Life prediction system for atherosclerotic-free heart disease and application

Abstract

The invention provides a life prediction system for atherosclerotic-free heart disease and application of the life prediction system. The prediction system adopts the application of a composition of reagent materials and / or instruments and equipment of individual risk factor information in preparation of the life prediction system for the atherosclerotic-free heart disease: age, gender, waistline, systolic pressure, whether antihypertensive drugs are used in two weeks, total cholesterol, high-density lipoprotein cholesterol, whether smoking is carried out, and whether the life prediction system for the atherosclerotic-free heart disease is used or not. Whether the patient suffers from diabetes mellitus, residence in cities / villages and residence in south / north, atherosclerotic cardiovascular disease family history, smoking cessation intervention conditions, blood pressure reduction intervention conditions and lipid reduction intervention conditions are determined. The system quantifies the benefit of the life span of the non-atherosclerotic heart disease after individuals in Chinese population receive intervention, reveals the importance of earlier and stricter intervention measures on prolonging the life span of the non-atherosclerotic heart disease, and provides a scientific basis and a practical tool for individualized preventive intervention and reasonable allocation of regional medical resources.

Description

Technical Field

[0001] This invention relates to the field of cardiovascular disease prediction technology. Specifically, this invention relates to the application of a composition of reagents, materials and / or instruments for taking individual risk factor information in the preparation of a life prediction system for patients with atherosclerotic heart disease, and also relates to a life prediction system for patients with atherosclerotic heart disease. Background Technology

[0002] The increased mortality rate and stroke disability rate from ischemic heart disease (IHD) are limiting factors for HLE in the Chinese population. 60The main reason for the increase. Numerous previous studies have also shown that cardiovascular disease (CVD) has consistently been the leading cause of disease burden in China. Therefore, appropriate interventions are needed to prolong the lifespan of individuals with atherosclerotic cardiovascular disease (ASCVD), reduce its disease burden, and alleviate societal medical pressure. Using the Prediction for ASCVD Risk in China (China-PAR) model to predict an individual's 10-year and lifetime risk of ASCVD, active drug treatment and lifestyle interventions can be implemented for high-risk individuals, while active lifestyle modifications should be recommended for low- to intermediate-risk individuals to prevent early onset or occurrence of ASCVD. However, existing ASCVD risk prediction models actually predict the risk of ASCVD in a population with similar risk factor levels over a period of time, failing to reflect changes in ASCVD risk before and after changes in risk factor exposure levels within the same individual. Taking the China-PAR risk scoring model as an example: A 60-year-old male living in a northern city has a systolic blood pressure (SBP) of 140 mmHg, a waist circumference of 100 cm, total cholesterol (TC) of 180 mg / mL, high-density lipoprotein cholesterol (HDL-C) of 60 mg / mL, and diabetes but no other risk factors. At this time, his 10-year ASCVD risk score is 9.7%, which belongs to the intermediate-risk group. However, if this individual wants to know how his 10-year ASCVD risk would change if he actively controlled his blood pressure and used antihypertensive drugs to control his SBP to 130 mmHg, and uses the China-PAR risk scoring model to calculate it, he will find that his 10-year ASCVD risk actually increases to 10.8%, which belongs to the high-risk group. The reason for this "anomaly" is that antihypertensive medication use typically fails to reduce the ASCVD risk in hypertensive individuals to the same level as those who maintain normal blood pressure without medication. Therefore, in models, antihypertensive medication use increases the 10-year or lifetime risk of ASCVD. Other 10-year ASCVD prediction models, such as pooled cohort equations (PCE), suffer from similar problems. Consequently, relying on current risk prediction models cannot effectively predict the lifetime benefits of individualized interventions.

[0003] To accurately assess the individualized benefits of interventions in populations, some scholars have proposed integrating traditional risk scoring models with existing RCT evidence. The intervention effects obtained from RCTs are directly added to the coefficients of the risk scoring model to estimate changes in risk scores after individualized medication. The probability of ASCVD or death in each age group before and after individual intervention is also estimated, thus estimating the ASCVD-free lifespan benefit after intervention. Researchers have used various tools based on European patient populations to estimate the lifetime benefits of individualized interventions, assessing the increase in ASCVD-free lifespan after implementing different intervention regimens. However, the above studies are mainly based on cohort studies and RCT evidence from European populations. Numerous studies have shown that the epidemiological characteristics of ASCVD, the distribution of risk factors, and the effectiveness of interventions in European and American populations differ from those in the Chinese population, limiting the extrapolation of their models to the Chinese population. Currently, there is still a lack of predictive models for extending ASCVD-free lifespan based on evidence from Chinese population studies. Summary of the Invention

[0004] To address the aforementioned problems, the present invention aims to provide a lifespan prediction system and application for atherosclerotic heart disease (ASCVD) free disease, which can quantify the ASCVD-free lifespan benefit of individuals in the Chinese population after intervention.

[0005] Based on large-scale prospective cohort population data, this invention utilizes the guideline-recommended China-PAR ASCVD 10-year and lifetime risk prediction model, incorporates Chinese RCT studies or guideline-recommended intervention evidence, constructs a prediction model applicable to the Chinese population for individualized interventions to prolong ASCVD-free lifespan, and evaluates its ability to predict the risk of individual ASCVD morbidity or all-cause mortality.

[0006] One objective of this invention is to establish a lifespan prediction system and method for patients with atherosclerotic heart disease (ASCVD). This invention uses individual data from the China-PAR cohort study, incorporating intervention parameters such as smoking cessation, lipid-lowering, and blood pressure reduction obtained from randomized controlled trials (RCTs) or meta-data analysis, to establish a lifespan prediction system for ASCVD to quantify the lifetime benefits of different intervention strategies and intervention durations on extending ASCVD-free lifespan.

[0007] Specifically, on one hand, the present invention provides the use of a composition of reagent materials and / or instruments for taking information on the following individual risk factors in the preparation of a life prediction system for patients without atherosclerotic heart disease:

[0008] Age, gender, waist circumference, systolic blood pressure, whether antihypertensive medication has been used within the past two weeks, total cholesterol, high-density lipoprotein cholesterol, smoking status, diabetes status, whether residing in urban / rural areas, whether residing in the south / north, family history of atherosclerosis and cardiovascular disease, smoking cessation interventions, blood pressure reduction interventions, and lipid-lowering interventions.

[0009] According to a specific embodiment of the present invention, in the application of the present invention, the individual is a male individual or a female individual, and the risk factors of the male individual include: age, sex, waist circumference, systolic blood pressure, whether antihypertensive drugs have been used within the past two weeks, total cholesterol, high-density lipoprotein cholesterol, whether smoking, whether having diabetes, whether living in urban / rural areas, whether living in the south / north, family history of atherosclerotic cardiovascular disease, smoking cessation intervention, blood pressure reduction intervention, and lipid reduction intervention;

[0010] The risk factors for the female individuals include: age, sex, waist circumference, systolic blood pressure, whether antihypertensive medication has been used in the past two weeks, total cholesterol, high-density lipoprotein cholesterol, smoking status, diabetes status, whether they live in the south / north, smoking cessation interventions, blood pressure interventions, and lipid-lowering interventions.

[0011] According to specific embodiments of the present invention, in its application, any feasible technology in the art can be used to collect individual risk factor information. For example, measurement methods, or inquiries or questionnaires can be used. In the present invention, the composition of reagents and / or instruments used to collect individual risk factor information includes combinations of various testing reagents and / or instruments used in the process of collecting individual risk factor information, and / or questionnaires used for investigation, and may also include virtual materials and / or instruments (e.g., obtaining relevant information through manual inquiry).

[0012] On the other hand, the present invention also provides a life prediction system for patients with non-atherosclerotic heart disease, which includes a data acquisition unit and a data analysis unit.

[0013] The data acquisition unit is used to collect individual risk factor information; wherein the individual is either male or female; the risk factors for male individuals include: age, sex, waist circumference, systolic blood pressure, whether antihypertensive medication has been used within the past two weeks, total cholesterol, high-density lipoprotein cholesterol, smoking status, diabetes status, whether they live in urban / rural areas, whether they live in the south / north region, family history of atherosclerosis and cardiovascular disease, smoking cessation intervention, blood pressure reduction intervention, and lipid-lowering intervention; the risk factors for female individuals include: age, sex, waist circumference, systolic blood pressure, whether antihypertensive medication has been used within the past two weeks, total cholesterol, high-density lipoprotein cholesterol, smoking status, diabetes status, whether they live in the south / north region, smoking cessation intervention, blood pressure reduction intervention, and lipid-lowering intervention;

[0014] The data analysis unit is used to analyze and process the information collected by the data collection unit to obtain the life expectancy benefit of individuals without atherosclerotic heart disease.

[0015] According to a specific embodiment of the present invention, in the prediction system of the present invention, the data analysis unit uses the Fine-Gray partial distribution risk model to consider the competing risks of non-atherosclerotic cardiovascular disease death and atherosclerotic cardiovascular disease events, and integrates blood pressure lowering intervention parameters, lipid lowering intervention parameters, and smoking cessation intervention parameters, converting the lifetime risk probability of individualized intervention into lifelong benefit.

[0016] The parameters for blood pressure reduction, lipid reduction, and smoking cessation interventions were derived from the results of randomized controlled trials and meta-analysis of the benefits of blood pressure reduction, lipid reduction, and smoking cessation interventions.

[0017] According to a specific embodiment of the present invention, in the prediction system of the present invention, the blood pressure intervention parameter is set with three target values: <130 mmHg; 130 mmHg to <140 mmHg; 140 mmHg to <150 mmHg, wherein the blood pressure parameter of the study subjects whose blood pressure has been controlled within the blood pressure target is set to 1;

[0018] The lipid-lowering intervention parameters were set with two target values: 1.8 mmol / L to <2.6 mmol / L and 2.6 mmol / L to <3.4 mmol / L. Among them, the lipid-lowering parameter for subjects whose LDL-C was already controlled within the lipid-lowering target was set to 1.

[0019] The smoking cessation intervention parameter settings include: setting the intervention parameter for smoking cessation on atherosclerotic cardiovascular disease to 0.60, and setting the parameter for mortality from non-atherosclerotic cardiovascular disease to 0.73.

[0020] According to a specific embodiment of the present invention, in the prediction system of the present invention, the analysis process of the data analysis unit includes the following steps:

[0021] S1: Applying Equation 1 to calculate the lifetime risk benefit of an individual's atherosclerotic heart disease before intervention:

[0022]

[0023] In the lifetime risk model for non-atherosclerotic cardiovascular disease (NACHD) mortality, F(a,t;Z) in Equation 1 refers to the cumulative NACHD mortality risk for an individual with a statistical variable level of Z from age a to age t, where t is at most 85 years old; βZ is the sum of the products of each individual's risk factor Z and its corresponding regression coefficient β, and βZ0 is the sum of the products of the average level of each statistical variable Z0 in the population and its corresponding regression coefficient β; F(a;Z0) and F(t;Z0) are the cumulative incidence rates of NACHD mortality at ages a and t, respectively, when the average level of each risk factor in the population is Z0; βZ0, F(a;Z0), and F(t;Z0) are fixed parameters estimated based on the cohort population; and / or,

[0024] In the lifetime risk model for atherosclerotic cardiovascular disease, F(a,t;Z) in Equation 1 refers to the cumulative risk of developing atherosclerotic cardiovascular disease for an individual with a statistical variable level of Z from age a to age t, where t is at most 85 years old; βZ is the sum of the products of each individual's risk factor Z and the corresponding regression coefficient β, and βZ0 is the sum of the products of the average level of each statistical variable Z0 in the population and the corresponding regression coefficient β; F(a;Z0) and F(t;Z0) are the cumulative incidence rates of atherosclerotic cardiovascular disease at ages a and t, respectively, when the average level of each risk factor in the population is Z0. βZ0, F(a;Z0), and F(t;Z0) are fixed parameters estimated based on the cohort population.

[0025] The prediction results of the lifetime risk model for non-atherosclerotic cardiovascular disease and the lifetime risk model for atherosclerotic cardiovascular disease are added together to obtain the predicted probability of atherosclerotic heart disease or all-cause mortality risk before individual intervention.

[0026] S2: Applying Equation 2 to calculate the lifetime risk of atherosclerotic heart disease after individual intervention:

[0027]

[0028] Among them, the intervention parameters for lowering blood pressure, lowering blood lipids, and smoking cessation include HR. smoke Multiplying these values ​​yields the HR (risk reduction) of the three intervention combinations in reducing the risk of atherosclerotic cardiovascular disease. combine By incorporating the parameters into the model, we obtained lifetime risk models for atherosclerotic cardiovascular disease after intervention and lifetime risk models for non-atherosclerotic cardiovascular disease mortality after intervention, respectively. HR int The parameter HR was incorporated into the lifetime risk model for atherosclerotic cardiovascular disease after intervention. combine Introducing the parameter HR into the lifetime risk model for death from non-atherosclerotic cardiovascular disease smokeThe prediction results of the lifetime risk model for non-atherosclerotic cardiovascular disease and the lifetime risk model for atherosclerotic cardiovascular disease are added together to obtain the predicted probability of atherosclerotic heart disease or all-cause mortality risk after individual intervention.

[0029] S3: Calculate the lifetime risk probability P at age Age. By comparing the difference between the lifetime risk probability of atherosclerotic heart disease after individual intervention and the lifetime risk probability of atherosclerotic heart disease before intervention, calculate the lifetime risk benefit of individualized intervention.

[0030] S4: Convert the lifetime risk probability of individualized intervention into lifelong benefit: Based on Equation 1, let t = a + 1, calculate the 1-year probability of atherosclerotic heart disease p1 and the 1-year probability of death from non-atherosclerotic heart disease p2 for an individual in a given age group; p1 + p2 is the predicted probability of an individual's risk of atherosclerotic heart disease or all-cause mortality; where the baseline age of a healthy individual has a 100% survival probability of non-atherosclerotic heart disease, and the calculation method for the survival probability of non-atherosclerotic heart disease for the next age group is: 100% × (1 - p1 - p2); replace lifetime risk model Equation 1 with lifetime risk model Equation 2 after intervention, and calculate the survival probability of non-atherosclerotic heart disease for an individual after intervention in different age groups; calculate the survival probability of non-atherosclerotic heart disease for each subsequent year in turn, obtain the cumulative survival probability of non-atherosclerotic heart disease for the two age groups before and after intervention, plot Kaplan-Meier curves respectively, calculate the difference in the area under the curves of the two groups, and obtain the lifelong benefit result of non-atherosclerotic heart disease for the individual.

[0031] In this invention, the cohort population consists of four sub-cohorts of the China-PAR project. Specifically, the sub-cohorts include: the China Multi-Center Collaborative Study of Cardiovascular Epidemiology (ChinaMUCA) (1992-1994), which conducted baseline surveys in 1998; ChinaMUCA (1998), which conducted baseline surveys in 1998 and 2000-2001; the International Collaborative Study of Cardiovascular Disease in Asia (InterASIA), which conducted baseline surveys in 1998 and 2000-2001; and the Community Intervention of Metabolic Syndrome in China & Chinese Family Health Study (CIMIC), which conducted baseline surveys in 2007-2008.

[0032] According to a specific embodiment of the present invention, in the prediction system of the present invention, the calculation method of the lifetime risk probability P in step S3 is as follows:

[0033] Let t = Age_Int, calculate the predicted probability of atherosclerotic heart disease or all-cause mortality from the current age Age to the intervention age Age_Int. Predicted probability of lifetime atherosclerotic heart disease or all-cause mortality after intervention age The lifetime risk probability at age Age is calculated, and the formula for calculating the lifetime risk probability P is shown in Equation 3:

[0034]

[0035] The lifespan prediction system for atherosclerotic heart disease of the present invention can be a virtual device, as long as it can realize the functions of the data acquisition unit and the data analysis unit. The data acquisition unit can include various testing reagents and / or testing instruments and equipment; the data analysis unit can be any computing instrument, module, or virtual device that can analyze and process the information from the data acquisition unit to obtain a risk score, or it can be a system that pre-defines data charts and graphs based on various possible score results and corresponding risk levels and / or corresponding health guidance.

[0036] On the other hand, the present invention also provides an electronic device for predicting life expectancy in patients with atherosclerotic heart disease, comprising a first memory, a first processor, and a computer program stored in the first memory and executable on the first processor. When the first processor executes the program, it implements a scoring process including the following steps:

[0037] Receive individual gender information and determine male or female;

[0038] If the individual is determined to be male, the following risk factors information will be obtained: age, sex, waist circumference, systolic blood pressure, whether antihypertensive drugs have been used within the past two weeks, total cholesterol, high-density lipoprotein cholesterol, smoking status, diabetes status, whether the individual lives in an urban / rural area, whether the individual lives in the south / north, family history of atherosclerosis and cardiovascular disease, smoking cessation intervention, blood pressure reduction intervention, and lipid reduction intervention.

[0039] If the individual is identified as female, the following risk factors information will be obtained: age, sex, waist circumference, systolic blood pressure, whether antihypertensive medication has been used within the past two weeks, total cholesterol, high-density lipoprotein cholesterol, smoking status, diabetes status, whether the individual resides in the south / north region, smoking cessation intervention status, antihypertensive intervention status, and lipid-lowering intervention status. Risk factors for male individuals include:

[0040] Based on the obtained individual risk factor information, Equation 1 is applied to calculate the lifetime risk benefit of atherosclerotic heart disease before individual intervention:

[0041]

[0042] In the lifetime risk model for non-atherosclerotic cardiovascular disease (NACHD) mortality, F(a,t;Z) in Equation 1 refers to the cumulative NACHD mortality risk for an individual with a statistical variable level of Z from age a to age t, where t is at most 85 years old; βZ is the sum of the products of each individual's risk factor Z and its corresponding regression coefficient β, and βZ0 is the sum of the products of the average level Z0 of each statistical variable in the population and its corresponding regression coefficient β; F(a;Z0) and F(t;Z0) are the cumulative incidence rates of NACHD mortality at ages a and t, respectively, when the average level of each risk factor in the population is Z0, and are fixed parameters estimated based on the cohort population; and / or,

[0043] In the lifetime risk model for atherosclerotic cardiovascular disease, F(a,t;Z) in Equation 1 refers to the cumulative risk of developing atherosclerotic cardiovascular disease for an individual with a statistical variable level of Z from age a to age t, where t is at most 85 years old; βZ is the sum of the products of each individual's risk factor Z and the corresponding regression coefficient β, and βZ0 is the sum of the products of the average level of each statistical variable Z0 in the population and the corresponding regression coefficient β; F(a;Z0) and F(t;Z0) are the cumulative incidence rates of atherosclerotic cardiovascular disease at ages a and t, respectively, when the average level of each risk factor in the population is Z0, and are fixed parameters estimated based on the cohort population.

[0044] The prediction results of the lifetime risk model for death from non-atherosclerotic cardiovascular disease and the lifetime risk model for atherosclerotic cardiovascular disease are added together to obtain the lifetime risk probability of atherosclerotic heart disease before individual intervention.

[0045] Based on the obtained individual risk factor information, Equation 2 is applied to calculate the lifetime risk of atherosclerotic heart disease after individual intervention:

[0046]

[0047] Among them, the intervention parameters for lowering blood pressure, lowering blood lipids, and smoking cessation include HR. smoke Multiplying these values ​​yields the HR (risk reduction) of the three intervention combinations in reducing the risk of atherosclerotic cardiovascular disease. combine By incorporating the parameters into the model, we obtained lifetime risk models for atherosclerotic cardiovascular disease after intervention and lifetime risk models for non-atherosclerotic cardiovascular disease mortality after intervention, respectively. HR int The parameter HR was incorporated into the lifetime risk model for atherosclerotic cardiovascular disease after intervention. combine Introducing the parameter HR into the lifetime risk model for death from non-atherosclerotic cardiovascular disease smoke The prediction results of the lifetime risk model for death from non-atherosclerotic cardiovascular disease and the lifetime risk model for atherosclerotic cardiovascular disease are added together to obtain the lifetime risk probability of atherosclerotic heart disease after individual intervention.

[0048] The lifetime risk probability P at age Age is calculated as follows:

[0049] Let t = Age_Int, calculate the predicted probability of atherosclerotic heart disease or all-cause mortality from the current age Age to the intervention age Age_Int. Predicted probability of lifetime atherosclerotic heart disease or all-cause mortality after intervention age The lifetime risk probability at age Age is calculated, and the formula for calculating the lifetime risk probability P is shown in Equation 3:

[0050]

[0051] The lifetime risk benefit of individualized intervention is calculated by comparing the difference between the lifetime risk probability of atherosclerotic heart disease after individual intervention and the lifetime risk probability of atherosclerotic heart disease before intervention.

[0052] The lifetime risk probability of individualized intervention is converted into life expectancy benefit: Based on Equation 1, let t = a + 1, calculate the 1-year probability of atherosclerotic heart disease p1 and the 1-year probability of non-atherosclerotic heart disease death p2 for an individual in a certain age group; p1 + p2 is the predicted probability of atherosclerotic heart disease or all-cause mortality risk for an individual; where the baseline age of a healthy individual has a 100% probability of survival without atherosclerotic heart disease, and the calculation method for the survival probability of the healthy individual without atherosclerotic heart disease in the next age group is: 100% × (1 - p1 - p2); replace lifetime risk model Equation 1 with lifetime risk model Equation 2 after intervention, and calculate the survival probability of the individual without atherosclerotic heart disease after intervention in different age groups; calculate the survival probability of the individual without atherosclerotic heart disease for each subsequent year in turn, and obtain the cumulative survival probability of the individual without atherosclerotic heart disease for the two age groups before and after intervention, respectively draw Kaplan-Meier curves, calculate the difference in the area under the curve of the two curves, and obtain the life expectancy benefit of the individual without atherosclerotic heart disease.

[0053] On the other hand, the present invention also provides a non-transitory computer-readable storage medium storing a computer program that, when executed by a processor, implements a prediction system based on the above-described prediction system to predict life expectancy without atherosclerotic heart disease.

[0054] On the other hand, the present invention also provides a computer program product comprising computer instructions that, when executed by a processor, enable the aforementioned prediction system to predict life benefits without atherosclerotic heart disease.

[0055] This invention offers the following beneficial technical effects: The ASCVD-free lifespan prediction system of this invention demonstrates better applicability in the Chinese population compared to the European LIFE-CVD model. Applying this system to a cohort population allows for the quantification of the lifetime benefits of different intervention strategies and intervention durations on extending ASCVD-free lifespan, exploring regional or urban-rural differences in the lifetime benefits brought by different intervention measures. Furthermore, packaging this system into a website provides a user-friendly calculation tool for individuals and physicians to assess the benefits of individual interventions. This invention constructs and validates an individualized intervention lifetime benefit prediction system for the general Chinese population, providing a practical tool for selecting primary prevention intervention strategies and evaluating intervention effects for ASCVD in China. This invention quantifies the changes in ASCVD-free lifespan under different intervention measures, strategies, and intervention durations, suggesting that appropriate intervention measures should be selected in different regions or urban-rural populations, and that timely, early, and strict primary prevention can bring greater lifetime benefits. This finding provides important scientific evidence for guiding the allocation of medical resources in China and achieving precise prevention of ASCVD. This invention is expected to provide more quantitative and intuitive evidence to support the assessment of whether individuals need to receive primary prevention of ASCVD, help to rationally allocate ASCVD prevention and control resources, and promote precise prevention of ASCVD. Attached Figure Description

[0056] Figure 1 Flowchart for including and excluding research subjects.

[0057] Figure 2 The calibration curves of the non-ASCVD lifetime mortality risk prediction model on the training and validation sets.

[0058] Figure 3 Calibration curves for the lifetime risk assessment model for ASCVD morbidity or all-cause mortality on the training and validation sets.

[0059] Figure 4 This is the calibration curve of the lifetime risk assessment model for ASCVD morbidity or all-cause mortality in the LIFE-CVD model on the training and validation sets.

[0060] Figure 5 Forest plot of ASCVD risk for every 10 mmHg reduction in SBP in a Chinese RCT study.

[0061] Figure 6 The probability of survival without ASCVD for 1 year before and after spline smoothing.

[0062] Figure 7 This is an example of using a personalized intervention lifetime benefit prediction model; where, Figure 7 A represents the results of an intervention program starting at age 40, targeting blood pressure <130 mmHg and blood lipids <1.8 mmol / L. Figure 7 B is in Figure 7 Based on A, the results of delaying the intervention start age to after 50 years of age. Figure 7 C is the result of adjusting the blood pressure reduction target to <140mmHg based on A in 7. Detailed Implementation

[0063] In order to provide a clearer understanding of the technical features, objectives and beneficial effects of the present invention, the technical solution of the present invention will now be described in detail below, but it should not be construed as limiting the scope of implementation of the present invention.

[0064] Example

[0065] This embodiment provides a life prediction model for patients without atherosclerotic heart disease and its construction method. The life prediction model for patients without atherosclerotic heart disease was established using individual data from the China-PAR cohort study subjects.

[0066] 1. Model Construction

[0067] 1.1 Baseline Survey

[0068] This embodiment utilizes the baseline survey of 106,281 participants aged 35-74 years at baseline and without ASCVD from four sub-cohorts of the large prospective cohort study China-PAR project (see Section 1.2.3 for the specific inclusion and exclusion process). The specific sub-cohorts include: the China Multi-Center Collaborative Study of Cardiovascular Epidemiology (ChinaMUCA) (1992-1994) with baseline surveys conducted in 1992-1994; ChinaMUCA (1998) with baseline surveys conducted in 1998; the International Collaborative Study of Cardiovascular Disease in Asia (InterASIA) with baseline surveys conducted in 1998 and 2000-2001; and the Community Intervention of Metabolic Syndrome in China & Chinese Family Health Study (CIMIC) with baseline surveys conducted in 2007-2008. When CIMIC conducted its baseline survey, the other three sub-cohorts underwent their first follow-up. Subsequent follow-up was conducted on all four sub-cohorts between 2012 and 2015.

[0069] 1.1.1 Modeling Queue and Validation Queue

[0070] This embodiment uses ChinaMUCA (1998) and InterASIA as modeling cohorts, and validates the constructed models using the ChinaMUCA (1992-1994) and CIMIC cohorts, respectively. The research protocols for all the above sub-cohorts have been reviewed and approved by the Ethics Committee of Fuwai Hospital, Chinese Academy of Medical Sciences. All research subjects fully understood the research objectives and methods before the project began and signed informed consent.

[0071] 1.2 Baseline Survey and Variable Definition

[0072] The baseline surveys for the four cohorts included in this embodiment were conducted using a standardized protocol, with data collection completed by uniformly trained survey teams at hospitals, village clinics, or community health service centers in the study participants' locations. The survey content covered three modules: questionnaire survey, physical examination, and laboratory testing.

[0073] 1.2.1 Questionnaire Survey

[0074] All four cohorts employed a standardized field survey procedure, with questionnaires completed by systematically trained field investigators through one-on-one interviews. The questionnaires primarily collected basic information about the study participants and their exposure to common cardiovascular disease risk factors, specifically including basic demographic characteristics (such as gender, age, and education level), lifestyle risk factors (such as smoking, alcohol consumption, diet, and physical activity), and personal and family medical history (such as a history or family history of hypertension, diabetes, coronary heart disease, stroke, etc.).

[0075] 1.2.2 Physical Examination

[0076] The physical examination was conducted by trained investigators according to standard procedures, including measuring the study subjects' height, weight, blood pressure, and waist circumference.

[0077] 1.2.3 Laboratory Testing

[0078] Laboratory tests mainly include blood glucose, total cholesterol (TC), HDL-C, and triglycerides (TG). This embodiment uses disposable vacuum-sealed EDTA anticoagulant tubes and ordinary serum tubes for peripheral venous blood sample collection. The investigator should instruct subjects to avoid eating after 8 PM the night before sampling to ensure a fasting period of at least 10 hours before blood collection. Two tubes of peripheral venous blood are collected (6 ml each, one tube of anticoagulated blood and one tube of non-anticoagulated blood). These samples should be centrifuged within 3 hours at a local laboratory to separate serum and plasma, and then temporarily stored at -20°C. The ChinaMUCA cohort was tested under strict adherence to the standardized operating procedures established by the Fuwai Hospital Central Laboratory of the Chinese Academy of Medical Sciences. This laboratory provided standardized materials, laboratory consumables, and testing reagents, and established a comprehensive quality supervision system. The InterASIA and CIMIC sub-cohorts, on the other hand, adopted a centralized testing model. Peripheral blood samples from study participants were regularly transported to the Fuwai Hospital Central Laboratory of the Chinese Academy of Medical Sciences via a professional cold chain logistics system. This laboratory has passed the lipid standardization quality assessment and certification of the US Centers for Disease Control and Prevention. Blood samples were analyzed by laboratory technicians using a unified testing platform.

[0079] In this embodiment, fasting blood glucose was measured using the glucose oxidase method, while TC, TG, and HDL-C levels were measured using enzymatic methods. LDL-C was calculated using the Friedewald formula; that is, if the TG concentration of the study subject was <400 mg / dL, then LDL-C was further calculated using the Friedewald formula.

[0080] LDL-C = TC - HDL-C - TG / 5 (Formula 1);

[0081] All blood lipid units are expressed in mg / dL.

[0082] 1.2.4 Definition of Baseline Variables

[0083] In this embodiment, in addition to the laboratory test indicators mentioned above, the main covariates used are consistent with the China-PAR model, including whether the participant is currently smoking (excluding those who have quit smoking, only considering those who continue to smoke), whether they have diabetes, whether they have taken antihypertensive medication within the past two weeks, whether they have a family history of ASCVD, and their place of residence (urban / rural and North / South). Specific definitions are as follows: Whether currently smoking: Participants who smoke at least one cigarette per day for a year or more, or whose cumulative smoking amount reaches 20 packs of cigarettes / at least one kilogram of tobacco leaves, and who still have a smoking habit at the time of the survey, are considered smokers; otherwise, they are defined as non-smokers. Whether they have diabetes: Participants whose fasting blood glucose is ≥126 mg / dL (i.e., 7.0 mmol / L), or who use insulin / oral hypoglycemic agents, are considered to have diabetes; otherwise, they are defined as not having diabetes. Whether they use antihypertensive medication: Participants who have taken antihypertensive medication within the past two weeks are defined as using antihypertensive medication; otherwise, they are defined as not using antihypertensive medication. Family history of ASCVD: Participants who self-report that at least one of their parents or siblings has been diagnosed with ASCVD are defined as having a family history of ASCVD; otherwise, they are defined as not having a family history of ASCVD. Urban and rural areas: Based on the definition of the permanent residence address of the research subjects, urban areas include municipalities directly under the central government and districts under prefecture-level cities, while rural areas include rural counties and county-level cities. North and South: Based on the definition of the permanent residence address of the research subjects.

[0084] 1.3 Cohort follow-up and inclusion / exclusion

[0085] 1.3.1 Follow-up survey

[0086] The follow-up surveys of the four sub-cohorts were conducted strictly according to a unified design and operating manual. Data on basic demographic characteristics, lifestyle risk factors, and personal and family medical history were collected from the study subjects in the same manner as at baseline. Physical examinations and laboratory biochemical tests were also performed using the same protocol as at baseline. The endpoint events were tracked and verified. All endpoint events were submitted to the endpoint review committee established by Fuwai Hospital of the Chinese Academy of Medical Sciences for verification and confirmation. This embodiment is based on the International Classification of Diseases, 10th Revision (1050-1060). thASCVD is defined in the International Classification of Diseases (ICD-10). Death due to IHD includes all fatal outcomes directly or indirectly caused by myocardial infarction or other coronary artery disease. Stroke is defined as having clinical features of subarachnoid hemorrhage, cerebral hemorrhage, or cerebral infarction, manifested as sudden focal or global cerebral dysfunction lasting more than 24 hours, excluding non-vascular precipitates. All participants who died from causes other than ASCVD were considered to have experienced competing risk events. For participants who experienced ASCVD events, the endpoint determination rules were as follows: for those diagnosed with ASCVD, the endpoint was the time of first onset; for those who did not experience ASCVD but died from other causes, the endpoint was the time of death; for those who did not experience ASCVD and survived, the endpoint was the time of the last follow-up. Follow-up duration was defined as the time interval from the baseline survey date to the observation endpoint.

[0087] 1.3.2 Inclusion and Exclusion

[0088] Of the 27,020 participants in the modeling cohorts ChinaMUCA (1998) and InterASIA, 24,334 (90.1%) participated in at least one follow-up. 418 participants with baseline ASCVD and 1,094 participants with substandard blood sample quality were excluded. To ensure the accuracy of follow-up outcomes, participants who were followed up to 2008 but did not participate in the second follow-up in 2012-2015 were further excluded, resulting in a final inclusion of 21,320 participants for the training set. In the validation cohorts ChinaMUCA (1992-1994) (N=14,392) and CIMIC (N=86,428), 14,125 (98.1%) and 80,929 (93.6%) participants, respectively, completed the 2012-2015 follow-up. Participants aged outside the 35-75 age range and those with baseline ASCVD were excluded. Ultimately, two cohorts included 14,123 and 70,838 participants, respectively, to validate the model. The inclusion and exclusion flowcharts for participants are shown below. Figure 1 As shown.

[0089] 1.4 Quality Control

[0090] The four sub-cohorts involved in this embodiment strictly adhered to the unified quality control system led by Fuwai Hospital, Chinese Academy of Medical Sciences, during baseline and follow-up surveys. At the project initiation phase, Fuwai Hospital, Chinese Academy of Medical Sciences, coordinated the development of standardized survey protocols, operation manuals, and disease diagnosis / cause-of-death classification standards. A two-tiered training system was employed during the training phase to ensure standardized implementation. During on-site execution, the quality control team supervised and inspected the standardization of questionnaire surveys, physical examinations, and blood sample collection. Data management employed a dual-person independent data entry system, tracing and verifying inconsistencies in data through original questionnaires, contacting study participants for confirmation when necessary, and integrating hospital medical record information for multi-source cross-validation of morbidity and mortality data. Regarding laboratory quality control, blood sample collection, processing, and transportation strictly followed the standardized operating procedures established by the central laboratory of Fuwai Hospital, Chinese Academy of Medical Sciences. For endpoint event determination, two independent reviewers made initial judgments based on questionnaires, medical records, and death certificates. In case of disagreements, a third reviewer was introduced for joint discussion, and the accuracy of morbidity and mortality was further verified based on medical data.

[0091] 1.5 Statistical Analysis

[0092] In this embodiment, continuous variables were described using mean and standard deviation, while categorical variables were described using frequency (n) and percentage (%). Differences between groups were compared using two-sample t-tests and chi-square tests. All p-values ​​were two-tailed tests, and the statistical significance threshold was set at 0.05. All analyses were performed using R 4.0.2 (R Foundation for Statistical Computing, Vienna, Austria).

[0093] This embodiment is a prospective cohort design with a training set of 21,320 participants and a mean follow-up time of 12.3 years. To determine whether the sample size is sufficient for constructing a lifetime risk prediction model for non-ASCVD mortality as an outcome, this embodiment calculates the minimum sample size required to achieve statistical power based on the sample size calculation formula recommended for clinical prediction models:

[0094]

[0095] Where P represents the predictive factors to be included in the model. Previous studies have considered 12 variables during the modeling process, including age, gender, waist circumference, SBP, hypertension medication, TC, HDL-C, smoking status, diabetes status, north / south orientation, urban / rural location, and family history of ASCVD. Interaction terms between age and the remaining 11 variables were also included. Therefore, this embodiment sets the number of predictive factors to 23; S is the regularization / penalty coefficient (i.e., the number of parameters reduced during modeling to avoid overfitting, typically set to 0.9). It is Cox-Snell R 2The C-statistic is used to represent the expected model performance. A search revealed that the C-statistic (Concordance index) for previous lifetime risk models with ASCVD as the outcome and death from non-ASCVD causes was 0.776 for men and 0.801 for women. Using the model proposed by Riley et al. (Riley RD, Snell KI, Ensor J, et al. Minimum sample size for developing a multivariable prediction model: PART II-binary and time-to-event outcomes. Stat Med, 2019, 38(7): 1276-1296), the sample sizes (10334 men, 10986 women) and the number of events (645 men, 403 women) in previous studies were used, and the C-statistic was converted to... It can calculate the scores for males and females separately. The values ​​are 0.047 and 0.033, respectively. Furthermore, based on Formula 2, the minimum sample sizes required for men and women in this embodiment are calculated to be 4243 and 6227, respectively. In this embodiment, the sample sizes for both men and women meet and exceed the minimum sample sizes required for constructing the prediction model.

[0096] 1.5.1 Construction of ASCVD or All-Cause Mortality Lifetime Risk Model

[0097] To further estimate the ASCVD-free lifespan benefit based on existing ASCVD lifetime risk models, a lifetime risk model for non-ASCVD mortality needs to be constructed. These two models are then combined to form a lifetime risk prediction model for ASCVD or all-cause mortality. To ensure consistency in the prediction models, this embodiment uses the same predictors as previous lifetime ASCVD risk models, namely age, systolic blood pressure, use of antihypertensive medication, smoking status, diabetes status, TC, HDL-C, waist circumference, family history of ASCVD, and residency status (urban / rural, southern / northern). Family history of ASCVD and urban / rural residency are not included in the female model. All continuous variables are transformed using the natural logarithm. The lifetime risk of non-ASCVD mortality is defined as the cumulative risk of developing the disease from baseline age to the endpoint age (set as 85 years). Considering the mutually exclusive relationship between non-ASCVD mortality and ASCVD events, this embodiment uses a Fine-Gray subdistribution hazards model to account for competing risks, constructing a lifetime non-ASCVD mortality model on an age-scale. The following function is constructed for the lifetime risk of non-ASCVD death for each study subject:

[0098]

[0099] Here, F(a,t;Z) refers to the cumulative non-ASCVD mortality risk of a study subject with risk factor level Z from age a to age t, where t is at most 85 years old. βZ is the sum of the products of each risk factor Z and its corresponding regression coefficient β for each study subject, and βZ0 is the sum of the products of the average level Z0 of each risk factor in the population and its corresponding regression coefficient β. F(a;Z0) and F(t;Z0) are the cumulative non-ASCVD mortality rates at ages a and t, respectively, when the average level of each risk factor in the population is Z0. These are fixed parameters estimated based on the cohort population. After determining the risk factor levels of a particular subject, βZ can be calculated, thereby obtaining the lifetime non-ASCVD mortality risk prediction value for that subject.

[0100] 1.5.2 Evaluation of ASCVD or All-Cause Mortality Lifetime Risk Model

[0101] Since the vast majority of participants in the cohort are still alive, it is impossible to directly assess the predictive effectiveness of the lifetime risk model. This embodiment references the method used in the Q-Risk Index (QRISK) study to validate the predictive effectiveness of the lifetime risk model, transforming the assessment of lifetime risk into a model evaluation based on a given follow-up period. Specifically, lifetime risk assesses an individual's risk from baseline age to age 85. In this embodiment, age 85 is replaced with baseline age plus a follow-up period (e.g., 5, 10, or 15 years), thus evaluating the model's predictive effectiveness based on the actual follow-up results of the cohort. This embodiment evaluates the model's predictive ability from two dimensions: discrimination and calibration. Discrimination is the model's ability to accurately predict events from a population, typically measured by the C-statistic. A C-statistic of 0.5 indicates that the predicted probability is similar to random prediction and has no predictive value; the closer the C-statistic is to 1, the higher the model's discrimination. Calibration evaluates the consistency between the model's predicted risk and the actual risk. This embodiment uses a Cox model with corrected competing risks to estimate the actual risk of non-ASCVD mortality. Furthermore, the population is divided into 10 groups according to the predicted non-ASCVD mortality risk from the model. The consistency between the predicted and actual non-ASCVD mortality risks in each group is compared, and the chi-square value and calibration slope are calculated. A smaller chi-square value and a less significant difference (larger P-value) indicate better consistency between the predicted and actual non-ASCVD mortality risks, and a higher model calibration accuracy.

[0102] In the modeling cohort, the model's predictions were compared with actual observed 10-year non-ASCVD mortality to evaluate the discriminative and calibrated properties of the lifetime risk model for non-ASCVD mortality. Simultaneously, a 10×10 cross-validation method was used to assess the robustness of the model's predictions. This involved randomly dividing the cohort population into 10 equal parts, each containing 10% of the original sample. Nine parts were used as the training set to build the model, and one part was used as the validation set to verify the model's discriminative and calibrated properties. This process was repeated 10 times, resulting in a total of 100 model validations. The C-statistic, calibration chi-square, and calibration slope were recorded for each of these 100 validations. The distribution of the mean, standard deviation, and 5th, 50th, and 95th quantiles was used to evaluate the model's internal validity.

[0103] During external validation, the model’s discrimination and calibration were evaluated using 10-year and 15-year observed non-ASCVD deaths in the ChinaMUCA (1992-1994) cohort and 5-year observed non-ASCVD deaths in the CIMIC cohort, respectively.

[0104] Adding the predictions from the non-ASCVD lifetime risk model and the ASCVD lifetime risk model yields either the ASCVD or all-cause mortality lifetime risk model. To test the applicability of the model developed in this embodiment to the Chinese population, it is compared with a LIFE-CVD model for the European population. The LIFE-CVD model is calibrated using age-specific baseline survival rates for the Chinese population, and the calibration degree of the two models is calculated using the Kaplan-Meier multiplication limit method in both the training and validation cohorts, comparing the applicability of the two models in the Chinese population.

[0105] 1.5.3 Setting Individualized Intervention Benefit Parameters and Introducing Intervention Benefit Parameters into the Lifetime Risk Model

[0106] Hypertension, unhealthy diet, high LDL-C, air pollution, and smoking are significant risk factors for the morbidity and mortality of ASCVD. Numerous randomized controlled trials (RCTs) and multinational prospective cohort studies have demonstrated that interventions such as lowering blood pressure, reducing lipids, and quitting smoking can effectively reduce the risk of cardiovascular disease. Therefore, this embodiment introduces key parameters for individualized intervention based on three measures: lowering blood pressure, reducing lipids, and quitting smoking.

[0107] For blood pressure reduction intervention, a combination of keyword and free term search was used to retrieve relevant clinical trials or meta-analyses on blood pressure reduction benefits in the Chinese population from CNKI, Wanfang Data Knowledge Service Platform, PubMed, and Web of Science databases. The search period was from database inception to February 10, 2025. The PubMed database search query was "bloodpressure[Title / abstract]AND China[Title / abstract]", and clinical trials, meta-analyses, randomized controlled trials, and systematic reviews were selected. The latest Chinese hypertension guidelines, including the "Chinese Hypertension Clinical Practice Guidelines 2024", "Chinese Hypertension Prevention and Treatment Guidelines (2024 Revised Edition)", and "Chinese Cardiovascular Disease Primary Prevention Guidelines", were also consulted to further supplement relevant studies in the guideline references. The inclusion criteria for relevant studies included: (1) the study type was a clinical trial or systematic review; (2) the study subjects included the Chinese population; and (3) the study content clearly defined the blood pressure difference between the intervention group and the control group and their corresponding cardiovascular benefits. Exclusion criteria included: (1) studies targeting cardiovascular patients rather than the general population; (2) studies that were not the primary report of an RCT but were post-hoc analyses; and (3) studies where the full text was unavailable. During the literature screening process, the titles and abstracts were read first to exclude obviously irrelevant literature, and then the full text was read to finalize the list of included literature. Data extraction included: publication year, journal, project name, first author, country, population characteristics, age, percentage of males, sample size and number of events in the intervention and control groups, blood pressure in the intervention and control groups, blood pressure difference, median follow-up years, hazard ratio (HR), and 95% confidence interval (CI). The blood pressure (SBP) of the control group after intervention in each RCT study was calculated. ct Blood pressure (SBP) in the intervention group tr The HR and 95% CI of blood pressure reduction were calculated, and their HR and 95% CI (with the upper limit of the interval being UCI and the lower limit being LCI) were converted into the blood pressure reduction effect for every 10 mmHg decrease:

[0108]

[0109] HR after unification 10The 95% CI was aggregated using a meta-analysis, and the blood pressure intervention parameters for the Chinese population were obtained using a random-effects model. This embodiment, referencing the recommendations of the "Chinese Guidelines for the Prevention and Treatment of Hypertension 2024," provides three target blood pressure values ​​for users to choose from: <130 mmHg; 130 mmHg to <140 mmHg; and 140 mmHg to <150 mmHg. For subjects whose blood pressure is already controlled within the target range, the blood pressure parameter is set to 1, and their blood pressure benefit is not estimated further.

[0110] For lipid-lowering intervention, this embodiment refers to the "Chinese Guidelines for Lipid Management 2023" and takes LDL-C control as the main goal of lipid-lowering intervention. Due to limited RCT evidence for lipid-lowering intervention in the general Chinese population, this embodiment sets the intervention parameter to 0.78 for every 1 mmol / L reduction in LDL-C, based on the recommended targets in the lipid guidelines and the results of previous RCT studies used in European population modeling. Three lipid-lowering target values ​​are provided for users to choose from: <1.8 mmol / L (recommended for very high-risk groups, this value is not used in this simulated intervention); 1.8 mmol / L to <2.6 mmol / L; 2.6 mmol / L to <3.4 mmol / L. For subjects whose LDL-C is already controlled within the lipid-lowering target, the lipid-lowering parameter is set to 1, and their lipid-lowering benefit is not estimated.

[0111] For smoking cessation intervention, since smoking cessation affects not only ASCVD but also all-cause mortality, this embodiment refers to the meta-analysis results of previous prospective cohort studies in multiple countries, and sets the intervention parameter for smoking cessation on ASCVD to 0.60 and the parameter for non-ASCVD mortality to 0.73.

[0112] This embodiment combines lowering blood pressure, lowering blood lipids, and smoking cessation (HR). smoke The parameters of the three intervention combinations were multiplied to obtain the impact (HR) on reducing the risk of ASCVD after the three intervention combinations were obtained. combine Incorporating parameters into the model yields lifetime mortality risk models for individuals with and without ASCVD after intervention. The lifetime risk prediction model after intervention is as follows:

[0113]

[0114] Among them, HR int Incorporating the parameter HR into the ASCVD lifetime risk model combine In the non-ASCVD lifetime mortality risk model, the parameter HR is introduced. smoke .

[0115] 1.5.4 Predicting Lifetime Risk and Benefit of Individualized Intervention for ASCVD

[0116] To calculate the lifetime risk benefit of ASCVD after intervention, we need to use the lifetime risk model without intervention (Equation 3) and the lifetime risk model after intervention (Equation 7) respectively. Let t = Age_Int, and calculate the ASCVD risk from the current age Age to the intervention age Age_Int. and lifetime ASCVD risk after intervention age The lifetime risk at age Age can be calculated as 1 minus the probability that no event occurs in either of the two time periods, i.e.:

[0117]

[0118] The lifetime risk benefit of individualized intervention can be calculated by comparing the difference between the lifetime risk after intervention and the original lifetime risk.

[0119] 1.5.5 Predicting ASCVD-free lifespan benefits through individualized intervention

[0120] Using two lifetime risk models—ASCVD and non-ASCVD mortality—based on Equation 3, with t = a + 1, we can calculate the 1-year ASCVD probability p1 and the 1-year non-ASCVD mortality probability p2 for an individual in a given age group. Adding these two probabilities yields the predicted ASCVD or all-cause mortality risk for the individual. For a healthy individual, the ASCVD-free survival probability at baseline age is 100%. Therefore, for the next age group, the ASCVD-free survival probability is calculated as: 100% × (1 - p1 - p2). This process continues until age 85, calculating the ASCVD-free survival probability. Similarly, by replacing the lifetime risk model with the post-intervention lifetime risk model (Equation 7), we can calculate the ASCVD-free survival probability for an individual after intervention in different age groups.

[0121] To illustrate this process more intuitively, we use data from a 40-year-old individual as an example to calculate the 1-year ASCVD probability and 1-year non-ASCVD mortality probability of this individual at different ages. The cumulative ASCVD-free survival probability decreases annually starting from 1 at the current age (40 years old). The calculation process for the ASCVD-free survival probability of each age group using this 40-year-old individual as an example is shown in Table 1. It is important to note that estimating the 1-year ASCVD probability p1 and the 1-year non-ASCVD mortality probability p2 of an individual in a certain age group is affected by the actual number of people in that age group. When the number of people in the age group is small, the estimation results may be inaccurate. This embodiment uses spline functions to smooth the 1-year survival rate of the model to reduce the impact of the small sample size in the older age group on the stability of the results.

[0122] Table 1

[0123]

[0124]

[0125] After obtaining the age-specific cumulative ASCVD-free survival probabilities for both groups before and after individual intervention, Kaplan-Meier curves can be plotted. The difference in the area under the curve between the two groups represents the ASCVD-free lifespan benefit. This embodiment encapsulates this algorithm, allowing users to input actual risk factor exposure levels and intervention goals to calculate the lifetime risk and ASCVD-free lifespan benefit after intervention.

[0126] 2. Data Results

[0127] 2.1 Baseline characteristics and follow-up status

[0128] This embodiment is a prospective cohort design. The training set consisted of 21,320 participants (10,334 males, 48.5%), with a mean follow-up time of 12.3 years. A total of 1,048 ASCVD events were observed (645 males and 403 females), and 1,304 non-ASCVD deaths were observed (841 males and 463 females). The per-year incidence rates of ASCVD for males and females were 514.24 / 100,000 person-years and 293.39 / 100,000 person-years, respectively, while the per-year incidence rates of non-ASCVD deaths were 670.51 / 100,000 person-years and 337.08 / 100,000 person-years, respectively. In the validation cohort ChinaMUCA (1992-1994), there were 14,123 participants (6,565 males, 46.5%), with a mean follow-up time of 17.1 years. Of these, 916 ASCVD events were observed (511 in men and 405 in women), and 1403 non-ASCVD deaths were observed (852 in men and 551 in women). The per-year incidence rates of ASCVD in men and women were 472.54 / 100,000 person-years and 314.41 / 100,000 person-years, respectively, while the per-year incidence rates of non-ASCVD deaths were 787.87 / 100,000 person-years and 427.75 / 100,000 person-years, respectively. In the validation cohort CIMIC, there were 70,838 participants (26,872 in men, or 37.9%). The mean follow-up time was 5.9 years. Of these, 2446 ASCVD events were observed (1204 in men and 1242 in women), and 2370 non-ASCVD deaths were observed (1351 in men and 1019 in women). The per-year incidence rates of ASCVD for men and women were 775.17 / 100,000 and 476.67 / 100,000, respectively, while the per-year mortality rates of non-ASCVD were 869.81 / 100,000 and 391.08 / 100,000, respectively. The gender-segregated baseline characteristics of the modeling and validation cohorts are shown in Table 2. The baseline characteristics of the training and validation cohorts are consistent with previous findings.

[0129] Table 2

[0130]

[0131]

[0132] Table 3 compares the baseline characteristics of individuals with and without ASCVD in the modeling and validation cohorts. Compared to those without ASCVD, the ASCVD-affected individuals generally had an older baseline age, a higher proportion of males, higher SBP, TC, and waist circumference levels, and lower HDL-C levels. They were mostly from rural areas in northern China, had a higher prevalence of smoking, and also had a higher prevalence of diabetes and had taken antihypertensive medication within the past two weeks. Many also had a family history of ASCVD. Table 4 compares the baseline characteristics of individuals with and without non-ASCVD deaths in the modeling and validation cohorts. This shows that the non-ASCVD-affected individuals had an older baseline age, a higher proportion of males, mostly lived in rural areas in southern China, had a higher prevalence of smoking, and also had a higher prevalence of diabetes and had taken antihypertensive medication within the past two weeks. A smaller proportion of them had a family history of ASCVD.

[0133] Table 3

[0134]

[0135] Table 4

[0136]

[0137] 2.2 Construction of a Non-ASCVD Lifetime Mortality Risk Prediction Model

[0138] In this embodiment, the risk factors selected for constructing the lifetime risk prediction model for non-ASCVD mortality are consistent with those selected in the China-PAR model. Therefore, variable screening is not repeated; a Cox proportional hazards model is directly constructed with age as the time scale, non-ASCVD mortality as the outcome, and ASCVD onset as a competing event. The parameters of each risk factor in the male and female models of the China-PAR lifetime risk prediction model for non-ASCVD mortality are shown in Table 5. The model results show that current smokers, those with diabetes, and those living in southern regions have a higher risk of non-ASCVD mortality. Although the effect size is small or statistically insignificant in some genders, overall, SBP is positively correlated with non-ASCVD mortality, while TC and waist circumference are negatively correlated. Among men, those with a family history of ASCVD and those living in urban areas have a lower risk of non-ASCVD mortality.

[0139] Table 5

[0140]

[0141]

[0142] Note: ln, natural logarithm. βZ0, the sum of the products of the regression coefficients of the risk factors in the model and the average levels of each risk factor in the population; NA, the variable was not included in the model.

[0143] 2.3 Validation of the Non-ASCVD Lifetime Mortality Risk Prediction Model

[0144] Table 6 shows the predictive performance of the lifetime risk prediction model for non-ASCVD mortality in the modeling and validation cohorts. During the 10-year follow-up period from baseline, 567 non-ASCVD mortality events were observed in men and 331 in women in the modeling cohort, respectively. After adjusting for competing risks, these figures were 580.4 and 338.3, respectively. The number of events predicted by the lifetime risk model was 574.5 and 315.0, respectively, which is close to the actual observed results. The model C-statistics were 0.734 (95% CI: 0.714–0.754) and 0.753 (95% CI: 0.727–0.779) in men and women, respectively, indicating good discrimination of the model. The model's calibration chi-square values ​​were 5.2 (P = 0.815) and 5.6 (P = 0.778) for men and women, respectively, indicating that the model training results were close to the actual observation results, suggesting good model calibration. The calibration curves of the non-ASCVD lifetime mortality risk prediction model on the training and validation sets are shown below. Figure 2 As shown in the figure. After further simulations of 100 modeling and validations, the results of 10×10 cross-validation show that the mean and median of the C-statistic, calibration chi-square value, and calibration slope of the multiple simulations are close to the initial values ​​directly constructed using the modeling queue, and the standard deviation and quantiles are relatively stable, demonstrating that the lifetime risk model has internal consistency and the estimation results are relatively stable. The 10×10 cross-validation results of the non-ASCVD mortality lifetime risk model modeling queue are shown in Table 7.

[0145] In the external validation cohorts, the model's discrimination and calibration were assessed using non-ASCVD mortality events observed over 10 and 15 years at ChinaMUCA (1992-1994) and over 5 years at CIMIC. Although the model's discrimination decreased in each external validation cohort, the C-statistic remained above 0.6, and the points on the calibration curves were still distributed near the diagonal, indicating good consistency between the model's predicted ASCVD incidence risk and the actual ASCVD incidence observed in the cohorts (Table 6). Figure 2 ).

[0146] Table 6

[0147]

[0148]

[0149] Note: 95% CI, 95% confidence interval.

[0150] Table 7

[0151] Male (N=10034) initial value mean Standard deviation <![CDATA[P5]]> median <![CDATA[P 95 ]]> C statistic 0.734 0.730 0.034 0.674 0.730 0.781 Calibration Chi-square 5.2 8.3 3.6 3.4 7.4 16.4 Calibration slope 1.000 0.954 0.311 0.448 0.926 1.594 Females (N=10986) initial value mean Standard deviation <![CDATA[P5]]> median <![CDATA[P 95 ]]> C statistic 0.753 0.748 0.044 0.670 0.747 0.814 Calibration Chi-square 5.6 6.0 2.9 2.3 5.2 10.5 Calibration slope 1.000 0.885 0.400 0.188 0.890 1.577

[0152] Note: P5, 5th percentile; P95, 95th percentile; ASCVD, atherosclerotic cardiovascular disease.

[0153] 2.4 Validation of ASCVD or All-Cause Mortality Lifetime Risk Prediction Model and Comparison with LIFE-CVD Model

[0154] This embodiment adds the constructed non-ASCVD risk prediction model to the China-PAR ASCVD lifetime risk prediction model to obtain a lifetime risk prediction model for ASCVD or death, and compares this model with the LIFE-CVD model trained on the European population. In terms of model discrimination, the model trained on China-PAR significantly outperforms the LIFE-CVD model on all datasets (P<0.05). The discrimination and calibration of the China-PAR-trained model and the LIFE-CVD model are compared in Table 8.

[0155] Regarding model calibration, the calibration curves in this embodiment, based on the China-PAR cohort, show that the points on the calibration curve are distributed near the diagonal. The calibration curves of the ASCVD morbidity or all-cause mortality lifetime risk assessment model on the training and validation sets are as follows: Figure 3 As shown, the LIFE-CVD model parameters have been calibrated based on the China-PAR cohort. However, even with this calibration, the LIFE-CVD model still shows a significant overestimation of the risk of ASCVD or all-cause mortality in the Chinese population. The calibration curves of the lifetime risk assessment model for ASCVD morbidity or all-cause mortality in the LIFE-CVD model on the training and validation sets are shown below. Figure 4 As shown, SE is the standard error; HR is the hazard ratio; 95% CI is the 95% confidence interval; RCT is the randomized controlled trial; and SBP is the systolic blood pressure. From the numerical values ​​of the calibration curve slopes, the calibration slope of the China-PAR training model is close to 1 in all data, while the calibration slope of the LIFE-CVD model is significantly lower than 1 (Table 8).

[0156] Table 8

[0157]

[0158] 2.5 Individualized Intervention Parameter Estimation and Model Building

[0159] In this embodiment, a total of 702 articles related to antihypertensive research were retrieved. After the inclusion and exclusion process, 8 randomized controlled trials (RCTs) on antihypertensive blood pressure in the general population in China were finally included (the 8 studies include: ① Zhang W, Zhang S, Deng Y, et al. Trial of Intensive Blood-Pressure Control in Older Patients with Hypertension. N Engl J Med, 2021, 385(14): 1268-1279. ② He J, Ouyang N, Guo X, et al. Effectiveness of a non-physician community health-care provider-led intensive blood pressure intervention versus usual care on cardiovascular disease (CRHCP): an open-label, blinded-endpoint, cluster-randomised trial. Lancet, 2023, 401(10380): 928-938. ③ Liu J, Li Y, Ge J, et al. Lowering systolic blood pressure to less than 120 mm Hg versus less than 140mm Hg inpatients with high cardiovascular risk with and without diabetes or previous stroke: an open-label, blinded-outcome, randomized trial. Lancet, 2024, 404 (10449): 245-255. ④ Yuan Y, Jin A, Neal B, et al. Salt substitution and salt-supplyrestriction for lowering blood pressure in elderly care facilities: a cluster-randomized trial. Nat Med,2023,29(4):973-981.⑤Lonn EM,Bosch J,López-JaramilloP,et al.Blood-Pressure Lowering in Intermediate-Risk Persons withoutCardiovascular Disease.N Engl J Med,2016,374(21):2009-2020.⑥Beckett NS,Peters R,Fletcher AE,et al.Treatment of hypertension in patients 80years ofage or older.N Engl J Med,2008,358(18):1887-1898.⑦Wang JG,Staessen JA,GongL,et al.Chinese trial on isolated systolic hypertension in theelderly.Systolic Hypertension in China(Syst-China)Collaborative Group.ArchIntern Med,2000,160(2):211-220.⑧Gong L, Zhang W, Zhu Y, et al.Shanghai trial ofnifedipine in the elderly(STONE).J Hypertens, 1996, 14(10):1237-1245), and quantified the results of each study as the protective effect of a 10 mmHg reduction in SBP on the risk of ASCVD. Due to the heterogeneity among blood pressure reduction studies (I). 2 =73.2%), this embodiment uses a random-effects model to pool the evidence from the RCT. The final results show that in the Chinese population, for every 10 mmHg decrease in SBP, the risk of ASCVD decreases by 26%. The forest plot of ASCVD risk for every 10 mmHg decrease in SBP in the Chinese population RCT study is shown below. Figure 5 As shown, SE is the standard error; HR is the hazard ratio; 95% CI is the 95% confidence interval; RCT is a randomized controlled trial; and SBP is systolic blood pressure.

[0160] In estimating the 1-year ASCVD probability and the 1-year non-ASCVD mortality probability for each age group, this embodiment uses a spline function to smooth the above probabilities for each age group. The 1-year ASCVD-free survival probability before and after spline smoothing is as follows: Figure 6As shown, after smoothing, the mortality probabilities of ASCVD and non-ASCVD in both men and women tend to be more stable across age groups. The smoothed model was used to calculate the cumulative ASCVD-free survival probability for individuals. Combined with individualized intervention parameters such as blood pressure reduction, lipid reduction, and smoking cessation, an individualized intervention lifelong benefit model was constructed, and the algorithm was encapsulated for direct application.

[0161] 3. Model Usage Examples

[0162] Examples of using personalized intervention lifetime benefit prediction models include: Figure 7 As shown. Figure 7 This study simulated the effects of different interventions and intervention goals on prolonging ASCVD-free lifespan in a 40-year-old male with a waist circumference of 90 cm, SBP of 160 mmHg (not taking antihypertensive medication), total cholesterol of 260 mg / dL, HDL-C of 40 mg / dL, LDL-C of 150 mg / dL, residing in a rural area of ​​southern China, currently a non-smoker, and without a family history of diabetes or ASCVD. Among these interventions, Figure 7 A represents the results of interventions starting at age 40, with the target blood pressure <130 mmHg and blood lipids <1.8 mmol / L. Figure 7 B is in Figure 7 Based on A, the results of delaying the intervention start age to after 50 years of age. Figure 7 C is the result of adjusting the blood pressure reduction target to <140mmHg based on A in 7.

[0163] 4. Conclusion

[0164] This embodiment, based on the China-PAR ASCVD lifetime risk model, utilizes the same dataset and incorporates the same variables to predict an individual's lifetime risk of non-ASCVD mortality. Although the risk factors for non-ASCVD mortality and ASCVD differ to some extent, the model developed in this embodiment still exhibits a certain degree of discrimination and calibration on the validation set, maintaining consistency with previous studies. Furthermore, this embodiment merges the two models to form a predictive model for ASCVD morbidity or all-cause mortality. This model demonstrates good calibration in all four sub-cohorts of the China-PAR model. Compared to the European LIFE-CVD model, this embodiment additionally incorporates two variables that are significant in the Chinese population: residence in the north / south region and urban / rural areas. Applying the European model to the Chinese population for risk prediction, despite using age-specific risk calibration in China, shows that the European model performs significantly worse than the model built in this embodiment based on the China-PAR cohort, both in terms of discrimination and calibration. This result highlights the necessity of constructing a lifetime ASCVD benefit model specifically for the Chinese population, helping to clarify the benefits of primary prevention of ASCVD in the Chinese population. Although ASCVD mostly occurs in middle age and later, early progression of ASCVD or subclinical atherosclerosis can also occur in young and middle-aged adults. Studies have shown that preventative interventions at a young age may reduce the probability of ASCVD development. This example quantitatively demonstrates the effect of early intervention on prolonging ASCVD-free life by comparing the differences in interventions received at different ages, suggesting that strengthening primary prevention in young people is a more public health-beneficial measure. Similarly, this example also found that stricter primary prevention standards (intensive blood pressure and lipid-lowering) can also yield greater benefits, which may help in the selection of future primary prevention goals.

[0165] Overall, the predictive model in this embodiment has the following advantages: First, based on the China-PAR ASCVD ten-year and lifetime risk prediction recommended in the guidelines, this embodiment further constructs a non-ASCVD mortality lifetime risk prediction model using the same modeling and validation cohort data, and integrates the two models into an ASCVD morbidity or all-cause mortality prediction model. This not only leverages the advantages of large sample size, long duration, and high-quality follow-up in cohort studies to construct a stable predictive model, but also ensures the continuity of research. It allows for further extension of predictive outcomes to ASCVD morbidity or all-cause mortality without altering existing mature models, and its predictive performance in the Chinese population is significantly better than the LIFE-CVD model used in the European population. Second, this embodiment is the first to construct an assessment framework for individualized intervention lifetime benefits in the Chinese population. Although this embodiment currently only considers three interventions—blood pressure reduction, lipid reduction, and smoking cessation—within this framework, the effect size of the interventions can be updated or more interventions can be included in the future without changing the original model, demonstrating the model's high flexibility. Furthermore, this embodiment transforms the model into a user-friendly measurement tool, which may help improve individual adherence to primary prevention medication or lifestyle interventions, and provide assistance to clinicians in selecting intervention measures. Finally, this embodiment also simulates the lifetime benefits of individuals receiving different interventions, choosing different intervention ages, and adopting different intervention goals at the population level, as well as regional or urban-rural differences, demonstrating that the constructed model has the function of identifying major risk factors, emphasizing the importance of early intervention, and strengthening intervention. This finding suggests that this model can provide a scientific reference for setting goals for primary prevention and the rational allocation of medical resources in China. This embodiment constructs and validates an individualized intervention lifetime benefit prediction model for the general Chinese population, providing a practical tool for selecting primary prevention intervention strategies and evaluating intervention effects for ASCVD in China. The study quantifies the changes in ASCVD-free lifespan under different intervention measures, intervention strategies, and different intervention times, suggesting that appropriate intervention measures should be selected in different regions or urban-rural populations, and that timely, early, and strict primary prevention can bring more lifetime benefits. This finding provides important scientific evidence for guiding the allocation of medical resources in China and achieving precise prevention of ASCVD.

Claims

1. Use of a combination of reagent materials and / or apparatus for taking the following risk factor information of an individual in the preparation of a non-atherosclerotic heart disease life expectancy prediction system: age, gender, waist circumference, systolic blood pressure, whether antihypertensive drugs are used within two weeks, total cholesterol, high-density lipoprotein cholesterol, whether smoking, whether suffering from diabetes, whether living in urban / rural areas, whether living in the south / north, family history of atherosclerotic cardiovascular disease, smoking cessation intervention, antihypertensive intervention and lipid-lowering intervention. The individual is a male individual or a female individual, the risk factors of the male individual include: age, gender, waist circumference, systolic blood pressure, whether antihypertensive drugs are used within two weeks, total cholesterol, high-density lipoprotein cholesterol, whether smoking, whether suffering from diabetes, whether living in urban / rural areas, whether living in the south / north, family history of atherosclerotic cardiovascular disease, smoking cessation intervention, antihypertensive intervention and lipid-lowering intervention; 2. Use according to claim 1, wherein, The risk factors of the female individual include: age, gender, waist circumference, systolic blood pressure, whether antihypertensive drugs are used within two weeks, total cholesterol, high-density lipoprotein cholesterol, whether smoking, whether suffering from diabetes, whether living in the south / north, smoking cessation intervention, antihypertensive intervention and lipid-lowering intervention.

3. A non-atherosclerotic heart disease life expectancy prediction system, comprising a data taking unit and a data analysis unit; The data taking unit is used for taking the risk factor information of an individual; Wherein the individual is a male individual or a female individual; The risk factors of the male individual include: age, gender, waist circumference, systolic blood pressure, whether antihypertensive drugs are used within two weeks, total cholesterol, high-density lipoprotein cholesterol, whether smoking, whether suffering from diabetes, whether living in urban / rural areas, whether living in the south / north, family history of atherosclerotic cardiovascular disease, smoking cessation intervention, antihypertensive intervention and lipid-lowering intervention; The risk factors of the female individual include: age, gender, waist circumference, systolic blood pressure, whether antihypertensive drugs are used within two weeks, total cholesterol, high-density lipoprotein cholesterol, whether smoking, whether suffering from diabetes, whether living in the south / north, smoking cessation intervention, antihypertensive intervention and lipid-lowering intervention; The data analysis unit is used for analyzing and processing the information taken by the data taking unit to obtain the non-atherosclerotic heart disease life expectancy benefit result of the individual. The data analysis unit uses the Fine-Gray partial distribution risk model to consider the competitive risk of non-atherosclerotic cardiovascular disease death and atherosclerotic cardiovascular disease event occurrence, integrates the antihypertensive intervention parameter, the lipid-lowering intervention parameter and the smoking cessation intervention parameter, and converts the life-long risk probability of individualized intervention into life expectancy benefit; 4. The prediction system of claim 3, wherein, The antihypertensive intervention parameter, the lipid-lowering intervention parameter and the smoking cessation intervention parameter come from the randomized controlled trial research results and Meta analysis results of the benefits of antihypertensive intervention, lipid-lowering intervention and smoking cessation intervention. ​ 5. The prediction system of claim 4, wherein, The antihypertensive intervention parameter sets three target values: < 130 mmHg; 130 mmHg ~ < 140 mmHg; 140 mmHg ~ < 150 mmHg, wherein the blood pressure has been controlled within the antihypertensive target, and the antihypertensive parameter of the subject is set to 1; The lipid-lowering intervention parameter sets two target values: 1.8 mmol / L ~ < 2.6 mmol / L; 2.6 mmol / L ~ < 3.4 mmol / L, wherein the LDL-C has been controlled within the lipid-lowering target, and the lipid-lowering parameter of the subject is set to 1; The smoking cessation intervention parameter setting includes: setting the intervention parameter of smoking cessation on atherosclerotic cardiovascular disease to 0.60, and setting the parameter on non-atherosclerotic cardiovascular disease death to 0.

73.

6. The prediction system of any one of claims 3-5, wherein, The analysis process of the data analysis unit includes the following steps: S1: Calculate the individual's atherosclerotic heart disease lifetime risk benefit before intervention by applying formula 1: Wherein, in the non-atherosclerotic cardiovascular disease death lifetime risk model, F(a, t; Z) in formula 1 refers to the cumulative non-atherosclerotic cardiovascular disease death risk of an individual with a statistical variable level of Z from age a to age t, and t is at most 85 years old; βZ is the sum of the product of each individual risk factor Z and the corresponding regression coefficient β, and βZ0 is the sum of the product of the average level of each statistical variable Z0 of the population and the corresponding regression coefficient β; F(a; Z0) and F(t; Z0) are the cumulative incidence rates of non-atherosclerotic cardiovascular disease death at ages a and t, respectively, when the average level of each risk factor of the population is Z0, which are fixed parameters estimated based on the cohort population; and / or, In the atherosclerotic cardiovascular disease lifetime risk model, F(a, t; Z) in formula 1 refers to the cumulative atherosclerotic cardiovascular disease incidence risk of an individual with a statistical variable level of Z from age a to age t, and t is at most 85 years old; βZ is the sum of the product of each individual risk factor Z and the corresponding regression coefficient β, and βZ0 is the sum of the product of the average level of each statistical variable Z0 of the population and the corresponding regression coefficient β; F(a; Z0) and F(t; Z0) are the cumulative incidence rates of atherosclerotic cardiovascular disease at ages a and t, respectively, when the average level of each risk factor of the population is Z0, which are fixed parameters estimated based on the cohort population; Add the prediction results of the non-atherosclerotic cardiovascular disease death lifetime risk model and the atherosclerotic cardiovascular disease lifetime risk model to obtain the prediction probability of the individual's atherosclerotic heart disease or all-cause mortality risk before intervention; S2: Calculate the individual's atherosclerotic heart disease lifetime risk after intervention by applying formula 2: wherein the blood pressure intervention parameter, the lipid intervention parameter, and the smoking cessation intervention parameter HR smoke are multiplied to obtain the effect of the three intervention combinations on reducing the risk of atherosclerotic cardiovascular disease HR combine The parameters are incorporated into a model to obtain an individual's atherosclerotic cardiovascular disease lifetime risk model after intervention and a non-atherosclerotic cardiovascular disease mortality lifetime risk model after intervention, respectively, HR int The parameters are incorporated into a model to obtain an individual's atherosclerotic cardiovascular disease lifetime risk model after intervention and a non-atherosclerotic cardiovascular disease mortality lifetime risk model after intervention, respectively, HR combine The parameters are incorporated into a model to obtain an individual's atherosclerotic cardiovascular disease lifetime risk model after intervention and a non-atherosclerotic cardiovascular disease mortality lifetime risk model after intervention, respectively, HR smoke The non-atherosclerotic cardiovascular disease mortality lifetime risk model and the atherosclerotic cardiovascular disease lifetime risk model are added to obtain an individual's atherosclerotic heart disease or all-cause mortality risk prediction probability after intervention. S3: Calculate the lifetime risk probability P at age Age, by comparing the difference between the individual's atherosclerotic heart disease lifetime risk probability after intervention and the atherosclerotic heart disease lifetime risk probability before intervention, to calculate the individualized intervention lifetime risk benefit; S4: Converting the lifetime risk probability of individualized intervention into life span gain: based on formula 1, let t=a+1, respectively calculate the 1-year atherosclerotic heart disease probability p1 and the 1-year non-atherosclerotic heart disease probability p2 of the individual in an age group; p1+p2 is the predicted probability of the individual atherosclerotic heart disease or all-cause death risk; wherein the atherosclerotic heart disease-free survival probability of a healthy individual at baseline age is 100%, and the atherosclerotic heart disease-free survival probability of the healthy individual at the next age group is calculated as: 100% x (1-p1-p2); replace the lifetime risk model formula 1 with the lifetime risk model formula 2 after intervention to calculate the atherosclerotic heart disease-free survival probability of the individual after intervention in different age groups; calculate the atherosclerotic heart disease-free survival probability of each subsequent year in turn, obtain the age-specific cumulative atherosclerotic heart disease-free survival probability of the individual before and after intervention, respectively draw the Kaplan-Meier curve, calculate the area difference of the two curves, and obtain the atherosclerotic heart disease life span gain result of the individual.

7. The prediction system of claim 6, wherein, The calculation method of the lifetime risk probability P in step S3 is as follows: Let t = Age lnt, compute the predicted probability of atherosclerotic heart disease or all-cause mortality risk from current age, Age, to intervention age, Age lnt and the predicted probability of atherosclerotic heart disease or all-cause mortality risk for life after intervention age and compute the lifetime risk probability at age, Age, the formula for the lifetime risk probability, P, is shown in Equation 3:

8. An atherosclerotic heart disease life span prediction electronic device, comprising a first memory, a first processor, and a computer program stored on the first memory and executable on the first processor, wherein the first processor implements a scoring process comprising the following steps when executing the program: Receiving individual gender information to determine male or female; If male, obtaining the following risk factor information of the individual: age, gender, waist circumference, systolic blood pressure, whether to use antihypertensive drugs within two weeks, total cholesterol, high-density lipoprotein cholesterol, whether to smoke, whether to have diabetes, whether to live in urban / rural areas, whether to live in southern / northern areas, family history of atherosclerotic cardiovascular disease, smoking cessation intervention, antihypertensive intervention, and lipid-lowering intervention; If female, obtaining the following risk factor information of the individual: age, gender, waist circumference, systolic blood pressure, whether to use antihypertensive drugs within two weeks, total cholesterol, high-density lipoprotein cholesterol, whether to smoke, whether to have diabetes, whether to live in southern / northern areas, smoking cessation intervention, antihypertensive intervention, and lipid-lowering intervention; the risk factors of the male individual include: Based on the obtained individual risk factor information, applying formula 1 to calculate the atherosclerotic heart disease lifetime risk gain of the individual before intervention: In the non-atherosclerotic cardiovascular disease death lifetime risk model, F(a, t; Z) in formula 1 refers to the cumulative non-atherosclerotic cardiovascular disease death risk of an individual with a statistical variable level of Z from age a to age t, and t is at most 85 years old; βZ is the sum of the product of each risk factor Z of each individual and the corresponding regression coefficient β; βZ0 is the sum of the product of the average level Z0 of each statistical variable of the population and the corresponding regression coefficient β; F(a; Z0) and F(t; Z0) are the cumulative incidence rates of non-atherosclerotic cardiovascular disease death at ages a and t, respectively, when the average level of each risk factor of the population is Z0; βZ0, F(a; Z0) and F(t; Z0) are fixed parameters estimated based on the cohort population; and / or, In the atherosclerotic cardiovascular disease lifetime risk model, F(a, t; Z) in formula 1 refers to the cumulative atherosclerotic cardiovascular disease incidence risk of an individual with a statistical variable level of Z from age a to age t, and t is at most 85 years old; βZ is the sum of the product of each risk factor Z of each individual and the corresponding regression coefficient β; βZ0 is the sum of the product of the average level Z0 of each statistical variable of the population and the corresponding regression coefficient β; F(a; Z0) and F(t; Z0) are the cumulative incidence rates of atherosclerotic cardiovascular disease at ages a and t, respectively, when the average level of each risk factor of the population is Z0; βZ0, F(a; Z0) and F(t; Z0) are fixed parameters estimated based on the cohort population; The atherosclerotic cardiovascular disease lifetime risk model and the atherosclerotic cardiovascular disease lifetime risk model are added to obtain the atherosclerotic heart disease or all-cause death risk prediction probability of the individual before intervention; Based on the obtained individual risk factor information, the atherosclerotic heart disease lifetime risk of the individual after intervention is calculated by applying formula 2: wherein the blood pressure intervention parameter, the lipid intervention parameter, and the smoking cessation intervention parameter HR smoke are multiplied to obtain the effect of the three intervention combinations on reducing the risk of atherosclerotic cardiovascular disease HR combine The parameters are incorporated into a model to obtain an individual's atherosclerotic cardiovascular disease lifetime risk model after intervention and a non-atherosclerotic cardiovascular disease mortality lifetime risk model after intervention, respectively, HR int The parameters are incorporated into the atherosclerotic cardiovascular disease lifetime risk model after intervention HR combine The parameters are incorporated into the non-atherosclerotic cardiovascular disease mortality lifetime risk model HR smoke The non-atherosclerotic cardiovascular disease mortality lifetime risk model and the atherosclerotic cardiovascular disease lifetime risk model are added to obtain the individual's atherosclerotic heart disease or all-cause mortality risk prediction probability after intervention. The lifetime risk probability P at age Age is calculated, and the calculation method of the lifetime risk probability P is as follows: Let t = Age lnt, compute the predicted probability of atherosclerotic heart disease or all-cause mortality risk from the current age, Age, to the intervention age, Age lnt and the predicted probability of atherosclerotic heart disease or all-cause mortality risk for life after the intervention age and compute the lifetime risk probability at age, Age, the formula for the lifetime risk probability, P, is shown in Equation 3: By comparing the difference between the atherosclerotic heart disease lifetime risk probability of the individual after intervention and the atherosclerotic heart disease lifetime risk probability before intervention, the lifetime risk benefit of individualized intervention is calculated. Converting the life-long risk probability of individualized intervention into life gain: based on formula 1, let t=a+1, respectively calculate the 1-year atherosclerotic heart disease probability p1 and the 1-year non-atherosclerotic heart disease probability p2 of the individual in an age group; p1+p2 is the predicted probability of the individual's atherosclerotic heart disease or all-cause death risk; wherein the atherosclerotic heart disease-free survival probability of a healthy individual at baseline age is 100%, and the atherosclerotic heart disease-free survival probability of the healthy individual to the next age group is calculated as: 100% x (1-p1-p2); replace the life-long risk model formula 1 with the life-long risk model formula 2 after intervention, calculate the atherosclerotic heart disease-free survival probability of the individual after intervention in different age groups; calculate the atherosclerotic heart disease-free survival probability of each subsequent year in turn, obtain the two groups of age-specific cumulative atherosclerotic heart disease-free survival probability of the individual before and after intervention, respectively draw the Kaplan-Meier curve, calculate the area difference of the two curves, and obtain the atherosclerotic heart disease life gain result of the individual. 9.A non-transitory computer readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the prediction system of any one of claims 3-7 to predict the atherosclerotic heart disease life. 10.A computer program product comprising computer instructions, wherein the computer instructions are executed by a processor to implement the prediction system of any one of claims 3-7 to predict the atherosclerotic heart disease life gain.

Images