Method of predicting weight loss

Abstract

To provide a technique that enables prediction of weight loss of a subject through increased exercise.SOLUTION: A method of the present invention for quantitatively predicting weight loss of a subject through increased exercise comprises an acquisition step of acquiring at least one indicator selected from a group consisting of: physical indicators indicative of at least either of a physique and body composition of the subject, consisting of the abdominal circumference, systolic blood pressure, and pulse rate; and predetermined answer indicators computed from answers of the subject to a questionnaire.SELECTED DRAWING: None

Description

[Technical Field]

[0001] The present invention relates to a technique for predicting the weight loss effect of an increased amount of exercise in a subject. [Background technology]

[0002] Methods to enhance weight loss are being sought for the purposes of health promotion, obesity relief, beauty, etc. For example, while it is known that increased exercise and dietary management can contribute to weight loss, it is also known that the effects of these measures vary from person to person. Therefore, factors involved in the weight loss effects of increased exercise and dietary management are being investigated.

[0003] For example, Non-Patent Document 1 describes the results of an investigation into the effects of body size, blood biochemistry tests, dietary intake, and physical activity on the amount of weight loss before and after dietary improvement guidance (weight loss intervention). The document suggests that BMI and physical activity level before the weight loss intervention are related to the amount of weight loss.

[0004] Furthermore, for example, Non-Patent Document 2 describes a study in which plasma leptin levels, body weight, body composition, etc. were measured in overweight and obese subjects before and after a weight loss program involving dietary management, aerobic exercise, and strength training. This document suggests that plasma leptin levels before weight loss intervention can be used as a predictor of weight loss results. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] The effects of body size, dietary intake, and physical activity before weight loss intervention on weight loss, Obesity Research, Vol. 13, No. 2, Pages 154-163 (August 25, 2007) [Non-patent document 2] Leptin, superoxide dismutase, and weight loss: Initial leptin predicts weight loss, Obesity, 2006 Vol. 14 (12) p2184-2192 Summary of the Invention [Problem to be solved by the invention]

[0006] The above-mentioned literature discloses factors related to weight loss in a subject due to a combination of dietary management and increased physical activity, but does not disclose specific predictive markers or methods for predicting the weight loss effect in a subject due to increased physical activity alone.

[0007] The present invention relates to a technique for predicting the weight loss effect of an increased amount of exercise in a subject. [Means for solving the problem]

[0008] A method for predicting a weight loss effect according to one embodiment of the present invention includes at least one step selected from the group consisting of the following steps (A), (B), and (C): The step (A) is a step of measuring at least one physical index selected from the group consisting of MQP (Muscle Quality Point), total body water percentage, body weight, trunk lean mass, and trunk muscle mass, which represents at least one of the physique or body composition of the subject. The step (B) is a step of measuring at least one blood indicator selected from the group consisting of red blood cell count, hemoglobin level, hematocrit level, platelet count, free fatty acid concentration, serine concentration, glutamine concentration, citrulline concentration, creatine kinase (CK) level, 3-hydroxybutyric acid concentration, free triiodothyronine concentration, free thyroxine concentration, and magnesium concentration from a blood sample of the subject. Step (C) is a step of calculating at least one response index calculated from the subject's responses to the questionnaire, selected from the group consisting of an index of mental fatigue calculated from the responses to the Chalder Fatigue Scale, an index calculated from the responses to questions asking about awareness and behavior regarding lack of exercise, an index of agreeableness personality calculated from the responses to the Big Five Scale, an index of openness personality calculated from the responses to the Big Five Scale, and an index of conscientiousness personality calculated from the responses to the Big Five Scale.

[0009] A method for quantitatively predicting a weight loss effect of an increased amount of exercise in a subject according to another aspect of the present invention includes: Waist circumference, systolic blood pressure, and pulse rate as physical indicators representing at least one of the subject's physique or body composition; As response indices calculated from the subject's responses to the questionnaire, an index calculated from the response to the question item "I felt calm and relaxed" in the WHO-5 (Mental Health Status Questionnaire), an index of mental fatigue calculated from the responses to the Chalder Fatigue Scale, an index calculated from the responses to a questionnaire evaluating the stage of change of rest behavior, an index calculated from the responses to a question asking about motivation to increase the amount of exercise, an index calculated from the responses to a question asking about awareness and / or behavior regarding irregular mealtimes, an index calculated from the responses to a question asking about awareness and behavior regarding lack of exercise, an index about proxy eating calculated from the responses to the eating behavior questionnaire, an index about dietary content calculated from the responses to the eating behavior questionnaire, and an index of conscientiousness personality calculated from the responses to the Big Five scale. The method includes an acquisition step of acquiring at least one index selected from the group consisting of: [Effects of the Invention]

[0010] According to the present invention, a technique for predicting the weight loss effect of an increased amount of exercise in a subject can be provided. [Brief explanation of the drawings]

[0011] [Figure 1]FIG. 1 shows the Chalder Fatigue Scale questionnaire used in the examples of the present invention. [Figure 2] FIG. 1 shows a dietary and lifestyle questionnaire relating to obesity (35-question dietary questionnaire) used in an example of the present invention. [Figure 3] FIG. 1 is a diagram showing the first half of a questionnaire for the Big Five scale used in an embodiment of the present invention. [Figure 4] FIG. 1 is a diagram showing the second half of the Big Five scale questionnaire used in an embodiment of the present invention. [Figure 5] FIG. 1 is a diagram showing questions included in an eating behavior questionnaire used in an example of the present invention. [Figure 6] FIG. 1 is a diagram showing the eating behavior self-efficacy scale used in an example of the present invention. [Figure 7] FIG. 1 is a diagram showing the rest behavior self-efficacy scale and the exercise behavior self-efficacy scale used in an example of the present invention. [Figure 8] FIG. 1 is a diagram showing the WHO-5 (Mental Health Status Questionnaire) used in an example of the present invention. [Figure 9] FIG. 1 shows a questionnaire for assessing the stages of health behavior change used in an example of the present invention. [Figure 10] FIG. 1 is a diagram showing a simple occupational stress questionnaire used in an example of the present invention. [Figure 11] FIG. 1 is a diagram showing a simple occupational stress questionnaire used in an example of the present invention. [Figure 12] FIG. 1 is a diagram showing the Athens Insomnia Scale used in an example of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0012] The present invention is described in detail below. All patents, non-patent documents, and other publications cited herein are hereby incorporated by reference in their entirety.

[0013] [Summary of the Invention] The present invention relates to a method for predicting the weight loss effect of an increased amount of exercise in a subject. The present invention can predict and / or assist in the prediction of the weight loss effect of an increased amount of exercise in a subject. An advantage of the present invention is that it makes it possible to predict the weight loss effect of an increased amount of exercise alone, thereby providing information for determining whether an increased amount of exercise is an effective means of weight loss for each subject.

[0014] In the present invention, the term "subject" refers to a person who is the subject of the weight loss effect prediction according to the present invention. The gender of the subject is not particularly limited. The age group of the subject is also not particularly limited, but from the viewpoint of improving the accuracy of prediction, the subject is preferably an adult, preferably 18 to 80 years old, more preferably 20 to 60 years old.

[0015] The obesity level of the subject is not particularly limited, but preferably the subject is, for example, of normal weight and / or of high obesity. A normal weight subject is, for example, a subject with a Body Mass Index (BMI) of 18.50 or more and less than 25.00. A high obesity subject is, for example, a subject with a BMI of 25.00 or more. An example of the subject of the present invention is a subject with a BMI of 23.00 or more. In addition to the above-mentioned BMI, "obesity level" can be evaluated using, for example, visceral fat mass, subcutaneous fat mass, body fat mass, body fat percentage, waist circumference, abdominal circumference, weight, body mass index (Rohrer index, Kaup index, etc.), obesity index calculated from the body mass index, subcutaneous fat thickness in the triceps, subcutaneous fat thickness in the subscapularis, basal metabolic rate, blood markers correlated with visceral fat mass (cytokines such as adiponectin, leptin, TNFα, and IL-6, inflammatory markers such as CRP, blood lipids such as LDL-cholesterol and triglycerides, liver function markers, blood glucose levels, etc.).

[0016] In the present invention, the term "fat mass" such as "visceral fat mass" refers to a numerical value that quantitatively represents the amount of fat in the body of a subject, and includes, for example, the mass of fat (at a specific site or the whole body), the volume of fat (at a specific site or the whole body), the cross-sectional area of ​​fat at a specific site, etc. For example, "visceral fat mass" is an expression that includes the mass of visceral fat, the visceral fat area which is the cross-sectional area of ​​the abdomen, and the volume of visceral fat.

[0017] In the present invention, the term "amount of exercise" refers to the amount of exercise determined based on the intensity, duration, and frequency of exercise. "Exercise" refers to physical activity that consumes more energy than a resting state, and is intentionally and continuously performed for the purpose of weight loss. In the present invention, "increasing the amount of exercise" refers to increasing the amount of exercise so as to increase energy expenditure. Specifically, an increase in a subject's amount of exercise can be achieved by, in addition to the subject's current physical activity, for example, increasing the intensity, duration, and frequency of specific types of exercise, such as sports, training, and aerobic exercise, or physical activity such as work or housework. The target amount of exercise can be set appropriately depending on the subject's current amount of exercise, age, health condition, etc., and can be set by referring to the WHO's "Optimal Physical Activity Guidelines for Health Promotion." Examples of increased exercise include exercise that increases energy expenditure to achieve a target weight (a 1-10% reduction from current weight, preferably a 1-5% reduction, and more preferably a 3% reduction), as well as moderate-intensity (3.0-6.0 METs) physical activity (e.g., brisk walking, dancing, mowing the lawn) for a set period of time per week (e.g., at least 150-300 minutes), high-intensity (6.0 METs or higher) physical activity (e.g., running, swimming, climbing stairs) for a set period of time per week (e.g., at least 75-150 minutes), and prescribed strength training (e.g., 15 minutes of multiple strength training exercises such as squats) a set number of times per week (e.g., three times).

[0018] In the present invention, "increasing the amount of exercise" preferably refers to a continuous increase in the amount of exercise. A continuous increase in the amount of exercise refers to maintaining an increased state of exercise for a predetermined period of time. The predetermined period is preferably one week or more, more preferably ten days or more, and even more preferably one month or more. Furthermore, during the predetermined period, the frequency of consciously increasing the amount of exercise is preferably three days or more per week, and more preferably every day.

[0019] In the present invention, the "weight loss effect due to an increase in exercise volume" refers to an effect evaluated by weight loss (body weight loss), changes in physique associated with weight loss, or changes in body composition associated with weight loss before and after an increase in exercise volume. In the present invention, indicators for evaluating the weight loss effect (hereinafter referred to as "weight loss evaluation indicators") include, for example, body weight, visceral fat mass, subcutaneous fat mass, body fat mass, body fat percentage, waist circumference, abdominal circumference (trunk muscle mass, etc.), muscle mass, BMI, body mass index (Rohrer index, Kaup index, etc.), obesity index calculated from the body mass index, triceps subcutaneous fat thickness, and subscapular subcutaneous fat thickness. For example, the "weight loss effect due to an increase in exercise volume" can be evaluated by the change in the weight loss evaluation indicator before and after an increase in exercise volume. In this specification, the "change in the weight loss evaluation indicator" refers to the difference between before and after an increase in exercise volume, and the "change rate of the weight loss evaluation indicator" refers to the ratio of the difference to the value before the increase in exercise volume.

[0020] In the present invention, "prediction of weight loss effect" includes at least one of prediction of overall weight loss effect or prediction of changes in a specific weight loss evaluation index. Prediction of overall weight loss effect refers to a qualitative or quantitative prediction of weight loss effect without referring to a specific weight loss evaluation index. Qualitative prediction of overall weight loss effect is expressed, for example, by expressions such as "high / low weight loss effect" or "easy / difficult to lose weight." Note that qualitative weight loss effect is not limited to an example expressed in two stages, but may be expressed in three or more stages. When weight loss effect is expressed in three or more stages, it may be expressed by letters, words, symbols, etc., such as numbers (1, 2, 3, etc.) or alphabets (A, B, C, etc.). Quantitative prediction of overall weight loss effect includes an example of expressing weight loss effect numerically.

[0021] The prediction of the change in the specific weight loss evaluation index may be a qualitative and / or quantitative prediction of the change in the specific weight loss evaluation index. The specific weight loss evaluation index may be one or more. As described below, the weight loss evaluation index used in the present invention preferably includes at least one of body weight and visceral fat mass.

[0022] Qualitative predictions of changes in weight loss evaluation indexes include, for example, predictions of how easily the weight loss evaluation index will change (e.g., "it is easy / difficult to lose weight"), predictions of the magnitude of changes in the weight loss evaluation index (e.g., "the amount of weight loss is large / small"), etc. Note that changes in qualitative weight loss evaluation indexes are not limited to examples expressed in two stages, but may be expressed in three or more stages. Quantitative predictions of changes in weight loss evaluation indexes include, for example, predictions of the amount of change, rate of change, and ranges thereof of the weight loss evaluation index.

[0023] In the present invention, the prediction of weight loss effect preferably includes at least one of a prediction of body weight loss or a prediction of visceral fat mass loss. In one embodiment of the present invention, the prediction of body weight loss may be a qualitative or quantitative prediction of weight loss. Qualitative predictions of weight loss include predictions of the ease of weight loss (e.g., "easy / difficult to lose weight") and the magnitude of weight loss (e.g., "large / small amount of weight loss"). Quantitative predictions of weight loss include at least one prediction selected from the amount of weight loss, the rate of weight loss, and a range thereof. Similarly, the prediction of visceral fat mass loss is preferably a qualitative or quantitative prediction of visceral fat mass loss. Qualitative predictions of visceral fat mass loss include predictions of the ease of visceral fat mass loss (e.g., "easy / difficult to lose weight") and the magnitude of visceral fat mass loss (e.g., "large / small amount of visceral fat mass loss"). The quantitative prediction of the reduction in visceral fat mass includes at least one prediction selected from the amount of visceral fat reduction, the reduction rate, and the range thereof.

[0024] In this specification, the "amount of reduction" of body weight and visceral fat refers to the difference obtained by subtracting the value after the increase from the value before the increase in exercise volume, and the "rate of reduction" refers to the ratio of the above difference to the value before the increase in exercise volume.

[0025] In the present invention, the "prediction" of a weight loss effect can also be expressed by terms such as "determination," "detection," "measurement," "evaluation," etc. Note that in the present invention, the terms "prediction," "determination," "detection," "measurement," and "evaluation" of a weight loss effect do not include a doctor's diagnosis of a disease such as obesity.

[0026] [Methods for predicting weight loss effects] A method for predicting a weight loss effect according to one embodiment of the present invention includes at least one step selected from the group consisting of steps (A), (B), and (C) below. The method of the present invention may include one step selected from steps (A), (B), and (C), or may include multiple steps selected from these. When the method includes multiple steps, the order in which the steps are performed is not particularly limited.

[0027] The indices listed in steps (A), (B), and (C) of the present invention are collectively referred to as "predictive markers" for predicting weight loss effects. Predictive markers are indices used to predict weight loss effects, and specifically include at least one indice selected from the group consisting of physical indices, blood indices, and response indices described below. As shown in the Examples described below, the predictive markers of the present invention are indices selected based on test results from physical examinations, blood tests, and questionnaire tests conducted before and after intervention for a group of subjects who underwent an intervention to increase their exercise volume. More specifically, the predictive markers are indices that showed significant differences between the responder and non-responder groups selected from the group of subjects in Test Example 1 described below based on the presence or absence of a weight loss effect. The gender of the subjects to be predicted by the predictive markers obtained by steps (A), (B), and (C) is not particularly limited, but males are preferred from the viewpoint of improving prediction accuracy.

[0028] Step (A) is a step of measuring at least one physical index representing at least one of the subject's physique or body composition, selected from the group consisting of muscle quality point (MQP), total body water percentage, body weight, trunk lean mass, and trunk muscle mass. Of these physical indexes, total body water percentage shows a positive correlation with weight loss effects, while MQP, body weight, trunk lean mass, and trunk muscle mass show a negative correlation with weight loss effects.

[0029] In the present invention, the physical indices may be measurements obtained by general physical measurements, or may be measurements obtained by analyzing image information of the subject's body. Measurement of each physical indices is performed by a person performing the measurement in order to predict the weight loss effect. In this case, all or part of the measurement may be outsourced to a testing institution, research institution, medical institution, etc. When multiple physical indices are measured, the timing of these measurements may be different.

[0030] Among the physical indices, body weight is the mass of the entire body (e.g., unit: kg) and can be measured using a general weighing scale, for example. Body water percentage refers to the ratio of the mass of water in the body to body weight (unit: %). Trunk lean mass refers to the mass of the entire trunk (thorax, abdomen, pelvis, and back) minus the mass of fat in the trunk, and is expressed, for example, as the mass of that fat (unit: kg). Trunk muscle mass refers to the amount of muscle contained in the trunk, and is expressed, for example, as the mass of that muscle (unit: kg).

[0031] Body composition, such as total body water percentage, trunk lean mass, and trunk muscle mass, can be measured by density methods such as underwater weighing, dual energy X-ray absorptiometry (DEXA), bioelectrical impedance analysis, ultrasound, skinfold thickness (caliper method), etc. Among these, the bioelectrical impedance analysis is preferred because it can be easily measured using a body composition analyzer.

[0032] MQP is an index of muscle quality obtained by applying alternating current at two frequencies, high frequency (e.g., 250 kHz) and low frequency (e.g., 5 kHz), to a subject's body using bioelectrical impedance analysis. Specifically, MQP can be an index that reflects the ratio of muscle fiber to interstitial tissue in muscle tissue. MQP can be calculated using the "Muscle Quality Score (registered trademark)" (1 to 100) measured using a body composition analyzer manufactured by Tanita Corporation (e.g., the "Innerscan Dual" series, the "Professional Multi-Frequency Body Composition Analyzer MC-980A-N plus," etc.). MQP can be calculated, for example, as the "muscle quality score" described in JP 2015-29832 A. The muscle quality score according to this document is calculated by acquiring at least one of a first parameter expressed as the ratio between the resistance component and the reactance component of bioelectrical impedance, and a second parameter expressed as the ratio between the first impedance measured by supplying a low-frequency alternating current and a second impedance measured by supplying a predetermined high-frequency alternating current to a living organism, and a physique parameter, multiplying the statistically calculated contribution rate of each parameter to muscle quality by the corresponding parameter, and adding up all the items (see paragraphs

[0075] to

[0086] of the above document, etc.).

[0033] Step (B) involves measuring at least one blood indicator selected from the group consisting of red blood cell count, hemoglobin level, hematocrit level, platelet count, free fatty acid concentration, serine concentration, glutamine concentration, citrulline concentration, creatine kinase (CK) level, 3-hydroxybutyric acid concentration, free triiodothyronine concentration, free thyroxine concentration, and magnesium concentration from a blood sample from the subject. Among these blood indicators, serine concentration, glutamine concentration, citrulline concentration, and magnesium concentration show a positive correlation with weight loss effects. Red blood cell count, hemoglobin level, hematocrit level, platelet count, free fatty acid concentration, CK level, 3-hydroxybutyric acid concentration, free triiodothyronine concentration, and free thyroxine concentration show a negative correlation with weight loss effects. Hereinafter, free triiodothyronine will also be referred to as "FT3" and free thyroxine will also be referred to as "FT4."

[0034] In the present invention, the blood sample for measuring blood indicators is selected from whole blood, serum, and plasma, and an appropriate sample can be used depending on the indicator. Each blood indicator is measured by a person who measures it to predict the weight loss effect, but in this case, all or part of the measurement process may be outsourced to a testing institution, research institution, medical institution, etc. Note that when multiple blood indicators are measured, the timing of these measurements may be different.

[0035] Red blood cell count (e.g., units: × 10 4 / μL), hemoglobin level (e.g., unit: g / dL), hematocrit level (e.g., unit: %), platelet count (e.g., unit: × 10 4 / μL) is an index related to blood cell components in blood and can be measured, for example, by a blood cell count test using the subject's whole blood.

[0036] Free fatty acid concentration (e.g., unit: mEq / L) is a value measured in lipid-related tests and can be measured, for example, using serum samples using established methods such as enzymatic methods. Serine concentration (e.g., unit: nmol / mL), glutamine concentration (e.g., unit: nmol / mL), and citrulline concentration (e.g., unit: nmol / mL) are values ​​measured in amino acid analysis of plasma samples and can be measured, for example, by HPLC. CK level (e.g., unit: U / L) can be measured using serum samples using established methods such as the JSCC standardized method. 3-Hydroxybutyric acid concentration (e.g., unit: nmol / L) is a ketone body and can be measured, for example, using serum samples using enzymatic methods. FT3 concentration (e.g., unit: pg / mL) and FT4 concentration (e.g., unit: ng / dL) are values ​​measured in thyroid function tests and can be measured, for example, using serum samples using established methods such as ECLIA. The magnesium concentration (e.g., unit: mg / dL) can be measured by an established method such as the xylidyl blue method using a serum sample. Note that the units of each indicator described above are examples and are not limiting.

[0037] Step (C) is a step of calculating at least one response index calculated from the subject's responses to the questionnaire, selected from the group consisting of an index of mental fatigue calculated from responses to the Chalder Fatigue Scale, an index calculated from responses to questions asking about perceptions and behaviors regarding lack of exercise, an index of agreeableness personality calculated from responses to the Big Five Scale, an index of openness personality calculated from responses to the Big Five Scale, and an index of conscientiousness personality calculated from responses to the Big Five Scale. Among these response indexes, the index of mental fatigue shows a positive correlation with weight loss effects. The index calculated from responses to questions asking about perceptions and behaviors regarding lack of exercise, the index of agreeableness personality, the index of openness personality, and the index of conscientiousness personality show a negative correlation with weight loss effects.

[0038] In the present invention, the questionnaire for obtaining the response indicators may be provided in paper form or electronic form. The response format is not particularly limited, and examples thereof include handwritten responses, responses input using a computer input device, and oral responses in interviews.

[0039] The calculation of the response index may be performed using a calculation device such as a calculator or computer as a calculation means, if necessary. The calculation of each response index is performed by a person who performs the calculation in order to predict the weight loss effect, but in this case, all or part of the calculation process may be outsourced to a testing institution, research institution, medical institution, etc. When multiple response indexes are calculated, the calculations may be performed at different times.

[0040] The Chalder Fatigue Scale is a measure of mental fatigue and is a questionnaire used in the treatment of conditions such as chronic fatigue syndrome. This questionnaire includes, for example, the 14 questions shown in Figure 1. The index of mental fatigue calculated from the responses to the Chalder Fatigue Scale can be one of the following: a total score (0-18) calculated from the subject's answers to six of the 14 questions (questions 9-14) that inquire about mental fatigue; an average score; or a score percentage, which is the ratio of the total score to the highest score. For example, the index can be the total score.

[0041] The questions inquiring about the subject's awareness and behavior regarding lack of exercise include questions about the subject's awareness of lack of exercise, their preference for exercise, and behaviors and preferences related to lack of exercise. The questions inquiring about awareness and behavior regarding lack of exercise preferably include at least one question selected from the group consisting of a question about preference for exercise, a question about the frequency of exercise, and a question about the subject's awareness of lack of exercise. Specifically, such questions include, for example, question 24 "I would use an elevator or escalator if one was available," question 26 "I don't like walking or cycling," question 27 "I don't exercise much," and question 29 "I think it's more that I'm not exercising than that I'm eating too much" from the dietary and lifestyle questionnaire related to obesity (hereinafter also referred to as the "35-question dietary questionnaire") illustrated in FIG. 2. The index calculated from the responses to the questions inquiring about awareness and behavior regarding lack of exercise can be, for example, one selected from the average score, total score, and score percentage, which is the ratio of the total score to the highest score, corresponding to the answers to the above four questions, and can be, for example, the average score. The questionnaire shown in Figure 2 was disclosed in "Objective Creation of a Dietary Questionnaire and the Relationship Between Evaluation in a Workplace Survey and Visceral Fat Accumulation" (Yoshiko Yanagisawa et al., Ningen Dock 32:611-617, 2017) (hereinafter referred to as "Yanagisawa et al., 2017"). Specifically, the questionnaire was created by comprehensively collecting a wide range of information on so-called dieting, creating questions, conducting a nationwide online survey, and using statistical methods to create the questionnaire from the response data. Furthermore, the four questions above correspond to the fifth factor (exercise) out of five factors (diet, health awareness, mealtimes, dietary restrictions, and exercise) derived from factor analysis (maximum likelihood method, oblique rotation) of all 35 questions, and can be used as a measure of physical inactivity.

[0042] The Big Five Scale (BFS) is a questionnaire for calculating five indicators of personality traits based on the Big Five personality traits: openness, conscientiousness, extraversion, agreeableness, and emotional stability. The Big Five personality traits are a method for classifying an individual's personality traits based on five factors: openness, conscientiousness, extraversion, agreeableness, and emotional stability. The Big Five Scale can be implemented, for example, as a questionnaire including 60 questions, as shown in Figures 3 and 4. In this case, the subject selects the degree to which each question on the Big Five Scale applies to their perceived personality traits from a seven-point scale (1: not applicable at all, 2: not applicable at all, 3: not applicable very much, 4: neither applicable nor unapplicable, 5: somewhat applicable, 6: very applicable, and 7: very applicable). As shown in "Subjective Fulfillment and the Big Five" (Kazuya Horike, Journal of Contemporary Behavioral Science, Vol. 15, pp. 1-8 (1999)) (hereafter referred to as "Horike, 1999"), the agreeableness, openness, and conscientiousness personality indices can be calculated based on the scores of the three items with the highest loadings on each factor extracted by factor analysis (items marked with a solid line in Table 4 of the same reference). Referring to the same reference, the three items used to calculate the agreeableness personality index can be "kind," "gentle," and "generous." The three items used to calculate the openness personality index can be "quick-witted," "insightful," and "quick learner." The three items used to calculate the conscientiousness personality index can be "careless," "responsible," and "thoughtless." As a specific method for calculating each index, the scale scores can be reversed as necessary so that higher scores on the three items related to each index indicate stronger agreeableness, openness, or conscientiousness, and then the total score can be calculated, and the average score divided by the number of items can be calculated as each index.

[0043] Furthermore, the method according to one embodiment of the present invention may further include step (D) after steps (A), (B), and / or (C). Step (D) is a step of determining whether at least one index selected from a physical index, a blood index, and a response index satisfies a condition corresponding to the level of weight loss effect set for that index. In step (D), the weight loss effect may be predicted from one predictive marker, or the weight loss effect may be predicted from multiple predictive markers. Note that the determination of whether the condition in this step is satisfied may be performed using a computer or the like, as necessary.

[0044] In step (D), the "weight loss effect level" is a level that qualitatively or quantitatively represents the weight loss effect, and is expressed, for example, as an indication (words, numbers, letters, symbols, etc.) that indicates the degree of the weight loss effect in stages, a numerical value that indicates the degree of the weight loss effect, etc. Specific examples of weight loss effect levels include an indication that qualitatively represents the overall weight loss effect (e.g., "high / low weight loss effect," "easy / difficult to lose weight," etc.), a numerical value that quantitatively represents the overall weight loss effect, an indication that indicates the degree of change in a specific weight loss evaluation index (e.g., "large / small amount of weight loss"), a numerical value that indicates the amount of change in a specific weight loss evaluation index, etc.

[0045] Examples of "conditions corresponding to the level of weight loss effect" include the range of values ​​that can be taken by each index set corresponding to the level of weight loss effect, or the numerical value and range of a value calculated by substituting the numerical value of an index into a predetermined mathematical formula. The predetermined mathematical formula may be a mathematical formula that can calculate a numerical value indicating the level of weight loss effect from the numerical values ​​of one or more indexes.

[0046] When the condition is a range of values ​​that each index can take, step (D) may include comparing at least one index selected from the physical index, blood index, and response index with a reference value of the index set corresponding to the level of weight loss effect. The reference value in step (D) is a threshold value set for each index (predictive marker) for determining the level of weight loss effect. The reference value can be set, for example, by classifying a group of subjects who have undergone dietary management intervention according to the level of weight loss effect, and using statistical values ​​(such as mean, median, quartile, 95% confidence interval) of the predictive marker calculated for each group.

[0047] Specifically, for an index in which a predictive marker has a positive correlation with weight loss effect, if the value of the predictive marker is greater than or equal to the reference value, it can be determined that the level of weight loss effect is relatively high. On the other hand, for an index in which a predictive marker has a negative correlation with weight loss effect, if the value of the predictive marker is less than or equal to the reference value, it can be determined that the level of weight loss effect is relatively high.

[0048] For example, if the weight loss effect is expressed in two levels, the level of the weight loss effect can be determined by comparing the value of the predictive marker with one reference value. If the weight loss effect is expressed in three or more levels, the level of the weight loss effect can be determined by comparing the value of the predictive marker with multiple reference values.

[0049] In the above example, when a determination is made for multiple predictive markers, the weight loss effect level is determined for each indicator, and the overall weight loss effect level can be determined based on these multiple prediction results. As a method for determining the overall weight loss effect level, for example, the weight loss effect level predicted for the most indicators may be used as the predicted weight loss effect result. As another example, the multiple weight loss effect levels determined for multiple predictive markers may each be converted into numerical values ​​to calculate a representative value (such as the average, median, or mode), and the weight loss effect level corresponding to this representative value may be used as the predicted overall weight loss effect result.

[0050] As another example of predicting the weight loss effect, a computer may be used to predict the weight loss effect using a prediction model for predicting the weight loss effect from the predictive markers. In this case, the weight loss effect may be either a qualitative weight loss effect or a predicted value indicating the weight loss effect. Prediction using a prediction model is suitable for predicting the predicted value of a weight loss evaluation index or for predicting the weight loss effect from multiple predictive markers.

[0051] Hereinafter, specific examples of steps (A), (B), and (C) when the weight loss effect prediction is a prediction of a weight loss or a prediction of a visceral fat mass will be described. Note that the present invention is not limited to the following embodiments.

[0052] [Example of a method for predicting weight loss] (Embodiment 1) In the method according to embodiment 1 of the present invention, the prediction of the weight loss effect is a prediction of body weight loss, and includes at least one step selected from the group consisting of steps (A1) and (B1) below. The predictive marker in this embodiment is an index that showed a significant difference between a responder group and a non-responder group, selected based on the rate of body weight loss from a group of subjects who underwent a training program to increase the amount of exercise in Test Example 1 of the Examples described below.

[0053] Step (A1) is a step of measuring at least one physical index selected from the group consisting of MQP and total body water percentage. Of these physical indexes, total body water percentage shows a positive correlation with the amount of weight loss, and MQP shows a negative correlation with the amount of weight loss.

[0054] Step (B1) is a step of measuring at least one blood index selected from the group consisting of red blood cell count, hemoglobin level, hematocrit level, platelet count, free fatty acid concentration, serine concentration, glutamine concentration, citrulline concentration, creatine kinase (CK) level, 3-hydroxybutyric acid concentration, and free thyroxine (FT4) concentration. Among these blood indexes, serine concentration, glutamine concentration, and citrulline concentration show a positive correlation with the amount of weight loss. Red blood cell count, hemoglobin level, hematocrit level, platelet count, free fatty acid concentration, CK level, 3-hydroxybutyric acid concentration, and FT4 concentration show a negative correlation with the amount of weight loss.

[0055] (Embodiment 1-1) As a specific example of the above-mentioned embodiment 1, the method according to embodiment 1-1 preferably includes at least one step selected from the group consisting of the following steps (A11) and (B11). The predictive marker in this embodiment is an index that showed a significant difference between a responder group and a non-responder group selected based on the weight loss rate from all subjects who underwent a training program to increase the amount of exercise in Test Example 1 of the Examples described below. The predictive marker according to this embodiment can be used to predict weight loss due to increased exercise for a variety of subjects. Note that the prediction of weight loss in this embodiment is preferably a qualitative prediction of weight loss.

[0056] Step (A11) is a step of measuring at least one physical index selected from the group consisting of MQP and total body water percentage.

[0057] Step (B11) is a step of measuring at least one blood index selected from the group consisting of red blood cell count, hemoglobin level, hematocrit level, platelet count, free fatty acid concentration, serine concentration, glutamine concentration, and citrulline concentration.

[0058] In this embodiment, the above-mentioned predictive markers preferably include at least one selected from MQP, total body water percentage, red blood cell count, hemoglobin value, hematocrit value, platelet count, and free fatty acid concentration, and more preferably include at least one selected from red blood cell count, hemoglobin value, hematocrit value, platelet count, and free fatty acid concentration.

[0059] When qualitatively predicting weight loss using each predictive marker in this embodiment, examples of reference values ​​for each predictive marker for predicting a responder, i.e., "large weight loss," are shown below. Note that each reference value is the numerical value of each predictive marker measured by the method shown in the Examples.

[0060] The MQP is preferably 64.0 or less, more preferably 60.0 or less, and even more preferably 56.0 or less. The body water percentage (%) is preferably 48.7% or more, more preferably 51.1% or more, and even more preferably 53.5% or more.

[0061] The red blood cell count is preferably 515.7 x 10 4 / μL or less, preferably 503.2 × 10 4 / μL or less, more preferably 490.6 × 10 4 / μL or less. The hemoglobin level is preferably 15.7 g / dL or less, more preferably 15.3 g / dL or less, and even more preferably 14.9 g / dL or less. The hematocrit level is preferably 47.7% or less, more preferably 46.5% or less, and even more preferably 45.3% or less. The platelet count is preferably 29.2 × 10 4 / μL or less, preferably 26.6 × 10 4 / μL or less, more preferably 24.1 × 10 4 The free fatty acid concentration is preferably 0.60 mEq / L or less, preferably 0.51 mEq / L or less, and more preferably 0.42 mEq / L or less.

[0062] The serine concentration is preferably 110.2 nmol / mL or more, preferably 117.5 nmol / mL or more, more preferably 124.7 nmol / mL or more. The glutamine concentration is preferably 559.3 nmol / mL or more, more preferably 591.8 nmol / mL or more, more preferably 624.2 nmol / mL or more. The citrulline concentration is preferably 30.8 nmol / mL or more, more preferably 34.1 nmol / mL or more, more preferably 37.4 nmol / mL or more.

[0063] (Embodiment 1-2) As another specific example of the above-mentioned embodiment 1, the method according to embodiment 1-2 preferably includes the following step (B12). The predictive marker in this embodiment is an index that shows a significant difference between a responder group and a non-responder group selected based on the weight loss rate from a group of subjects who completed a training program to increase the amount of exercise in the examples described below, and who were recognized as having complied with the training in the program. The predictive marker according to this embodiment can be used, for example, to predict weight loss due to increased exercise for subjects who are able to comply with a set program to increase the amount of exercise. Note that the prediction of weight loss in this embodiment is preferably a qualitative prediction of weight loss.

[0064] Step (B12) is a step of measuring at least one blood indicator selected from the group consisting of red blood cell count, platelet count, free fatty acid concentration, glutamine concentration, citrulline concentration, creatine kinase (CK) level, 3-hydroxybutyric acid concentration, and free thyroxine (FT4) concentration.

[0065] In this embodiment, the above-mentioned predictive markers preferably include at least one selected from red blood cell count, platelet count, free fatty acid concentration, and creatine kinase (CK) value, and more preferably include at least one selected from red blood cell count and free fatty acid concentration.

[0066] When qualitatively predicting weight loss using each predictive marker in this embodiment, examples of reference values ​​for each predictive marker for predicting a responder, i.e., "large weight loss," are shown below. Note that each reference value is the numerical value of each predictive marker measured by the method shown in the Examples.

[0067] The CK value is preferably 182.9 U / L or less, more preferably 147.3 U / L or less, and even more preferably 111.8 U / L or less. The red blood cell count is preferably 521.0 × 10 4 / μL or less, preferably 505.1 × 10 4 / μL or less, more preferably 489.1 × 10 4 The platelet count is preferably 29.8 x 10 4 / μL or less, preferably 27.2 × 10 4 / μL or less, more preferably 24.5 × 10 4 / μL or less. The 3-hydroxybutyric acid concentration is preferably 97.3 nmol / L or less, more preferably 74.4 nmol / L or less, and even more preferably 49.4 nmol / L or less. The free fatty acid concentration is preferably 0.65 mEq / L or less, more preferably 0.55 mEq / L or less, and even more preferably 0.45 mEq / L or less. The FT4 concentration is preferably 1.07 ng / dL or less, more preferably 1.04 ng / dL or less, and even more preferably 1.01 ng / dL or less. The glutamine concentration is preferably 545.5 nmol / L or more, more preferably 585.3 nmol / L or more, and even more preferably 625.1 nmol / L or more. The citrulline concentration is preferably 29.6 nmol / L or more, more preferably 33.7 nmol / L or more, and even more preferably 37.8 nmol / L or more.

[0068] [Example of a method for predicting the reduction of visceral fat mass] (Embodiment 2) In the method according to embodiment 2 of the present invention, the prediction of the weight loss effect is a prediction of the reduction in visceral fat mass, and includes at least one step selected from the group consisting of steps (A2), (B2), and (C2) below. The predictive marker in this embodiment is an index that showed a significant difference between a responder group and a non-responder group, selected based on the amount of reduction in visceral fat mass from a group of subjects who underwent a training program to increase exercise volume in Test Example 1 of the Examples described below. The "visceral fat mass" in this embodiment is preferably visceral fat area.

[0069] Step (A2) is a step of measuring physical indices including at least one selected from the group consisting of MQP, body weight, trunk lean mass, and trunk muscle mass. All of these physical indices show a negative correlation with the reduction in visceral fat mass.

[0070] Step (B2) is a step of measuring at least one blood index selected from the group consisting of glutamine concentration, free triiodothyronine (FT3) concentration, and magnesium concentration. Among these blood indexes, glutamine concentration and magnesium concentration show a positive correlation with the reduction in visceral fat mass, and FT3 concentration shows a negative correlation with the reduction in visceral fat mass.

[0071] Step (C2) is a step of calculating at least one response index selected from the group consisting of an index calculated from answers to questions asking about awareness and behavior regarding lack of exercise, an index of mental fatigue calculated from answers to the Chalder Fatigue Scale, an index of agreeableness personality calculated from answers to the Big Five Scale, an index of openness personality calculated from answers to the Big Five Scale, and an index of conscientiousness personality calculated from answers to the Big Five Scale. Among these response indexes, the index of mental fatigue shows a positive correlation with the amount of reduction in visceral fat mass, while the index calculated from answers to questions asking about awareness and behavior regarding lack of exercise, the index of agreeableness personality, the index of openness personality, and the index of conscientiousness personality show a negative correlation with the amount of reduction in visceral fat mass.

[0072] (Embodiment 2-1) As a specific example of the above-mentioned embodiment 2, the method according to embodiment 2-1 preferably includes at least one step selected from the group consisting of the following steps (A21), (B21), and (C21). The predictive marker in this embodiment is an index that showed a significant difference between a responder group and a non-responder group selected based on the amount of visceral fat reduction from a group of subjects who underwent a training program to increase the amount of exercise in Test Example 1 of the Examples described below. The predictive marker of this embodiment can be used to predict the reduction in visceral fat due to an increase in the amount of exercise for a variety of subjects. The prediction of the reduction in visceral fat in this embodiment is preferably a qualitative prediction of the reduction in visceral fat.

[0073] Step (A21) is a step of measuring at least one physical index selected from the group consisting of MQP, body weight, trunk lean mass, and trunk muscle mass.

[0074] Step (B21) is a step of measuring at least one blood index selected from the group consisting of glutamine concentration, free triiodothyronine (FT3) concentration, and magnesium concentration.

[0075] Step (C21) is a step of calculating at least one response index selected from the group consisting of an index of mental fatigue calculated from responses to the Chalder Fatigue Scale, an index of agreeableness personality calculated from responses to the Big Five Scale, an index of openness personality calculated from responses to the Big Five Scale, and an index of conscientiousness personality calculated from responses to the Big Five Scale.

[0076] In this embodiment, it is preferable that the above-mentioned predictive markers include at least one selected from MQP, body weight, trunk lean mass, trunk muscle mass, an index of mental fatigue calculated from responses to the Chalder Fatigue Scale, an index of agreeableness personality calculated from responses to the Big Five scale, an index of openness personality calculated from responses to the Big Five scale, and an index of conscientiousness personality calculated from responses to the Big Five scale, and it is more preferable that the above-mentioned predictive markers include at least one selected from body weight, an index of mental fatigue calculated from responses to the Chalder Fatigue Scale, an index of agreeableness personality calculated from responses to the Big Five scale, an index of openness personality calculated from responses to the Big Five scale, and an index of conscientiousness personality calculated from responses to the Big Five scale.

[0077] When qualitatively predicting the reduction of visceral fat mass using the above-mentioned predictive markers, examples of reference values ​​for each predictive marker for predicting a responder, i.e., "a large reduction in visceral fat mass," are shown below. Each reference value is the numerical value of each predictive marker measured or calculated by the method shown in the Examples.

[0078] The body weight is preferably 83.4 kg or less, more preferably 79.5 kg or less, and even more preferably 75.7 kg or less. The MQP is preferably 65.5 or less, more preferably 60.5 or less, and even more preferably 55.5 or less. The trunk lean mass is preferably 31.8 kg or less, more preferably 30.8 kg or less, and even more preferably 29.7 kg or less. The trunk muscle mass is preferably 30.3 kg or less, more preferably 29.4 kg or less, and even more preferably 28.4 kg or less.

[0079] The magnesium concentration is preferably 1.95 mg / dL or more, more preferably 1.96 mg / dL or more, and even more preferably 2.08 mg / dL or more. The FT3 concentration is preferably 3.26 pg / mL or less, more preferably 3.16 pg / mL or less, and even more preferably 3.07 pg / mL or less. The glutamine concentration is preferably 546.7 nmol / mL or more, more preferably 590.5 nmol / mL or more, and even more preferably 634.2 nmol / mL or more.

[0080] The mental fatigue index is preferably 4.3 or higher, more preferably 5.9 or higher, and even more preferably 7.5 or higher. The agreeableness personality index is preferably 5.37 or lower, preferably 4.96 or lower, and even more preferably 4.55 or lower. The openness personality index is preferably 4.86 or lower, more preferably 4.09 or lower, and even more preferably 3.49 or lower. The conscientiousness personality index is preferably 4.96 or lower, more preferably 4.36 or lower, and even more preferably 3.75 or lower.

[0081] (Embodiment 2-2) As a specific example of the above-mentioned embodiment 2, the method according to embodiment 2-2 preferably includes at least one step selected from the group consisting of the following steps (A22), (B22), and (C22). The predictive marker in this embodiment is an index that showed a significant difference between a responder group and a non-responder group, selected based on the amount of visceral fat reduction from a group of subjects who completed a training program for increasing physical activity and were certified as having complied with the training in the program, in Test Example 1 of the Examples described below. The predictive marker of this embodiment can be used, for example, to predict the reduction in visceral fat due to an increase in physical activity in subjects who can comply with a set program for increasing physical activity. The prediction of the reduction in visceral fat in this embodiment is preferably a qualitative prediction of the reduction in visceral fat.

[0082] Step (A22) is a step of measuring a physical index consisting of MQP.

[0083] Step (B22) is a step of measuring at least one blood index selected from the group consisting of glutamine concentration and magnesium concentration.

[0084] Step (C22) is a step of calculating at least one response index selected from the group consisting of an index calculated from responses to questions asking about awareness and behavior regarding lack of exercise, an index of openness personality calculated from responses to the Big Five scale, and an index of conscientiousness personality calculated from responses to the Big Five scale.

[0085] In this embodiment, it is preferable that the above-mentioned predictive markers include at least one selected from MQP, body weight, trunk lean mass, trunk muscle mass, an index of mental fatigue calculated from responses to the Chalder Fatigue Scale, an index of agreeableness personality calculated from responses to the Big Five scale, an index of openness personality calculated from responses to the Big Five scale, and an index of conscientiousness personality calculated from responses to the Big Five scale, and it is more preferable that the predictive markers include body weight, an index of mental fatigue calculated from responses to the Chalder Fatigue Scale, an index of agreeableness personality calculated from responses to the Big Five scale, an index of openness personality calculated from responses to the Big Five scale, and an index of conscientiousness personality calculated from responses to the Big Five scale.

[0086] When qualitatively predicting a reduction in visceral fat mass using the above-mentioned predictive markers, examples of reference values ​​for each predictive marker for predicting a responder, i.e., a "high effect of reducing visceral fat mass," are shown below. Each reference value is the numerical value of each predictive marker measured or calculated by the method shown in the Examples.

[0087] The MQP is preferably 65.3 or less, more preferably 60.4 or less, and even more preferably 55.6 or less.

[0088] The magnesium concentration is preferably 1.9 mg / dL or higher, more preferably 2.0 mg / dL or higher, and even more preferably 2.1 mg / dL or higher. The glutamine concentration is preferably 537.4 nmol / mL or higher, more preferably 582.3 nmol / mL or higher, and even more preferably 627.3 nmol / mL or higher.

[0089] The index calculated from the responses to questions asking about awareness and behavior regarding lack of exercise is preferably 3.7 or less, more preferably 3.3 or less, and even more preferably 2.9 or less. The index of openness personality is preferably 4.68 or less, more preferably 4.14 or less, and even more preferably 3.61 or less. The index of conscientiousness personality is preferably 4.75 or less, more preferably 4.31 or less, and even more preferably 3.86 or less.

[0090] [Method for quantitatively predicting weight loss effects] (Embodiment 3) Next, as embodiment 3 of the present invention, a method for quantitatively predicting the weight loss effect due to increased exercise volume will be described. The predictive markers used in the method according to this embodiment are predictive markers selected from Test Example 2, which had an even larger number of subjects than Test Example 1, from which the predictive markers of the above-mentioned embodiment were selected. In the following description, explanations of terms and predictive markers that overlap with those of the above-mentioned embodiment will be omitted as appropriate.

[0091] As described above, "quantitatively predicting the weight loss effect" includes quantitatively predicting the overall weight loss effect and quantitatively predicting changes in specific weight loss evaluation indexes. Quantitative predictions of changes in specific weight loss evaluation indexes preferably include at least one of quantitative predictions of body weight loss or quantitative predictions of visceral fat mass reduction. Quantitative predictions of weight loss include at least one prediction selected from the amount of weight loss, the rate of weight loss, and ranges thereof, and are preferably predictions of the rate of weight loss. Quantitative predictions of visceral fat mass reduction include at least one prediction selected from the amount of visceral fat reduction, the rate of visceral fat reduction, and ranges thereof, and are preferably predictions of the amount of visceral fat area reduction.

[0092] The method for quantitatively predicting the weight loss effect due to increased exercise volume in this embodiment includes an acquisition step (E1) of acquiring at least one index selected from the group consisting of abdominal circumference, systolic blood pressure, and pulse rate as physical indices representing at least one of the subject's physique or body composition, and response indices calculated from the subject's answers to the questionnaire, including an index calculated from the answer to the question item ``I felt calm and relaxed'' in the WHO-5 (Mental Health Status Questionnaire), an index of mental fatigue calculated from answers to the Chalder Fatigue Scale, an index calculated from answers to a questionnaire evaluating the stage of change of rest behavior, an index calculated from answers to questions asking about motivation to increase exercise volume, an index calculated from answers to questions asking about awareness and / or behavior regarding irregular mealtimes, an index calculated from answers to questions asking about awareness and behavior regarding lack of exercise, an index of proxy eating calculated from answers to an eating behavior questionnaire, an index of dietary content calculated from answers to an eating behavior questionnaire, and an index of conscientiousness personality calculated from answers to the Big Five scale. The indices acquired in the acquisition step (E1) function as "prediction markers" for quantitatively predicting the weight loss effect. The acquisition step (E1) is a step of acquiring at least one predictive marker listed as a physical index and a response index.

[0093] The above-mentioned predictive markers are high-priority indicators classified as "Priority 1" among the indicators used to construct a linear regression model for calculating the weight loss rate and / or a linear regression model for calculating the amount of visceral fat area reduction from the data of all subjects in a group of subjects who underwent a training program for increasing the amount of exercise in Test Example 2 of the Examples described below. Indicators classified as Priority 1 are indicators that are minimally invasive and can be easily obtained, specifically, physical indicators as body measurement values ​​and indicators belonging to response indicators. These predictive markers can be used to predict the weight loss effect of increased exercise for a variety of subjects. The gender of the subject to be predicted using the predictive markers obtained by the acquisition step (E1) is not particularly limited, but from the viewpoint of improving prediction accuracy, it is preferable that the subject be male.

[0094] In this embodiment, "acquisition" of a predictive marker refers to acquisition of information about the predictive marker. Here, "acquisition of information" includes not only acquisition of computer-processable data but also acquisition of information via paper media. Specifically, "acquisition of physical indices" refers to both measuring physical indices and acquiring information about their measurement values, as described in step (A), and acquiring information about physical indices from other devices. The "acquisition of blood indices" described below refers to both measuring blood indices from a subject's blood sample and acquiring information about their measurement values, as described in step (B), and acquiring information about blood indices from other devices. The "acquisition of response indices" refers to both calculating response indices from the subject's responses to the questionnaire and acquiring information about the calculated values, as described in step (C), and acquiring information about response indices from other devices. In other words, the "acquisition" in the acquisition step (E1) refers to a concept that includes the measurement in step (A), the measurement in step (B), and the calculation in step (C).

[0095] More specifically, "acquiring physical information" preferably includes measuring a physical indicator and storing information on the measurement value in a measuring device, and / or receiving information on the measurement value. "acquiring blood indicators" preferably includes measuring a blood indicator and storing information on the measurement value in a measuring device, and / or receiving information on the measurement value. "acquiring answer indicators" preferably includes calculating answer indicators and storing information on the calculated values ​​in a computer, and / or receiving information on the calculated values.

[0096] Hereinafter, predictive markers not described in the above embodiments will be described.

[0097] The abdominal circumference as a physical index is the size around the waist measured at the height of the navel, and can be measured by measuring the waist horizontally at the height of the navel with a tape measure. The systolic blood pressure and pulse rate can be measured by known methods, for example, using an upper arm digital blood pressure monitor (e.g., "Automatic Blood Pressure Monitor HEM-104" manufactured by Omron Corporation).

[0098] The WHO-5 (Mental Health Questionnaire) is a commonly used questionnaire for investigating mental health status, and is a questionnaire consisting of five questions as shown in FIG. 8. In this embodiment, the response index is calculated from the response to question 2 of these five questions, which asks, "I felt calm and relaxed." This index can be a number assigned to the selected option.

[0099] The "rest behavior change stage" refers to a stage of change for rest behavior based on the stage of behavior change theory (reference: Prochaska, J.O. & Velicer, W.F.: The transtheoretical model of health behavior change, American Journal of Health Promotion, 12, 38-48, 1997). A questionnaire for assessing the rest behavior change stage can be the "Health Behavior Change Stage Test" shown in Figure 1 of "Study of Self-Efficacy Scales and Social Support Scales Useful for Supporting Health Behavior Change" (Kiyoshi Moriya and Mari Shimizu, Bulletin of Tenshi University, Vol. 9, June 30, 2009) (hereafter referred to as "Moritani et al., 2009"), where "activity and exercise behavior" is replaced with "stress management and rest behavior." The "Health Behavior Change Stage Test" described in the same publication is shown in Figure 9. The index calculated from the answers to this test paper can be calculated by referring to the same literature, and the numbers of the options for the question I. about the awareness of change and stage of health behavior can be used as the index.

[0100] The question asking about motivation to increase the amount of exercise, for example, asks about the subject's motivation (motivation) to increase the amount of exercise before starting to increase the amount of exercise. For example, the answer options for motivation are expressed on an 11-point scale from 0 to 10 (very motivated to not motivated at all). The index calculated from the question can be the score corresponding to the answer.

[0101] The questions inquiring about perceptions and / or behaviors regarding irregular mealtimes preferably include at least one question selected from the following: a question about eating a late-night snack after dinner, a question about irregular mealtimes, a question about the interval between dinner and bedtime, and a question about lifestyle rhythms (night owl or morning person). Specifically, such questions include, for example, questions 13 ("I often eat a late-night snack after dinner"), 17 ("My meal times are irregular"), 19 ("I eat a late-night snack"), 23 ("I often eat dinner late because of work"), 25 ("I finish dinner more than two hours before bedtime"), and 30 ("I'm a night owl and have trouble catching up in the mornings") in the "35-Question Dietary Habits Questionnaire" described in Yanagisawa et al., 2017, as illustrated in Figure 2. The index calculated from the responses to the questions inquiring about perceptions and behaviors regarding irregular mealtimes may be, for example, a numerical value selected from the average score, total score, and score percentage, which is the ratio of the total score to the highest score, corresponding to the answers to the six questions. Furthermore, the six questions above correspond to the third factor (meal time) of the five factors (diet, health awareness, mealtime, dietary restrictions, and exercise) derived from factor analysis (maximum likelihood method, oblique rotation) of all 35 questions using the method described in "Yanagisawa et al., 2017," and can be used as a measure suggesting irregular mealtimes. Furthermore, as mentioned above, the question asking about awareness and behavior regarding lack of exercise corresponds to the fifth factor (exercise) of the five factors (diet, health awareness, mealtime, dietary restrictions, and exercise).

[0102] The Eating Behavior Questionnaire, cited in the "Obesity Treatment Guidelines 2006" (Japan Society for the Study of Obesity), is a questionnaire provided by Sakata et al. (Reference: Hironobu Yoshimatsu (1996) Initial Operation "Obesity Treatment Manual," edited by Toshiie Sakata, Ishiyaku Shuppan, Tokyo, pp. 17-38). The questions in this questionnaire are categorized into seven areas (perception of constitution and weight, eating motivation, surrogate eating, hunger, satiety, eating habits, meal content, and regularity of eating habits), as illustrated in Figure 5. The indices for surrogate eating and meal content calculated from this questionnaire can be numerical values ​​selected from the total score, average score, and percentage score, which is the ratio of the total score to the highest score, of the questions in each category. For example, the response options for each question include four levels: "Not at all," "Sometimes this happens," "Tends to be like that," and "Exactly true," and each response option can be assigned a predetermined score (e.g., 0 to 3 points).

[0103] Furthermore, the prediction method according to this embodiment may include a prediction step (F) after the acquisition step (E1). The prediction step (F) is a step of quantitatively predicting the weight loss effect of the subject by applying at least one index acquired in the acquisition step (E1) to a prediction model for deriving a predicted value indicating the weight loss effect from the acquired index. The prediction step (F) may be executed, for example, by a computer having a control unit. The control unit may be composed of a processor such as a CPU (Central Processing Unit).

[0104] The predictive model can be generated by acquiring a large amount of data from a sample population of subjects. For example, the predictive model can be generated by machine learning using the values ​​of indicators in the population of subjects as explanatory variables and values ​​indicating the weight loss effect in the population of subjects as objective variables. The explanatory variables are selected from the training data using an appropriate explanatory variable selection algorithm, such as lasso regression or elastic net. Algorithms used in machine learning models are preferably linear regression models, such as linear regression models, regularized linear regression models (lasso regression, elastic net, etc.), generalized linear models, etc., and nonlinear algorithms are also possible.

[0105] The optimal prediction model can be determined by inputting verification data into a prediction model constructed using the algorithm described above to calculate predicted values, and then selecting the model whose predicted values ​​best match the actual measured values, for example, the model with the highest accuracy rate or the model with the smallest prediction error.

[0106] As described above, the predictive markers of this embodiment are all minimally invasive and can be easily obtained, which can reduce the burden on the subject associated with obtaining the indicators. Therefore, according to this embodiment, the weight loss effect can be quantitatively and accurately predicted using easily obtainable indicators.

[0107] (Embodiment 3-1) When the quantitative prediction of the weight loss effect is a quantitative prediction of body weight loss, in the acquisition step (E1), in addition to the indicators described in embodiment 3, it is preferable to acquire at least one indicator selected from the group consisting of non-HDL cholesterol concentration, γ-GT value, HDL cholesterol concentration, and hemoglobin value as a blood indicator measured from the subject's blood sample.

[0108] The above-mentioned predictive markers are high-priority indicators classified as "priority 2" among indicators selected by an appropriate explanatory variable selection algorithm for predicting weight loss rate from the data of all subjects in a group of subjects who underwent a training program for increasing exercise volume in Test Example 2 of the Examples described below. Indicators classified as priority 2 are blood indices included in the test items of general health checkups. In this embodiment, weight loss can be quantitatively predicted with greater accuracy by using priority 2 indicators that can be measured at general medical institutions in addition to the priority 1 indicators described in Embodiment 3.

[0109] Below, we will explain predictive markers not described in the above embodiments. HDL cholesterol concentration (e.g., unit: mg / dL) can be measured by an established method such as an enzymatic method using a serum sample. Non-HDL cholesterol concentration (e.g., unit: mg / dL) is the value obtained by subtracting the HDL cholesterol concentration from the total cholesterol concentration. γ-GT (γ-glutamyl transpeptidase) level can be measured by an established method such as the JSCC standardized method using a serum sample.

[0110] (Embodiment 3-2) When the quantitative prediction of the weight loss effect is a quantitative prediction of body weight loss, in the acquisition step (E1), in addition to the indicators described in embodiments 3 and 3-1, it is preferable to further acquire at least one indicator selected from the group consisting of hydroxyproline concentration, threonine concentration, glutamine concentration, alanine concentration, citrulline concentration, valine concentration, β-alanine concentration, 1-methylhistidine concentration, tryptophan concentration, and glycine concentration as a blood indicator.

[0111] The above-mentioned predictive markers are indices classified as "Priority 3" among indices selected by an appropriate explanatory variable selection algorithm for predicting weight loss rates from the data of all subjects in a group of subjects who underwent a training program for increasing exercise volume in Test Example 2 of the Examples described below. Indicators classified as Priority 3 are values ​​measured in amino acid analysis (amino acid fractions) of plasma samples. The blood indices of this embodiment are all expressed in units such as "nmol / mL" and can be measured, for example, by HPLC or the like. In this embodiment, weight loss can be quantitatively predicted with greater accuracy by using a Priority 3 indicator that can be measured as an amino acid fraction in addition to the Priority 1 and 2 indicators described in Embodiments 3 and 3-1.

[0112] (Embodiment 3-3) When the quantitative prediction of the weight loss effect is a quantitative prediction of body weight loss, in addition to the indicators described in embodiments 3, 3-1, and 3-2, the acquisition step (E1) preferably further acquires at least one indicator selected from the group consisting of physical indicators such as visceral fat area and MQP (Muscle Quality Point), and blood indicators such as total protein concentration, potassium concentration, chloride concentration, inorganic phosphorus concentration, platelet count, DHEA-S (dehydroepiandrosterone sulfate) concentration, total ketone body concentration, acetoacetic acid concentration, 3-hydroxybutyric acid concentration, high-sensitivity CRP concentration, alkaline phosphatase activity, growth hormone concentration, and blood ammonia concentration.

[0113] The predictive markers are those classified as "Priority 4" among those selected by an appropriate explanatory variable selection algorithm for predicting weight loss rates from the data of all subjects in a group of subjects who underwent a training program for increasing exercise volume in Test Example 2 of the Examples described below. The indicators classified as Priority 4 are those selected above, excluding those classified as Priorities 1 to 3, and are more specialized than those classified as Priorities 1 to 3. Examples of indicators classified as Priority 4 include those measured using specific equipment, test items with higher precision than those in general periodic health checkups (such as test items performed in comprehensive medical checkups), and specialized blood test items. In this embodiment, weight loss can be predicted with even greater accuracy by using a Priority 4 indicator in addition to the Priority 1 to 3 indicators described in Embodiments 3, 3-1, and 3-2.

[0114] Predictive markers not described in the above embodiments will now be described. Total protein (e.g., g / dL) can be measured using established methods such as the Biuret method using serum samples. Potassium concentration (e.g., mEq / L) and chloride concentration (e.g., mEq / L) can be measured using established methods such as an electrode method using serum samples. Inorganic phosphorus concentration (e.g., mg / dL) can be measured using established methods such as an enzymatic method using serum samples. DHEA-S (dehydroepiandrosterone sulfate) concentration (e.g., μg / dL) can be measured using established methods such as the CLEIA method using serum samples. Total ketone body concentration, 3-hydroxybutyric acid concentration, and acetoacetic acid concentration (e.g., μmol / L) can be measured using established methods such as an enzymatic method using serum samples. High-sensitivity CRP (C-reactive protein) concentrations (e.g., mg / dL) can be measured using serum samples by established methods such as the latex agglutination assay. Alkaline phosphatase activity (e.g., U / dL) can be measured using serum samples by established methods such as the IFCC-standardized method. Growth hormone (GH) concentrations (e.g., ng / mL) can be measured using serum samples by established methods such as the ECLIA method. Blood ammonia concentrations (e.g., μg / dL) can be measured using deproteinized supernatants by established methods such as the modified Fujii-Okuda method.

[0115] (Embodiments 3-4) When the quantitative prediction of the weight loss effect is a quantitative prediction of the reduction in visceral fat mass, in the acquisition step (E1), in addition to the indicators described in embodiment 3, it is preferable to acquire at least one indicator selected from the group consisting of HDL cholesterol concentration, γ-GT value, and non-HDL cholesterol concentration as a blood indicator measured from the subject's blood sample.

[0116] The above-mentioned predictive markers are high-priority indexes classified as "Priority 2" among indexes selected by an appropriate explanatory variable selection algorithm for predicting the amount of visceral fat area reduction from the data of all subjects in a group of subjects who underwent a training program for increasing the amount of exercise in Test Example 2 of the Examples described later. In this embodiment, in addition to the Priority 1 indexes described in Embodiment 3, Priority 2 indexes measurable at general medical institutions are used, making it possible to quantitatively and accurately predict the reduction in visceral fat mass.

[0117] (Embodiments 3-5) When the quantitative prediction of the weight loss effect is a quantitative prediction of the reduction in visceral fat mass, in the acquisition step (E1), in addition to the indicators described in embodiments 3 and 3-4, it is preferable to further acquire at least one indicator selected from the group consisting of hydroxyproline concentration, threonine concentration, alanine concentration, isoleucine concentration, Fischer ratio, α-amino-n-butyric acid concentration, and methionine concentration as a blood indicator.

[0118] The predictive markers are the indices classified as "Priority 3" among the indices selected by an appropriate explanatory variable selection algorithm for predicting the amount of visceral fat area reduction from all of the subjects who underwent the training program for increasing the amount of exercise in Test Example 2 of the Examples described later. In this embodiment, in addition to the indices of Priorities 1 and 2 described in Embodiments 3 and 3-4, a Priority 3 indice measurable as an amino acid fraction is used, thereby making it possible to quantitatively predict the reduction in visceral fat mass with higher accuracy.

[0119] Among the predictive markers not described in the above embodiments, the isoleucine concentration, α-amino-n-butyric acid concentration, and methionine concentration are all expressed in units such as "nmol / mL" and can be measured, for example, by HPLC. The Fischer ratio is a value indicating the ratio of branched-chain amino acids to aromatic amino acids contained in blood, and can be determined by calculating the ratio of these amino acid concentrations measured, for example, by HPLC.

[0120] (Embodiments 3-6) When the quantitative prediction of the weight loss effect is a quantitative prediction of a reduction in visceral fat mass, in addition to the indices described in embodiments 3, 3-4, and 3-5, the acquisition step (E1) preferably further acquires at least one index selected from the group consisting of MQP, total body water content, and trunk fat mass as physical indices, and potassium concentration, total bile acid concentration, mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), platelet count, acetoacetic acid concentration, free testosterone concentration, high-sensitivity CRP concentration, growth hormone (GH) concentration, and serum iron concentration as blood indices.

[0121] The above-mentioned predictive markers are indices classified as "Priority 4" among indices selected by an appropriate explanatory variable selection algorithm for predicting the amount of visceral fat area reduction from all of the subjects who underwent the training program for increasing the amount of exercise in Test Example 2 of the Examples described later. In this embodiment, by using such indices in addition to the indices described in Embodiments 3, 3-4, and 3-5, it is possible to quantitatively predict the reduction in visceral fat mass with higher accuracy.

[0122] Predictive markers not described in the above embodiments will now be described. Total body water (TBW) (e.g., unit: kg) refers to the mass of water in the body. Trunk fat mass refers to the amount of fat contained in the trunk portion of the body, and is expressed, for example, as the mass of the muscle (unit: kg). As mentioned above, these body compositions can be measured by density methods such as underwater weighing, dual energy X-ray absorptiometry (DEXA), bioelectrical impedance analysis, ultrasound, skinfold thickness analysis (caliper method), and the like. These body composition indicators can also be measured, for example, using a body composition analyzer manufactured by Tanita Corporation (e.g., the "Innerscan Dual" series, the "Professional Multi-Frequency Body Composition Analyzer MC-980A-N plus," etc.).

[0123] Total bile acids (e.g., in μmol / L) can be measured using established methods, such as enzymatic methods, using serum samples. MCH (e.g., in pg) and MCHC (e.g., in %) can be measured, for example, by a blood cell count test using the subject's whole blood. Free testosterone levels (e.g., in pg / mL) can be measured using established methods, such as RIA solid-phase methods, using serum samples. Serum iron levels (e.g., in μg / dL) can be measured using established methods, such as the nitroso-PSAP method, using serum samples.

[0124] (Embodiment 4) In the above-mentioned embodiment 3, a method was described in which an index selected from the data of all subjects in a group of subjects who underwent a training program to increase the amount of exercise in Test Example 2 of the Examples described later, but in the following embodiment 4, a method is described in which an index (predictive marker) selected from a group of subjects who were certified as having complied with the training program to increase the amount of exercise in Test Example 2 is used. Hereinafter, terms and explanations that overlap with those in the above-mentioned embodiments will be omitted as appropriate.

[0125] The method for quantitatively predicting the weight loss effect due to increased exercise volume in this embodiment includes an acquisition process (E2) for acquiring at least one index selected from the group consisting of abdominal circumference and systolic blood pressure as physical indicators, and an index calculated from responses to questions asking about awareness and behavior regarding lack of exercise as response indicators, an index calculated from responses to the question item ``I felt calm and relaxed'' in the WHO-5 (Mental Health Status Questionnaire), an index calculated from responses to the question item ``There were many things that interested me in my daily life'' in the WHO-5 (Mental Health Status Questionnaire), an index calculated from responses to questions about stress responses in the Occupational Stress Short Questionnaire, an index calculated from responses to questions asking about motivation to increase exercise volume, an index of conscientiousness personality calculated from responses to the Big Five scale, an index calculated from responses to the Rest Behavior Self-Efficacy Scale, and an index calculated from responses to questions asking about awareness of dieting.

[0126] The predictive markers are selected from data of a group of subjects who completed a training program for increasing the amount of exercise in Test Example 2 of the Examples described below and who were certified as having complied with the training in the program by a linear regression model for predicting a weight loss rate and / or an appropriate explanatory variable selection algorithm for predicting a reduction in visceral fat area, and are classified as having priority 1, as in Embodiment 3. The predictive markers can be used, for example, to predict the weight loss effect of increased exercise for subjects who can comply with a set program for increasing the amount of exercise.

[0127] Predictive markers not described in the above embodiments will now be described.

[0128] The index calculated from the response to the question "There were many things that interested me in my daily life" in the WHO-5 (Mental Health Status Questionnaire) is the index calculated from the response to the question "There were many things that interested me in my daily life" in question 5 of the five questions shown in Figure 8. This index can be the number assigned to each option.

[0129] The Simple Occupational Stress Questionnaire is a questionnaire provided by the Ministry of Health, Labor and Welfare to investigate occupational stress levels, as shown in Figures 10 and 11. The questions about stress reactions are a group of 29 questions in Area B. The index calculated from the answers to the questions can be, for example, the total score of the answers to the questions. These indices can be the total score, or the score percentage, which is the ratio of the total score to the average score or the highest score.

[0130] The rest behavior self-efficacy scale can be the one described in "A Study of Self-Efficacy Scales and Social Support Scales Useful for Supporting Health Behavioral Change" (Kiyoshi Moriya and Mari Shimizu, Bulletin of Tenshi University, Vol. 9, June 30, 2009) (hereinafter referred to as "Moritani et al., 2009") (see Figure 7). For example, the index calculated from the responses to the rest behavior self-efficacy scale can be the total score of the answers selected from the same literature and Figure 7, ranging from "I think I can do it well" (-3 points) to "I think I can't do it at all" (+3 points). Note that these indices may be the total score or the score percentage, which is the ratio of the total score to the average score or the highest score.

[0131] A question asking about attitudes toward dieting, for example, asks about the subject's attitude toward dieting (weight loss and obesity reduction) before implementing dietary management. A specific question might be, "Please rate your current attitude toward dieting (weight loss and obesity reduction) on a scale of 1 to 10." The options are, for example, expressed as 11 levels from 0 to 10 (very high, none to very low, none). The index calculated from the question can be the score corresponding to this answer.

[0132] The predictive markers of this embodiment are all minimally invasive and easily obtainable indicators, which can reduce the burden on subjects associated with obtaining the indicators. Therefore, according to this embodiment, it is possible to quantitatively and accurately predict the weight loss effect using easily obtainable indicators, particularly for subjects who are able to comply with an exercise increase program.

[0133] Furthermore, the method of this embodiment may also include the above-mentioned prediction step (F) in addition to the acquisition step (E2).

[0134] (Embodiment 4-1) When the quantitative prediction of the weight loss effect is a quantitative prediction of body weight loss, in the acquisition step (E2), it is preferable to acquire at least one index selected from the group consisting of γ-GT concentration, triglyceride concentration, and HDL cholesterol concentration as a blood index measured from a blood sample of the subject, in addition to the indexes described in embodiment 4. Triglyceride (TG) concentration (e.g., unit: mg / dL), which is not described in the above embodiments, can be measured by an established method such as an enzymatic method using a serum sample.

[0135] The predictive markers are high-priority indicators classified as "priority 2" above, selected by an appropriate explanatory variable selection algorithm for predicting weight loss rates from data on a group of subjects who underwent a training program for increasing exercise volume in Test Example 2 of the Examples described below and who were certified as having complied with the training in the program. In this embodiment, weight loss can be quantitatively predicted with higher accuracy by using priority 2 indicators that can be measured at general medical institutions in addition to the priority 1 indicators described in Embodiment 4.

[0136] (Embodiment 4-2) When the quantitative prediction of the weight loss effect is a quantitative prediction of body weight loss, in addition to the indicators described in embodiments 4 and 4-1, the acquisition step (E2) preferably further acquires at least one indicator selected from the group consisting of MQP and total body water (TBW) as physical indicators, and sodium concentration, chloride concentration, total bile acid concentration, white blood cell count, platelet count, total ketone body concentration, acetoacetic acid concentration, 3-hydroxybutyric acid concentration, free fatty acid concentration, alkaline phosphatase activity, growth hormone concentration, and tumor necrosis factor-α concentration as blood indicators.

[0137] The predictive markers are those classified as "Priority 4" among those selected by an appropriate explanatory variable selection algorithm for predicting weight loss rates from data on a group of subjects who completed the instruction program for increasing exercise volume in Test Example 2 and were certified as having complied with the instruction in the program. In this embodiment, by using an index of Priority 4 in addition to the indexes of Priorities 1 and 2 described in Embodiments 4 and 4-1, weight loss can be quantitatively predicted with higher accuracy. Note that the indexes selected by an appropriate explanatory variable selection algorithm for predicting weight loss rates from data on a group of subjects who complied with the instruction program for increasing exercise volume in Test Example 2 in the Examples described below did not include any indexes classified as Priority 3.

[0138] Predictive markers not described in the above embodiments will now be described. The sodium concentration (e.g., unit: mEq / L) can be measured by an established method such as the electrode method using a serum sample. The white blood cell count can be measured by, for example, a blood cell count test using the subject's whole blood, and can be calculated by, for example, "×10 2 Tumor necrosis factor (TNF)-α concentrations (e.g., in pg / mL) can be measured using established methods such as EIA using serum samples.

[0139] (Embodiment 4-3) When the quantitative prediction of the weight loss effect is a quantitative prediction of the reduction in visceral fat mass, in the acquisition step (E2), in addition to the indicators described in embodiment 4, it is preferable to acquire at least one indicator selected from the group consisting of triglyceride concentration, HDL cholesterol concentration, and non-HDL cholesterol concentration as a blood indicator measured from the subject's blood sample.

[0140] The predictive markers are high-priority indicators classified as "Priority 2" among indicators selected by an appropriate explanatory variable selection algorithm for predicting the amount of visceral fat area reduction from data of a group of subjects who underwent a training program for increasing exercise volume in Test Example 2 of the Examples described below and who were certified as having complied with the training in the program. In this embodiment, in addition to the Priority 1 indicators described in Embodiment 4, Priority 2 indicators that can be measured at general medical institutions and the like are used to quantitatively and more accurately predict the reduction in visceral fat mass.

[0141] (Embodiment 4-4) When the quantitative prediction of the weight loss effect is a quantitative prediction of the reduction in visceral fat mass, in addition to the indices described in embodiments 4 and 4-3, the acquisition step (E2) preferably further acquires at least one blood index selected from the group consisting of hydroxyproline concentration, threonine concentration, valine concentration, methionine concentration, β-alanine concentration, 1-methylhistidine concentration, and histidine concentration. All of these indices are expressed in units such as "nmol / mL" and can be measured, for example, by HPLC.

[0142] The predictive markers are the indices classified as "Priority 3" above, selected by an appropriate explanatory variable selection algorithm for predicting the amount of visceral fat area reduction from data of a group of subjects who underwent a training program for increasing exercise volume in Test Example 2 of the Examples described below and who were certified as having complied with the training in the program. In this embodiment, in addition to the indices of Priorities 1 and 2 described in Embodiments 4 and 4-3, a Priority 3 index measurable as an amino acid fraction is used, thereby making it possible to quantitatively predict the reduction in visceral fat mass with higher accuracy.

[0143] (Embodiments 4-5) When the quantitative prediction of the weight loss effect is a quantitative prediction of the reduction in visceral fat mass, in addition to the indicators described in embodiments 4, 4-3, and 4-4, the acquisition step (E2) preferably further acquires at least one indicator selected from the group consisting of MQP and total body water (TBW) as physical indicators, and total bilirubin concentration, potassium concentration, inorganic phosphorus concentration, magnesium concentration, serum iron concentration, platelet count, acetoacetic acid concentration, growth hormone concentration, tumor necrosis factor-α concentration, total ketone body concentration, and cortisol concentration as blood indicators.

[0144] The predictive markers are the indices classified as "Priority 4" above, selected by an appropriate explanatory variable selection algorithm for predicting the amount of visceral fat area reduction from data of a group of subjects who were certified as having complied with the instruction in the program among a group of subjects who underwent an instruction program for increasing the amount of exercise in Test Example 2 of the Examples described later. In this embodiment, by using the index of Priority 4 in addition to the indexes of Priorities 1 to 3 described in Embodiments 4, 4-3, and 4-4, it is possible to more accurately predict the reduction in visceral fat mass.

[0145] Predictive markers not described in the above embodiments will now be described. Total bilirubin concentration (e.g., mg / dL) can be measured using a serum sample by an established method such as the vanadate oxidation method. Cortisol concentration (e.g., μg / dL) can be measured using a serum sample by an established method such as the ECLIA method.

[0146] [Additional remarks] Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the spirit of the present invention. For example, in the acquisition step of Embodiment 3 and / or Embodiment 4, each predictive marker used in Embodiments 1 and 2 may be further acquired.

[0147] In the above embodiments, a method for predicting a weight loss effect has been described, but as another embodiment of the present invention, a method for predicting a weight loss effect can be provided. For example, a prediction method according to one embodiment of the present invention is a prediction method for quantitatively predicting a weight loss effect of a subject due to an increase in the amount of exercise, and includes, as physical indices representing at least one of the subject's physique or body composition, abdominal circumference, systolic blood pressure, and pulse rate, and as response indices calculated from the subject's answers to a questionnaire, an index calculated from the answer to the question item "I felt calm and relaxed" in the WHO-5 (Mental Health Status Questionnaire), an index of mental fatigue calculated from the answer to the Chalder Fatigue Scale, an index calculated from the answer to a questionnaire evaluating the stage of change of rest behavior, and an index calculated from the answer to a question asking about the willingness to increase the amount of exercise. and a prediction step of quantitatively predicting the weight loss effect of the subject by applying at least one index selected from the group consisting of an index calculated from answers to a questionnaire, an index calculated from answers to questions asking about awareness and / or behavior regarding irregular mealtimes, an index calculated from answers to questions asking about awareness and behavior regarding lack of exercise, an index regarding proxy eating calculated from answers to the questionnaire, an index regarding dietary content calculated from answers to the questionnaire, and an index of conscientiousness personality calculated from answers to the Big Five scale to a prediction model for deriving a predicted value indicating the weight loss effect from the acquired index. The prediction step may further use at least one of the predictive markers described in embodiments 3-1 to 3-6, or may further use other predictive markers.

[0148] Alternatively, a prediction method according to another embodiment of the present invention is a prediction method for quantitatively predicting the weight loss effect of a subject due to an increase in the amount of exercise, and includes, as physical indices, abdominal circumference and systolic blood pressure, and as response indices, an index calculated from answers to questions asking about awareness and behavior regarding lack of exercise, an index calculated from answers to a question item in the WHO-5 (mental health status questionnaire) such as "I felt calm and relaxed," an index calculated from answers to a question item in the WHO-5 (mental health status questionnaire) such as "There were many things that interested me in my daily life," an index calculated from an occupational status, and an index calculated from answers to questions regarding awareness and behavior regarding lack of exercise, and ... and a prediction step of quantitatively predicting the weight loss effect of the subject by applying at least one index selected from the group consisting of an index calculated from answers to questions about stress response in a brief stress questionnaire, an index calculated from answers to questions asking about motivation to increase exercise, an index of conscientiousness personality calculated from answers to the Big Five scale, an index calculated from answers to the rest behavior self-efficacy scale, and an index calculated from answers to questions asking about attitudes toward dieting to a prediction model for deriving a predicted value indicating the weight loss effect from the acquired index. The prediction step may further use at least one of the predictive markers described in embodiments 4-1 to 4-5, or may further use other predictive markers.

[0149] The prediction step can be carried out in the same manner as the prediction step (F) described above.

[0150] In addition, in the present invention, for example, a plurality of the above-mentioned predictive markers may be combined to predict the weight loss effect, or the above-mentioned predictive markers may be combined with other indicators to predict the weight loss effect.

[0151] In another embodiment, the present invention may provide a predictive marker for predicting weight loss effects, the predictive marker including at least one of the physical indicators, blood indicators, and response indicators described above. Furthermore, the present invention may provide a use of the predictive marker. [Example]

[0152] <Test Example 1> [Exercise intervention trial] BMI of 23 kg / m 2 Fifty-five adult men (aged 20 to 60 years) who met the eligibility criteria for the study were selected as the subject group for Test Example 1. The subject group underwent a physical examination, blood tests, and questionnaire tests as pre-intervention tests before the exercise intervention. The tests in this test example were conducted with the consent of the subjects.

[0153] Physical examinations were conducted by gathering a group of subjects at a designated venue. Physical examination items included height, weight, body composition (fat mass, lean mass, muscle mass, MQP, estimated bone mass, total body water (TBW), total body water percentage (TBW%), body fat percentage, SMI (appendicular muscular index), trunk body fat percentage, trunk body fat mass, trunk lean mass, trunk muscle mass, etc.), visceral fat area, abdominal circumference, blood pressure, pulse, and body temperature. Height was measured using a standard height measuring device. Weight and body composition were measured using a body composition analyzer ("Professional Multi-Frequency Body Composition Analyzer MC-980A-N plus" manufactured by Tanita Corporation). Visceral fat area and abdominal circumference were measured using a Panasonic EW-FA90 visceral fat analyzer (medical device approval number 22500BZX00522000) wrapped around the subject's abdomen. Blood pressure and pulse were measured using an upper arm digital blood pressure monitor ("Automatic Blood Pressure Monitor HEM-104", manufactured by Omron Corporation). Body temperature was measured on the forehead using a non-contact thermometer.

[0154] Blood tests were performed by drawing venous blood from the subjects and requesting a testing institution to measure the following items using the blood samples obtained. General blood test items included white blood cell count, red blood cell count, hemoglobin content, hematocrit value, platelet count, mean corpuscular volume, mean corpuscular hemoglobin content (hemoglobin value), and mean corpuscular hemoglobin concentration. Blood biochemistry test items included creatinine, uric acid, urea nitrogen, aspartate aminotransferase, alanine aminotransferase, γ-glutamyltranspeptidase, alkaline phosphatase, lactate dehydrogenase, creatine kinase, total cholesterol, HDL cholesterol, non-HDL cholesterol, LDL cholesterol, triglycerides, total bilirubin, total bile acids, sodium, potassium, chloride, calcium, magnesium, inorganic phosphorus, iron, and total Protein, albumin, albumin / globulin ratio, free fatty acids, blood glucose, HbA1c, insulin, NEFA, total ketone bodies, acetoacetate, 3-hydroxybutyrate, high-sensitivity CRP (hsCRP), leptin, growth hormone, cortisol, dehydroepiandrosterone (DHEA), dehydroepiandrosterone sulfate (DHEA-S), testosterone, androgens, TNF-α, IGF-1, T3, T4, TSH, FT3, FT4, ammonia, and amino acid fractions.

[0155] The questionnaire test was conducted by distributing questionnaires to the subjects, who completed them themselves. The questionnaires included the BDHQ (Brief Self-Administered Diet History Questionnaire), the International Standardized Physical Activity Questionnaire, the Obesity-Related Diet and Lifestyle Questionnaire (35-question dietary questionnaire), the WHO-5 Mental Health Status Questionnaire, the Brief Occupational Stress Questionnaire, the Chalder Fatigue Scale, the Athens Insomnia Scale, the Polyphenol Questionnaire, the Big Five Scale, self-efficacy scales (eating behavior self-efficacy scale, exercise behavior self-efficacy scale, and rest behavior self-efficacy scale), the Exercise Behavior Stage of Change Questionnaire, the Rest Behavior Stage of Change Questionnaire, the Eating Behavior Stage of Change Questionnaire, the Eating Behavior Questionnaire, a questionnaire on motivation for dietary management, and a questionnaire on attitudes toward dieting.

[0156] The BDHQ (Brief Self-Administered Dietary History Questionnaire) is a dietary survey developed by the Department of Social Preventive Epidemiology (Laboratory), Department of Public Health / Department of Health Sciences and Nursing, Graduate School of Medicine, The University of Tokyo, and the BDHQ questionnaire was used with permission from the lab. Analysis of the responses was requested from the lab. As a result of the analysis, estimates of nutrient intake by food group, such as energy intake, water intake, protein, lipids, saturated fatty acids, carbohydrates, dietary fiber, alcohol, salt, potassium, calcium, vitamin C, and folic acid, were obtained.

[0157] The International Standardized Physical Activity Questionnaire is a commonly used questionnaire on physical activity, and the short version was used. Calculation of indices from the scores was performed with reference to the literature (Murase et al., "International Standardization of Physical Activity - Evaluation of the Reliability and Validity of the Japanese Version of the IPAQ," Welfare Indicators, 49(11), 1-9, 2002).

[0158] The 35-question dietary questionnaire used the questionnaire shown in Figure 2 in "Yanagisawa et al., 2017," and calculated the average score for each question group that constitutes each of the factors described in the same publication: Factor 1 (appetite), Factor 2 (health awareness), Factor 3 (meal time), Factor 4 (dietary restrictions), and Factor 5 (exercise). The questions that make up Factor 3 (meal time) correspond to the above-mentioned "questions asking about awareness and / or behavior regarding irregular mealtimes." The questions that make up Factor 5 (exercise) correspond to the above-mentioned "questions asking about awareness and behavior regarding lack of exercise."

[0159] The WHO-5 Mental Health Questionnaire is a commonly used questionnaire for investigating mental health status, as shown in Figure 8, and the sum of the scores of all responses was calculated as an index. In addition, the scores of the responses to each question shown in Figure 8 were also calculated as an index.

[0160] The Simple Occupational Stress Questionnaire is a questionnaire provided by the Ministry of Health, Labor and Welfare that investigates occupational stress levels, and the sum of the scores of all responses was calculated as an index. This questionnaire is shown in Figures 10 and 11. In addition, the sum of the scores of the responses to each of the questions in Area A (questions about work stressors), Area B (questions about stress reactions), and Area C (questions about modifying factors, including support from those around you and satisfaction with work or family life) as well as the sum of the scores of the responses to the questions in Areas A and C were calculated.

[0161] The Chalder Fatigue Scale was prepared using a questionnaire containing the 14 questions shown in Figure 1. The total score of the eight items asking about physical fatigue was calculated as an index of physical fatigue, and the total score of the six items asking about mental fatigue was calculated as an index of mental fatigue.

[0162] The Big Five scale was created using a questionnaire containing 60 questions as shown in Figures 3 and 4. Referring to "Horike et al., 1999," the three items with the highest loadings for each of the five factors (openness, conscientiousness, extraversion, agreeableness, and emotional stability) (items marked with a solid line in Table 4 of the same document) were reversed as necessary to determine the total score so that a higher score indicates stronger extraversion, openness, emotional stability, conscientiousness, and agreeableness. The average score divided by the number of items was used to determine the individual personality scores.

[0163] The Athens Insomnia Scale is a commonly used questionnaire on insomnia tendency, and the total score calculated from all responses was used as an index. Specifically, we used the questionnaire shown in Figure 12, referring to the literature on the Japanese version of the Athens Insomnia Scale (Okajima I, Nakajima S, Kobayashi M, Inoue Y. Psychiatry Clin Neurosci 2013; 67: 420-425.).

[0164] The eating behavior questionnaire, which includes the questions shown in Figure 5, was provided by Sakata et al. and cited in the "Obesity Treatment Guidelines 2006" (Japan Society for the Study of Obesity) (reference: Hironobu Yoshimatsu (1996) Initial Operation "Obesity Treatment Manual," edited by Toshiie Sakata, Ishiyaku Publishing, Tokyo, pp. 17-38). The "Men's" eating behavior questionnaire, described in the same publication, included seven domains (perception of body type and weight, eating motivation, vicarious eating, hunger, satiety, eating habits, meal content, and regularity of eating habits). Responses of "Not at all," "Sometimes at all," "Tendency to be like that," and "Exactly true" were rated on a four-point scale from 0 to 3. The total score for each domain was calculated as a percentage of the maximum possible score.

[0165] The self-efficacy scale for eating behavior, the self-efficacy scale for exercise behavior, and the self-efficacy scale for rest behavior were the same as those described in "Study of Self-Efficacy Scales and Social Support Scales Useful for Supporting Health Behavioral Change" (Kiyoshi Moriya and Mari Shimizu, Bulletin of Tenshi University, Vol. 9, June 30, 2009) (hereinafter referred to as "Moritani et al., 2009"). Figure 6 shows the self-efficacy scale for eating behavior taken from the same publication, and Figure 7 shows the self-efficacy scale for exercise behavior and the self-efficacy scale for rest behavior taken from the same publication. Referring to the publication, participants were asked to select one answer on a seven-point scale, and the sum of the scores for each scale was calculated as the index for each scale.

[0166] The exercise behavior stage of change questionnaire used the "Health Behavior Change Stage Test Paper" shown in Figure 1 of "Moritani et al., 2009" mentioned above. Similarly, the rest behavior change stage questionnaire and the eating behavior change stage questionnaire were prepared by replacing "activity and exercise behavior" with "stress management and rest behavior" and "dietary habits and eating behavior," respectively, from the "Health Behavior Change Stage Test Paper." The "Health Behavior Change Stage Test Paper" taken from the same publication is shown in Figure 9. The indices calculated from the responses to these tests were calculated by referring to Figure 9, and the numbers of the options for questions regarding I. health behavior change awareness and stage were used as indices.

[0167] The questionnaire regarding motivation to increase the amount of exercise consisted of the question, "Please rate your current motivation to participate in this program (to increase the amount of exercise) on a scale of 1 to 10." Participants were asked to respond on an 11-point scale from 0 to 10 (very motivated to not motivated at all). The score corresponding to this response was used as an "index calculated from the responses to the motivation questions."

[0168] The questionnaire about attitudes toward dieting consisted of the question, "Please rate your current attitude toward dieting (weight loss and obesity reduction) on a scale of 1 to 10." Participants were asked to respond on an 11-point scale from 0 to 10 (very high to very low to none). The score corresponding to this response was used as an "index calculated from the responses to questions about attitudes toward dieting."

[0169] Subsequently, each subject in the subject group underwent a two-month exercise intervention. Specifically, a fitness instructor provided each subject with exercise instruction to increase energy expenditure in order to achieve a target weight (a 3% reduction from their current weight). The fitness instructor provided exercise instruction once every two weeks. Each subject was instructed to keep a daily exercise log (diary) and to measure and record their daily activity using a provided activity monitor. In addition, each subject was instructed to install an application program (Asuken (registered trademark)) on their personal information terminal, which allows them to record their dietary habits, and to record their daily dietary habits without consciously changing their dietary habits.

[0170] After two months of exercise intervention, the subjects underwent a post-intervention test, which consisted of the same physical examination, blood tests, and questionnaire tests as the pre-intervention test (excluding the self-efficacy scale and polyphenol questionnaire). After that, the subjects were given no exercise intervention for one month, and then the same tests as the post-intervention test were administered as a follow-up test.

[0171] [Selection of participants for overall analysis and homogeneous analysis] All 44 subjects who completed the two-month exercise intervention and one-month follow-up were selected as subjects for the overall analysis. Furthermore, 36 subjects who were judged to have achieved sufficient exercise volume (more than 50% of the target exercise volume over the entire two-month period) and not have made significant changes in dietary intake were selected as subjects for a homogeneous analysis, which is an analysis of a homogeneous group. Daily exercise records (diaries) were used primarily to determine exercise volume, with activity measurement results used as secondary evidence. Dietary intake changes were determined using the food records in the above application program.

[0172] [Overall analysis of weight loss rate] The weight loss rate was calculated from the data of the subjects (44 people) in the overall analysis using the following formula. Weight loss rate (%) = (weight before intervention (kg) - weight after intervention (kg)) / weight before intervention × 100 The analysis subjects whose weight loss rate was in the top 25% were defined as the responder group (11 subjects), and the analysis subjects whose weight loss rate was in the bottom 25% were defined as the non-responder group (11 subjects).

[0173] For each of the responder and non-responder groups, the average values ​​and standard deviations of the results of the physical examination, blood test, and questionnaire test in the pre-test were calculated. Student's t-tests were performed on the average values ​​of the responder and non-responder groups for each item, and items with p<0.05 were extracted as items with significant differences between the groups. The extracted items correspond to the physical and blood indices in embodiment 1-1. The average values ​​(ave) and standard deviations (sd) of the extracted items for the responder and non-responder groups are shown in Table 1.

[0174] [Table 1]

[0175] [Homogenous analysis of weight loss rate] The weight loss rate was calculated from the data of the subjects (36 subjects) in the homogeneous analysis described above, in the same manner as in the overall analysis. The group of subjects whose weight loss rate was in the top 25% was defined as the responder group (9 subjects), and the group of subjects whose weight loss rate was in the bottom 25% was defined as the non-responder group (9 subjects). Next, as in the overall analysis, the average value and standard deviation of the results for each item were calculated for the responder group and the non-responder group, and items with significant differences in average value between the groups were extracted. The extracted items correspond to the blood indices in embodiments 1-2. The average value (ave) and standard deviation (sd) of the extracted items for the responder group and the non-responder group are shown in Table 2.

[0176] [Table 2]

[0177] [Overall analysis of visceral fat area reduction] The visceral fat area reduction was calculated from the data of the subjects (44 subjects) in the overall analysis by subtracting the visceral fat area value after the intervention from the visceral fat area value before the intervention. The analysis subjects in the top 25% of the visceral fat area reduction amount were defined as the Responder group (11 subjects), and the analysis subjects in the bottom 25% were defined as the Non-Responder group (11 subjects). Next, as in the overall analysis of weight loss rate, the mean and standard deviation of each item were calculated for the Responder group and the Non-Responder group, and items with significant differences in mean values ​​between the groups were extracted. The extracted items correspond to the physical index, blood index, and response index of embodiment 2-1. The mean (ave) and standard deviation (sd) of the extracted items for the Responder group and the Non-Responder group are shown in Table 3.

[0178] [Table 3]

[0179] [Homogenous analysis of visceral fat area reduction] The data from the subjects (36 subjects) in the homogeneous analysis were used to calculate the reduction in visceral fat area, as in the overall analysis. The subjects in the top 25% of subjects with a reduction in visceral fat area were defined as the Responder group (9 subjects), and the subjects in the bottom 25% were defined as the Non-Responder group (9 subjects). Next, as in the overall analysis, the mean and standard deviation of the results for each item were calculated for the Responder group and the Non-Responder group, and items with significant differences in mean values ​​between the groups were extracted. The extracted items correspond to the physical indicators, blood indicators, and response indicators in embodiment 2-2. The mean values ​​(ave) and standard deviations (sd) of the extracted items for the Responder group and the Non-Responder group are shown in Table 4.

[0180] [Table 4]

[0181] <Test Example 2> [Exercise intervention trial] Next, BMI is 23 kg / m 2 Fifty-seven adult males (aged 20 to 60 years) who met the same eligibility criteria as in Test Example 1 were selected, and a physical examination, blood test, and questionnaire test were conducted as pre-intervention tests before the exercise intervention. These tests were the same as those in Test Example 1. The tests in this test example were conducted with the consent of the subjects.

[0182] Subsequently, each newly selected subject underwent a two-month exercise intervention identical to that in Test Example 1. After the two-month exercise intervention, the subjects underwent post-intervention tests similar to those in the pre-intervention tests, including physical examinations, blood tests, and questionnaire tests (excluding the self-efficacy scale and polyphenol questionnaire tests).

[0183] [Selection of participants for overall analysis and homogeneous analysis] The 44 subjects selected in Test Example 1 and the 53 subjects who completed the two-month exercise intervention in Test Example 2 were combined to form the subject group for Test Example 2. All of this subject group were selected as subjects for the overall analysis in Test Example 2. Furthermore, from the subject group for the overall analysis, 77 subjects who were judged to have achieved sufficient exercise volume (more than 50% of the target exercise volume over the entire two-month period) and not have significantly changed their dietary intake were selected as subjects for a homogeneous analysis, which is an analysis of a homogeneous group. Daily exercise records (diaries) were primarily used to determine the amount of exercise, with activity measurement results used as secondary evidence. Changes in dietary intake were determined using the dietary records in the application program.

[0184] [Overall analysis of weight loss rate] Weight loss rates were calculated from the data of the subjects (97 people) in the overall analysis, as in Test Example 1. Using the data from the results of each test item as explanatory variables and the calculated weight loss rate data as the objective variable, an explanatory variable selection algorithm was constructed to generate a machine learning model that predicts weight loss rates. The Elastic Net algorithm was specified as the method and constructed, and the optimal values ​​of the hyperparameters were tuned using 5-fold cross-validation to minimize the error between the predicted value and the actual measured value (e.g., root mean square error (RMSE)). The explanatory variables (prediction markers) of the prediction model selected in this way are shown in Table 5.

[0185] [Table 5]

[0186] The selected indicators were classified into Priorities 1 to 4 as shown in Table 5. Priority 1 was assigned to indicators that belong to the category of physical indicators and response indicators as physical measurements that are minimally invasive and easily obtainable. Priority 2 was assigned to blood indicators included in test items in regular health checkups. Priority 3 was assigned to values ​​measured in amino acid analysis (amino acid fractions) of plasma samples. Priority 4 was assigned to indicators classified as "other," such as items measured using specific equipment, test items with higher precision than those in general regular health checkups (such as test items performed in comprehensive medical checkups), and special blood test items.

[0187] Next, indicators were selected according to priority, and linear regression models 1 to 3 were similarly constructed using the selected indicators as explanatory variables. Model 4 was a model using all the indicators shown in Table 5 above as explanatory variables. In Model 1, systolic blood pressure and "Chalda: Mental fatigue" (an index calculated from the Chalda mental fatigue scale) were selected from Priority 1 shown in Table 5 as explanatory variables. Model 2 selected all indicators of Priority 1 and 2 shown in Table 5 as explanatory variables. Model 3 selected all indicators of priorities 1, 2, and 3 shown in Table 5 as explanatory variables.

[0188] A correlation analysis of the predicted and measured weight loss rates using each model was performed using Spearman's correlation test. The correlation coefficients were -0.261 for Model 1, 0.427 for Model 2, 0.558 for Model 3, and 0.594 for Model 4, all of which had absolute values ​​of 0.2 or greater. These results demonstrate that weight loss rates can be accurately predicted by generating a prediction model using the indicators shown in Table 5.

[0189] [Overall analysis of visceral fat area reduction] The reduction in visceral fat area was calculated from the data of the subjects (97 people) in the overall analysis, as in Test Example 1. Using the data from the results of each test item as explanatory variables and the calculated data on the reduction in visceral fat area as the objective variable, an explanatory variable selection algorithm was constructed to generate a machine learning model that predicts the reduction in visceral fat area. The Elastic Net algorithm was specified as the method and constructed, and the optimal values ​​of the hyperparameters were tuned using 5-fold cross-validation to minimize the error between the predicted value and the actual measured value (e.g., root mean square error (RMSE)). The explanatory variables (prediction markers) of the prediction model selected in this way are shown in Table 6.

[0190] [Table 6]

[0191] The selected indicators were classified into priorities 1 to 4 as shown in Table 6. Next, indicators were selected according to the priority, and linear regression models 5 to 10 were similarly constructed using the selected indicators as explanatory variables. Note that a model using all of the indicators shown in Table 6 above as explanatory variables was designated Model 11. In Model 5, pulse rate and "Factor 5: Exercise, 35 Dietary Questions" (an index calculated from responses to questions asking about awareness and behavior regarding lack of exercise) were selected from Priority 1 shown in Table 6 as explanatory variables. For Model 6, waist circumference, "Factor 3 of 35 Dietary Habits: Mealtime" (an index calculated from responses to questions asking about awareness and / or behavior regarding irregular mealtimes), and "Factor 5 of 35 Dietary Habits: Exercise" were selected from Priority 1 shown in Table 6 as explanatory variables. In Model 7, based on Priority 1 shown in Table 6, in addition to the explanatory variables selected in Model 6, the following were selected as explanatory variables: "Eating Behavior Questionnaire: Dietary Content" (an index of dietary content calculated from responses to the Eating Behavior Questionnaire), "Chalda: Mental Fatigue," and "Big 5: Conscientiousness Personality Score" (an index of conscientiousness personality calculated from responses to the Big Five scale). Model 8 selected all indicators of priority 1 shown in Table 6 as explanatory variables. Model 9 selected all indicators of Priority 1 and 2 shown in Table 6 as explanatory variables. Model 10 selected all indicators of priorities 1, 2, and 3 shown in Table 6 as explanatory variables.

[0192] A correlation analysis of the predicted and measured values ​​of visceral fat area reduction predicted by each model was performed using Spearman's correlation test. As a result, the correlation coefficient values ​​were 0.204 for model 5, 0.291 for model 6, 0.293 for model 7, 0.378 for model 8, 0.344 for model 9, 0.402 for model 10, and 0.594 for model 11, all of which had absolute values ​​of 0.2 or greater. From these results, it was found that by generating a prediction model using the indicators shown in Table 6, it is possible to accurately predict the amount of visceral fat area reduction.

[0193] [Homogenous analysis of weight loss rate] Weight loss rates were calculated from the data of the subjects (77 people) in the homogeneous analysis described above, in the same way as in the overall analysis. Using the results of each test item as explanatory variables and the calculated weight loss rate data as the target variable, an explanatory variable selection algorithm was constructed to generate a machine learning model for predicting weight loss rates. The Elastic Net algorithm was specified as the method and constructed, and the optimal values ​​of the hyperparameters were tuned using five-fold cross-validation to minimize the error between the predicted value and the actual measured value (e.g., root mean square error (RMSE)). The explanatory variables (predictive markers) of the prediction model selected in this way are shown in Table 7.

[0194] [Table 7]

[0195] The selected indicators were classified into priority levels 1 to 4 as shown in Table 7. However, there were no indicators that fell into priority level 3. Next, indicators were selected according to the priority level, and linear regression models 12 to 15 were similarly constructed using the selected indicators as explanatory variables. Model 16 was a model that used all of the indicators shown in Table 7 above as explanatory variables. For Model 12, waist circumference, "Factor 5 of the 35 Dietary Habits Questions: Exercise," and systolic blood pressure were selected from Priority 1 shown in Table 7 as explanatory variables. In Model 13, in addition to the explanatory variables selected in Model 12 from Priority 1 shown in Table 7, "Occupational Stress: B" (an index calculated from responses to a group of questions about stress reactions in the Occupational Stress Simple Questionnaire) was selected as an explanatory variable. Model 14 selected all indicators of priority 1 shown in Table 7 as explanatory variables. Model 15 selected all indicators of Priority 1 and 2 shown in Table 7 as explanatory variables.

[0196] A correlation analysis of the predicted and measured weight loss rates using each model was performed using Spearman's correlation test. The correlation coefficients were 0.318 for Model 12, 0.387 for Model 13, 0.540 for Model 14, 0.693 for Model 15, and 0.713 for Model 16, all of which had absolute values ​​of 0.2 or greater. These results demonstrate that weight loss rates can be accurately predicted by generating a predictive model using the indicators shown in Table 7.

[0197] [Homogenous analysis of visceral fat area reduction] The reduction in visceral fat area was calculated from the data of the subjects (77 people) in the homogeneous analysis described above, in the same way as in the overall analysis. Using the results of each test item as explanatory variables and the calculated reduction in visceral fat area as the objective variable, an explanatory variable selection algorithm was constructed to generate a machine learning model to predict reduction in visceral fat area. The model was constructed using the Elastic Net algorithm as the method, and the optimal values ​​of the hyperparameters were tuned using five-fold cross-validation to minimize the error between the predicted value and the actual measured value (e.g., root mean square error (RMSE)). The explanatory variables (predictive markers) of the prediction model selected in this way are shown in Table 8.

[0198] [Table 8]

[0199] The selected indicators shown in Table 8 were classified into priorities 1 to 4 in the same way as in Table 5. Next, indicators were selected according to the priority, and linear regression models 17 to 21 were similarly constructed using the selected indicators as explanatory variables. Note that a model using all of the indicators shown in Table 8 above as explanatory variables was designated Model 22. For Model 17, from Priority 1 shown in Table 8, "Dietary 35 Questions Factor 5: Exercise" and "WHO-5: Feeling calm and relaxed" (an index calculated from the response to the question "Feeling calm and relaxed" on the WHO-5 (Mental Health Status Questionnaire)) were selected as explanatory variables. For Model 18, from Priority 1 shown in Table 8, "Factor 5 of 35 Diet Questions: Exercise," "Big 5: Conscientiousness Personality Score," and "Self-Efficacy: Stress Coping and Rest Behavior" (an index calculated from the Rest Behavior Self-Efficacy Scale) were selected as explanatory variables. Model 19 selected all indicators of priority 1 shown in Table 8 as explanatory variables. Model 20 selected all indicators of Priority 1 and 2 shown in Table 8 as explanatory variables. Model 21 selected all indicators of priorities 1, 2, and 3 shown in Table 8 as explanatory variables.

[0200] A correlation analysis of the predicted and measured values ​​of visceral fat area reduction predicted by each model was performed using Spearman's correlation test. As a result, the correlation coefficient values ​​were 0.288 for model 17, 0.457 for model 18, 0.408 for model 19, 0.485 for model 20, 0.565 for model 21, and 0.712 for model 22, all of which had absolute values ​​of 0.2 or more. From these results, it was found that by generating a prediction model using the indicators shown in Table 8, it is possible to accurately predict the amount of visceral fat area reduction.

[0201] [Summary] From the above, it was found that the above physical indices, blood indices, and response indices shown in Test Example 1 were indices in which significant differences were confirmed between the responder group and the non-responder group, and that they can be used as predictive markers for predicting the weight loss effect of exercise intervention. Furthermore, since these indices can predict the weight loss effect of exercise intervention individually, it is thought that these indices can be combined or combined with other indices related to weight loss effect to predict the weight loss effect. Furthermore, it was found that the above physical indices, blood indices, and response indices shown in Test Example 2 function as predictive markers that can quantitatively predict the weight loss effect. Furthermore, it was found that the prediction accuracy can be improved by combining these predictive markers.

Claims

1. A method for quantitatively predicting a weight loss effect of an increased amount of exercise in a subject, comprising: Waist circumference, systolic blood pressure, and pulse rate as physical indicators representing at least one of the subject's physique or body composition; As response indices calculated from the subject's responses to the questionnaire, an index calculated from the response to the question item "I felt calm and relaxed" in WHO-5 (Mental Health Questionnaire), an index of mental fatigue calculated from the responses to the Chalder Fatigue Scale, an index calculated from the responses to a questionnaire evaluating the stage of change of rest behavior, an index calculated from the responses to a question asking about motivation to increase the amount of exercise, an index calculated from the responses to a question asking about awareness and / or behavior regarding irregular mealtimes, an index calculated from the responses to a question asking about awareness and behavior regarding lack of exercise, an index about proxy eating calculated from the responses to the eating behavior questionnaire, an index about dietary content calculated from the responses to the eating behavior questionnaire, and an index of conscientiousness personality calculated from the responses to the Big Five scale. An acquisition step of acquiring at least one index selected from the group consisting of: A method for quantitatively predicting weight loss effects.

2. The quantitative prediction of the weight loss effect includes at least one of a quantitative prediction of a weight loss or a quantitative prediction of a visceral fat mass loss, The method for quantitatively predicting the weight loss effect according to claim 1.

3. The quantitative prediction of the weight loss effect is a quantitative prediction of weight loss, The obtaining step further comprises: non-HDL cholesterol concentration, γ-GT value, HDL cholesterol concentration, and hemoglobin value as blood indices measured from a blood sample of the subject; obtaining at least one indicator selected from the group consisting of: The method for quantitatively predicting the weight loss effect according to claim 2.

4. The obtaining step further comprises: hydroxyproline concentration, threonine concentration, glutamine concentration, alanine concentration, citrulline concentration, valine concentration, β-alanine concentration, 1-methylhistidine concentration, tryptophan concentration, and glycine concentration as the blood indicators; obtaining at least one indicator selected from the group consisting of: The method for quantitatively predicting the weight loss effect according to claim 3.

5. The obtaining step further comprises: The physical indicators include visceral fat area, MQP (Muscle Quality Point), and the blood indices, including total protein, potassium concentration, chloride concentration, inorganic phosphorus concentration, platelet count, DHEA-S concentration, total ketone body concentration, acetoacetic acid concentration, 3-hydroxybutyric acid concentration, high-sensitivity CRP concentration, alkaline phosphatase activity, growth hormone concentration, and blood ammonia concentration; obtaining at least one indicator selected from the group consisting of: The method for quantitatively predicting the weight loss effect according to claim 4.

6. The quantitative prediction of the weight loss effect is a quantitative prediction of a reduction in visceral fat mass, The obtaining step further comprises: HDL cholesterol concentration, γ-GT value, and non-HDL cholesterol concentration as blood indices measured from a blood sample of the subject; obtaining at least one indicator selected from the group consisting of: The method for quantitatively predicting the weight loss effect according to claim 2.

7. The obtaining step further comprises: hydroxyproline concentration, threonine concentration, alanine concentration, isoleucine concentration, Fischer ratio, α-amino-n-butyric acid concentration, and methionine concentration as the blood indicators; obtaining at least one indicator selected from the group consisting of: The method for quantitatively predicting the weight loss effect according to claim 6.

8. The obtaining step further comprises: As the physical index, MQP, total body water content, and trunk fat mass; the blood indices, potassium concentration, total bile acid concentration, mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), platelet count, acetoacetic acid concentration, free testosterone concentration, high-sensitivity CRP concentration, growth hormone concentration, and serum iron concentration; obtaining at least one indicator selected from the group consisting of: The method for quantitatively predicting the weight loss effect according to claim 7.

9. The method further includes a prediction step of quantitatively predicting the weight loss effect of the subject by applying at least one index acquired in the acquisition step to a prediction model for deriving a predicted value indicating the weight loss effect from the acquired index. A method for quantitatively predicting a weight loss effect according to any one of claims 1 to 8.

10. A method for quantitatively predicting a weight loss effect of an increase in exercise volume in a subject, comprising: Waist circumference, systolic blood pressure, and pulse rate as physical indicators representing at least one of the subject's physique or body composition; As response indices calculated from the subject's responses to the questionnaire, an index calculated from the response to the question item "I felt calm and relaxed" in WHO-5 (Mental Health Questionnaire), an index of mental fatigue calculated from the responses to the Chalder Fatigue Scale, an index calculated from the responses to a questionnaire evaluating the stage of change of rest behavior, an index calculated from the responses to a question asking about motivation to increase the amount of exercise, an index calculated from the responses to a question asking about awareness and / or behavior regarding irregular mealtimes, an index calculated from the responses to a question asking about awareness and behavior regarding lack of exercise, an index about proxy eating calculated from the responses to the eating behavior questionnaire, an index about dietary content calculated from the responses to the eating behavior questionnaire, and an index of conscientiousness personality calculated from the responses to the Big Five scale. and a prediction step of quantitatively predicting the weight loss effect of the subject by applying at least one index selected from the group consisting of: How to predict weight loss results.

Images