JP2006119967A - Medical information system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To hold a patient database of data throughout patients' lives to allow on-line data access for a long time, to enable a plurality of users to simultaneously read/write a chart of one patient, and to efficiently back up data in a short time, while holding high speed and convenience of a conventional relational database. <P>SOLUTION: A medical information system includes a recorder using a relational database, a data input device being well-known as itself, and a data output device being well-known as itself and is provided with a database file (patient file) wherein document records about a certain patient are integrated, and a database file (group file) wherein document records extracted from the patient file as records meeting a designated condition are recorded together with document records extracted from other patient files in the same manner. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to information processing related to medical treatment such as an ordering system for creating, transmitting, and recording medical instructions in medical care, and an electronic medical record that describes information related to a patient in a medical profession. The present invention relates to a medical information system technology performed using a network.

In medical care, it is necessary to issue a large amount of documents such as medical records for examinations and nursing (so-called electronic medical records) and prescriptions for examinations and treatments (so-called ordering). In the past, handwritten medical records and slips have been used, but now it is an electronic record on a computer because it is difficult to read characters and errors are likely to occur. In many cases, so-called electronic medical records, ordering, and medical information systems that integrate them are introduced.
In most cases, a conventional medical information system recording apparatus uses a relational database. A relational database forms a database as a collection of records, but it is necessary to specify a format and a data length in advance for each element (cell) constituting a document record. Due to such restrictions, it is possible to quickly calculate the position (address) of the target data from a long recorded data file and read / write it. In addition, operations such as selecting (selecting) a condition to extract a corresponding record and combining two databases using key items can be executed by a standardized query language called SQL. System construction is easy. For this reason, it is not an exaggeration to say that not only medical treatment but also paperwork called core business is almost done in a relational database.

The types of orders and recorded documents that are created are usually several tens, and more than one hundred when there are many. Each document is different in all items described. Therefore, in a conventional medical information recording apparatus using a relational database, the structure of a document record is defined for each type of document, and a separate independent database file is used.
When a patient is specified, the document record is extracted from each document database using an index or the like by using an ID number or the like as a key, and they are combined to reconstruct the entire chart. It was supposed to be displayed to the user.
A technique described in Japanese Patent Application Laid-Open No. 2004-164196 is known as a technique related to such an electronic medical record.
JP 2004-164196 A

JP-A-2004-164196 includes a treatment database, a prescription database, an examination database, an image diagnosis database, and the like, which are databases storing medical information divided into categories, and extracts specific patient data therefrom. It is the composition to do.
In the medical information system using a conventional relational database such as the above publication, the following problems have occurred.
(1) As data is accumulated, the speed of reading and writing gradually decreases, and the work efficiency is significantly reduced.
Consider drug prescription as an example. A database file is created for each type of document, but even a single patient will see hundreds to thousands of drug prescriptions over a long period of time. In addition, all drug prescriptions for tens of thousands of patients will be recorded in one database file. In large hospitals, it is not uncommon to exceed millions in a few years. In this case, a reduction in processing speed is unavoidable even if an acceleration method such as an index is used. If the processing speed is remarkably lowered, a great deal of trouble occurs in business, and all old data is saved to another recording medium, and the online state is shifted to the offline state. When it becomes necessary to refer to an old record, it must be restored online using a time-consuming method. Despite the dramatic increase in storage capacity of hard disks and the like, it is a big problem that past information is virtually impossible to see in just a few years. In the treatment of diseases that require long-term data reference such as lifestyle-related diseases, even fatal defects can be said.

(2) When a user opens a patient chart, other users cannot open the patient chart.
The medical information system is usually operated in a local area network. However, if multiple users write requests simultaneously, the previous user's write is overwritten by the later user's write, resulting in inconsistent data. There is danger. In order to prevent this, a relational database system provides a technique called record lock. In this case, when a user opens a medical record of a patient, a measure is taken to prohibit writing from the other user to the document record of the patient. In a recording apparatus in a conventional medical information system, since all patient data is recorded in one database file related to a certain category of document, it is necessary to lock the record for all the documents related to the patient. I do not get.
For this reason, there is a problem that the information on the patient cannot be accessed until other people use it, which may cause a significant delay in work.

(3) It takes a long time to back up data.
The loss of medical information due to machine troubles such as hard disk crashes can cause serious problems. For this reason, it is necessary to back up data frequently. It is not uncommon for the hospital where the inventor is working to have a medical record data of over 200 GB, or to reach several terabytes in a large hospital. If you want to back up this much data, you will need several hours to half a day of work, which will cause problems with costs and business problems.

On the other hand, there is a medical information system that uses a variable length database that does not use a relational database and does not predetermine the attributes and data lengths of individual cells. This case does not have the above-mentioned drawback, but has a serious disadvantage that reading and writing of data is slow because the address calculation of the data position becomes very complicated. In addition, since a unique data structure and development environment are used, there is a drawback that various development tool groups that are currently distributed in large quantities and at low cost cannot be used.
The present invention has been made to solve such conventional problems. The object of the present invention is to maintain the high speed and convenience of a conventional relational database, and (1) a patient for a lifetime. Maintain a database and enable long-term online data access; (2) Enable multiple users to read and write the same patient's charts simultaneously; (3) Efficient data backup in a short time Is to make it possible.

  In the medical information system according to claim 1, as a means for solving such a problem, a recording device using a relational database, a known data input device per se, and a known data output device per se, (1) A database file (patient file) in which document records related to a patient are aggregated, and (2) a document record extracted from the patient file as matching a specified condition is extracted in the same manner from other patient files. A database file (group file) is recorded along with the document record.

  3. The medical information system according to claim 2, wherein the server comprises (1) a document related to a patient individual, comprising a server having a data recording device and a client connected to the server via a network system. A database file (patient file) that aggregates records, and (2) a database file that records document records extracted from the patient file as matching the specified conditions together with document records extracted in the same manner from other patient files. (Group file) and (3) A copy function for copying updated document record data to the patient file to the group file is provided.

  The medical information system according to claim 3 is configured such that, in the medical information system according to 1 or 2, the access right in the edit mode is set for each document record of the patient file.

  In the medical information system according to claim 4, the medical information system according to any one of claims 1 to 3, further comprising a management unit for an access log or an update log to the patient file, and updated based on the log It has an automatic backup means that picks up only the patient files that have sex and backs them up.

  According to a fifth aspect of the present invention, in the medical information system according to any one of the first to fourth aspects, an XML text or XML object is used in a cell that records the contents of a document in the document record.

By adopting the above configuration, the following effects can be obtained.
In the medical information system according to the first aspect of the present invention, since the database file (patient file) in which document records related to a certain individual patient are collected is provided, a database system centered on the patient file is constructed.
Therefore, necessary data is collected and managed in a patient file.
In the past, data was compiled by document category, not by patient. For example, prescription database, inspection database, etc. For this reason, there is a problem that databases are crossed between categories, the databases are complicated, the database is enlarged, and the burden at the time of backup is large. However, in the present invention, since the patient file is the center, centralized management is possible with a single database, and the database can be downsized and managed easily.
In addition, it has a database file (group file) that records document records extracted from the patient file as matching specified conditions together with document records extracted in the same manner from other patient files. This makes it possible to create data (group file) for business execution as in the past.

  The medical information system according to claim 2 is provided with a copy function for copying the document record data updated and inputted to the patient file to the group file in addition to the claim 1, so that it is inputted to the patient file. Data is also recorded in the group file, ensuring data consistency between the patient file and the group file.

In the medical information system according to claim 3, since the access right in the edit mode is set for each document record of the patient file, the access right can be set for each specific document record in the patient file. Become.
In the conventional method of grouping data by document category, document records of multiple patients are mixed, so the entire document record belonging to a patient must be locked and other access rights must be denied. . Otherwise, a plurality of update information is overwritten after a plurality of accesses, and data consistency is lost.
As a result, the other person had to wait for input until one person's writing process was completed, and the work efficiency was significantly hindered.
However, since the patient file contains only the document records of the same patient, by setting access rights for each document record unit, many people can run at the same time to open the items in the patient file and update / overwrite them. It becomes possible to improve business efficiency.

In the medical information system according to the fourth aspect, since a means for managing an access log or an update log for a patient file is provided, it is possible to grasp which patient file may have been overwritten.
Further, since the automatic backup means can pick up and back up only the patient file that may be updated, the backup can be efficiently executed with a small amount of data.

  In the medical information system according to the fifth aspect, since the XML text or the XML object is used in the cell in which the document content is recorded in the document record, the input content can be set flexibly.

The best mode for carrying out the present invention will be described below with reference to the drawings.
As shown in FIG. 1, the medical information system of the present invention includes a server that executes a specific process and a client that receives a service from the server, and they constitute a network system. An input device, a control device, a memory device, and an external storage device are provided.
The input device is a keyboard, a mouse, a tablet, a touch panel, or the like, and character string data and various commands are input by this input device.
The display device is a CRT or a liquid crystal display, and input data information and search results are displayed by this display device.
The control device is a CPU and controls the entire device.
The memory device is a ROM or a RAM, and includes a program unit that stores programs necessary for the control device to execute various controls and processes, and a buffer unit that stores data necessary for the processing.
The external storage device is a hard disk, a magneto-optical disk, a flash memory, or the like.
The external communication device is also called a network adapter board and is connected to the LAN. In the LAN, other similar computers and servers for recording large volumes of data are connected. The LAN is connected to the Internet via a modem, a router, and the like.
The client PC requests the server to record a document record to the patient chart and send a prescription record that has already been entered. The server searches the patient chart database, records the prescription contents in the electronic chart, and the result is sent to the client PC. Reply to

In FIG. 1, the server records individual patient file groups and group file groups in which document records extracted from patient files according to designated conditions are recorded together.
The patient file stores specific personal data and is indexed to form an independent file.
As will be described later, the group data file is composed of data extracted from a patient file according to specific conditions.
The client PC requests the server to open the designated patient file. The server returns the specified patient file to the client PC. In the client PC, the document record in the patient file is newly created or edited. The changed document record is automatically copied and updated in the corresponding group file.
When a group file is opened from a client PC, documents corresponding to target patients such as by document category or by hospital ward can be listed.

First, a medical information system according to a conventional example will be described with reference to FIGS.
2 to 5 are recording formats of a conventional medical information system using a relational database. FIG. 2 is a patient ledger database. First, there is an ID number called a key item that does not allow duplication in the first column. Normally, the number is numbered from 1 and a unique number is given to each patient.
The second column is the patient name. The third column is gender and the fourth column is the date of birth. In this way, the patient-specific information is collected. An area for recording individual data is called a cell. These four item cells form a set to form a record.
FIG. 3 shows a drug prescription database. The first column is an ID number as a key item. Hereinafter, prescription department, prescription date, prescribing doctor, drug name, daily prescription amount of the drug, unit of quantity, number of doses, usage, and prescription days are defined.

FIG. 4 shows an X-ray prescription database. Although it is the same as FIG. 3 until the middle, prescription contents are only an imaging | photography site | part and irradiation number.
In any of FIGS. 2 to 4, attributes of each column, that is, data types such as characters, numerical values, and dates, and storage capacity (number of bytes) to be used are defined in advance. As a result, the position (address) of the data of an arbitrary row and column in the file can be easily calculated. From another point of view, records with different record structures must be stored in separate database files.

  FIG. 5 is a combination of a patient ledger database and a drug prescription database using an ID number as a key. It can be seen that the contents of the patient ledger database are combined with the drug prescription database. In this way, relational databases are widely used because join operations, projection, extraction, union and difference sets, and other set operations can be performed on a database very easily using a language called SQL. It became.

FIG. 6 shows the structure of a recording file in a conventional medical information system. FIG. 6A shows a drug prescription database. All drug prescriptions for all patients will be recorded in this file. Here, data of ID number 2031 is picked up. FIG. 6B is an XP prescription database. Similarly, only the data of ID number 2031 is picked up. FIG. 6C is a blood test database. Similarly, only data with ID number 2031 is picked up.
FIG. 6D shows a case where a document record of a patient with ID No. 2031 is extracted from these databases and dynamically reconstructed. Since the number of items varies depending on the document category, the number of items matches the maximum number of items. Depending on the category of the document, some of them will be used. Some medical information systems even record actual data in this type of relational database from the beginning.

In a conventional medical information system using a relational database, records corresponding to a specified ID number are collected from each database, a patient chart is reconstructed, and displayed to a user.
Thus, conventionally, data recording and storage has been a database by document category including data of all patients. Patient charts were temporarily reorganized on demand. For this reason, (1) If the operation is prolonged, the database file becomes huge and the response speed becomes slow. (2) When someone opens the chart in edit mode, the entire patient chart is locked. (3) Various problems such as long time for data backup occurred.

Next, the medical information system according to the first embodiment will be described with reference to FIGS.
FIG. 7 shows a record structure of a patient file according to the present invention. In the patient file, all records of all document categories related to the patient are collected in one file. In this embodiment, the ID number itself is used as the file name. The first column is the document category. This indicates which category of document this document record represents. The second column is the creator name, the third column is the creation date and time of the record, the fourth column is the execution designation date and time of the instruction document, and the fifth column is the contents of the document.

FIG. 8 shows the content of the drug prescription. In the present embodiment, the same contents as in FIG. 3 are expressed using an expression format called XML. XML is an expression format in which contents related to a tag name are usually written between a pair of <tag name> and </ tag name>. Since tags can be nested in tags, data having a hierarchical structure can be expressed easily.
FIG. 9 is an example of XP prescription, and the contents of FIG. 4 are expressed in XML format.
As shown here, if the XML format is used, differences in the structure of individual documents are absorbed by the configuration of the tag, so if you prepare one storage place for plain text with variable length, all Documents of different categories can be stored with the same record structure.
In recent relational databases, not only plain text format XML data but also plain text data can be analyzed and recorded and processed as structured objects. If it is converted into an object, the analysis (parsing) of the XML structure has already been completed, so there is an advantage that processing such as retrieval can be performed at high speed.
FIG. 10 shows the relationship between files in a new medical information system using a relational database. A corresponding patient file is created for each patient. There are also group databases by document category and ward.

Here, a group file is created according to specific extraction conditions for the data input to the patient file. After the group file is created, a copy function is provided to automatically copy the patient file input to the group file.
In the patient file, an access right can be set for each document record (for example, for each input item of drug prescription and XP prescription).
In the conventional method of grouping data by document category, document records of multiple patients are mixed, so the entire document record of the patient must be locked and other access rights must be denied. . Otherwise, a plurality of update information is overwritten after a plurality of accesses, and data consistency is lost.
As a result, the other person had to wait for input until one person's writing process was completed, and the work efficiency was significantly hindered.
However, with patient files, setting access rights for each document record allows many people to open and update different document records in the patient file at the same time, improving work efficiency. To do.

When a new document is created or modified, it is performed on the patient file in principle. Each document category defines in advance which group file the copy should be created. Here, the corresponding group file may be single or may be copied to a plurality of group databases. When a new document is created, a copy of the document is created in the group database corresponding to the document category. Similarly, when a document is modified, it is in the corresponding group database. The corresponding document is corrected (copy function).
Here, the main recording is performed by the patient file. The group file is a so-called working file, and is rarely referenced after several months. Therefore, when the group file becomes large to some extent, it is possible to maintain the access speed by deleting the old part or moving it to the archive. In other words, unlike the conventional medical information system in which only the group file exists as an entity and the patient chart is dynamically reconfigured, the patient file also exists as an entity. There is no reference. Individual patient databases are at most several hundred MB even in hospitalized medical records over 10 years, and delays in reading speed are hardly a problem. Thus, lifelong patient records are possible without problems.

As described above, in the conventional medical information system, since all documents of the corresponding category of all patients exist in the same file, the record lock must be applied to all the documents of the patient at the same time. In the present invention, all the documents related to the patient are already collected into one file. For this reason, the record lock when someone opens the patient file is one step and fine. That is, record lock can be performed in document records. When the physician describes the patient's observation record, only that document is locked. Meanwhile, since the nursing record is not locked, the nurse can write the nursing record in parallel without affecting the work of the doctor. Being able to use this simultaneous access is a great advantage when conducting medical treatment as a team.
Assuming that there are tens of thousands of patient charts in a medical institution, the number of charts whose contents have changed at the end of the day is less than 1% of the total. Because no one touches the outpatients, inpatients, patients who opened the chart for the purpose of creating any document, and the patients other than this, the contents cannot be changed. Therefore, each time the patient database is opened, an ID number log is recorded, and at the end of the day, only the chart with the ID number described in the log record is backed up. At the hospital where the inventor is working, backups that previously took more than 6 hours can now be completed in just 10 minutes.

Although the embodiment of the present invention has been described above, the specific configuration of the present invention is not limited to the present embodiment, and even if there is a design change or the like without departing from the gist of the invention, Included in the invention.
For example, although it is a structure of a document record, it is not restricted to an Example, You may add and change an item according to the actual condition of a medical institution. In the examples, for convenience of explanation, individual data is represented by character strings. However, for practical purposes, a separate code table, doctor list, etc. are prepared, and actual data is recorded using drug numbers, doctor numbers, etc. May be.
In the embodiment, the description was made using plain text or object format XML as the recording format of the content portion. The reason is that various utilization programs are provided for XML, and recording using these programs is the most practical. However, in some cases, headwords and text data as in a normal document may be written as they are without using tags such as XML.
When extracting a document record from a patient file and specifying a single document category such as a drug prescription in the extraction designation conditions constituting the group file, as shown in the embodiment, Xp prescription, MRI prescription, CT There are various methods such as collecting a plurality of document categories such as prescriptions as a radiation group file, and collecting all prescriptions of patients admitted to the ward as a ward group file.
In the embodiment, the patient file of a certain patient or each group file has been described as one. However, depending on the database configuration method, the file may be divided into a plurality of files. This method is also useful when using heavy data such as an image file.
In the embodiment, the number of cells for storing the contents in the document record is one, but depending on the situation, it may be divided into a plurality of cells and recorded.
In the embodiment, all the creation and editing of the document is performed on the patient file, and this content is transferred to the group file. However, when a group file is regarded as a working working file, a mechanism that creates an implementation report document corresponding to an instruction document on the group file and transfers it to the patient file is also useful in some cases. .
This medical information system usually runs on a local area network (LAN) in a hospital, but it can be operated on a wide area network (WAN), or on a stand-alone PC that does not use a network in a small clinic. Also good.

It is a block diagram which shows the hardware constitutions of this invention. It is explanatory drawing of the patient ledger database which concerns on a prior art example. It is explanatory drawing of the chemical | medical agent prescription database which concerns on a prior art example. It is explanatory drawing of the XP prescription database which concerns on a prior art example. This database combines a patient register database and a drug prescription database. It is explanatory drawing of the (a) patient register database which concerns on another prior art example, (b) XP prescription database, (c) blood test database, (d) dynamically reconfigure | reconstructed patient data. It is explanatory drawing which shows the record structure of the patient file of this invention. In FIG. 7, it is explanatory drawing of the data structure stored in the item of the content of a medicine prescription. In FIG. 7, it is explanatory drawing of the data structure stored in the item of the content of XP prescription. It is explanatory drawing which shows the relationship between the patient file and group file of this invention.

Claims (5)

A recording device using a relational database, a known data input device per se, and a known data output device per se,
(1) A database file (patient file) in which document records related to an individual patient are aggregated;
(2) It is provided with a database file (group file) for recording a document record extracted from the patient file as matching a specified condition together with a document record extracted in the same manner from another patient file. Medical information system. In a medical information system comprising a server equipped with a data recording device and a client connected to the server via a network system,
The server
(1) A database file (patient file) in which document records related to an individual patient are aggregated;
(2) a database file (group file) for recording a document record extracted from the patient file as matching the specified condition together with a document record extracted in the same manner from another patient file;
(3) A medical information system provided with a copy function for copying inputted document record data to a patient file to a group file.   The medical information system according to claim 1 or 2, wherein an access right in an edit mode is set for each document record of a patient file.   An access log or update log management means for a patient file is provided, and an automatic backup means for picking up and backing up a patient file that may have been updated based on the log is provided. The medical information system in any one of Claims 1-3. The medical information system according to any one of claims 1 to 4, wherein an XML text or an XML object is used in a cell for recording the contents of a document in the document record.

Images