / output terminal 120 is installed and used in an insurer such as a medical institution (health care provider) such as a hospital or clinic, a country, a health insurance association, or a private insurance company. On the other hand, the data server 100 is installed in the data center.

As described above, by installing the data server 100 in the data center, the user's personal information and privacy information such as the data collected from the user can be centrally managed, so that security management such as information leakage prevention can be simplified.

As a user (hereinafter referred to as a user) of the input / output terminal 120, a doctor, manager or manager of a medical institution or insurer is assumed.

The user operates the input / output terminal 120 and uses the information system shown in this embodiment to analyze and visualize the effects, side effects, and medical costs of prescription drugs.

The data server 100 includes a control unit 101, a memory 102, a communication unit 103, a medical data acquisition unit 104, a medical cost data acquisition unit 105, an output unit 107, a disease information selection unit 108, a case data extraction unit 109, The prescription data extraction unit 111, the examination data extraction unit 112, the effect determination unit 113, the side effect determination unit 114, the medical cost calculation unit 115, the screen generation unit 116, the case database 117, and the knowledge database 119 are configured.

The control unit 101 is a processor that executes a program stored in the memory 102, for example, and controls each unit of the data server 100. The memory 102 is a storage device such as a DRAM (Dynamic Random Access Memory), for example, and stores data (for example, a program executed by the control unit 101) referred to by each unit of the data server 100.

The case database 117 and the knowledge database 119 may be stored in the memory 102 or may be stored in another storage device (not shown) in the data server 100 such as an HDD (Hard Disk Drive).

The communication unit 103 is connected to the network 140 and communicates with the input / output terminal 120.

The output unit 107 is a device that outputs a result of processing by the data server 100, and may be a display device, for example.

Medical data acquisition unit 104, medical cost data acquisition unit 105, disease information selection unit 108, case data extraction unit 109, prescription data extraction unit 111, examination data extraction unit 112, effect determination unit 113, side effect determination unit 114, medical cost calculation The unit 115 and the screen generation unit 116 are processing units that execute processing for realizing the functions of the data server 100, and each may be realized by dedicated hardware or may be realized by software. In the latter case, the processing executed by each processing unit in the following description is actually executed by the control unit 101 in accordance with an instruction described in a program stored in the memory 102. Details of the processing executed by each processing unit will be described later.

The data server 100 and the input / output terminal 120 are connected to the network 140. The data server 100 communicates with the input / output terminal 120 via the network 140.

The network 140 uses wired communication using a LAN (Local Area Network) cable or wireless communication using a wireless LAN.

The network 140 can also use other wide area networks such as the Internet, VPN, mobile phone communication network, and PHS communication network.

FIG. 2 is an explanatory diagram of an example 200 of the case database 117 according to the embodiment of this invention.

The case database example 200 includes a case information table 210 for managing basic information for each case, a prescription information table 220 for storing prescription information for each case, a clinical test information table 230 for storing clinical test information for each case, , Image examination information table 240 for storing image examination information for each case, and cost information such as insurance claim costs and costs for each medical practice for medical practice such as prescription, clinical examination, image examination, and surgery. And a cost information table 250 to be stored.

The case information table 210 includes a field 211 for storing a patient ID for identifying a patient, a field 212 for storing a hospitalization date of a medical institution, a field 213 for storing a discharge date, a field 214 for storing a disease name, and discharge. And a field 215 for storing time outcome information. Information on one case is stored in each record (case information record) of the case information table 210. One hospitalization of one patient corresponds to one case. For example, in the case information record example 210A of the case information table 210 shown in FIG. 2, the case identified by the patient ID “Pt-0001” is the hospitalization date “December 12, 2012” and the discharge date “2012”. December 20, 2012, with a disease name of “lung cancer” and an outcome of “cure” (ie, a patient identified with patient ID “Pt-0001” is treated on December 12, 2012 for treatment of lung cancer. To December 20, 2012, and as a result, lung cancer was cured.

The prescription information table 220 includes a field 221 for storing a patient ID, a field 222 for storing information for identifying a prescription drug, a field 223 for storing a prescription date, and a field 224 for storing the length of a prescribed period. And a field 225 for storing a usage / dose. For example, in the example of the prescription information table 220 shown in FIG. 2, in the case of the patient ID “Pt-0001”, the medicine A and the medicine B are each 5 days from December 12, 2012 (that is, from December 12 to 16). This shows that medication C was prescribed for 3 days from December 15, 2012 (ie, during the period from December 15 to 17).

The clinical test information table 230 includes a field 231 for storing a patient ID, a field 232 for storing information for identifying a test item, a field 233 for storing a test date, and a field 234 for storing a test result. The For example, in the example of the clinical test information table 230 shown in FIG. 2, in the case of the patient ID “Pt-0001”, the values of AST (GOT) and ALT (GPT), which are test items indicating the liver function state, On December 12, 2012, they are “28” and “29”, respectively, and on December 15, 2012, they are “49” and “71”, respectively.

The image examination information table 240 includes a field 241 for storing a patient ID, a field 242 for storing information for identifying an examination item, a field 243 for storing an examination date, and a field for storing observation information as a result of image diagnosis. 244. For example, in the example of the image inspection information table 240 shown in FIG. 2, in the case of the patient ID “Pt-0001”, 2012 regarding the examination findings of lung CT (Computed Tomography), which is an examination item for diagnosing the lung state. It is shown that “Tumor A was 10.2 mm” on December 12, and “Tumor A was 9.1 mm” on December 15, 2012.

The above-described clinical examination information table 230 and image examination information table 240 are examples of information indicating the results of examinations actually performed on each patient. In this example, a clinical test information table 230 mainly including blood test results and an image test information table 240 including test results based on images such as CT are independent. May be included. In addition, the case database 117 may include information indicating the results of other types of examinations performed on each patient, and further information related to the patient's subjective symptoms such as “nausea” or “vomiting” May be included.

The cost information table 250 includes a field 251 for storing a medical practice ID for identifying a medical practice, a field 252 for storing the name of the medical practice, and the number of insurance points that can be claimed when the medical practice is performed, that is, cost information that can be claimed. And a field 254 for storing the cost of the medical practice, that is, the cost when the medical practice is performed. Information on one medical practice is stored in each record (medical practice record) of the cost information table 250. For example, in the case of the medical practice record 250A shown in FIG. 2, the medical practice “drug A” identified by the medical practice ID “P0010001” has an insurance score “a” and a cost “x” yen. In Japan, insurance claim costs are converted at 10 yen per insurance point. Therefore, the cost “x” yen is the insurance claim cost when the insurance score is “a”, that is, “a × 10” yen.

The cost assumes the purchase cost of prescription drugs (drugs) necessary for medical practice, reagents used for testing, and medical materials used for surgery or treatment in medical institutions that use this system. It can also include personnel costs of medical staff such as doctors when performing medical practice, and depreciation costs of medical equipment such as diagnostic imaging devices.

FIG. 3 is an explanatory diagram of an example 300 of the knowledge database 119 according to the embodiment of this invention.

The knowledge base example 300 includes a prescription drug knowledge table 310 and a side effect knowledge table 320.

The prescription drug knowledge table 310 specifies a field 311 for storing information for identifying a prescription drug, a field 312 for storing information for identifying a disease to which the prescription drug can be applied, and a test result or finding showing the effect of the prescription drug. A field 313 for storing information to be stored, and a field 314 for storing information for specifying test results or findings of side effects of prescription drugs.

Information about one prescription drug is stored in each record (prescription drug knowledge record) of the prescription drug knowledge table 310. For example, in the case of the prescription drug knowledge record 310A, the prescription drug “drug A” can be applied to the treatment of “lung cancer” (field 312), the effect is the improvement of “tumor size” (field 313), and the side effect is “liver function”. Of these, the inspection item “ALT (GPT)” is deteriorated (field 314).

The side effect knowledge table 320 includes information indicating a side effect determination condition based on a test result or the like. Specifically, the side effect knowledge table 320 stores a field 321 for storing information for identifying an item indicating a side effect such as a test or a subjective symptom, and a side effect. The field 322 stores knowledge for determining an abnormal state for each item, and the field 323 stores knowledge for determining a dangerous state for each item indicating a side effect.

Information related to one examination item is stored in each record (side effect knowledge record) of the side effect knowledge table 320. For example, in the case of the side effect knowledge record 320A, when the inspection item “ALT (GPT)” is 45 or more, it indicates an abnormal value indicating “abnormal condition”, and when it is 100 or more, it indicates a dangerous value indicating “dangerous condition”. Yes.

Also, for example, in the case of the side effect knowledge record 320B, it indicates that the test item “digestive symptom” is “abnormal state” when “nausea”, and “dangerous state” when “vomiting”.

The abnormal state and the dangerous state are both states in which abnormalities in test results suspected to be caused by the side effects of prescription drugs have occurred, but the dangerous state is a state in which an abnormality greater than the abnormal state has occurred. It is. In the above example, the side effect determination condition based on the test result of one test item is stored in one side effect knowledge record. Specifically, a criterion for determining an abnormal state is stored in the field 322, and a criterion for determining a dangerous state is stored in the field 323, respectively.

In this way, for each item of the test result, having two knowledge (standards) for determining the abnormal state and the dangerous state, the prescription of the medicine can be continued as it is, or the test result showing a side effect is displayed. It can assist the judgment of the prescription medical process, such as whether to continue the prescription while monitoring, or to stop the prescription. For example, if a doctor prescribes a drug and the patient does not become abnormal, the prescription is continued as it is, and if the patient becomes abnormal, the prescription is continued while monitoring test results that show side effects. A determination may be made that the prescription is interrupted immediately when the condition is reached.

Next, the operation of this system will be described using a flowchart.

FIG. 4 is a flowchart showing the operation of the data server 100 according to the embodiment of this invention.

First, the control unit 101 executes step 401 in which case data is read from the case database 117 and stored in the memory 102. Here, data related to cases, for example, information for identifying cases, information on prescription drugs corresponding to cases, information on test results corresponding to cases, and the like are collectively referred to as case data. For example, the control unit 101 may read all data included in the case database 117 and store it in the memory 102 in step 401.

Next, the control unit 101 reads out the knowledge data from the knowledge database 119 and executes step 402 for storing it in the memory 102. For example, the control unit 101 may read all data included in the knowledge database 119 and store it in the memory 102 in step 402.

Next, the control unit 101 activates the disease information selection unit 108. The disease information selection unit 108 executes step 403 that allows the user to select a disease to be analyzed and visualized. Hereinafter, the disease selected in step 403 may be described as a target disease. When the disease information selection unit 108 executes step 403, a disease selection screen is displayed on the input / output terminal 120.

FIG. 6 is an explanatory diagram of an example 600 of a disease selection screen displayed by the input / output terminal 120 when the user selects a disease to be analyzed and visualized in the embodiment of the present invention.

The screen example 600 includes a disease selection box 611, an execution button 613, and a prescription drug display area 620.

The user operates a drop-down button 612 attached to the disease selection box 611, and selects a disease to be analyzed and visualized from a plurality of diseases displayed in a drop-down list (not shown). Alternatively, the user can directly input a disease name in the disease selection box 611.

Here, when the user selects the execute button 613, the control unit 101 activates the case data extraction unit 109. The case data extraction unit 109 executes Step 404 for extracting a case corresponding to the disease selected in Step 403 from the case database 117 stored in the memory 102. For example, the case data extraction unit 109 refers to the case information table 210, and the case described in the case information record in which the disease name selected in step 403 is stored in the field 214 (that is, a combination of the patient and the hospitalization period) Are all extracted.

Next, the control unit 101 activates the prescription data extraction unit 111. The prescription data extraction unit 111 executes step 405 for extracting prescription data corresponding to the case extracted in step 404. For example, the prescription data extraction unit 111 identifies the patient ID and hospitalization period of each case from each case information record extracted in step 404, and stores the same patient ID as the identified patient ID in the field 221. The record (prescription information record) of the prescription information table 220 in which the prescription date within the hospitalization period that has been stored in the field 223 may be extracted.

This makes it possible to extract all prescription drugs prescribed for the target disease, and to perform analysis and visualization easily and comprehensively.

In step 405, the prescription data extraction unit 111 may extract prescription data by a method different from the above method. For example, the prescription data extraction unit 111 uses the prescription drug knowledge table 310 of the knowledge database 300 stored in the memory 102 in step 402, and sets the applicable disease field 312 based on the disease (target disease) selected in step 403. One or more prescription drugs stored in the prescription drug field 311 of one or more prescription drug knowledge records having a value in the applicable disease field 312 that matches the target disease can be extracted as prescription data.

This makes it possible to extract only prescription drugs set in advance, and to efficiently perform analysis and visualization.

Next, the control unit 101 determines whether or not the analysis processing from step 407 to step 409, which will be described later, is completed for all prescription drugs for each prescription drug described in the prescription data extracted in step 405, If there are one or more unprocessed prescription drugs, step 406 is performed to select one of them as an analysis target.

When it is determined in step 406 that there is a prescription drug whose analysis processing has not been completed, the control unit 101 determines one of unprocessed prescription drugs (hereinafter also referred to as a target prescription drug). The analysis processing from step 407 to step 409 is executed. In step 407, the control unit 101 activates the inspection data extraction unit 112. The test data extraction unit 112 uses the prescription drug knowledge table 310 to extract test data showing effects or side effects related to the prescription drug based on the cases extracted in step 404.

For example, if “lung cancer” is selected as the disease in step 403 and it is determined in step 406 that the analysis process for the prescription drug “drug A” has not been completed, the test data extraction unit 112 stores the prescription drug knowledge record 310A. By reference, it is specified that the effect of drug A appears in “tumor size” and the side effect appears in “ALT (GPT)”. Then, the examination data extraction unit 112 extracts a record storing the tumor size of the lung CT from the image examination information table 240 and a record storing ALT (GPT) from the clinical examination information table.

Next, in step 408, the control unit 101 activates the effect determination unit 113. The effect determination unit 113 calculates the effect index of the target prescription drug. Here, as an example, the target prescription drug is “medicine A”.

For example, the effect determination unit 113 calculates the effect index DEI using Equation (1).

Here, the effect determination unit 113 determines the number of improved cases based on the difference between the time of hospitalization and the time of discharge, such as a test result that shows the effect. For example, the effect determination unit 113 identifies a case corresponding to each record based on the patient ID and the examination date of each record related to the tumor size extracted in step 407, and the tumor size at the time of hospitalization and discharge at each case. Compare For example, when the “tumor size” of a case is “10.2 mm” at the time of hospitalization and “2.0 mm” at the time of discharge, there is a difference in tumor size before and after the prescription. Since this difference coincides with the decrease in tumor size, which is the effect stored in the field 313 of the prescription drug knowledge record 310A of the drug “A”, the effect determination unit 113 counts the case as an improved case.

In addition, the difference in the test results that show the effect may be calculated between the pre-prescription time and the post-prescription time point, and the above "on hospitalization" and "discharge time" are examples of those time points. is there. For example, in a case where the patient has not been discharged, the latest test result may be used instead of the test result at the time of discharge.

In addition, the number of cases APN prescribed for the target disease is the number of cases prescribed the target prescription drug “drug A” among the cases extracted in step 404.

The effect index DEI calculated by this is the probability that the effect of the prescription drug will appear by prescribing the prescription drug to the patient with the target disease. A large effect index DEI of a prescription drug indicates a high rate of occurrence of cases where it is estimated that the effect of the prescription drug has appeared. As described above, by using the effect index obtained by quantifying the effect using the prescription information and the test information in a plurality of cases, the comparison with other prescription drugs can be analyzed.

In the above example, the effect determination unit 113 calculates the effect index by comparing the difference such as the test result with the effect stored in the prescription drug knowledge table 310, but this method has a high effect of the prescription drug. This is an example of a method for calculating an effect index, which is an index indicating the degree, and the effect determination unit 113 may calculate the effect index by another method. For example, the effect determination unit 113 can also calculate the effect index using the “hospitalization period” from the hospitalization date to the discharge date of each patient for whom the prescription drug is prescribed. Specifically, for example, the effect determination unit 113 may calculate the effect index DEI_A using the mathematical formula (1A).

The smaller the value of the effect index DEI_A calculated for a given disease / prescription drug pair, the smaller the average value of the patient's hospital stay when prescribing the prescription drug for the disease, that is, the greater the effect of the prescription drug. It is estimated that This is based on the assumption that the greater the effect of the prescription drug, the faster the patient's symptoms will improve and the shorter the hospital stay will result. In this way, it is possible to analyze the comparison of prescription drugs with other prescription drugs while considering the case where the hospitalization period changes due to the effect of treatment effects.

Next, in step 409, the control unit 101 activates the medical cost calculation unit 115. The medical cost calculation unit 115 calculates the medical cost of the target prescription drug.

For example, the medical cost calculation unit 115 calculates the medical cost PRC using Equation (2).

The medical cost PRC calculated in this way is an expected value of the total medical cost of the prescription drug prescribed during the hospitalization period of the patient with the target disease. Thus, by calculating the medical cost of a prescription drug using prescription information or the like in a plurality of cases, it is possible to analyze a comparison with other prescription drugs.

The cost described in the cost information table 250 can be used as the unit price PUP of the prescription drug. Thereby, since the medical cost can be calculated using the cost for each medical institution, the comparison with other prescription drugs in the medical institution can be accurately analyzed.

Also, as the unit price PUP of the prescription drug, the insurance score or insurance claim cost described in the cost information table 250 can be used. This makes it easy to analyze the comparison with other prescription drugs even if the cost is unclear. In addition, since insurance scores and insurance claim costs are standardized in each country including Japan, it is possible to analyze comparisons with other medical institutions using these.

For the medical costs of prescription drugs, if the formula (2) is not used, only the unit price of prescription drugs can be used.

If it is determined in step 406 that the analysis process has been completed for all prescription drugs, the control unit 101 then activates the side effect determination unit 114. The side effect determination unit 114 executes step 410 for calculating a side effect index.

FIG. 5 is a flowchart showing an operation when the side effect determination unit 114 according to the embodiment of the present invention executes an analysis process for calculating a side effect index in Step 410.

FIG. 8 is an explanatory diagram of a screen example 800 on which the input / output terminal 120 according to the embodiment of the present invention displays case data.

The screen example 800 is an example of a screen that the input / output terminal 120 displays on an output unit (not shown) such as a display. The screen 810 displays an area 810 that displays clinical indicators, an area 820 that displays prescription information, and medical cost information. Display area 830. Here, the screen generation unit 116 generates data of the screen example 800, the communication unit 103 transmits the data via the network 140, and the input / output terminal 120 receiving the data outputs the screen example 800 to the output unit (illustrated). An example of display in (omitted) will be described. However, this is an example of the output form of the processing result by the data server 100. For example, the output unit 107 may display the screen example 800, or the data server 100 sends the processing result of FIG. The input / output terminal 120 may receive the processing result via the network 140, generate the screen example 800 based on the result, and display the screen example on an output unit (not shown).

In the clinical index display area 810, the horizontal axis represents the number of days elapsed from the date of hospitalization to the date of discharge, and the vertical axis represents the size of the clinical index. Plots of test results that show effects and test results that show side effects are plotted. Will be displayed. Further, on this graph, the abnormal value stored in the abnormal condition knowledge field 322 of the side effect knowledge table 320 and the dangerous value stored in the dangerous condition knowledge field 323 are respectively shown as an abnormal value determination line 811 and a dangerous value determination line 812. Is displayed.

In the prescription information display area 820, the prescription continuation days of prescription drugs prescribed from the hospitalization date to the discharge date of the target case are displayed. In the example of FIG. 8, the prescription period of each prescription drug is displayed on a graph in which the elapsed time from the hospitalization date to the discharge date is assigned to the horizontal axis.

In the medical cost information display area 830, a value obtained by adding the unit prices of prescription drugs for each day for the prescription drugs prescribed from the hospitalization date to the discharge date of the target case is displayed. In the example of FIG. 8, a graph in which the elapsed time from the hospitalization date to the discharge date is assigned to the horizontal axis and the value obtained by adding the unit prices of prescription drugs for each day is assigned to the vertical axis.

As shown in FIG. 8, the above three graphs are displayed side by side on the same screen, and by making the scales of the horizontal axes the same, the prescription period of each prescription drug, the time of occurrence of an effect or a side effect, and , It becomes possible to easily grasp the relationship with the cost.

First, the side effect determination unit 114 determines whether or not the side effect analysis processing from step 502 to step 508, which will be described later, has been completed for all cases extracted in step 404, and there are one or more unprocessed cases In step S501, one of unprocessed cases is selected as a target for the side effect analysis process.

If it is determined in step 501 that the side effect analysis processing is to be executed for any case, the side effect determination unit 114 executes step 502 for extracting a side effect occurrence period. For example, as shown in the screen example 800, the side effect determination unit 114 indicates that the test data corresponding to the side effect has changed from less than the abnormal value to more than the abnormal value as the “side effect occurrence start date”, and the abnormal value after the side effect occurrence start date. From the above, when it returns to less than the abnormal value, it is determined as the “side effect occurrence end date”, and from the side effect occurrence start date to the side effect occurrence end date is determined as the “side effect occurrence period”.

In the following explanation, the period from when the inspection data changes from less than the abnormal value to more than the abnormal value until when it returns from the abnormal value to less than the abnormal value is described as the “abnormal value occurrence period” In some cases, the period from when the inspection data changes from less than the dangerous value to more than the dangerous value until when the inspection data returns from more than the dangerous value to less than the dangerous value is referred to as the “dangerous value occurrence period”.

Next, the side effect determination unit 114 executes step 503 for extracting a prescription drug prescribed during the side effect occurrence period as a side effect causative drug candidate. Specifically, the side effect determination unit 114 extracts prescription drugs whose prescription period overlaps with the side effect occurrence period among prescription drugs prescribed during the hospitalization period of one case as a side effect cause drug candidate. In the case data of screen example 800, “drug A”, “drug B”, and “drug D” are extracted from the prescription drugs as drug candidates that cause side effects, excluding “drug C” that is not prescribed during the side effect occurrence period. The

Next, the side effect determination unit 114 analyzes the variation of the test data and executes Step 504 for determining the date of occurrence of the dangerous value. In the case data of the screen example 800, the side effect determination unit 114 determines the risk value occurrence date 813 when the test data corresponding to the side effect changes from less than the risk value to more than the risk value.

Next, the side effect determination unit 114 determines whether or not the prescription information has changed before and after the risk value occurrence date, and prescription information that has not changed, that is, prescription drugs that are continuously prescribed before and after the risk value occurrence date. Step 505 of excluding from the side effect cause drug candidate is executed. This is because, in general, when a risk value is generated by prescribing a certain medicine, it is considered that the doctor stops prescribing the medicine at that time. In other words, a drug that has been prescribed before the date of occurrence of the risk value and has been prescribed continuously after the date of occurrence of the risk value is not the cause of the occurrence of the risk value by the doctor who prescribes the drug. It is estimated that it was determined. For this reason, for example, the side effect determination unit 114 may exclude a drug prescribed for a predetermined period after the risk value occurrence date from the side effect causative drug candidate.

In the case data of the screen example 800, “drug A” prescribed continuously before and after the risk value occurrence date 813 is excluded from the side effect cause drug candidates.

Next, the side effect determination unit 114 determines whether or not the prescription information has changed before and after the risk value occurrence date, there is a change in the prescription information, there is no prescription information before the risk value occurrence date, and prescription information after the risk value occurrence date. In other words, step 506 is executed to exclude prescription drugs prescribed after the risk value occurrence date from the side effect cause drug candidates. This is because prescription medicines that are not prescribed before the risk value occurrence date and that are prescribed after the risk value occurrence date cannot cause the risk value. In the case data of the screen example 800, “drug D” that is not prescribed before the risk value occurrence date 813 and is prescribed after the risk value occurrence date 813 is excluded from the side effect cause drug candidates. .

Next, the side effect determination unit 114 executes step 507 of extracting prescription drugs that are not excluded in either step 505 or step 506 among the side effect cause drug candidates extracted in step 503 as side effect cause drugs. In the case data of the screen example 800, “drug B” is extracted as a side effect-causing drug.

Eventually, the side effect determination unit 114 stores test values in the side effect knowledge table 320 for each case corresponding to the disease selected in step 403 (ie, each patient hospitalized for the disease selected in step 403). Based on the relationship between the time when the determined determination condition is satisfied and the prescription period of the drug to the patient, the drug estimated to be the cause of the side effect is extracted. Specifically, the side effect determination unit 114 is prescribed during the side effect occurrence period (step 503) and is not prescribed for a predetermined period after the test value changes from a value lower than the dangerous value to a higher value (step 505). And the medicine prescribed before the test value changes from a lower value to a higher value than the dangerous value (step 506) is extracted as a side effect-causing drug (step 507).

In the above example, each determination condition stored in each record of the side effect knowledge table 320 is used to determine a first criterion for determining an abnormal value (that is, a criterion stored in the field 322) and a dangerous value. The second criterion (that is, the criterion stored in the field 323) is included, but each determination condition may include only one criterion. For example, each determination condition may include only the second reference for determining the dangerous value. In any case, the side effect occurrence period is a predetermined period including a period in which the determination condition is satisfied. Specifically, in the above example in which each determination condition includes the first criterion and the second criterion, the side effect occurrence period is the same period as the abnormal value occurrence period (that is, including the abnormal value occurrence period), It is also a period including the dangerous value occurrence period. On the other hand, for example, when each determination condition includes only the second criterion, the side effect occurrence period may be a predetermined period longer than the danger value occurrence period, including the danger value occurrence period.

In this way, when determining a side effect-causing drug, the change in test data corresponding to the side effect is determined, and by using the continuity of prescription before and after the change (for example, the occurrence of a dangerous value), the cause of the side effect is determined. Can be narrowed down efficiently.

Next, the side effect determination unit 114 executes Step 508 for setting the side effect flag “1” to the prescription drug extracted as the side effect cause drug.

In step 507, when a plurality of prescription drugs are extracted as side effect causative agents based on the side effect determination condition related to one inspection item (that is, stored in one side effect knowledge record), the side effect determination unit 114 Using equation (3), side effect flags are allocated and set according to the number of extracted prescription drugs.

This makes it possible to set a side effect flag that evenly distributes the possibility of a side effect to a plurality of causative agents even when a plurality of side effect causative agents are extracted.

When the side effect determination unit 114 sets a side effect flag for a plurality of side effect causative agents in step 508, the side effect determination unit 114 may perform slope distribution according to a predetermined criterion instead of using the above formula (3). For example, the side effect determination unit 114 is based on the estimation that the prescription drug having a longer number of days from the prescription end date to the subsequent risk value occurrence date is more likely to be unrelated to the occurrence of the risk value. In order to increase the value of the side effect flag allocated to the prescription drug with a shorter number of days from the risk value occurrence date to the prescription end date, the side effect flag is inclined and distributed for each prescription drug using Equation (4). You can also set it.

As a result, the side effect flag can be weighted so that the closer the two are, the closer the two are closer to the risk value occurrence date. It is possible to set a side effect flag according to the condition.

In this way, the prescription continuation state before and after the time when the inspection data shows an abnormal state or a dangerous state, that is, the prescription process (prescription continuity) and the variability of the inspection data cause side effects. Drugs can be narrowed down more efficiently. This makes it possible to conduct a more accurate side effect analysis for individual drugs in a wide variety of medications.

If it is determined in step 501 that the side effect analysis processing has been completed for all cases, the side effect determination unit 114 counts the side effect flags set in step 508 for all cases for each prescription drug, thereby determining the side effect index. Step 509 is calculated.

For example, the side effect determination unit 114 calculates the side effect index DAI of the prescription drug m using Equation (5).

The side effect index DAI calculated by this is the probability that a side effect will appear when a prescription drug m is prescribed to a patient with the target disease. A large side effect index DAI of the prescription drug m indicates a high rate of occurrence of cases suspected of causing side effects due to the prescription drug m. In this way, it is possible to analyze the comparison with other prescription drugs by using the side effect index that quantifies the degree of influence of side effects using hospitalization periods and examination information on multiple cases, etc. it can.

Next, the control unit 101 activates the screen generation unit 116. The screen generator 116 outputs the results of all the analysis processes based on the effect index generated in step 408, the medical expenses generated in step 409, and the side effect index generated in step 410. Step 411 is generated.

Next, the control unit 101 executes Step 412 for outputting the screen generated in Step 411 to the input / output terminal 120.

FIG. 7 is an explanatory diagram of an example screen 700 on which the input / output terminal 120 according to the embodiment of the present invention displays the result of the analysis process after step 412 of FIG. 4 is executed.

A screen example 700 has a prescription drug display area 620 similar to that shown in FIG. 6, an effect index on the vertical axis, a side effect index on the horizontal axis, and the size of a circle (bubble) as medical expenses, and drugs A and B Drug C is displayed in a bubble chart on the same axis.

From the example of FIG. 7, it can be seen that drug A has a high therapeutic effect and few side effects, but has a high medical cost. It can also be seen that medicine B has moderate side effects, but has many side effects. It can also be seen that medicine C has a low medical cost, but has a low effect.

In this way, the therapeutic effects, side effects, and medical cost indicators of multiple prescription drugs prescribed for one disease are presented on the same axis of the same graph. It is possible to grasp the therapeutic effect, side effects, and the position of medical expenses (drag position) from a bird's-eye view. Thereby, since the user can easily grasp whether or not an appropriate drug is selected, it is possible to support optimization of a medical treatment process such as an appropriate medication treatment.

In the screen example 700, the horizontal axis indicates side effects (for example, 3%) similar to those in the attached document, and the average value of all side effect indexes calculated from data collected from other medical institutions (for example, the national average of 5%). The average value of all effect indexes calculated from data collected from other medical institutions (for example, the national average of 80%) is displayed on the vertical axis. In addition, the average value of all the side effect indexes and the average value of all the effect indexes calculated from the collected data in the medical institution are displayed by dotted lines as in-

hospital averages

710 and 720, respectively.

As a result, in medical institutions that use this system, the positioning of therapeutic effects and side effects (drug positions) can be objectively grasped compared to other medical institutions, so that medication treatment is supported while supporting more appropriate drug selection. Can help optimize the medical process.

In addition, when doctors use this system at the time of medical care, for example, when explaining the treatment policy to patients, that is, by using it for informed consent, while comparing the treatment effect, side effect, and medical cost for each prescription drug Since it can explain, it becomes possible to effectively explain the treatment policy to the patient and obtain consent.

In the above example, the screen generation unit 116 generates data for displaying the screen example 700, the data server 100 transmits the data to the input / output terminal 120 via the network 140, and the input / output terminal 120 converts the data into the data. Although an example in which the screen example 700 is displayed on an output device (not shown) such as a display device based on the above is shown, the output unit 107 may display the screen example 700.

FIG. 7 shows an example in which the effect index, the side effect index, and the medical cost of a plurality of prescription drugs are displayed in association with the vertical axis, the horizontal axis, and the size of the displayed circle, respectively. However, such a display method is an example, and these indicators may be displayed by other methods. For example, any one of the effect index, the side effect index, and the medical cost, or any combination of the two may be displayed, or a number indicating the value may be displayed.

As described above, the drug positioning visualization system according to the present invention provides an information system for visualizing the effects of individual drugs, side effects, and positioning of medical expenses (drug positions) in various types of medications and supporting appropriate medications. It becomes possible to do.

In addition, this invention is not limited to the Example mentioned above, Various modifications are included. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described. Further, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment.

The above-described configurations, functions, processing units, processing means, etc. may be realized in hardware by designing a part or all of them, for example, with an integrated circuit. Further, each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor. Information such as programs, tables, and files that realize each function is a memory, hard disk drive, storage device such as SSD (Solid State Drive), or computer-readable non-transitory data such as an IC card, SD card, or DVD. It can be stored in a storage medium.

Further, the drawings show control lines and information lines that are considered necessary for explaining the embodiments, and not necessarily all control lines and information lines included in an actual product to which the present invention is applied. Not necessarily. Actually, it may be considered that almost all the components are connected to each other.

Claims

An analysis system comprising a processor and a storage device connected to the processor,
The storage device
Case information showing the disease and hospitalization period of each patient,
Prescription information indicating the prescription period of each drug for each patient;
Test information indicating the test results of each patient,
Holding knowledge information indicating conditions for determining side effects based on the test results,
The processor is
For each disease, the drug that caused the side effect is determined based on the relationship between the time when the test result of each patient satisfies the determination condition for the side effect and the prescription period of each drug for each patient. Estimate
An analysis system that outputs the result of the estimation.
The analysis system according to claim 1,
The processor is prescribed for each patient within a predetermined period including a period in which the test result satisfies the determination condition for the side effect, and the test result is changed from a state in which the test result does not satisfy the determination condition for the side effect to a state in which the test result is satisfied. Presumed as a cause of the side effect is a prescription that is not prescribed continuously for a predetermined period of time after the test, and the test result is changed from a state that does not satisfy the determination condition of the side effect to a state that satisfies the condition. An analysis system characterized by that.
The analysis system according to claim 2,
The processor is
For each patient, a predetermined flag value is given to the presumed cause of the side effect,
For each drug, by calculating the flag values given for a plurality of the patients, calculating an index of side effects of each drug,
An analysis system, wherein an index of a side effect of each drug is output as the estimation result.
The analysis system according to claim 3,
The processor, for each patient, allocates the predetermined flag value to the plurality of medicines when a plurality of medicines are estimated to be the cause of the side effects based on the determination condition of the side effects concerning one examination item. The analysis system characterized by giving.
The analysis system according to claim 4,
The processor, for each of the patients, when the plurality of drugs are estimated to be the cause of the side effect based on the determination condition of the side effect for one test item, the prescription of each of the plurality of drugs is completed. The predetermined flag value is given to each of the plurality of medicines in an inclined manner based on the length of the period until the test result changes from a state not satisfying the determination condition for the side effect to a state satisfying An analysis system characterized by that.
The analysis system according to claim 2,
The determination condition of the side effect includes a first criterion for determining an abnormality of a test result and a second criterion for determining an abnormality that is greater than that determined by the first criterion,
The processor is prescribed for each patient within a period in which the test result satisfies the first criterion and the test result changes from a state not satisfying the second criterion to a state satisfying the second criterion. An analysis system characterized in that a prescription that is not continuously prescribed and that is prescribed before the test result changes from a state that does not satisfy the second standard to a state that satisfies the second criterion is estimated as the cause of the side effect.
The analysis system according to claim 3,
The storage device further holds cost information of each medicine,
The processor is
Based on at least the case information and the prescription information, calculate an index of the effect of each drug for each disease,
Based on the cost information, the case information and the prescription information, calculate an index of the cost of each drug for each disease,
An analysis system that outputs an index of an effect of each drug for each disease, an index of the cost, and an index of the side effect.
The analysis system according to claim 7,
The knowledge information further includes information associating each medicine, a disease to which each medicine can be applied, and an effect obtained by prescribing each medicine,
The processor compares, for each disease, a change in the test result before and after the prescription of each drug to each patient and an effect associated with each drug included in the knowledge information. An analysis system characterized in that an index of the effect of each drug is calculated.
The analysis system according to claim 7,
The analysis system characterized in that the processor calculates an index of the effect of each medicine based on the length of hospital stay of each patient prescribed for each medicine for each disease.
The analysis system according to claim 7,
A display device;
The analysis system, wherein the processor causes the display device to display a screen in which an index of effect of the plurality of drugs, an index of cost, and an index of side effects are plotted on the same graph for each disease.
An analysis method executed by a computer system comprising a processor and a storage device connected to the processor,
The storage device
Case information showing the disease and hospitalization period of each patient,
Prescription information indicating the prescription period of each drug for each patient;
Test information indicating the test results of each patient,
Holding knowledge information indicating conditions for determining side effects based on the test results,
The analysis method is:
For each disease, the drug that caused the side effect is determined based on the relationship between the time when the test result of each patient satisfies the determination condition for the side effect and the prescription period of each drug for each patient. A first step to estimate;
And a second procedure for outputting the result of the estimation.
The analysis method according to claim 11, comprising:
In the first procedure, for each patient, the test result is prescribed within a predetermined period including a period in which the side effect determination condition is satisfied, and the test result is satisfied from a state in which the side effect determination condition is not satisfied. A drug prescribed before the change from a state where the test result does not satisfy the determination condition for the side effect to a state where the test result does not satisfy the condition for determining the side effect An analysis method comprising an estimation procedure.
The analysis method according to claim 12, comprising:
The first procedure includes:
A procedure for giving a predetermined flag value to the medicine presumed to be the cause of the side effect for each patient,
Calculating a side effect index for each drug by counting the flag values given for a plurality of patients for each drug, and
The analysis method characterized in that the second procedure includes a procedure for outputting an index of a side effect of each drug as a result of the estimation.
The analysis method according to claim 13, comprising:
The storage device further holds cost information of each medicine,
The analysis method further includes:
A third procedure for calculating an index of the effect of each drug for each disease based on at least the case information and the prescription information;
A fourth step of calculating an index of the cost of each drug for each disease based on the cost information, the case information, and the prescription information,
The analysis method characterized in that the second procedure includes a procedure for outputting an index of the effect of each drug for each disease, an index of the cost, and an index of the side effect.
The analysis method according to claim 14, comprising:
The computer system further includes a display device,
The second procedure includes a procedure for causing the display device to display a screen in which the effect index, the cost index, and the side effect index of the plurality of drugs for each disease are plotted in the same graph. Method.