JP2011039653A - Medical information generation device, medical information generation method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide technology for properly grasping relationship between injury/disease and a medical expense when a patient is injured and diseased. <P>SOLUTION: The medical information generation device 1 includes: a part 100 for calculating an average medical expense for each injury/disease group based on a plurality of receipt information storing at least one piece of injury/disease information and the medical expenses; a part 110 for calculating an average difference medical expense that is a difference between the average medical expense of one injury/disease group and the average medical expense in another injury/disease medical group, based on the average medical expense of one injury/disease group and the average medical expense of the other injury/disease group comprising all injuries/diseases constituting the one injury/disease group and the other injury/disease, and extracting combinations between the injury/disease groups whose average difference medical expense is ranked high; and a direction target information generation part 120 generating direction target information including information for identifying an insured person who needs a direction for an increase of the medical expense, the current injury/disease of the insured person, and the injury/disease to be prevented of the insured person, based on the receipt information and the extracted combinations between the injury/disease groups. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a medical information generation device, a medical information generation method, and a program.

As measures to control the increase in medical expenses, early detection of injury and illness and prevention of serious illness through health guidance are considered. In insurance guidance by an insurer, targeting by narrowing down the target to a disease with higher preventive effect by grasping the relationship between the disease and medical expenses is effective.
Currently, there is a technique for grasping the relationship between a medical condition and medical expenses necessary for realizing targeting (see, for example, Patent Document 1). The technology described in Patent Document 1 prescribes a typical prescription for each injury and illness, and converts the medical treatment described in the medical fee remuneration statement (hereinafter referred to as “recept”) to a typical injury and illness. Dividing into prescriptions and other prescriptions, typical prescriptions are allocated to the corresponding injuries and diseases, and other prescriptions are allocated to each injuries and diseases by applying statistical analysis methods, and the medical costs for each injuries are estimated. Is.
In addition, information on patients and medical practices, such as names of wounds and medical treatments, names of drugs used, medical expenses for each medical practice, medical expenses for each chemical used, etc. are written on the receipt. Expenses can be calculated. The medical expenses are expressed in a score format, and one point of the medical fee is equivalent to 10 yen.

Japanese Patent Laid-Open No. 11-357437

However, the technique described in Patent Document 1 has a problem in that it is impossible to accurately grasp the relationship between the disease and the medical expenses when the disease and the disease occur together. Therefore, the technique described in Patent Document 1 has a problem in that effective health guidance cannot be performed because it is impossible to accurately grasp an increase in medical expenses in scenes where illnesses and illnesses occur simultaneously.
In other words, for example, patients with both diabetes and hypertension need treatment that takes into account the interaction between diabetes treatment and antihypertensive drugs. There may be cases where a new medical practice or a more advanced medical practice is required due to the concurrent occurrence. Since the combination of injury and illness imposes restrictions on the medical treatment as described above, it is generally said that the medical expenses at the time of the concurrent occurrence of injury and illness are larger than the sum of the individual medical expenses for each medical treatment. . However, the technique described in Patent Document 1 is described in Patent Document 1 because medical expenses are allocated to each wound based on the idea that wounds and illnesses occur independently and treatment is performed for each wound and illness. This technology has limited applicability at the time of complications.

  The present invention has been made in view of such a situation, and an object of the present invention is to provide a technique capable of accurately grasping the relationship between the wound and the medical expenses when the wound and the disease coexist.

  In order to solve the above-described problem, a medical information generation apparatus according to one aspect of the present invention is based on a plurality of pieces of receipt information describing at least one or more wound information and medical expenses. A medical cost calculation unit for each wound and disease group that calculates an average medical cost for each group, an average medical cost for one wound and disease group calculated by the medical cost calculation unit for each wound and disease group, and all the wounds and diseases that constitute the one wound and disease group And an average medical cost that is a difference between an average medical cost in the one wound group and the other wound group based on an average medical cost in the other wound group composed of the other wound and disease, By a medical expense rapid increase disease group combination extraction unit that extracts a combination between the injury group and the disease group whose average differential medical cost is higher from among the combinations between injury groups, a receipt information, and the medical cost rapid increase disease group combination extraction unit Extracted wound group Information identifying an insured person who needs guidance for an increase in medical expenses, information identifying the current injuries of the insured person, and information identifying injuries to be prevented by the insured person And a guidance object information generation unit for generating guidance object information.

  In the medical information generation apparatus, the medical cost calculation unit for each sick and sick group uses a similarity between the sick and sick combinations constituting each sick and sick group as a vector between each vector having the presence or absence of the sick and sick elements constituting the combination. It is also possible to calculate the average medical cost for each wound and disease group after aggregation, based on the similarity.

  In the medical information generating apparatus, the medical cost calculation unit for each wound and disease group has a relationship between nodes having the similarity of a wound disease and disease group as a node and a similarity equal to or greater than a predetermined threshold for the similarity of all combinations of the calculated wound and disease group. Adjacency matrix data as network data is created as an edge, and from the adjacency matrix data, one injury and illness group that constitutes a maximum clique having the maximum possible size among cliques having edges for all nodes You may make it collect as a group and may calculate the average medical expenses according to the disease group after aggregation.

  The medical information generation device is associated with information for identifying each wound and disease constituting one wound and disease group, and information for identifying each wound and disease constituting the one wound and disease group, A preventive content storage unit for storing preventive content information indicating the preventive content for the onset of the other one of the wounds during the onset of each injured disease and disease group, and the guidance generated by the guidance target information generating unit Based on the target information and the preventive content information stored in the preventive content storage unit, information for identifying the insured who needs the guidance, the current injury / illness name of the insured, the prevention of the insured You may make it further provide the guidance information output part which outputs the guidance information including the said injury content information which should be prevented, and the said prevention content information which concerns on the said injury and sickness which should be prevented.

  In order to solve the above-described problem, a medical information generation method according to another aspect of the present invention includes at least one injury and illness based on at least one injury and illness information and a plurality of receipt information describing medical expenses. Medical cost calculation means by wound group for calculating the average medical cost by wound group, average medical cost of one wound group calculated by the medical cost calculation unit by wound group, and all the constituents of the one wound group Based on the average medical cost of the other wound group composed of the wound and the other one, the average differential medical cost that is the difference between the average medical cost in the one wound group and the other wound group is calculated. A medical expense rapid increase disease group combination extraction means for extracting a combination between the injury group and sickness group having the highest average differential medical expenses from among the combinations between the injury and disease groups, receipt information, and the medical expense rapid increase disease group combination extraction means Extracted by Based on the combination of the injured disease group, information for identifying an insured person who needs guidance for an increase in medical expenses, information for identifying the injured person's current injured condition, and identifying injured / injured condition for the insured person Guidance target information generating means for generating guidance target information including information to be provided.

  In order to solve the above problem, a program according to another aspect of the present invention is based on a plurality of pieces of receipt information describing at least one or more injury information and medical expenses in a computer of a medical information generation apparatus that generates medical information. The medical cost calculation step by wound and disease group for calculating the average medical cost by wound and disease group composed of one or more wounds and sickness, and the average medical cost of one wound and disease group calculated by the medical cost calculation step by wound and disease group and , Based on the average medical cost of the other wound group consisting of all the wounds and other diseases that constitute the one wound group, the average medical cost of the one wound group and the other wound group An average difference medical cost that is a difference is calculated, and a medical cost rapid injury group combination extraction step for extracting a combination between wound groups having the highest average differential medical cost among combinations between wound groups, receipt information, and And information for identifying an insured person who needs guidance for an increase in medical expenses based on the combination between the injured disease groups extracted by the medical expense suddenly injured disease group combination extracting step, A guidance target information generating step for generating guidance target information including information for identifying and information for identifying the injury or illness to be prevented of the insured person is executed.

  According to the present invention, medical costs are calculated for each wound group composed of one or more wounds, so that it is possible to accurately grasp the relationship between the wounds and medical expenses when wounds occur simultaneously. Become. In addition, according to the present invention, since the difference between the medical expenses of a certain wound group and the medical expenses of the wound group when one wound is added to the wound group, N pieces (N is an integer) ), It is possible to accurately grasp the increase in medical expenses when one additional disease is found. Specifically, it is possible to appropriately grasp an increase in medical expenses even in a scene where a sickness and disease are added for the first time and a situation where a sickness and disease are already occurring and one more is added. Therefore, effective health guidance can be performed.

1 is a configuration diagram of a medical information generation device 1. FIG. It is an example of the information memorize | stored in a memory | storage means. It is an example of the information memorize | stored in a memory | storage means. It is an example of the information memorize | stored in a memory | storage means. It is an example of the information memorize | stored in a memory | storage means. It is a flowchart which shows an example of operation | movement of the medical expenses calculation part 100 by medical treatment group, the medical expenses rapid increase sickness group combination extraction part 110, and the guidance object information generation part 120. It is DFD (Data Flow Diagram) which shows the flow of various information in the medical expenses calculation part 100 according to a wound disease group, the medical expenses rapid injury group combination extraction part 110, and the guidance object information generation part 120. FIG. 5 is a flowchart illustrating an example of an operation of a medical cost totaling unit 102. 5 is a flowchart illustrating an example of an operation of a similarity calculation unit 104. 6 is an explanatory diagram for explaining an operation of a similarity calculation unit 104. FIG. 5 is a flowchart showing an example of the operation of a similar wound group extraction unit 106. 6 is an explanatory diagram for explaining the operation of a similar wound group extraction unit 106. FIG. 6 is an explanatory diagram for explaining the operation of a similar wound group extraction unit 106. FIG. 5 is a flowchart illustrating an example of an operation of a medical cost totaling unit 102. 5 is a flowchart illustrating an example of an operation of a differential medical cost calculation unit 112. It is a flowchart which shows an example of operation | movement of the guidance object information generation part 120. 10 is a flowchart showing an example of an operation of a guidance information output unit 130. It is a conceptual diagram of generation and application of a rapid increase in medical expenses. It is explanatory drawing explaining the concept of aggregation of a wound group.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. A medical information generation apparatus 1 according to an embodiment of the present invention is, for example, a server computer or a personal computer, and includes an input / output unit, a calculation unit, and a storage unit as shown in FIG. The medical information generation apparatus 1 includes, as storage means, a wound name master storage unit 10, a prevention content master storage unit 20, a receipt database 30, a receipt database 40, a wound disease group-specific record storage unit 50, and a wound disease group differential medical cost record storage unit. 60, a higher-order differential medical cost record storage unit 70 and a guidance target record storage unit 80 are provided. The medical information generation device 1 includes a medical cost calculation unit 100 for each wound and disease group, a medical expense rapid increase disease group combination extraction unit 110, a guidance target information generation unit 120, a guidance information output unit 130, and a memory 140 as calculation means. The medical cost calculation unit 100 for each wound group includes a medical cost totaling unit 102, a similarity calculation unit 104, and a similar wound group extraction unit 106. The medical expenses suddenly injured disease group combination extraction unit 110 includes a differential medical cost calculation unit 112 and an upper differential medical cost extraction unit 114.

  The wound and disease name master information storage unit 10 is an area for storing wound and disease name master information. Each record of wound name master information includes items of wound code and wound name, as shown in FIG. The wound code is identification information for identifying the wound. Note that the record A in FIG. 2A stores the wound name “A” in association with the wound code “S01”.

  The prevention content master information storage unit 20 is an area for storing prevention content master information. Each record of prevention content master information has each item of the 1st information group, the 2nd information group, and prevention content information. The item “first information group” stores each injury / illness code that identifies each injury / illness that constitutes an injury / illness group including one or more injury / illnesses. The item “second information group” includes the first Each wound code that identifies each wound and disease identified by the information group and another wound and disease code that identifies one other wound and disease are stored, and each wound and disease identified by the first information group is stored in the item “prevention content information”. Information indicating the content of prevention against the onset of the other one of the injuries when the disease is occurring is stored. For example, in the item “prevention content information”, text data such as “instruct a low-salt diet to avoid high blood pressure” or “instruct regular exercise (walking) to suppress weight gain”. Is memorized. In addition, the prevention content is guidance content to an insured person by an insurer or a medical institution.

  That is, each record of the prevention content master information includes, as shown in FIG. 2B, a current injury code list, information corresponding to the first information group, which is information for identifying each disease that is currently active, Corresponding to the information group of, the future wound code list that is information for identifying each wound that is presently affected and one other disease that is likely to develop in the future, each wound that is identified by the current wound code list In this case, each item of the prevention content information indicating the prevention content for the occurrence of the injury or illness identified by the injury or illness code that is present in the future injury or illness code list but is not present in the injury or illness code list is included. In the case where the record B in FIG. 2B is associated with the current wound code list “S01” and the future wound code list “S01, S02”, the wound is identified by the wound code “S01”. , The prevention content information indicating the prevention content for the onset of the injury and illness identified by the injury and illness code “S02” is stored.

  The receipt database 30 is an area for storing receipt information that is information based on receipts for a certain period in the past. The receipt information is information in which at least one or more injury information and medical expenses are described. As shown in FIG. 2C, each record of the receipt information based on the receipt for a certain period in the past has items of a receipt ID, an insured person name, a wound code list, and a medical fee score. The receipt ID is information for identifying individual receipt information. The record C in FIG. 2C stores the insured person name “Insurance Taro”, the wound code list “S02”, and the medical fee “500” in association with the receipt ID “R080001”.

  The receipt database 40 is an area for storing receipt information that is information based on the receipts for the current fiscal year. As shown in FIG. 2 (d), each record of the receipt information based on the receipt for the current year has items of a receipt ID, an insured person name, a wound code list, and a medical fee score. Note that the record D in FIG. 2D stores the insured person name “Insurance Taro”, the wound code list “S01”, and the medical fee “210” in association with the receipt ID “R090001”.

  The record storage unit 50 for each wound group, the differential medical cost record storage unit 60 for the wound group, the upper differential medical cost record storage unit 70, and the instruction target record storage unit 80 will be described later.

  The medical expenses calculation unit 100 for each wound and disease group calculates an average medical cost for each wound group composed of one or more wounds and diseases based on a plurality of pieces of receipt information. Specifically, the medical expenses calculation unit 100 for wounds and diseases group aggregates similar wounds and groups into one wound group, and calculates an average medical cost for each wound group after the consolidation. For example, the medical expenses calculation unit 100 for each wound group calculates the similarity between the combinations of wounds and diseases constituting each wound group as an angle between the vectors with the presence or absence of the wound elements constituting the combination as a vector, Based on the similarity, it is aggregated into one wound group, and an average medical cost for each wound group is calculated. More specifically, the medical cost calculation unit 100 for each wound and disease group uses the wound and disease group as a node and the relationship between nodes having a similarity equal to or higher than a predetermined threshold as an edge for the calculated similarity of all combinations of the wound and disease group. Create adjacency matrix data as data, and from the adjacency matrix data, out of the cliques that have edges for all nodes, aggregate the wound groups that constitute the maximum clique with the maximum possible size as one wound group. , Calculate the average medical expenses by injury group after aggregation.

  Hereinafter, the medical cost totaling unit 102, the similarity calculation unit 104, and the similar wound / disease group extraction unit 106 provided in the medical cost calculation unit 100 for each wound and disease will be described. First, the medical expense totaling unit 102 included in the medical expenses calculating unit 100 for wounds and diseases refers to the receipt database 30 and the record storage unit 50 for individual groups of wounds and totals the average medical expenses for each diseased group. Specifically, the medical expenses totaling unit 102 generates a record for each wound and disease group including at least the wound and disease code list and the average medical expenses for each wound and disease group, and stores the record in the record storage unit 50 for each wound and disease group. FIG. 3A is an example of a record for each wound group that is generated and stored by the medical cost totaling unit 102. Note that the wound and disease group-specific record shown in FIG. 3A includes items of a list ID for identifying each wound and disease code list, a wound and disease code list, the number of cases, a total medical cost, and an average medical cost. The record A in FIG. 3A stores the wound code list “S01”, the number of cases “800”, the total medical cost “160000”, and the average medical cost “200” in association with the list ID “L001”. ing.

  The similarity calculation unit 104 included in the medical expense calculation unit 100 for each wound and disease group refers to the record storage unit 50 for each wound and disease group, and calculates the similarity between each wound and disease combination constituting each wound and disease group. For example, the similarity calculation unit 104 calculates the similarity of the combination of wounds and diseases constituting each wound group using a preset similarity scale. The similarity calculation unit 104 that has calculated the similarity of the combination of injury and illness generates information indicating the similarity of the combination of injury and illness. However, when the difference in average medical expenses between combinations of wounds and diseases to be compared is equal to or greater than a threshold, the similarity calculation unit 104 sets the similarity between the combinations of wounds and diseases in the information indicating the similarity of the combinations of wounds and diseases. Write to. The similarity measure is, for example, cosine similarity, Hamilton distance, Hamming distance, edit distance, and the like.

  The similar wound / patient group extraction unit 106 included in the medical expenses calculation unit 100 for each wound / patient group includes information indicating the similarity of combinations of wounds and diseases calculated by the similarity calculation unit 104, and record storage for each wound / patient group. With reference to the unit 50, wounds and sickness groups in which the similarity of combinations of wounds and sickness is equal to or greater than a threshold are extracted as similar wounds and sickness groups. For example, the similar wound group extraction unit 106 generates an adjacency matrix, and extracts the wound groups having similarities of the combination of wounds and diseases as a similar wound group using a maximum clique extraction algorithm. By adjusting the threshold, it is possible to adjust the number of wound groups included in similar wound groups, that is, the degree of aggregation of wound groups. A graph having an edge between any two nodes is called a complete graph, but a partial graph that is a complete graph is called a clique. That is, when a certain graph H is a complete graph and a partial graph of the graph G, the graph H is a clique of the graph G. In addition, the maximal clique means a clique that is not a clique when one node is increased. It should be noted that by extracting a similar wound group using the maximum clique extraction algorithm, even if the wound code is different, it is uniformly treated as a similar wound group, and the difference between the wounds to be extracted can be easily obtained. In other words, doctors and clinics may have a blurring of the notation, and different wound codes may be assigned even for the same or similar wounds. It is desirable to grasp. As a means of grasping the relationship between injury and sickness combination and medical expenses after considering the blurring of the notation, focusing on the similarity between the injury and disease combination (node), extracting similar injury and sickness combinations as a group ing.

  Furthermore, the medical expenses totaling unit 102 refers to the information indicating the similar wounds and sick groups extracted by the similar wounds and disease group extraction unit 106, aggregates similar wounds and sickness groups as one wound and sickness group, and average medical expenses for each wound and sickness group. Recalculate In other words, the medical expenses totaling unit 102 regenerates the records for each wound and disease group stored in the records for each disease and disease group 50 based on the wound and disease groups after the aggregation, that is, aggregates similar wounds and disease groups. Generate records by injury group. FIG. 3B is an example of a record for each disease group after the aggregation generated and stored by the medical cost aggregation unit 102. Note that the record D in FIG. 3B is a record in which the records B and C in FIG.

  As described above, the medical cost calculation unit 100 for each wound group quantifies the degree of similarity in the combination of wounds and sickness constituting each wound group, and the similarity is equal to or greater than a threshold value and the average medical cost of each wound group is calculated. The similar wound groups having a difference equal to or less than a predetermined value are aggregated as one wound group, and the average medical cost for each wound group is calculated.

  The medical expenses rapid injury group combination extraction unit 110 calculates the average medical expenses of one wound group calculated by the medical expenses calculation unit 100 for each wound group, all wound diseases and other one wound disease that constitute the one wound group. Based on the average medical cost of the other wound group, the average differential medical cost, which is the difference between the average medical expenses in the one wound group and the other wound group, is calculated, The combination between the sick and sick groups having the highest average differential medical expenses is extracted from the above. In other words, the medical expenses rapid injury group combination extraction unit 110 extracts a combination of a wound group before the addition and an injury group after the addition in which the average medical expenses rapidly increase due to the addition of one wound, based on the average differential medical expenses. To do.

  More specifically, first, the differential medical cost calculation unit 112 included in the medical cost rapid increase injured disease group combination extraction unit 110 is a wound disease / disease group-specific record stored in the wound / disease group-specific record storage unit 50, that is, an aggregated injury / disease group. Refer to another record, based on the average medical cost of one wound group and the average medical cost of the other wound group consisting of all wounds and other wounds that constitute the one wound group. An average differential medical cost that is a difference between the average medical costs in one wound group and the other wound group is calculated. Specifically, the differential medical cost calculation unit 112 generates a differential medical cost record between wound and disease groups including the average differential medical cost, and stores it in the differential medical cost record storage unit 60 between wounds and disease. FIG. 4A is an example of a medical differential record between wound and sick groups generated and stored by the differential medical cost calculation unit 112. In addition, the sick and sick group differential medical expenses record shown to Fig.4 (a) has each item of a current sickness and disease code list, a future sickness and disease code list, and an average differential medical expenses. Note that the record A in FIG. 4A stores the average differential medical cost “600” in association with the current injury code list “S01” and the future injury code list “S01, S02”.

  The high-order differential medical cost extraction unit 114 included in the medical cost rapid increase injured disease group combination extraction unit 110 refers to the differential medical cost record storage unit 60 between wounds and sick groups, and the average differential medical treatment is used as a combination between the wound and disease groups in which the average medical cost increases rapidly. A combination between the upper and lower predetermined number of injuries and groups having a higher cost is extracted. Specifically, the high-order differential medical expenses extraction unit 114 extracts a top predetermined number of medical-difference medical expenses records having a large average differential medical expenses from the wound-and-disease differential medical expenses record storage section 60, It is stored in the rank difference medical cost record storage unit 70 as a record. FIG. 4B is an example of a higher-order differential medical cost record stored by the higher-level differential medical cost extraction unit 114. Note that the high-order differential medical cost record shown in FIG. 4B includes items of a current injury / disease code list, a future injury / disease code list, and an average differential medical cost.

  The guidance target information generation unit 120 refers to the receipt database 40 and the higher-order differential medical cost record storage unit 70 and extracts a guidance target that is a target of health guidance. Specifically, the guidance target information generation unit 120 generates a guidance target record including at least an insured person name, a current injury / illness code list, and a future injury / illness code list, and stores the instruction target record in the instruction target record storage unit 80. In other words, the guidance target information generation unit 120 performs medical treatment based on the receipt information for the current fiscal year and the combination between the wound and disease groups in which the average medical cost rapidly extracted by the medical expenses rapid increase and decrease disease group combination extraction unit 110 increases. Identify the insured who needs guidance on the increase in costs, in other words, the insured whose medical expenses may increase rapidly, the information identifying the current injuries of the insured, and the injuries to be prevented by the insured The guidance target record including the information to be generated is generated and stored in the guidance target record storage unit 80. FIG. 5A is an example of a guidance target record generated and stored by the guidance target information generation unit 120. Note that the guidance target record shown in FIG. 5A includes items of an insured name, a current injury code list, a future injury code list, and an average increased medical cost that is an average differential medical cost. Note that the record A in FIG. 5A is associated with the insured person name “Insurance Taro”, the current injury code list “S01”, and the future injury code list “S01, S02”, and the average differential medical cost “600”. Is remembered.

  The guidance information output unit 130 refers to the guidance target record storage unit 80, the sickness name master information storage unit 10, and the prevention content master information storage unit 20, and generates and outputs guidance information that is information related to health guidance. Specifically, the guidance information output unit 130 generates guidance information including at least the insured name, the current injury / illness name, the injury / illness name to be prevented, and the prevention content information, and supplies the instruction information to the input / output unit. In other words, the guidance information output unit 130 increases the medical expenses based on the guidance target record generated by the guidance target information generation unit 120 and the prevention content master information stored in the prevention content master information storage unit 20. Generate and output guidance information, including insured persons who need guidance for medical expenses, that is, guidance on rapidly increasing medical expenses, the name of the insured's current injury and illness, the name of injury and illness to be prevented by the insured, and prevention content information . 5B and 5C are examples of instruction information generated and output by the instruction information output unit 130. FIG.

  Note that the guidance information shown in FIG. 5B is output when the insured is not designated, and the insured name, current injury / illness name, injury / injury disease code, injury / morbidity name to be prevented, injury / injury disease code, Each item includes an average increased medical cost, which is an average differential medical cost, and prevention content information. In FIG. 5B, column B includes the insured name “Insurance Taro”, the current wound name “A”, the current wound code “S01”, the wound name “B” to be prevented, and the prevention. Injury / disease code “S02”, average increase medical cost “600”, and prevention content information are stored.

  The guidance information shown in FIG. 5 (c) is an output when an insured person is designated, and the insured person name, current injury / illness name, injury / injury disease code, injury / morbidity name to be prevented, injury / injury disease code, Each item includes an average increased medical cost, which is an average differential medical cost, and prevention content information. The column C in FIG. 5C shows the name of the injured person “Insurance Taro” who has developed the injured name “A” identified by the injured code “S01”. The wound code of the wound to be prevented is “S02”, the average medical cost is “600”, and the prevention content information is stored.

  Subsequently, using FIG. 6 to FIG. 16, the operation of the medical expenses calculation unit 100 for each wound and disease group, the medical expenses suddenly increased disease group combination extraction unit 110, the guidance target information generation unit 120, and the medical expenses calculation unit 100 for each wound and disease group The flow of various types of information in the medical expenses rapid injury group combination extraction unit 110 and the guidance target information generation unit 120 will be described. The flowchart shown in FIG. 6 starts when, for example, an instruction to generate a guidance target record is acquired via the input unit, or at a predetermined time. Further, in the DFD of FIG. 7, the process “medical cost totaling” is a process executed by the medical cost totaling unit 102 and corresponds to the process of step S <b> 100 in FIG. 7, and the process “similarity calculation” is the similarity calculation. This process is executed by the unit 104 and corresponds to the process of step S200, and the process “similar wound and disease group extraction” is a process executed by the medical cost totaling unit 106 and corresponds to the process of step S300. “Group Aggregation” is a process executed by the medical cost totaling unit 102 and corresponds to the process of Step S400, and the process “Average Medical Cost Calculation” is a process executed by the differential medical cost calculation unit 112 and is Step S500. The process “upper extraction” is a process executed by the upper differential medical cost extraction unit 114. Corresponds to the processing in step S600 Te, the process "guidance target extraction" corresponds to a process executed by the guidance object information generating unit 120 to the processing in step S700.

  In FIG. 6, the medical expense totaling unit 102 refers to receipt information for a certain period in the past stored in the receipt database 30, records for wounds and sickness groups stored in the records for wounds and sickness groups 50, Are averaged (step S100). Specifically, as shown in FIG. 7, the medical cost totaling unit 102 refers to the receipt information and records for each disease group for a certain past period, generates a record for each disease group, and records the record unit 50 for each disease group. To remember. In FIG. 7, “record for each disease group (before aggregation)” represents a record for each disease group before consolidation, and “record for each disease group (after aggregation)” represents a record for each disease group after aggregation.

  The process according to step S100 will be further described using the flowchart shown in FIG. 8 that details the process according to step S100. In FIG. 8, the medical expenses totaling unit 102 acquires one record from the receipt database 30 (step S110). The medical expense totaling unit 102 refers to the medical expense record storage unit 50 for each wound and disease group, and determines whether there is a medical expense record for each wound and disease group that is the same injury / illness code list as the acquired record. Step S120).

  When it is determined that there is no medical cost record for each wound and disease group that is the same wound and disease code list as the wound and disease code list of the acquired record (step S120: No), the medical expenses totaling unit 102 A medical expense record is generated (step S130). Specifically, the medical expenses totaling unit 102 writes the value of the item “wound and illness code list” of the acquired record in the item “wound and illness code list”, writes 1 in the item “number of cases”, the item “total medical expenses” and the item The value of the item “medical remuneration score” of the acquisition record is written in “average medical cost”, a medical cost record for each sick and sick group is generated, and stored in the medical cost record storage unit 50 for each sick and sick group.

  When it is determined that there is a medical expense record for each wound and disease group that is the same wound and disease code list as the wound and sickness code list of the acquired record (step S120: Yes), the medical expenses totaling unit 102 Another medical expense record is updated (step S140). Specifically, the medical expense totaling unit 102 adds 1 to the value of the item “number of cases” of the medical expense record for each disease group and the item “total medical expenses”, and the item “medical reward points” of the acquisition record. The average medical cost is calculated from the medical remuneration score after the addition and the number after the addition, and written in the item “Average medical cost”.

  Subsequent to step S130 or step S140, the medical expenses totaling unit 102 determines whether or not all records have been acquired from the receipt database 30 (step S150). When it is determined that the medical expenses totaling unit 102 has not acquired all the records from the receipt database 30 (step S150: No), the medical cost totaling unit 102 returns to step S110 and acquires the next one record from the receipt database 30 ( Step S110).

  When it is determined that all the records have been acquired from the receipt database 30 (step S150: Yes), the medical cost totaling unit 102 sorts the medical cost records by wound group according to the wound group (step S160). The medical expense totaling unit 102 writes the codes in ascending order in the item “list ID” of the medical expense record for each wound group after sorting. As a result, a medical expense record for each wound group as shown in FIG. And this flowchart is complete | finished.

  In FIG. 6, the similarity calculation unit 104 refers to the wound / symptom group record before aggregation stored in the wound / disease group record storage unit 50, and calculates the similarity of the combination of the wounds and diseases constituting each wound group ( Step S200). Specifically, as shown in FIG. 7, the similarity calculation unit 104 refers to the injury / disease group-specific record before aggregation, generates information indicating the similarity of injury / illness combinations, and stores the information in the memory 140.

  The process according to step S200 will be further described using the flowchart shown in FIG. 9 in which the process according to step S200 is detailed. In FIG. 9, the similarity calculation unit 104 acquires one record as a comparison source record in ascending order of the list ID from the record storage unit 50 by disease group (step S210). The similarity calculation unit 104 determines whether or not the last record has been acquired as the comparison source record from the record group 50 for each disease group (step S220). When the similarity calculation unit 104 determines that the last record has been acquired as the comparison source record from the record storage unit 50 for each disease group (step S220: Yes), this flowchart ends. When it is determined that the similarity calculation unit 104 has not acquired the last record as the comparison source record from the record storage unit 50 by disease group (step S220: No), the comparison source record acquired in the last executed step S210, That is, based on the last acquired comparison source record, another record is acquired as a comparison destination record in ascending order of the list ID from the wound disease group record storage unit 50 (step S230).

  Subsequent to step S230, the similarity calculation unit 104 calculates the similarity between the wound code list of the comparison source record and the wound code list of the comparison target record (step S240). Specifically, first, the similarity calculation unit 104 represents the injury / illness code list by a dummy variable for each injury / illness code included in the injury / illness code list, that is, as shown in FIG. To do. FIG. 10A is a 0-1 vector representation of each disease code list indicated by each list ID. For example, row A is a 0-1 vector representation of the wound code list “S01” indicated by the list ID “L001”, and 1 is written in the item “S01” and 0 is written in the other items. Also, for example, row B is a 0-1 vector representation of the injury / illness code list “S01, S03” indicated by the list ID “L011”. The items “S01” and “S03” are 1 and the other items are 1 Write 0.

  The similarity calculation unit 104 expressing the injury / illness code list as a vector of 0-1 uses a scale for measuring the similarity between the 0-1 vectors, that is, the similarity between the injury / illness code lists to be compared, that is, one injury / illness code list. The degree of similarity between the combination of injury / illness indicated by the above and the combination of injury / illness indicated by other injury / illness code lists is calculated. For example, the similarity calculation unit 104 calculates the similarity between the injury / illness code lists to be compared using the cosine similarity expressed by the following equation (1).

Cosine similarity s (v, w): = <v, w> / (| v || w |) (1)
However, <v, w> is an inner product of vectors, and | v |, | w | are vector norms.

  The similarity calculation unit 104 that has calculated the similarity between the comparison-target disease / disease code lists, that is, the similarity between the comparison-target injury / disease combinations, generates information indicating the similarity of the combination of injury / disease (step S250). Specifically, the similarity calculation unit 104 determines that the difference in average medical expenses between combinations of injury and sickness to be compared, that is, the difference in average medical expenses between injury and sickness groups to be compared is equal to or less than a predetermined value. The similarity calculated in step S240 is written as the similarity between the combinations, and if the difference in average medical expenses between the injury-and-disease groups to be compared exceeds a predetermined value, 0 is written as the similarity between the combinations of the injury and disease As shown in FIG. 10B, information indicating the similarity of the combination of injury and illness is generated. Note that the record A in FIG. 10B stores the similarity “0.2” in association with the similarities of the wound code list “L001” and the wound code list “L002”. That is, it stores that the similarity between the wound and disease combination identified by the wound and disease code list “L001” and the wound and disease combination identified by the wound and disease code list “L002” is 0.2.

  Subsequent to step S250, the similarity calculation unit 104 determines whether or not the last record is acquired as a comparison target record from the record group 50 for each disease group (step S260). When the similarity calculation unit 104 determines that the last record has not been acquired as the comparison target record from the record storage unit 50 by injury / disease group (step S260: No), the process returns to step S230 to return to the record storage unit 50 by injury / disease group. To acquire the next record as a comparison target record (step S230). When the similarity calculation unit 104 determines that the last record has been acquired as the comparison target record from the record storage unit 50 for each disease group (step S260: Yes), the similarity calculation unit 104 returns to step S210 and continues from the record storage unit 50 for each disease group. Is acquired as a comparison source record (step S210).

  In FIG. 6, the similar wound group extraction unit 106 includes information indicating the similarity of the wound combination stored in the memory 140, and the pre-consolidation record for the wound group stored in the wound group record storage unit 50. , The wound groups having similarities in the combination of wounds and diseases constituting each wound group are extracted as similar wound groups (step S300). Specifically, as shown in FIG. 7, the similarity calculation unit 104 refers to the information indicating the similarity of the combination of wounds and the records for each wound group before the aggregation, and generates information indicating the similar wound group, Store in memory 140.

  The process according to step S300 will be further described using the flowchart shown in FIG. 11 in which the process according to step S300 is detailed. In FIG. 11, the similar wound group extraction unit 106 acquires the similarity between the wound code list (i) and the wound code list (j) from the information indicating the similarity of the wound combination stored in the memory 140 ( Step S310). Specifically, for example, the similar wound group extraction unit 106 acquires, in order from the top, each record of information indicating the similarity of the combination of wounds, and acquires the similarity from the acquired record. For example, the similar wound group extraction unit 106 first acquires the record A in FIG. 10B, and selects the wound code list “L001” and wound code list (2) of the comparison source list ID as the wound code list (1). ) Of the comparison target list ID “L002” is acquired.

  The similar wound group extraction unit 106 that has acquired the similarity between the wound code list (i) and the wound code list (j) has a similarity (for example, 0.5) between the wound code list (i) and the wound code list (j). ) It is determined whether or not this is the case (step S320).

  If the similar wound group extraction unit 106 determines that the similarity between the wound code list (i) and the wound code list (j) is less than the threshold (step S320: No), the (i) of the N × N adjacency matrix (i) , J) and 0 are written in the element and the (j, i) element (step S330). For example, the similar wound group extraction unit 106 determines that the similarity “0.2” between the wound code list “L001” that is the wound code list (1) and the wound code list “L002” that is the wound code list (2) is the threshold value. If it is determined that it is less than “0.5”, as shown in FIG. 12, 0 is added to the element A that is the (1,2) element of the adjacency matrix and the element B that is the (2,1) element of the adjacency matrix. Write.

  If the similar wound group extraction unit 106 determines that the similarity between the wound code list (i) and the wound code list (i) is equal to or greater than the threshold (Yes in step S320), the (i) of the N × N adjacency matrix (i) , J) and 1 are written in the element and the (j, i) element (step S340). For example, the similar wound group extraction unit 106 determines that the similarity “0.8” between the wound code list “L003” as the wound code list (3) and the wound code list “L004” as the wound code list (4) is a threshold. When it is determined that the value is “0.5” or more, as shown in FIG. 12, 1 is added to the element C that is the (3, 4) element of the adjacency matrix and the element D that is the (4, 3) element of the adjacency matrix. Write. The similar wound group extraction unit 106 generates an adjacency matrix in the memory 140 when step S330 or step S340 is executed for the first time.

  Subsequent to step S330 or step S340, the similar wound group extraction unit 106 determines whether or not the similarities of all combinations of the wound code list have been acquired (step S350). When the similar wound group extraction unit 106 determines that the similarity of all combinations of the wound code list has not been acquired (step S350: No), the process returns to step S310, and the wound code list (i) is stored from the memory 140. The similarity of the injury / illness code list (j) is acquired (step S310).

  When the similar wound group extraction unit 106 determines that the similarity of all combinations of the wound code list has been acquired (step S350: Yes), the similar wound group is referred to by the maximum clique algorithm with reference to the N × N adjacency matrix. Extraction is performed to generate information indicating a similar wound group (step S360). And this flowchart is complete | finished. Specifically, the similar wound group extraction unit 106 creates a graph as illustrated in FIG. 13A from the N × N adjacency matrix, in which the wound code list is a node and the similar relationship between the wound code lists is an edge. The maximum clique algorithm is used to extract the maximum clique as a similar wound group as shown in FIG. 13B, and information indicating the similar wound group as shown in FIG. 13C is generated. Note that the record A in FIG. 13C stores that the wound code list “L003” and the wound code list “L004” are similar wound groups.

The extraction of the maximum clique will be described.
(Data input)
The similar wound group extraction unit 106 inputs a graph G = (V, E) as input data. V is a node set when {wound and disease combination, number of cases, total medical expenses} is a node, and E is an edge set when the similarity between nodes is an edge. In other words, the nodes in the node set represent wound and disease combinations, edges are set between the wound and disease combinations having a similar relationship (edges are present), and the adjacency relationship is expressed by an N × N adjacency matrix.
(Creek extraction)
The similar wound group extraction unit 106 extracts a clique from the graph G. There are various methods for extracting a clique. The similar wound group extraction unit 106 extracts a clique by the following method, for example.
Step 1: One node v is selected from V.
Step 2: Create a list of nodes with nodes v and edges, and store them in the table as cliques of size 2. At this time, combinations equivalent to the set are excluded from the table.
Step 3: For each clique listed in the size 2 list, if there are nodes with edges and all nodes in the clique, the list of nodes in the clique and the nodes in the clique will be displayed. As a table.
Step 4: Repeat step 3 until there are no cliques registered as size 3, 4, 5,. That is, at the end of step 4, cliques including the node v are listed.
Step 5: For each registered clique, when one node that is not included in the clique is added, a combination that is equivalent to the set of already registered cliques is excluded from the list.
By the above method, the maximal cliques including the node v can be enumerated, so that all maximal cliques can be enumerated by performing Step 1 to Step 5 for all the nodes included in the node set V.

  In FIG. 6, the medical expense totaling unit 102 refers to the information indicating the similar wound group stored in the memory 140 and the record for each wound group before the consolidation stored in the record storage unit 50 for each wound group, The similar wound groups are aggregated as one wound group, and the average medical expenses for each wound group are recounted (step S400). Specifically, as shown in FIG. 7, the medical expense counting unit 102 refers to the information indicating the similar wound group and the record for each wound group before the aggregation, and the wound group stored in the wound storage group 50. Another record is regenerated, that is, a wound-by-disease group-specific record in which similar wound-disease groups are aggregated is generated.

  The process according to step S400 will be further described using the flowchart shown in FIG. 14 that details the process according to step S400. In FIG. 14, the medical expense totaling unit 102 refers to the information indicating the similar wound group stored in the memory 140 and the record for each wound group before the consolidation stored in the wound group record storage unit 50. A wound code common to each wound group within each similar wound group is extracted (step S410). In other words, the medical expenses totaling unit 102 extracts combinations of wounds and diseases that frequently appear in the maximum clique. For example, the medical expenses totaling unit 102 extracts wound and disease codes that appear in common in a wound and disease group of a predetermined ratio (for example, 60%) or more among wound and disease groups in the similar wound and disease group.

The extraction of combinations of wounds and diseases that frequently appear in the creek will be described.
(Data input)
The medical expense totaling unit 102 inputs the node set Vc included in the clique C and the appearance frequency threshold f (0 ≦ f ≦ 1) as input data. The node is {wound and disease combination, number of cases, total medical expenses}.
(Creek extraction)
There are various methods for extracting frequently occurring keywords, that is, combinations of injuries and sicknesses, but the medical cost totaling unit 102 extracts keywords by, for example, the following method, which is a method for extracting keywords that appear at a certain frequency or higher. To do.
Step 1: One wound name s is selected from the wound name master information.
Step 2: Count the number of nodes including injury / sickness name s among the nodes included in the node set Vc, and set it as Ns.
Step 3: If the wound name appears at a certain frequency or higher, that is, if Ns> f × | Vc |, the wound name is output as a frequent wound name.
The above method is repeated for the number of wound names existing in the wound name master information.

  The medical expenses totaling unit 102 aggregates the wounds and sickness groups in the similar wounds and sickness groups in the records for the wounds and sickness groups (step S420). Specifically, the medical expense totaling unit 102, for example, in the wound and disease group record before aggregation stored in the wound and disease group record storage unit 50, the list ID having the smallest list ID among the wound and disease groups in the similar wound group. By integrating the contents of the records of other wounds in the similar wound group into the record, the wounds in the similar wound group are aggregated. Thereby, for example, a medical expense record classified by wound and disease group as shown in FIG. 3B is generated from a medical expense record classified by wound and disease group before aggregation shown in FIG. And this flowchart is complete | finished. In the example illustrated in FIG. 3, the records indicated by the list IDs “L003” and “L004” are aggregated to the list ID “L003”, and the records indicated by the list IDs “L011” and “L012” are aggregated to the list ID “L011”. The records shown in the list IDs “L020” and “L021” are collected in the list ID “L020”. In addition, the common wound code extracted in step S410 is written in the item “wound and disease code list” of the medical expenses record classified by wound and disease group after aggregation, and the number of common wound and disease codes corresponds to the medical expenses record classified by wound and disease group before aggregation. When the number of injury / illness codes in the record to be reduced decreases, the item “total medical expenses” and the item “average medical expenses” are recalculated and updated based on the common injury / illness codes.

  In FIG. 6, the differential medical expenses calculation unit 112 refers to the consolidated records of wounds and diseases stored in the wound and disease groups record storage unit 50, and calculates the average differential medical expenses between one wound and other groups. Is calculated (step S500). Specifically, as shown in FIG. 7, the differential medical cost calculation unit 112 refers to the consolidated records for wound and disease groups, generates a differential medical cost record between wound and disease groups, and stores the differential medical cost record between wound and disease groups. Store in unit 60.

  The process according to step S500 will be further described with reference to the flowchart shown in FIG. 15 which details the process according to step S500. In FIG. 14, the differential medical cost calculation unit 112 acquires one record as a comparison source record in ascending order of the list ID from the wound disease group record storage unit 50 (step S <b> 510). The differential medical cost calculation unit 112 determines whether or not the last record has been acquired as the comparison source record from the record storage unit 50 by injury and illness group (step S520). When it is determined that the differential medical cost calculation unit 112 has acquired the last record as the comparison source record from the record storage unit 50 by injury and illness group (step S520: Yes), this flowchart ends. If it is determined that the last medical record has not been acquired as the comparison source record from the record storage unit 50 by injury and illness group (Step S520: No), the differential medical cost calculation unit 112 determines the comparison source record acquired in the last executed Step S510. That is, based on the comparison source record acquired last, another record is acquired as a comparison destination record in ascending order of the list ID from the wound disease group record storage unit 50 (step S530).

  Subsequent to step S530, the differential medical cost calculation unit 112 includes each injury / illness code in the injury / illness code list of the comparison source record and each injury / illness code in the injury / illness code list of the comparison source record. It is determined whether or not the number of wound codes in the code list is one more than the number of wound codes in the wound code list of the comparison source record (step S540). That is, the differential medical cost calculation unit 112 determines whether the injury / illness code of the comparison target record is obtained by adding one injury / illness code to the injury / illness code of the comparison source record. If the difference medical cost calculation unit 112 determines that the injury / illness code of the comparison target record is obtained by adding one injury / illness code to the injury / illness code of the comparison source record (step S540: Yes), the comparison source record and the comparison destination An average differential medical cost is calculated from the record, a differential medical cost record between wounds and sick groups as shown in FIG. 4A is generated, and stored in the differential medical cost record storage unit 60 between wounds and sickness (step S550). In addition, the item “current injury / disease code list” of the difference medical cost record between wound and disease groups is the injury / illness code list of the comparison source record, the item “future injury / injury code list” is the injury / illness code list of the comparison target record, and the item “average difference medical cost” "Memorize | stores the calculated average difference medical expenses. If the difference medical cost calculation unit 112 determines that the injury / illness code of the comparison target record is not the addition of one injury / illness code to the injury / illness code of the comparison source record (step S540: No), step S550 is skipped and step S560 is skipped. Proceed to

  Subsequent to step S550 or step S540 (No), the differential medical expenses calculation unit 112 determines whether or not the last record has been acquired as the comparison target record from the wound disease group record storage unit 50 (step S560). When it is determined that the last medical record has not been acquired as the comparison target record from the record storage unit 50 by injury / disease group (step S560: No), the differential medical cost calculation unit 112 returns to step S530 and returns to the record storage unit by injury / disease group. The next record from 50 is acquired as a comparison target record (step S530). When it is determined that the last medical record is acquired as the comparison target record from the record storage unit 50 for each disease group by the difference medical cost calculation unit 112 (step S560: Yes), the record storage unit 50 for each disease group is returned to step S510. The next one record is acquired as a comparison source record (step S510).

  In FIG. 6, the higher-order differential medical cost extraction unit 114 refers to the differential medical cost record between wounds and sick groups stored in the differential medical cost record storage unit 60 between wounds and sick groups. As above, a combination between the upper and lower predetermined number of wound groups having a large average differential medical cost is extracted (step S600). Specifically, as shown in FIG. 7, the higher-order differential medical cost extraction unit 114 refers to the difference medical cost record between wounds and sick groups, and the upper predetermined number having a large average differential medical cost from the difference medical cost record between the wounds and sick groups. Are recorded in the higher-order differential medical expenses record storage unit 70 as higher-order differential medical expenses records. Note that the upper differential medical cost extraction unit 114 may sort the upper differential medical cost records in descending order of the average differential medical cost.

  In FIG. 6, the guidance target information generation unit 120 refers to the receipt information for the current year stored in the receipt database 40 and the higher-order differential medical cost record stored in the higher-level differential medical cost record storage unit 70. Then, the guidance target is extracted (step S700). Specifically, as shown in FIG. 7, the guidance target information generation unit 120 refers to the receipt information for the current year and the higher-order differential medical expenses record, generates a guidance target record, and stores the guidance target record in the guidance target record storage unit 80. Remember. Then, the flowchart shown in FIG. 6 ends.

  The process according to step S700 will be further described with reference to the flowchart shown in FIG. 16 which details the process according to step S700. In FIG. 16, the guidance target information generating unit 120 acquires a record of one insured person from the receipt database 40 (step S710). The guidance target information generation unit 120 refers to the higher-order differential medical cost record storage unit 70 and determines whether or not there is an upper-level differential medical cost record that is the same current wound code list as the acquired record. Step S720).

  When the guidance target information generation unit 120 determines that there is a higher-order differential medical expense record that is the same current sickness / illness code list as the acquired record's injury / defect code list (step S720: Yes), Based on this, a guidance target record is generated (step S730). Specifically, the guidance target information generation unit 120 transcribes the item “insured person name” of the acquisition record to the value of the item “insured person name”, and adds the upper difference to the value of the item “current injury code list”. Transfer the value of the item “current injury code list” of the medical expense record, transfer the value of the item “future injury code list” of the higher-order differential medical expense record to the value of the item “future injury disease code list”, and item “average” The value of the item “average differential medical cost” of the higher-order differential medical cost record is transcribed to the value of “differential medical cost”, a guidance target record is generated, and stored in the guidance target record storage unit 80.

  Subsequent to step S730 or step S720 (No), the guidance target information generation unit 120 determines whether or not all records have been acquired from the receipt database 40 (step S740). When it is determined that the guidance target information generation unit 120 has not acquired all the records from the receipt database 40 (step S740: No), the process returns to step S710 to acquire the record of one insured person from the receipt database 40. (Step S710). When the guidance target information generation unit 120 determines that all records have been acquired from the receipt database 40 (step S740: Yes), this flowchart ends.

  Next, the operation of the guidance information output unit 130 and the flow of various information in the guidance information output unit 130 will be described with reference to FIG. In FIG. 17A, the guidance information output unit 130 determines whether or not an instruction to output guidance information has been acquired from the input / output means (step S810). If the instruction information output unit 130 determines that an instruction to output instruction information from the input / output means has not been acquired (step S810: No), this flowchart ends.

  When the instruction information output unit 130 determines that an instruction to output instruction information from the input / output means has been acquired (step S810: Yes), the instruction information output unit 130 determines whether or not an insured person is specified in the instruction ( Step S820). When the guidance information output unit 130 determines that there is no designation of the insured (step S820: No), the guidance information output unit 130 refers to the guidance target record, the injury / illness master information, and the prevention content master information, as illustrated in FIG. Then, list information as guidance information for all the insurers is generated and output to the input / output means (step S830). And this flowchart is complete | finished. When the guidance information output unit 130 determines that the insured person is designated (step S820: Yes), the guidance information output unit 130 refers to the guidance target record, the injury and illness master information, and the prevention content master information, as illustrated in FIG. The guidance information of the insurer is generated and output to the input / output means (step S840). And this flowchart is complete | finished. Note that the DFD process “guidance information output” illustrated in FIG. 17B corresponds to the processing in steps S730 and S740.

  According to the medical information generating device 1, since the medical expenses are calculated for each wound group composed of one or more wounds, it is possible to accurately grasp the relationship between the wounds and the medical expenses when the wounds are combined. become able to. Further, according to the medical information generation apparatus 1, since the difference between the medical expenses of a certain wound group and the medical expenses of the wound group when one wound is added to the wound group, N ( It becomes possible to accurately grasp the increase in medical expenses when one more disease is added from N). Specifically, it is possible to appropriately grasp an increase in medical expenses even in a scene where a sickness and disease are added for the first time and a situation where a sickness and disease are already occurring and one more is added. Therefore, effective health guidance, mainly guidance for a surge in medical expenses, can be provided.

  In other words, the medical information generation apparatus 1 does not divide the medical practice, that is, medical expenses for each sickness and sickness, but regards all the sickness and medical practice names described in the reception as one combination, Since medical expenses can be accurately calculated and analyzed even in situations where there are concurrent occurrences of medical expenses, for example, as shown in FIG. It is possible to efficiently provide health guidance, mainly guidance for a surge in medical expenses, to the top insured.

  Furthermore, according to the medical information generation apparatus 1, similar wounds and sickness groups are aggregated into one wound and sickness group, and the average medical expenses for each wounded and sickness group are calculated. Variations in the description of wounds and illnesses can be absorbed, and rational processing can be performed according to the actual situation. For example, according to the medical information generating apparatus 1, as shown in FIG. 19, “Diabetes”, “High blood pressure”, “Neuropathy” are described in the wound name, and the medical treatment score is 160 points. The receptor B, which is described as “diabetes”, “hypertension”, “peripheral neuropathy” and the medical remuneration score of 165 points, can be identified. Further, in addition to the advantage of rational processing according to the actual situation, there is also an advantage of preventing an unnecessary increase in the type of wound group.

  In the above embodiment, the medical information generation device 1 includes the guidance information output unit 130 and outputs guidance information as illustrated in FIG. 1. However, the medical information generation device 1 provides guidance for generating guidance information. The information output unit 130 may not be provided inside. That is, the medical information generation apparatus 1 is in charge of generating and storing guidance target information, and an external device having the same function as the guidance information output unit 130 refers to the guidance target record storage unit to generate guidance information. May be.

  In addition, a program for executing each process of the medical information generation apparatus 1 according to an embodiment of the present invention is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read by a computer system. By executing, the above-described various processes related to the medical information generation apparatus 1 according to the embodiment of the present invention may be performed. Here, the “computer system” may include an OS and hardware such as peripheral devices. Further, the “computer system” includes a homepage providing environment (or display environment) if a WWW system is used. The “computer-readable recording medium” means a flexible disk, a magneto-optical disk, a ROM, a writable nonvolatile memory such as a flash memory, a portable medium such as a CD-ROM, a hard disk built in a computer system, etc. This is a storage device.

  Further, the “computer-readable recording medium” means a volatile memory (for example, DRAM (Dynamic DRAM) in a computer system that becomes a server or a client when a program is transmitted through a network such as the Internet or a communication line such as a telephone line. Random Access Memory)), etc., which hold programs for a certain period of time. The program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes designs and the like that do not depart from the gist of the present invention.

1 ... Medical information generation device
DESCRIPTION OF SYMBOLS 10 ... Wound and illness name master memory | storage part 20 ... Prevention content master memory | storage part 30 ... Receptacle database 40 ... Receipt database 50 ... Wound and illness group record storage part 60 ... Injury and illness difference medical expenses record storage part 70 ... High-order difference medical expenses record storage part 80 ... Guidance target record storage unit 100 ... Medical cost calculation unit 102 for each wound group ... Medical cost totaling unit 104 ... Similarity calculation unit 106 ... Similar wound group extraction unit 110 ... Medical cost rapid injury group combination extraction unit 112 ... Differential medical cost Calculation unit 114 ... upper difference medical cost extraction unit 120 ... guidance target information generation unit 130 ... guidance information output unit 140 ... memory

Claims (6)

Based on a plurality of receipt information that describes at least one or more wound information and medical expenses, a medical expenses calculation unit for each wound group that calculates an average medical cost for each wound group composed of one or more wounds,
The average medical cost of one wound and disease group calculated by the medical expenses calculation unit for each wound and disease group, and the average medical cost of another wound and disease group composed of all wounds and other diseases that constitute the one wound and disease group Based on the above, the average difference medical cost that is the difference between the average medical costs in the one wound group and the other wound group is calculated, and the average differential medical cost is higher in the combination between the wound groups A medical expenses suddenly injured disease group combination extractor for extracting combinations between groups;
Based on the receipt information and the combination between the wound and disease groups extracted by the medical expenses suddenly injured disease group combination extraction unit, information for identifying the insured who needs guidance for an increase in medical expenses, the current state of the insured A medical information generation apparatus comprising: a guidance target information generation unit that generates guidance target information including information for identifying a wound and illness and information for identifying a wound and illness to be prevented by the insured. The medical cost calculation unit for each disease group is
The degree of similarity between the combinations of wounds and diseases constituting each wound group is calculated as an angle between each vector with the presence / absence of the elements of the wounds constituting the combination as a vector, and based on the degree of similarity, a single wound group is obtained. The medical information generation apparatus according to claim 1, wherein the medical information is aggregated and an average medical cost for each disease group after the aggregation is calculated. The medical cost calculation unit for each disease group is
With regard to the similarity of all combinations of the calculated disease and disease groups, the adjacent matrix data is created as network data using the injury and disease groups as nodes and the relationship between nodes having similarities equal to or higher than a predetermined threshold as edges. From the cliques that have edges for all nodes, the wound groups that make up the maximum clique with the maximum size that can be taken are aggregated as one wound group, and the average medical cost for each wound group is calculated. The medical information generation apparatus according to claim 2. Each of the constituents of the one wound and disease group is associated with information for identifying each wound and disease constituting one wound and disease group, and information for identifying each wound and disease constituting the one wound and disease group A preventive content storage unit for storing preventive content information indicating the preventive content for the onset of the other one of the wounds during the onset of the wound;
Information identifying the insured person who needs the instruction based on the instruction object information generated by the instruction object information generating unit and the prevention content information stored in the prevention content storage unit, the insured person And a guidance information output unit for outputting guidance information including the current injury and illness name, the injury and illness name to be prevented by the insured, and the prevention content information related to the illness to be prevented. The medical information generation apparatus according to any one of claims 1 to 3. Based on a plurality of receipt information in which at least one or more wound information and medical expenses are described, a medical cost calculation means for each wound group that calculates an average medical cost for each wound group composed of one or more wounds,
The average medical cost of one wound and disease group calculated by the medical cost calculation means for each wound and disease group, and the average medical cost of another wound and disease group composed of all wounds and other diseases that constitute the one wound and disease group Based on the above, the average difference medical cost that is the difference between the average medical costs in the one wound group and the other wound group is calculated, and the average differential medical cost is higher in the combination between the wound groups Medical cost rapid injury group extraction means for extracting combinations between groups;
Information identifying the insured person who needs guidance for an increase in medical expenses based on the receipt information and the combination between the wounded disease and disease groups extracted by the medical expenses suddenly injured disease group combination extracting unit; A medical information generation method comprising: guidance target information generation means for generating guidance target information including information for identifying a wound and illness and information for identifying a wound and illness to be prevented by the insured. To the computer of the medical information generation device that generates medical information
Based on a plurality of receipt information describing at least one or more wound information and medical costs, a medical cost calculation step for each wound group that calculates an average medical cost for each wound group composed of one or more wounds,
The average medical cost of one wound and disease group calculated by the medical cost calculation step for each wound and disease group, and the average medical cost of another wound and disease group composed of all wounds and other diseases that constitute the one wound and disease group Based on the above, the average difference medical cost that is the difference between the average medical costs in the one wound group and the other wound group is calculated, and the average differential medical cost is higher in the combination between the wound groups A medical expense rapid injury group combination extraction step for extracting a combination between groups;
Based on the receipt information and the combination between the wound and disease groups extracted by the medical expense rapid injury group combination extraction step, information for identifying an insured person who needs guidance for an increase in medical expenses, the current state of the insured person A program for executing instruction target information generating step for generating instruction object information, including information for identifying a person's injury and illness, and information for identifying an injured person to be prevented.

Images