JP2012208669A - Medical checkup result output system and medical checkup result output program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a medical checkup result output system for more accurately determining a health condition on the basis of the history of an inspection value in comparison with a conventional configuration.SOLUTION: Disclosed is a medical checkup result output system including a database for storing the medical checkup result information of a plurality of subjects and medical checkup result output means for outputting the medical checkup result information of each of those subjects stored in the database. The database is made to store the medical checkup results of each subject over the several times in association, and the medical checkup result output means is made to calculate a regression equation from the history of the inspection value of one inspection item about the medical checkup result information of one subject, and to output an inspection value history graph in which the history of the inspection value, a future predicted inspection value calculated from the regression equation and the reference range of the inspection value are displayed on a time axis.

Description

  The present invention relates to a health check result output system and a health check result output program for outputting a result of a health check.

  As a result of the health check, the test value and the reference range (normal range) are usually written together for each test item. This reference range is a range in which test values of 95% excluding the upper 2.5% and the lower 2.5% are included in the distribution of test values of the healthy population, and doctors and subjects Whether the inspection value is good or bad can be determined based on the relationship between the reference range and the inspection value (see Non-Patent Document 1).

  In recent years, it is preferable that the health check result is determined not only by the relationship between the reference range and the test value but also by the transition of the past test value to determine the deterioration or improvement of the health condition. Since the reference range depends on the measurement method and the analysis device, it is not appropriate to directly compare the actual measurement values (actual test values) in the present situation where a large number of measurement methods and analysis devices are mixed. There has already been proposed a conversion program that can appropriately compare different test values of measurement methods and analyzers (see Patent Document 1).

JP 2009-176023 A

Kunihide Ino, how to set the standard range, Gifu QA workshop report in 1995, p. 7-13

  By the way, by using the program described in Patent Document 1, it is possible to properly compare test values measured under different conditions, but that alone sufficiently evaluates the transition of the test value of the subject. I could not. That is, the history of inspection values rarely becomes the same numerical value even under the same measurement conditions, and usually increases or decreases more or less. For this reason, even if a doctor or a subject displays a history of test values, it may be difficult to determine an overall trend if the test values are repeatedly increased or decreased. Moreover, even if an increasing trend or a decreasing trend is seen in the transition of the inspection value, it is difficult to determine whether the tendency is small enough to be an error or notable.

  The present invention has been made in view of the current situation, and an object of the present invention is to provide a medical examination result output system and a medical examination result output program that can more accurately determine the health state based on the history of test values. .

  The present invention provides a medical examination result comprising a database storing medical examination result information of a plurality of subjects, and a medical examination result output means for outputting the medical examination result information of each subject stored in the database. In the output system, the database stores a plurality of medical examination results of each subject in association with each other, and the medical examination result output unit is configured to obtain information on medical examination results of one subject. The regression equation is calculated from the history of the inspection value of one inspection item, and the history of the inspection value, the predicted future inspection value obtained from the regression equation, and the reference range of the inspection value are displayed on the time axis. It is a medical examination result output system characterized by having a function of creating a test value history graph.

  Here, the “inspection value” according to the present invention is not limited to an inspection value measured by an inspection device or the like (hereinafter referred to as an actual inspection value), but an inspection value (hereinafter referred to as a standardized value obtained by variable conversion of the actual inspection value). , Called standardized inspection value).

  In such a configuration, the doctor or the subject can easily determine the importance of the test value history of the subject by referring to the predicted test value displayed on the test value history graph. For example, even if the latest test value is within the reference range, if the predicted test value after one year is a value that is out of the reference range, the subject preferably performs some measures in advance. The subject can obtain appropriate advice early from the doctor.

  In the present invention, a configuration is proposed in which the regression equation is a linear regression equation that is obtained by reducing the weighting of an inspection value with an older inspection date. Since the reliability of the test value is higher for the new test date and lower for the old test date, the accuracy of the regression equation can be improved by weighting according to the test date as in this configuration. Thus, it is possible to obtain a more accurate predicted inspection value.

  In the present invention, a configuration is proposed in which the medical examination result output means displays a confidence interval of a regression equation on a test value history graph. In such a configuration, the reliability of the predicted test value can be confirmed from the confidence interval of the regression equation, and thereby, the predicted test value can be used more appropriately.

  Further, in the present invention, the test value displayed on the test value history graph is based on an average of the distribution of the actual test values of the healthy population based on the reference range of the test values of one subject. A configuration is proposed which is an intra-group normalized inspection value converted to a dimensionless number indicating whether or not there is a deviation. That is, the “in-group standardized test value” is a deviation value, a Z value, or a value equivalent to the test value of the healthy population as a population. By displaying such in-group standardized test values on the test value history graph, it is possible to properly compare different test values such as measurement methods and analyzers, and high reliability based on the history of test values It becomes possible to show a predicted test value. It should be noted that the method of converting the actual test value into the in-group standardized test value is preferably selected in accordance with the distribution type of the actual test value as in the case of the test value conversion in Patent Document 1.

  Further, in the present invention, for one test item, an analysis of variance is performed between test value histories of a plurality of subjects, and based on the distribution of test values of one subject estimated from the analysis of variance, A configuration is proposed that has a function of determining an individual reference range of the subject. In such a configuration, a variance analysis is performed using the test value history of a plurality of subjects, instead of estimating the test value fluctuation of one subject only from the test value history of the subject. This makes it possible to estimate the distribution of test values of one subject with higher accuracy. The “intra-individual reference range” referred to here is a range in which most (usually 95%) of test values are included in the distribution of test values of individual subjects. It is different from what is described as “reference range”. That is, the “reference range” is calculated based on the distribution of the test values of the healthy population, whereas the “intra-individual reference range” is calculated based on the distribution of the test values of the individual subject. . Such “intra-individual reference range” can eliminate variations in test values due to differences in constitution among individuals, and therefore, it is possible to set a reference specific to the individual constitution of the subject as compared to a mere “reference range”.

  Further, in the present invention, the medical examination result output means is a deviation of an actual test value of one subject from an average of the distribution of the actual test values of the healthy population based on the reference range. The in-group normalized test value converted into a dimensionless number indicating the first coordinate axis, and how much the actual test value of the one subject is deviated from the average of the test value history distribution of the one subject A configuration having a function of creating a scatter diagram in which the inspection values of the plurality of inspection items of the one subject are plotted on the coordinates having the in-person standardized inspection value converted into a dimensionless number indicating whether or not the second coordinate axis Is proposed. Here, the “individualized standardized test value” is a deviation value, a Z value, or a value based on the test value of the subject individual as a population. Such an individual standardized test value indicates the degree of bias in the test value distribution of the individual of the subject. Therefore, in such a scatter diagram, the test values over a plurality of test items are included in the healthy population and individuals. It is possible to grasp at a glance the degree of deviation of each inspection value.

  Moreover, in this invention, the structure which calculates the Mahalanobis distance about the test value of one test subject's several test item, and outputs the determination result based on this Mahalanobis distance is proposed. In such a configuration, it is possible to make a comprehensive judgment on the inspection values over a plurality of inspection items.

  Further, another aspect of the present invention is obtained from the function of calculating the regression equation from the history of the inspection value of one inspection item of one subject, the history of the inspection value on the time axis, and the regression equation. Medical examination result output for causing a computer to realize a function of creating a test value history graph displaying future predicted test values and a reference range of the test values, and a function of outputting the test value history graph It is a program. According to such a program, the medical examination result output system of the present invention can be suitably realized.

  As described above, according to the medical examination result output system and the medical examination result output program of the present invention, in addition to the test value history and the reference range, the future predicted test value is displayed on the test value history graph. Thus, the doctor and the subject can judge the health state more accurately than before. For example, even if the test value history remains within the reference range, if the predicted test value in the near future (for example, one year later) is out of the reference range, the subject should be careful to take preventive measures. It is possible to prevent deterioration of the health condition by receiving advice.

It is a conceptual diagram which shows the structure of the medical examination result output system. It is a chart which shows the contents of a subject attribute table. It is explanatory drawing which shows the content of the personal history table. It is a chart which shows the contents of a distribution type decision table. It is explanatory drawing which shows the program structure of a medical examination result output program. It is an example of a test value history graph. It is an example of a scatter diagram. It is an example of a test value list.

  Embodiments of the present invention are described according to the following examples.

  As shown in FIG. 1, the medical examination result output system 1 of this embodiment includes a database server 2 that stores and stores medical examination result information of a subject, and a health server connected to the database server 2 via a network 3. A diagnosis result output terminal 4 and a printer 5 for printing the checkup result information are provided. The database server 2 constitutes a database according to the present invention, and the medical examination result output terminal 4 and the printer 5 constitute medical examination result output means according to the present invention.

  The medical examination result information of the subject stored in the database server 2 includes a subject attribute table that stores the subject's attribute information and a personal history table created for each subject. In the subject attribute table, as shown in FIG. 2, “subject ID”, “subject name”, “birth date”, “sex” of all subjects who have undergone medical examination at the facility And other attributes are stored. On the other hand, as shown in FIG. 3, the personal history table includes a plurality of examination value history tables 6 created for each examination item of each subject. Each inspection value history table 6 stores “inspection date” and “actual inspection value” for inspecting the inspection item, and “reference value” (upper and lower limit values of the reference range) corresponding to the actual inspection value. The Furthermore, in each test value history table, “in-group standardized test value” obtained by standardizing each actually measured test value is also stored. The in-group standardized inspection value will be described later.

  The database server 2 sequentially updates the health check result information of the subject by taking in the health check result (actual test value) of each subject measured by the test / analysis apparatus in the health check facility. In addition, the database server 2 can take in the results of the medical examination that the subject has received at an external medical examination facility and incorporate it into the personal history table of the subject.

  In-group standardized test values are obtained by standardizing each actual test value based on a reference value and variable-converting it to a dimensionless number indicating how much the test value of the healthy population is deviated from the average. . Specifically, the in-group standardized test value is a variable conversion of the actual test value of the healthy population so that the average value is “0” and the reference value is “± 1.96”. . When the medical examination result information is input from inside or outside the medical examination facility, the database server 2 converts the actual test value included in the medical examination result information into an intra-group standardized test value, and the test value history table together with the actual test value. Remember. In this way, by standardizing the actually measured inspection values, it becomes easy to compare inspection values measured under different conditions and comprehensively analyze inspection values over a plurality of inspection items.

  The method of variable-converting the actual test value into the in-group standardized test value differs depending on the distribution type of the test value to be converted. Specifically, in the same way as the inspection value conversion method described in Patent Document 1, when the distribution of actually measured inspection values is a normal distribution type, a distribution type (parametric distribution type) that can be converted into a normal distribution type distribution. ), Normalization of actually measured inspection values is performed separately in the case of a distribution type (non-parametric distribution type) that cannot be converted to a normal distribution type distribution. As shown in FIG. 4, the database server 2 includes a distribution type determination table that stores the distribution types of the test values of the healthy population for each test item, and the above three variable conversion methods include the distribution type determination table. To be determined.

When the distribution of the actually measured inspection values is a normal distribution type, first, the average value XB and the standard deviation SD of the actually measured inspection values are calculated by the following formulas based on the reference range lower limit value RL and the upper limit value RH.
XB = (RL + RH) / 2
SD = | RL−RH | / (2 × 1.96)

Next, based on the calculated average value XB and standard deviation SD, the in-group normalized test value ZXi is calculated from the actual test value Xi to be converted by the following formula, and stored in the test value history table.
ZXi = (Xi−XB) / SD
The in-group standardized inspection value ZXi in this case is the Z value itself of the actual inspection value Xi.

If the distribution of the actual inspection values is the parametric distribution type, refer to the distribution type table and select the conversion formula corresponding to the distribution type, and use the conversion formula to change the following values to values that conform to the normal distribution. Convert.
Xi: Actual inspection value
RL: Lower limit of reference range of actual inspection values
RH: Upper limit of reference range of actual inspection values

Specifically, as described below, logarithmic transformation is performed for the lognormal distribution type, and power transformation is performed for the power distribution type. In the following formula, p is an exponent in the power distribution type.
For lognormal distribution: Xi, RL, RH → cXi = ln (Xi), cRL = ln (RL), cRH = ln (RH)
Power distribution: Xi, RL, RH → cXi = Xi ^p , cRL = RL ^p , cRH = RH ^p

Next, based on the variable-converted reference range lower limit value cRL and the reference range upper limit value cRH, an average value cXB and a standard deviation cSD of inspection values after variable conversion are calculated. Specifically, the average value cXB and the standard deviation cSD are obtained from the following mathematical formula.
cXB = (cRL + cRH) / 2
cSD = | cRL−cRH | / (2 × 1.96)

Next, based on the actually measured test value cXi after the variable conversion, the average value cXB, and the standard deviation cSD, the in-group normalized test value ZXi is calculated by the following formula and stored in the test value history table.
ZXi = (cXi−cXB) / cSD

When the distribution of the actual inspection values is the non-parametric distribution type, the actual inspection value Xi to be converted is converted into the in-group standardized inspection value ZXi by the following formula.
When Xi ≧ RM
ZXi = 1.9 x (Xi-RM) / (RH-RM)
When Xi <RM
ZXi = 1.9 x (Xi-RM) / (RM-RL)

Xi: Actual inspection value
RL: Lower limit of reference range of actual inspection values
RH: Upper limit of reference range of actual inspection values
RM: Median reference range of actual inspection values

Note that this variable conversion requires not only the reference values RL and RH of the actual inspection value under the inspection conditions but also the median value RM of the actual inspection value. When the median RM is not known, the median RM is estimated from the reference range lower limit oRL and median oRM of the actual test values under different test conditions for the same test item by the following equation.
RM = RL + (oRM-oRL) x (TR / oTR)

  The medical examination result output terminal 4 is installed with the medical examination result output program according to the present invention, whereby the medical examination result output terminal 4 reads the medical examination result information of the subject from the database server 2. The medical examination result information can be displayed on the monitor in various patterns, or can be printed on the printer 5 connected to the network 3.

  Specifically, the medical examination result output program, as shown in FIG. Medical examination information reading process to be read, intra-personal distribution analysis process for determining the intra-individual reference range of the subject subject, and an individual base normalization process for converting the test value of the subject subject into an in-person standardized test value And a result output process for outputting the health check result in various modes.

  In the medical examination information reading process, the medical examination result output terminal 4 accesses the database server 2 and performs the medical examination of the subject stored in the database server 2 based on the ID and name of the subject subject. The result information is read and stored in the storage device. The medical examination result information of the subject stored in the storage device is output to a monitor or the like in the result output process.

  In the individual distribution analysis process, for one test item, the test value distribution of the target subject is analyzed from the test value history of the target subject, and the average value, standard deviation, 95% confidence interval (individual) Inner reference range) is calculated and stored in the storage device. Since the intra-individual reference range can ignore individual differences compared to the reference range based on the test value distribution of the healthy population, it is possible to indicate a reference range specialized for the constitution of the subject individual.

  In the intra-individual distribution analysis process, the intra-individual reference range can be selectively calculated by two types of methods. The first method is a method that calculates only from the test value history of the subject subject, and the second method is a method that uses the test value history other than the subject subject to calculate.

  When calculating the intra-individual reference range based only on the test value history of the subject individual, in order to obtain a more accurate result, an abnormal value is removed in advance using a rejection test (such as Smirnov Grubbs rejection). . Then, after the abnormal value is removed, the average value and the standard deviation are calculated from the test value history of the subject individual, and the intra-individual reference range (95% confidence interval) is calculated therefrom. Since the distribution of in-group standardized test values can be regarded as a normal distribution, the intra-individual reference range can be easily calculated from the in-group standardized test values.

  When using the test value history other than the subject, the test value history of other subjects is read from the database server 2. And by carrying out a one-way analysis of variance among the test value histories of a plurality of subjects together with the test value history of the subject subject, fluctuations in test values between subjects, and fluctuations in test values within individuals Analyze. Then, from the analysis result, an average value and a standard deviation in the test value distribution of the subject subject are calculated, and further, an intra-individual reference range (95% confidence interval) is calculated therefrom. In such a case, the accuracy of the calculated intra-individual reference range or the like can be improved by using the test value history of a plurality of subjects.

In the personal base standardization process, the test value of the subject subject is converted into a personalized standardized test value. The in-person standardized test value is obtained by standardizing the test value of the subject by variable conversion to a dimensionless number indicating how much the test value of the subject is deviated from the average value of the test value of the subject individual. As specific processing, the in-group standardized test value ZXi is converted into a test value pZXi standardized on an individual basis based on the following formula, and stored in the storage device.
pZXi = (ZXi−pM) / pSD
Here, pM and pSD are the average (pM) and standard deviation (pSD) of the intra-individual distribution of the test values of the subject calculated by the intra-individual distribution analysis process.

  In the result output process, as shown in FIG. 5, test value history graph creation processing for creating test value history graph data, scatter diagram creation processing for creating scatter diagram data, and test value list diagram data creation Inspection value list creation processing to be performed, comprehensive determination processing for comprehensively determining the health check result, monitor output processing for outputting graphs and figures to the monitor, and printer output processing for printing the graphs and figures on the printer 5 ing.

  FIG. 6 is an example of a test value history graph output by the medical examination result output terminal 4. In such an inspection value history graph, the horizontal axis represents a time axis representing the inspection date, and the vertical axis represents the inspection value (in-group standardized inspection value) of one inspection item (for example, ALB). On the vertical axis, a reference value indicating a reference range and five levels of determination values (determination values 1 to 5) are displayed. The five-stage judgment value is set for each inspection item as an index for judging the risk of the inspection value outside the reference range. In this inspection value history graph, the inspection value history is plotted with circles. The circles indicating the inspection value history are plotted from 2001 to 2008, and the inspection values as of 2008 are those of the latest inspection date. The test value history graph displays the linear regression equation calculated from the test value history over the entire time axis. This linear regression equation is calculated by reducing the weighting of the test value that is older on the test date. Specifically, in the inspection value history, the weight of the inspection value (2001) with the oldest inspection date is set to about 1/10 of the latest inspection value (2008). The weighting is changed according to the ratio of days. Then, on the linear regression equation, the predicted test values on the test date are plotted with square marks after 2009, which is the future test date. In the test value history graph, lines indicating 95% confidence intervals of the linear regression equation are displayed above and below the linear regression equation.

  According to the test value history graph, the doctor or the subject can analyze the tendency of the test value history in more detail by following the transition of the predicted test value indicated by the linear regression equation. For example, in the test value history graph of FIG. 6, the increase / decrease tendency of the test value is not obvious only by the test value history plotted with circles, but according to the predicted test value shown on the linear regression equation, It can be clearly seen that the inspection value of the inspection item tends to increase. In addition, since the predicted test value after several years (2012) shows a dangerous numerical value exceeding the judgment value 1, doctors and subjects are not careful about the current increase trend. It can be judged that it is worthy. As described above, in such a test value history graph, even if the latest test value is not a numerical value to be noted, if the predicted test value in the near future (for example, one year later) is a numerical value to which attention is to be paid, The examiner can detect the future risk based on the predicted test value, and can take an early countermeasure before the test value is out of the reference range. Conversely, even if the latest test value is outside the standard range of the group or individual, if the predicted test value in the near future is within the standard range, the doctor or the subject may have a tendency for the test value to improve. You know, you do n’t have to worry too much about the test value.

  In particular, since such a test value history graph displays a 95% confidence interval of the linear regression equation, the doctor or the subject can also grasp the reliability of the predicted test value displayed on the graph.

  In addition, the linear regression equation displayed in the test value history graph is calculated with lighter weight for the test values that are relatively unreliable and the older the test date, so display a highly accurate linear regression equation. This has the advantage that the certainty of the predicted test value can be improved.

  In addition, because the test values and reference ranges displayed in the test value history graph use standardized test values within the group, it is possible to comprehensively handle past test values with different measurement methods and analytical instruments. There is an advantage that a predicted test value with higher reliability can be shown based on the test value.

  FIG. 7 is an example of a scatter diagram output by the medical examination result output terminal 4. The scatter diagram is formed by plotting the inspection values of a plurality of inspection items on a two-dimensional coordinate having the in-group normalized inspection value on the horizontal axis and the in-person normalized inspection value on the vertical axis. In such a scatter diagram, the inspection values of three inspection items (TB, ALB, CRE) on a specific inspection date are plotted, and the vertical and horizontal axes indicate in-group standardized test values and individual standardized tests. A 95% confidence interval for the value is specified respectively. According to such a scatter diagram, doctors and subjects can not only individually determine test values for each test item, but also comprehensive and multifaceted understanding of test values for a plurality of test items. .

  FIG. 8 is an example of a list of test values output by the medical examination result output terminal 4. The inspection value list diagram displays the inspection values of a plurality of inspection items on a specific inspection date side by side. In the inspection value list, the vertical axis indicates the in-group standardized inspection value, and the inspection value is plotted with a circle for each inspection item. In addition, the reference range (± 1.96) is indicated by a line that crosses each inspection item, and the upper limit value and the lower limit value of the subject individual reference range are indicated by a short horizontal line for each inspection item. Has been. In this list of test values, the test values for each test item are displayed in parallel as standardized test values, so doctors and subjects can determine how far the test values for multiple test items are out of the reference range. It can be grasped at a glance.

  Moreover, the lower right part of FIG. 8 is an example of the comprehensive determination which the medical examination result output terminal 4 outputs. This comprehensive determination indicates the result of comprehensive determination of inspection values over a plurality of inspection items. The determination result is made in five stages of A to E, and the determination result is closer to A as the health check result is closer to the average value of healthy persons. Specifically, in the comprehensive determination, the Mahalanobis distance is calculated with respect to the inspection values (in-group standard inspection values) of a plurality of inspection items, and the comprehensive determination is performed based on the calculated Mahalanobis distance. The larger the calculated Mahalanobis distance, the greater the deviation of the test value from the distribution of the healthy population for the plurality of test items that are subject to comprehensive determination. The Mahalanobis distance can be calculated by a conventional method. In this way, if the data is normalized by the distribution within a group or individual, calculating the Mahalanobis distance for test values across multiple test items makes it easy to obtain a comprehensive index that includes multiple test items. Can be derived.

  The medical examination result output system according to the present invention is not limited to the embodiment described above, and various modifications can be made without departing from the scope of the present invention.

  For example, the medical examination result output system of the present invention is not limited to the configuration of the embodiment, and the system configuration of a server, a terminal, or the like can be changed as appropriate. In addition, the medical examination result output system of the present invention can be integrated with inspection equipment in a medical examination facility, a medical examination service system, and the like.

  Further, the output mode of the medical examination result information is not limited to the configuration of the embodiment, and can be changed as appropriate. For example, the regression equation displayed on the test value history graph (see FIG. 6) of the embodiment is not limited to a linear regression equation, and may be a regression curve.

1 Checkup result output system 2 Database server 3 Network 4 Checkup result output terminal 5 Printer

Claims (8)

A medical examination result output system comprising a database for storing medical examination result information of a plurality of subjects and a medical examination result output means for outputting the medical examination result information of each subject stored in the database. And
The database stores a plurality of medical examination results of each subject in association with each other,
The medical examination result output means calculates a regression formula from the history of test values of one test item for the health examination result information of one subject, and on the time axis, the history of the test values and the regression formula A medical examination result output system characterized by having a function of creating a test value history graph displaying future predicted test values obtained from the above and a reference range of the test values.   The medical examination result output system according to claim 1, wherein the regression equation is a linear regression equation that is obtained by reducing the weighting of an examination value with an older examination date.   The medical examination result output system according to claim 1 or 2, wherein the medical examination result output means displays a confidence interval of a regression equation in a test value history graph.   The test values displayed on the test value history graph are dimensionless indicating how much the actual test value of one subject deviates from the average of the distribution of the actual test values of the healthy population based on the reference range. The medical examination result output system according to any one of claims 1 to 3, wherein the examination result is an in-group normalized test value converted into a number.   One test item is subjected to analysis of variance among test value histories of a plurality of subjects, and based on the distribution of test values of one subject estimated from the analysis of variance, The medical examination result output system according to any one of claims 1 to 4, further comprising a function of determining a reference range.   The medical examination result output means converts the measured test value of one subject into a dimensionless number indicating how much the measured test value of the healthy population is deviated from the average of the distribution of the measured test value of the healthy person group based on the reference range. Using the in-group standardized test value as the first coordinate axis, the actual test value of the one subject is converted into a dimensionless number indicating how much deviation from the average of the test value history distribution of the one subject. 2. A function of creating a scatter diagram in which the inspection values of the plurality of inspection items of the one subject are plotted on the coordinates having the in-person standardized inspection values as the second coordinate axis. The medical examination result output system of any one of Claim 5.   The health value according to any one of claims 1 to 6, wherein a Mahalanobis distance is calculated for test values of a plurality of test items of a subject, and a determination result based on the Mahalanobis distance is output. Diagnosis result output system. A function to calculate a regression equation from a history of test values of one test item of one subject;
On the time axis, a function of creating a test value history graph that displays a history of the test value, a future predicted test value obtained from the regression equation, and a reference range of the test value;
A health check result output program for causing a computer to realize the function of outputting the test value history graph.

Images