JP2004348271A - Clinical trial data outputting device, clinical trial data outputting method, and clinical trial data outputting program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a clinical trial data outputting device for extracting clinical trial data by selecting the clinical trial data from classified and correlated item candidates in order to prevent the omission of the recording of clinical trial actions, and to output an easily understandable case result. <P>SOLUTION: This clinical trial data outputting device is provided with an electronic medical record database 61, a means for correlating clinical trial data with case report items, a means for designating an clinical investigation patient from clinical trial information added to the electronic medical record database 61, a normal medical examination information inputting means, a retrieving means based on conditions obtained by the correlation means for extracting the clinical trial data, and a means for outputting the case information of the extracted clinical trial data. Thus, it is possible to solve the problems that (1) it is necessary for a doctor to prepare a case report as the report of medication or a test performed by a medical institute to the clinical investigation patient in a trial in a clinical test, and that (2) it is necessary to manually transfer the contents from medical records or a test result report to a specific report or to re-input the test result to a report terminal by using an exclusive terminal. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a clinical trial data output device for accurately recording and outputting a so-called clinical trial history of a drug and a test performed by a medical institution among clinical trials performed in the development of a new drug and a treatment method.
[0002]
[Prior art]
A database for improving efficiency and accuracy of a clinical trial, such as a clinical trial support database for assisting a medical institution with respect to a drug administered to a clinical trial patient and a so-called clinical history of a test from an information management side, as in Patent Document 1 shown in the figure. Etc. are known.
[Patent Document 1]
JP 2001-142978 A
[Problems to be solved by the invention]
When a medical institution administers medications or tests to a study patient, a physician must prepare a case report as a report. This report must include patient information (age, gender, body weight, etc.), clinical data (medication, treatment details, test results) and patient symptoms. there were.
[0004]
It cannot be used when there is no trial protocol having the tagged information disclosed in Patent Document 1, or when there is no database device such as the trial database 62 or PC that can read the tagged information. In addition, since the format of the clinical trial report varies depending on the pharmaceutical manufacturer and the target case, it cannot be used unless the tags in the clinical trial protocol and the data items in the electronic medical record match. Manual correction is required.
[0005]
In the manual correction and selection work, it is necessary to find out the trial item names from the number of electronic medical record items exceeding 100 items. There are problems such as incorrect correction.
[0006]
In particular, the patient's name and insurance information are not protected, and there is a problem such as leakage of personal information due to incorrect selection.
[0007]
Furthermore, if the device cannot be used, it is necessary to manually transfer the medical record from the medical record filled in for the target case to various forms of documents, or to re-enter the test results in the report terminal using a dedicated terminal, There are problems such as enormous effort, transcription errors, and data falsification.
[0008]
The present invention has been made in view of the above problems, and eliminates omission of a record of a clinical trial by selecting from candidate items classified and related in extraction of clinical trial data. It is an object to provide an output device.
[0009]
[Means for Solving the Problems]
In order to eliminate the omission of the record of the clinical trial, the clinical trial data is associated with the clinical trial item of the clinical trial and the electronic medical record item of the electronic medical record database 61 in which the clinical data of the patient is recorded, and the clinical trial data associated with the clinical trial item is output. Obtain case results without recording omissions from the device.
[0010]
The electronic medical record database 61 has trial patient identification information for identifying a trial subject patient and a general patient, and the medical data search means can retrieve clinical data of the trial patient based on the trial patient identification information.
[0011]
The clinical trial data output device includes: a clinical trial item name selecting unit that selects the displayed clinical trial item name; a corresponding item name selecting unit that extracts, as a candidate, an electronic medical record item name including the identifier for each clinical trial item name; Medical data retrieval means for retrieving medical data of a subject patient based on a plurality of electronic medical record items associated with each clinical trial item, and trial data storage for storing the retrieved medical data in a storage device as trial data arranged therein Means for outputting the stored clinical trial data.
[0012]
The clinical trial data output device includes statistical processing means for statistically processing clinical trial data related to a plurality of clinical trials, and the corresponding item name associating means further includes an electronic medical record item name related to the clinical trial item in the electronic medical record item. And a hierarchical display means for hierarchically displaying electronic medical record item names related to clinical trial items in electronic medical record items.
[0013]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention (hereinafter, referred to as embodiments) will be described with reference to the drawings.
[0014]
FIG. 1 is a schematic diagram of an embodiment of a trial data output device 10. In one embodiment, an electronic medical record database 61, a trial database 62, and a trial data output device 10 as a computer device including all inputs and outputs are shown.
[0015]
First, an outline of the trial will be described. The medical institution receives a request for clinical trials such as drug administration and tests from a drug developer such as a pharmaceutical company, and also receives a trial protocol prepared by the drug developer and recording a clinical trial plan.
[0016]
The study protocol specifies a case record format for recording the progress of medications and tests performed on study patients, and a doctor at a medical institution creates a case report form. The cases are transcribed from the electronic medical records based on the results, attached to the case report, and reported to the drug developer.
[0017]
Next, the internal configuration of the clinical trial data output device 10 will be described. The trial data output device 10 includes a CPU 20, a trial item name selection unit 30, an output unit 40, an external interface unit 50, and an external storage unit 60.
[0018]
The external storage unit 60 including the hard disk includes an electronic medical record database 61 storing medical data, a clinical trial database 62 storing clinical trial data, a case item correspondence table 63 storing association information between clinical trial items and electronic medical record items, and a clinical trial database. It has a case data output format 64 which is a case record format attached to the report.
[0019]
The external interface unit 50 includes a LAN communication interface 51 connected to the Internet, a keyboard 52 for input, a mouse 53, a display 54 for display, a printer 55 for outputting trial data, a scanner 56 for reading documents, and a readable and writable CD. -It has an R / W drive 57.
[0020]
The printer 55 prints out predetermined information in characters on a sheet according to a predetermined format. The document recorded on the paper by the scanner 56 is read as image information, and if it is a sentence, it is converted into text information and stored in the hard disk as reusable data.
[0021]
The CPU 20 includes a trial item name selection unit 30 and an output unit 40, and includes modules of a program executed on the CPU 20.
[0022]
The trial item name selection unit 30 includes a trial item name selection module 31 and a corresponding item name selection module 32. The output unit 40 includes a statistical analysis module 41, a medical care data search module 42, a trial data storage module 43, and a trial data output module 44.
[0023]
Next, the electronic medical record database 61 will be described. FIG. 2 is a schematic diagram of the electronic medical record database 61. The electronic medical record database 61 is composed of three databases. The first is a patient personal information database, the second is a medical care information database, and the third is a progress information database.
[0024]
The first patient personal information database records information on individuals, such as a trial patient identification flag for identifying a trial patient, age, gender, medical history and family medical history, and the like. The second medical treatment information database records medical treatment information, medical treatment contents, medication contents, examination contents / results, image diagnosis results, and the like. The third progress information database records information on the progress in addition to symptoms, blood pressure, body temperature, medicinal progress, and the like.
[0025]
The trial patient identification flag is in the first patient personal information database of the electronic medical record database 61, and the trial patient identification flag is recorded in order to distinguish between a trial patient and a patient unrelated to the trial, and specifies the trial patient. Information.
[0026]
Next, the procedure for designating a trial patient will be described. Normally, a medical institution conducts a clinical trial review, determines whether it is appropriate to receive a clinical trial request, and specifies the trial patient designation means in the electronic medical record database 61 according to the trial patient selection criteria described in the clinical trial protocol. By setting the trial patient identification flag described above, the trial patient is selected.
[0027]
In one embodiment, if not registered in the electronic medical record database 61, the patient's personal information is added, and if it is necessary to add individual trial information, it can be added to the medical information and the progress information. Has become.
[0028]
Next, an outline of the trial database 62 will be described. FIG. 3 is a schematic diagram of the trial database 62. The trial database 62 has case-related search information, case data output format, and patient personal information, medical information, and progress information associated with the patient ID, which are associated by the trial ID. This allows individual trials to be separated by trial ID. The case-related search information includes search conditions such as a patient suffering from a disease widely related to the clinical trial, gender, age, and the like.
[0029]
In particular, the criteria for selecting study patients include information on patients suffering from a particular disease, gender, age, etc. The data can be suitably separated, and data classification for statistical analysis described later can be easily performed. When the above operation is completed, the preparation work for the trial data output is completed.
[0030]
Next, input of normal medical care information will be described. Medical treatment is performed by a doctor or medical staff, and the medical treatment information and the progress information are input to the electronic medical record database 61. Therefore, the physician conducting the clinical trial generally performs the same work as the normal medical treatment without being conscious of preparing the clinical trial data. Focus on patient and study patient care.
[0031]
FIG. 4 is a flowchart showing the flow of clinical trial data output in one embodiment. FIG. 5 shows a corresponding item selection screen by the sequential selection unit in the first embodiment.
[0032]
The first embodiment will be described with reference to the flowchart of FIG. 4 based on the graphical user interface (hereinafter referred to as GUI) of FIG.
[0033]
As described above, a medical institution receives a request for a clinical trial such as a drug administration or a test from a drug developer, and also receives a trial protocol prepared by the drug developer and recording a trial plan. The study protocol specifies a case record format for recording the progress of medications and tests performed on study patients. Therefore, first, it is necessary to associate the item names described in the case record form with the electronic medical record items and register them in the case item correspondence table 63.
[0034]
In the initial setting 100 of the flowchart of FIG. 4, the trial item name selection unit 30 executed on the CPU 20 of FIG. 1 outputs the CD-R / XML-tagged case data output format recorded in the XML format attached to the trial report. The W media is read by the CD-R / W drive 57 or the case data output format information described in the clinical trial report is read by the scanner, and converted into text by OCR software (not shown), and the case data output format of the database 60 is read. 64.
[0035]
Of course, the case record form may be input from the LAN communication interface 51 connected to the Internet 11, or the case record form may be input from the keyboard 52. Thus, the preparation of the trial item name and the electronic medical record item name recorded in advance is completed.
[0036]
As the next step, the trial item name selection 110 in the flowchart of FIG. 4 displays the corresponding item selection screen 200 by the sequential selection unit in the first embodiment of FIG.
[0037]
The user wants to select the trial item name selection cursor 215 (the cursor is displayed in a color different from the background color to indicate that it is selected) of the trial item 210 displaying the trial item list. Go to the name and click the radio button to the left of it with the mouse to specify the trial item name. The designated trial item name is displayed in the field on the left side of the item selection display area 230, and the letters are displayed dimmed to indicate that it has been selected.
[0038]
Next, the corresponding item name selection 120 in the flowchart of FIG. 4 is executed. The electronic medical record item name selection cursor 225 in the electronic medical record item 220 displaying the electronic medical record data item list in FIG. 5 is moved to the electronic medical record item name to be selected, and the radio button to the left of the electronic medical record item name is clicked with the mouse. Specify the chart item name. By this operation, the designated electronic medical record item name is displayed in the field on the right side of the item selection display area 230, and similarly, the letters are displayed in light color to indicate that the item has been selected. In this series of operations, the association of one item name is completed.
[0039]
Next, in the item selection end determination 130 in the flowchart of FIG. 4, the end is determined. If the selection is not completed, the process returns to the trial item name selection 110 and the sequential selection is repeated. If the selection is completed, it is registered in the case item correspondence table 63 and the procedure proceeds to the medical care data search 140.
[0040]
Next, in the medical care data search 140 of the flowchart of FIG. 4, the electronic medical record database 61 is searched for the medical care information of the patient for whom the trial patient flag is set, with reference to the case item correspondence table 63 and the case-related search information. Of course, in this search, association information in which search conditions such as a patient suffering from a specific disease, gender, age, and the like may be used.
[0041]
Next, in the clinical trial data storage 150 in the flowchart of FIG. 4, the data searched by the medical care data search 140 is stored in the clinical trial database 62.
[0042]
Finally, the case data output format 64 stored in the trial database 62 is read, and if the output format needs to be changed, the format is changed, and the data stored in the trial data storage 150 is read, and the data is read out and arranged in chronological order. The output data is output as an electronic file and further output to the printer 55. This electronic file can be recorded on a CD-R / W medium, or can be transmitted by the LAN communication interface 51 connected to the Internet.
[0043]
As described above, the data is output from the clinical trial data output device 10 based on the case data output format 64, and is attached as a clinical trial data case to the clinical trial report prepared by the doctor.
[0044]
Next, confirmation of omission of normal medical information entry will be described. Indications, dosages, precautions, and other precautions for new drugs are determined based on clinical trial effects.However, if there are any errors or omissions in clinical trial results, new drug development May affect planning.
[0045]
In one embodiment, as a means for confirming whether or not there is no omission in the input trial case, from the patient's personal information after the entry of the medical information, the patient for whom the trial patient identification flag is set is set daily to once a week. In the trial data output terminal, a process of confirming whether or not there is no entry is performed.
[0046]
If any omission is found, it is possible to output a message about the omission item and prompt the user to fill in, thereby preventing omission into the electronic medical record database 61. Implement for a specified period based on the plan.
[0047]
Next, a description will be given of a second embodiment using a corresponding item selecting unit using a similar narrowing unit. FIG. 6 shows a corresponding item selection screen by the similar narrowing-down unit according to the second embodiment.
[0048]
The trial item selection module 31 of FIG. 1 is configured to add and multiply characters (such as kanji and symbols) in order to realize character logical expressions such as (Expression 1) to (Expression 3) shown below. ing.
[0049]
In the first embodiment, electronic medical record items corresponding to a large number of clinical trial item names are manually selected and determined one by one. In the present embodiment, this point is improved.
[0050]
The character logical expression calculates the characteristics of characters and kanji in clinical trial data item names and electronic medical record item names, and searches for similar item names including desired characters and kanji based on the calculation results.
[0051]
For example, when searching for an electronic medical record item name similar to the clinical trial item name “disease name”, in this embodiment, for example, as shown in (Equation 1), an electronic medical record that can factorize the kanji characters “disease” and “name” can be used. The search is performed as the item name of the chart item name, and the “current disease name” and “family medical history” that can be factorized into “disease” are extracted, and the number of detections and the search position are also stored as a weight function.
[0052]
Furthermore, the kanji “name” is searched as an item name of an electronic medical record item that can be factored, and the “current disease name” and “name” are extracted. It is extracted as “current disease name”, “family medical history” and “name” in descending order.
[0053]
(Equation 1)
Disease name = disease (current disease name + family medical history) + name (current disease name + name) (Formula 1).
[0054]
Similarly, when searching as an electronic medical record item name similar to the clinical trial item name “medical history”, in the present embodiment, as shown in (Equation 2), the kanji characters “disease” and “history” can be factored. The search is performed as the item name of the chart item name, and the “current disease name” and the “family medical history” that can be factorized into the “disease” are extracted, and the search position is also stored as a weight function.
[0055]
Furthermore, by searching for the item name of the electronic medical record item that can factor the kanji character “History”, searching for “History history” and “Family medical history”, the number of detections and the search position are similarly stored as a weight function. , And are extracted as “previous medical history”, “family medical history” and “current disease name” in descending order of the degree of confirmation.
[0056]
(Equation 2)
Medical history = disease (current disease name + family medical history) + medical history (previous medical history + family medical history) (Equation 2).
[0057]
In addition to this, in the present embodiment, it is also possible to perform a similar operation from a previously prepared array having a similar meaning and refer to the determined item name. To be more specific, from the sequence that includes the characteristic items in the meaning of “disease”, “disease”, “abnormality” (such as abnormal color vision), and “disability” (such as language disorder) corresponding to “illness,” Extracted as “disease”, “disease”, “abnormal”, and “disability”.
[0058]
[Equation 3]
f (disease name) = disease + disease + abnormality + disorder (formula 3).
[0059]
Next, a second embodiment incorporated in the clinical trial item selection module 31 that implements such an algorithm will be described below. FIG. 6 shows a corresponding item selection screen 200 according to the similar narrowing-down unit in the second embodiment.
[0060]
There are two points different from the first embodiment. The first point is that the patient's private information and unnecessary information are enclosed in a dotted line in the electronic medical record item 220 so that they cannot be designated. The second point is that the item name list of the clinical trial item 210 is described above. This is the point that similar items after similar narrowing-down processing are displayed near the trial item names.
[0061]
The electronic medical record database 61 records all the information of the patient. However, if the user can select personal information such as a name, address, insurance information, etc., personal information is leaked, which is a problem. Therefore, in the second embodiment, personal information cannot be selected as a trial item. On the screen, it is displayed surrounded by a dotted line. Then, all items including personal information are rearranged in the order of “aiueo” and displayed as the electronic medical chart item list 220.
[0062]
Next, a description will be given of a flow of processing for realizing a character logical expression such as (Expression 1) to (Expression 3) described above. In the corresponding item name selection 120 of the flowchart of FIG. 4, the trial item 210 extracted as a candidate by similarity narrowing down and the remaining electronic medical record item list 220 are displayed on the corresponding item selection screen 200 of FIG.
[0063]
For example, if the clinical trial item name is "disease name", after extracting "current disease name" and "name" including "disease" and "family medical history" including "name" by character logical expression, The duplicate name and the "name" are removed to prevent personal information leakage, and the "current disease name" and "family medical history" are displayed next to the trial item "disease name" as similar.
[0064]
In the associating operation, in the case of the trial item name “disease name”, for example, double-clicking on “current disease name” specifies the association, and the remaining family medical history is moved to the electronic medical record item 220 on the left side. Alternatively, you can click on an unrelated "family medical history" and click the left triangle to move it to the left and remove it from the trial.
[0065]
As described above, the names, addresses, insurance information, and the like of the patients unnecessary as the trial patient information surrounded by the dotted line are not included in the candidates from the beginning and are not selected, so that leakage of personal information can be prohibited. Of course, you can use this feature elsewhere.
[0066]
If there is no similar electronic medical record item name in the similarity determination based on the character logical expression, "None" is displayed, and a message window (not shown) for requesting the operator to make a determination is opened. The operator clicks the medical treatment content after selecting the “treatment” of the electronic medical record item and then clicks the right arrow to associate the medical treatment item with the medical treatment content.
[0067]
FIG. 7 shows a corresponding item selection screen 200 by the hierarchical display means in the third embodiment. The electronic medical record data includes a tabbed hierarchical display unit 221 for displaying grouped electronic medical record data and displaying item names in a hierarchical structure. The electronic medical record item list is classified according to group names described in tabs according to predetermined group correspondence information. The item names are made easier to find by displaying in a hierarchical manner.
[0068]
FIG. 8 is a list of tagged hierarchical display sections showing the levels of electronic medical record items. In the test results and the like, more detailed items are arranged in a hierarchy, and it is possible to set the test results more appropriately.
[0069]
FIG. 9 shows a corresponding item selection screen 200 based on the similarity determination of the meaning of the trial item name in the fourth embodiment. The similar tab 222 of the electronic medical record item searches the similar term table provided in advance, and when the trial item name is “disease name”, “disease” and “disease” which are similar to “disease name” are also used as identifiers. Used for search index.
[0070]
Further, by clicking "disease name" of the trial item 210, "disease" is set in the character logical expression search index, and "current disease name" including "disease", "family medical history", and "search name" in the search index. To search for "current disease name" and "name", and similarly use "disease" and "patient" as search indices to remove duplicates, "current disease name", "family medical history" and "name". Is displayed next to the trial item "Disease name" as similar.
[0071]
When extracting the electronic medical record data of the related items from the associated case item correspondence table 63, the electronic medical record processing unit includes a process of calculating a preset age group, number of genders, disease name, case progress list, drug case list, and the like. At the same time, a step of storing the data extracted under the above conditions in the clinical trial database 62 can be output as comma-separated CSV format data or XML format data so that analysis can be performed by other statistical processing software. For this reason, the processing that increases the CPU load due to the statistical processing can be appropriately performed by another CPU.
[0072]
FIG. 10 is an example of trial data output from the trial data output device. In order to list data searched based on the link information and the narrowing-down information, a case output format is designated. Thereafter, a search is executed at the trial data output terminal, and the data recorded in chronological order is output as an electronic file, which can be output as a case list by the printer 55. Of course, it may be directly output to the printer 55.
[0073]
In the present embodiment, the electronic medical record database 61, the clinical trial database 62, and the input / output terminal 50 may be configured as computer systems distributed via a network.
[0074]
【The invention's effect】
As described above, according to the present invention, a physician conducting a clinical trial can concentrate on medical treatment and treatment without being conscious of creating clinical trial data, and can create a list of case records when creating a case report. Efficiency is achieved and entry omissions and entry errors are prevented.
[Brief description of the drawings]
FIG. 1 is a schematic diagram of a clinical trial data output device.
FIG. 2 is a schematic diagram of an electronic medical record database.
FIG. 3 is a schematic diagram of a trial database.
FIG. 4 is a flowchart showing a flow of clinical trial data output in one embodiment.
FIG. 5 is a corresponding item selection screen by a sequential selection unit in the first embodiment.
FIG. 6 is a corresponding item selection screen by a similar narrowing-down unit according to the second embodiment.
FIG. 7 is a corresponding item selection screen by a hierarchical display unit according to the third embodiment.
FIG. 8 is a list of tagged hierarchical display sections showing the levels of electronic medical record items.
FIG. 9 is a corresponding item selection screen based on a similarity determination of the meaning of a trial item name in the fourth embodiment.
FIG. 10 is an example of clinical trial data output from the clinical trial data output device.
[Explanation of symbols]
10 trial data output device, 20 CPU, 30 trial item name selection unit, 40 output unit, 50 external interface unit, 60 external storage unit, 61 electronic medical record database.

Claims (6)

In a clinical trial data output device that associates clinical trial items with electronic medical record items of an electronic medical record database in which medical data of patients are recorded, and outputs clinical trial data associated with the clinical trial items,
Trial item name selecting means for the user to select a desired one from the displayed trial item names;
Corresponding item name selecting means for a user to select an electronic medical record item to be associated with the selected trial item name from the list of electronic medical record item names,
Medical data retrieval means for retrieving medical data of the subject patient from the electronic medical record database based on the electronic medical record items respectively associated with the plurality of clinical trial items,
Trial data storage means for storing the searched medical data in a storage device as trial data arranged in a given format;
And outputting the stored clinical trial data. The clinical trial data output device according to claim 1,
The clinical data output device, wherein the corresponding item name selecting means includes a similar narrowing-down display means for narrowing down and extracting an electronic medical record item name related to each clinical trial item and displaying it near the clinical trial item name. . The clinical trial data output device according to claim 1,
The trial data output device, wherein the corresponding item name selecting means includes a hierarchical display means for hierarchically displaying an electronic medical record item name related to the clinical trial item in the electronic medical record item. The clinical trial data output device according to claim 2 or 3,
The trial data output device, wherein the corresponding item name selection means displays patient personal information in a non-selectable manner. In a clinical trial data output method of associating clinical trials with clinical record items and electronic medical record items in an electronic medical record database in which patient medical data is recorded, and outputting clinical trial data associated with the clinical trial items,
A trial item name selecting step in which the user selects a desired one from the displayed trial item names;
A corresponding item name selecting step in which the user selects an electronic medical record item corresponding to the selected trial item name from the electronic medical record item names displayed in the list,
A medical data search step of searching medical data of the subject patient from the electronic medical record database based on the electronic medical record items respectively associated with the plurality of clinical trial items,
A trial data storage step of storing the searched medical data in a storage device as trial data arranged in a given format;
And outputting the stored clinical trial data. A trial data output program executed on a trial data output device for associating a clinical trial item with an electronic medical record item of an electronic medical record database in which medical information of a patient is recorded and outputting trial data associated with the trial item At
A trial item name selection process in which the user selects a desired one from the displayed trial item names;
A corresponding item name selection processing procedure in which the user selects an electronic medical record item to be associated with the selected trial item name from the electronic medical record item names displayed in the list,
A medical data search processing procedure for searching medical data of an investigation target patient from an electronic medical record database based on an electronic medical record item associated with each of a plurality of clinical trial items,
A trial data storage procedure for storing the searched medical data as trial data arranged in a given format in a storage device;
An output processing procedure for outputting the stored trial data;
A trial data output program characterized by executing the following.

Images