The Lancet Digital Health｜心血管风险评分变化与未来心血

　　The Lancet Digital Health ArticleJuly 01, 2021: 3 (7), e434-e444 https://doi.org/10.1016/S2589-7500(21)00079-0本文由“天纳”临床学术信息人工智能系统自动翻译点击文末“阅读原文”下载本文PDFEvaluation of cardiovascular disease risk in primary care, which is recommended every 5 years in middle-aged and older adults (typical age range 40–75 years), is based on risk scores, such as the European Society of Cardiology Systematic Coronary Risk Evaluation (SCORE) and American College of Cardiology/American Heart Association Atherosclerotic Cardiovascular Disease (ASCVD) algorithms. This evaluation currently uses only the most recent risk factor assessment. We aimed to examine whether 5-year changes in SCORE and ASCVD risk scores are associated with future cardiovascular disease risk.初级保健中的心血管疾病风险评估，建议中老年人（典型年龄范围40-75岁）每5年进行一次，以风险评分为基础，如欧洲心脏病学会系统冠状动脉风险评估（SCORE）和美国心脏病学会/美国心脏协会动脉粥样硬化性心血管疾病（ASCVD）算法。该评估目前仅使用最新的风险因素评估。我们的目的是检验评分和ASCVD风险评分的5年变化是否与未来心血管疾病风险相关。We analysed data from the Whitehall II longitudinal, prospective cohort study for individuals with no history of stroke, myocardial infarction, coronary artery bypass graft, percutaneous coronary intervention, definite angina, heart failure, or peripheral artery disease. Participants underwent clinical examinations in 5-year intervals between Aug 7, 1991, and Dec 6, 2016, and were followed up for incident cardiovascular disease until Oct 2, 2019. Levels of, and 5-year changes in, cardiovascular disease risk were assessed using the SCORE and ASCVD risk scores and were analysed as predictors of cardiovascular disease. Harrell's C index, continuous net reclassification improvement, the Akaike information criterion, and calibration analysis were used to assess whether incorporating change in risk scores into a model including only a single risk score assessment improved the predictive performance. We assessed the levels of, and 5-year changes in, SCORE and ASCVD risk scores as predictors of cardiovascular disease and disease-free life-years using Cox proportional hazards and flexible parametric survival models.我们分析了Whitehall II纵向前瞻性队列研究的数据，这些研究对象无中风、心肌梗死、冠状动脉旁路移植术、经皮冠状动脉介入治疗、明确的心绞痛、心力衰竭或外周动脉疾病史。参与者在1991年8月7日至2016年12月6日期间每隔5年接受一次临床检查，并对心血管疾病事件进行随访，直到2019年10月2日。使用评分和ASCVD风险评分评估心血管疾病风险水平和5年变化，并将其作为心血管疾病的预测因子进行分析。Harrell的C指数、持续净重新分类改进、Akaike信息标准和校准分析用于评估将风险分数的变化纳入模型（仅包括单个风险分数评估）是否提高了预测性能。我们使用Cox比例风险和灵活的参数生存模型，评估评分和ASCVD风险评分的水平和5年变化，作为心血管疾病和无病生存年的预测因子。7574 participants (5233 [69·1%] men, 2341 [30·9%] women) aged 40–75 years were included in analyses of risk score change between April 24, 1997, and Oct 2, 2019. During a mean follow-up of 18·7 years (SD 5·5), 1441 (19·0%; 1042 [72·3%] men and 399 [27·7%] women) participants developed cardiovascular disease. Adding 5-year change in risk score to a model that included only a single risk score assessment improved model performance according to Harrell's C index (from 0·685 to 0·690, change 0·004 [95% CI 0·000 to 0·008] for SCORE; from 0·699 to 0·700, change 0·001 [0·000 to 0·003] for ASCVD), the Akaike information criterion (from 17?255 to 17?200, change ?57 [95% CI ?97 to ?13] for SCORE; from 14?739 to 14?729, change ?10 [–28 to 7] for ASCVD), and the continuous net reclassification index (0·353 [95% CI 0·234 to 0·447] for SCORE; 0·232 [0·030 to 0·344] for ASCVD). Both favourable and unfavourable changes in SCORE and ASCVD were associated with cardiovascular disease risk and disease-free life-years. The associations were seen in both sexes and all age groups up to the age of 75 years. At the age of 45 years, each 2-unit improvement in risk scores was associated with an additional 1·3 life-years (95% CI 0·4 to 2·2) free of cardiovascular disease for SCORE and an additional 0·9 life-years (95% CI 0·5 to 1·3) for ASCVD. At age 65 years, this same improvement was associated with an additional 0·4 life-years (95% CI 0·0 to 0·7) free of cardiovascular disease for SCORE and 0·3 life-years (95% CI 0·1 to 0·5) for ASCVD. These models were developed into an interactive calculator, which enables estimation of the number of cardiovascular disease-free life-years for an individual as a function of two risk score measurements.从1997年4月24日到2019年10月2日，年龄在40-75岁之间的7574名参与者（5233名[69.1%]男性，2341名[30.9%]女性）被纳入了风险评分变化分析。在平均18.7年（标准差5.5）的随访期间，1441名（19.0%；1042名[72.3%]男性和399名[27.7%]女性）参与者患上了心血管疾病。根据Akaike信息标准（从17.3%到17.3%），根据Harrell的C指数（从0.685到0.690，分数变化为0.004[95%可信区间0.000到0.008]，从0.699到0.700，ASCVD变化为0.001[0.000到0.003]），将风险分数的5年变化添加到只包括单一风险分数评估的模型，从而提高了模型的性能?255至17?200，找零?57[95%可信区间?97至?13] 得分；从14分开始?739至14?729，零钱?ASCVD为10[–28至7]，以及连续净重新分类指数（得分为0.353[95%可信区间0.234至0.447]；ASCVD为0.232[0.030至0.344]）。评分和ASCVD的有利和不利变化均与心血管疾病风险和无病生存期相关。在75岁之前的性别和所有年龄组中都可以看到这种关联。在45岁时，风险评分每提高2个单位，ASCVD患者无心血管疾病的寿命增加1.3年（95%可信区间0.4至2.2），无心血管疾病的寿命增加0.9年（95%可信区间0.5至1.3）。在65岁时，同样的改善与ASCVD额外0.4寿命年（95%可信区间0.0至0.7）无心血管疾病评分和0.3寿命年（95%可信区间0.1至0.5）评分相关。这些模型被开发成一个交互式计算器，它能够根据两个风险评分测量值来估计个体的无心血管疾病寿命年数。Changes in the SCORE and ASCVD risk scores over time inform cardiovascular disease risk prediction beyond a single risk score assessment. Repeat data might allow more accurate cardiovascular risk stratification and strengthen the evidence base for decisions on preventive interventions.随着时间的推移，评分和ASCVD风险评分的变化将为心血管疾病风险预测提供信息，而不仅仅是单一风险评分评估。重复数据可能使心血管风险分层更加准确，并加强预防性干预决策的证据基础。UK Medical Research Council, British Heart Foundation, Wellcome Trust, and US National Institute on Aging.英国医学研究委员会、英国心脏基金会、威康信托基金会和美国国家老龄研究所。Guidelines for cardiovascular disease prevention recommend assessment of an individual's future risk to inform decisions on lifestyle and medical interventions.[1],? [2],? [3] The risk of a cardiovascular disease event is computed using risk scores that require entry of data for multiple risk factors, commonly including age, sex, blood pressure, smoking, cholesterol, and diabetes status.[1],? [2],? [3] Although various scores exist and risk thresholds vary between them, the basic principles of prevention are similar across published guidelines.[1],? [2],? [3] Individuals at low risk are advised to maintain a healthy lifestyle to remain at low risk, whereas lifestyle changes are recommended and preventive medication (eg, statins or blood pressure-lowering medication) considered for those at borderline risk. For people identified as being at high risk, both preventive medication and lifestyle changes are indicated.心血管疾病预防指南建议对个人未来风险进行评估，以告知有关生活方式和医疗干预的决定。[1]、[2]、[3]心血管疾病事件的风险是使用风险评分计算的，该评分要求输入多个风险因素的数据，通常包括年龄、性别、血压、，吸烟、胆固醇和糖尿病状态。[1]、[2]、[3]尽管存在不同的得分和风险阈值，但在已发布的指南中，预防的基本原则是相似的。[1]、[2]、[3]建议低风险个体保持健康的生活方式，以保持低风险，然而，生活方式的改变是推荐的，预防性药物（如他汀类药物或降压药物）被考虑用于边缘风险人群。对于被确定为高危人群，需要预防性药物治疗和改变生活方式。Evidence before this study本研究之前的证据Quantifying an individual's probability of developing cardiovascular disease using a risk score is central to disease prevention in middle-aged and older-aged populations (typical age range 40–75 years). Current prevention guidelines recommend health checks every 5 years to evaluate risk and, if needed, appropriate lifestyle interventions are recommended and drug therapies prescribed. Since lifestyle and medical interventions alter cardiovascular disease risk, changes in risk assessment scores following interventions could potentially be used to improve risk stratification compared with a single updated risk assessment alone, as well as to set targets for such interventions. We searched PubMed for research articles, with no language restriction, published up to Oct 30, 2020, using the following search terms: “cardiovascular disease”, “myocardial infarction”, “risk score”, “risk algorithm”, “change”, “improvement”, “reduction”, and “decline”, but found no studies examining the extent to which changes in cardiovascular risk scores are associated with changes in the number of cardiovascular disease-free life-years, or any studies examining an association between changes in risk scores and cardiovascular disease events.在中老年人群（典型年龄范围为40-75岁）中，使用风险评分量化个体罹患心血管疾病的概率是疾病预防的核心。目前的预防指南建议每5年进行一次健康检查，以评估风险，如果需要，建议采取适当的生活方式干预措施，并规定药物治疗。由于生活方式和医疗干预改变了心血管疾病的风险，与单独更新的风险评估相比，干预后风险评估分数的变化有可能用于改善风险分层，并为此类干预设定目标。我们在PubMed上搜索到截至2020年10月30日的研究文章，没有语言限制，使用以下搜索词：“心血管疾病”、“心肌梗死”、“风险评分”、“风险算法”、“改变”、“改善”、“减少”和“下降”，但是没有研究发现心血管风险评分的变化与无心血管疾病生命年数的变化之间的关联程度，也没有研究发现风险评分的变化与心血管疾病事件之间的关联。Added value of this study本研究的附加值In this longitudinal, prospective cohort study of UK adults, 5-year changes in cardiovascular risk scores were associated with future incidence of cardiovascular disease. Both favourable and unfavourable 5-year changes in the European Society of Cardiology Systematic Coronary Risk Evaluation (SCORE) and the American College of Cardiology/American Heart Association Atherosclerotic Cardiovascular Disease (ASCVD) risk scores predicted incident cardiovascular disease events and life-years free of cardiovascular disease. These findings were seen in analyses taking into account competing risk of death, and were observed in both men and women and in different age groups up to the age of 75 years, with the strongest associations apparent in participants aged 40–49 years. For both the SCORE and ASCVD risk scores, adding 5-year change in risk score to a model that included a single risk score assessment improved predictive performance. To facilitate translational research on the use of repeated risk factor measurements in health care, our predictive model was developed into an online extension to the SCORE and ASCVD risk scores, which enables estimation of the number of cardiovascular disease-free life-years for an individual as a function of two risk score measurements.在这项针对英国成年人的纵向前瞻性队列研究中，心血管风险评分的5年变化与未来心血管疾病的发病率相关。欧洲心脏病学会系统性冠状动脉风险评估（SCORE）和美国心脏病学会/美国心脏协会动脉粥样硬化性心血管病（ASCVD）的5年有利和不利变化风险评分可预测发生心血管疾病事件和无心血管疾病的生命年数。这些发现见于考虑到竞争性死亡风险的分析中，并在男性和女性以及75岁以下的不同年龄组中观察到，在40-49岁的参与者中，相关性最强。对于评分和ASCVD风险评分，将风险评分的5年变化添加到包含单一风险评分评估的模型中，可提高预测性能。为了促进在医疗保健中使用重复风险因素测量的转化研究，我们的预测模型被开发成分数和ASCVD风险分数的在线扩展，这使得能够根据两个风险分数测量来估计个体的无心血管疾病生命年数。Implications of all the available evidence所有现有证据的含义Risk assessment using the SCORE or ASCVD risk scores has traditionally been based on a single updated measurement of risk factors. In this longitudinal study of UK adults, 5-year changes in SCORE and ASCVD risk scores improved predictive performance for all age groups up to the age of 75 years. Future studies should examine the validity and generalisability of these findings in other populations and test in clinical practice whether use of risk assessment tools based on repeat measurements is practical, improves risk communication between general practitioners (primary care or family doctors) and patients, and informs setting of realistic target goals for lifestyle and pharmacological interventions. Our interactive online calculator will help to facilitate such translational research.使用评分或ASCVD风险评分的风险评估传统上是基于风险因素的单一更新测量。在这项针对英国成年人的纵向研究中，评分和ASCVD风险评分的5年变化改善了75岁以下所有年龄组的预测性能。未来的研究应检查这些发现在其他人群中的有效性和普遍性，并在临床实践中测试基于重复测量的风险评估工具的使用是否可行，是否改善全科医生（初级保健或家庭医生）与患者之间的风险沟通，并为生活方式和药物干预的现实目标设定提供信息。我们的交互式在线计算器将有助于促进这种转化研究。To facilitate monitoring of disease risk, physicians are recommended to reassess risk levels every 5 years in middle-aged and older adults (aged 40–75 years), with any decisions on preventive interventions based on only the latest risk factor measurement.[1],? [2],? [3] However, lifestyle and medical interventions are predicated on modifying cardiovascular disease risk, and an earlier risk factor measurement can contain useful information about an individual's risk history.[4] Whether changes in these risk scores improve risk stratification relative to a single updated risk assessment is unknown. It is also unclear whether changes in risk scores are associated with cardiovascular disease-free life-years (a measure that might be more useful for risk communication than relative risk estimates) and whether these associations attenuate with increasing age.[5] Better understanding of these issues might facilitate setting of individualised targets for risk factor levels in future health checks and take into account the effect of increasing age.为了便于监测疾病风险，建议医生每5年对中年和老年人（40-75岁）的风险水平进行一次重新评估，并仅根据最新的风险因素测量做出预防性干预决定。[1]、[2]、[3]，生活方式和医疗干预措施是以改变心血管疾病风险为基础的，早期的风险因素测量可以包含关于个体风险史的有用信息。[4]相对于单个更新的风险评估，这些风险评分的变化是否改善了风险分层尚不清楚。还不清楚风险评分的变化是否与无心血管疾病寿命年（一种可能比相对风险估计更有益于风险沟通的指标）相关，以及这些相关性是否随着年龄的增长而减弱。[5]更好地理解这些问题可能有助于在未来的健康检查中为风险因素水平设定个性化目标，并考虑到年龄增长的影响。Accordingly, we aimed to assess whether changes in risk scores computed with the European Society of Cardiology's Systematic Coronary Risk Evaluation (SCORE)[3] and the American College of Cardiology/American Heart Association's Atherosclerotic Cardiovascular Disease (ASCVD) risk algorithms[2]—the two most commonly used risk models—were associated with subsequent cardiovascular disease event rates and life-years free of cardiovascular disease, and whether any associations were affected by increasing age. On the basis of our results, we developed an interactive illustration-of-concept online tool to quantify the effect of changes in risk scores on life-years free of cardiovascular disease.因此，我们旨在评估风险评分的变化是否与欧洲心脏病学会的系统性冠状动脉风险评估（SCORE）[3]和美国心脏病学会/美国心脏协会的动脉粥样硬化性心血管病（ASCVD）风险算法[2]计算的结果一致-两种最常用的风险模型与随后的心血管疾病事件发生率和无心血管疾病的生命年有关，以及是否有任何关联受到年龄增长的影响。根据我们的研究结果，我们开发了一个交互式的在线概念说明工具，用于量化风险评分变化对无心血管疾病生命年的影响。We analysed data from the Whitehall II longitudinal, prospective cohort study, in which all civil servants (government employees) aged 35–55 years based in 20 departments in London, UK, were invited to participate between Sept 10, 1985, and March 29, 1988, and 10?308 (73·0%) of 14?121 agreed.[6] In accordance with current guidelines,[1],? [2],? [3] cardiovascular disease risk factors were assessed at 5-year intervals between: Aug 7, 1991, and May 10, 1993; April 24, 1997, and Jan 8, 1999; Oct 8, 2002, and Sept 10, 2004; Oct 10, 2007, and Nov 18, 2009; Jan 27, 2012, and Oct 30, 2013; and Feb 2, 2015, and Dec 6, 2016. Participants with no history of stroke, myocardial infarction, coronary artery bypass graft, percutaneous coronary intervention, definite angina, heart failure, or peripheral artery disease at baseline were eligible for inclusion in the analysis. As our outcomes were new (incident) cases of cardiovascular disease and life-years free of cardiovascular disease, individuals with prevalent cardiovascular disease were excluded from all analyses.我们分析了Whitehall II纵向前瞻性队列研究的数据，在该研究中，英国伦敦20个部门的所有35-55岁公务员（政府雇员）都被邀请参加1985年9月10日至1988年3月29日期间的研究?308人（73.0%），其中14人?121同意。[6]根据现行指南，[1]、[2]、[3]心血管疾病风险因素在1991年8月7日至1993年5月10日之间每隔5年进行评估；一九九七年四月二十四日及一九九九年一月八日;；二○○二年十月八日及二○○四年九月十日；2007年10月10日和2009年11月18日；2012年1月27日和2013年10月30日；2015年2月2日和2016年12月6日。在基线检查时没有中风、心肌梗死、冠状动脉搭桥术、经皮冠状动脉介入治疗、明确的心绞痛、心力衰竭或外周动脉疾病病史的参与者有资格纳入分析。由于我们的结果是心血管疾病的新（偶发）病例和无心血管疾病的生命年，所有分析均排除患有流行性心血管疾病的个体。For analyses of changes in cardiovascular disease risk scores, we used the clinical assessments in 1991–93 and 1997–99 for individuals aged 40–59 years at baseline, change between 1997–99 and 2002–04 for individuals aged 60–69 years, and change between 2007–09 and 2012–13 for those aged 70–75 years (figure 1). In all analyses, cardiovascular disease event surveillance started after the second risk score assessment and ended on Oct 2, 2019. Research ethics approval was given by the University College London Hospital (London, UK) Committee on the Ethics of Human Research, and participants provided written informed consent at each examination.为了分析心血管疾病风险评分的变化，我们使用了1991-93年和1997-99年对基线年龄为40-59岁的个体的临床评估，1997-99年和2002-04年对60-69岁个体的变化，以及2007-09年和2012-13年对70-75岁个体的变化（图1）。在所有分析中，心血管疾病事件监测在第二次风险评分评估后开始，并于2019年10月2日结束。研究伦理由伦敦大学学院医院（英国伦敦）人类研究伦理委员会批准，参与者在每次检查时提供书面知情同意书。Study designFigure 1Schematic showing analyses of baseline risk scores and changes in risk scores, by age group, and follow-up for cardiovascular disease events.示意图显示按年龄组对基线风险评分和风险评分变化的分析，以及心血管疾病事件的随访。Standard self-administered questionnaires provided data on age, sex, ethnicity, socioeconomic status (high, intermediate, or low), medication type, smoking, physical activity (ideal, intermediate, or poor), and diet (ideal, intermediate, or poor). Ideal physical activity was defined as 150 min or more moderate activity, 75 min or more vigorous activity, or a combination of moderate and vigorous activity for 150 min or more, per week; intermediate activity was defined as 1–149 min moderate activity, 1–74 min vigorous activity, or a combination of moderate and vigorous activity for 1–149 min or more, per week; and poor activity was defined as being physically inactive. Ideal diet was defined as four or more, intermediate as two or three, and poor as none or one of the following dietary items: 4·5 cups or more of fruit and vegetables per day, 3·5 or more 100 g servings of fish per week, three or more 85 g servings of fibre-rich foods per day, sodium consumption less than 1500 mg per day, and 450 kcal or less of sugar-sweetened beverages per week. At baseline, socioeconomic position was approximated by the British civil service occupational grade: a three-level variable representing high (administrative), intermediate (professional or executive), and low (clerical or support) grades. This measure is a comprehensive marker of socioeconomic circumstances and is related to salary, social status, level of responsibility at work, and future pension. Administrative grades in the British civil service represent the highest grades; administrators run the different government departments. Experienced clinical nurses measured height, weight, and systolic and diastolic blood pressure, and took blood samples for lipid and glucose assays.[6] Additionally, participants were instructed to bring all of their medications to each 5-yearly clinical examination.标准的自填问卷提供了有关年龄、性别、种族、社会经济地位（高、中、低）、药物类型、吸烟、体力活动（理想、中等或差）和饮食（理想、中等或差）的数据。理想的体力活动定义为每周150分钟或以上的中等强度活动、75分钟或以上的剧烈活动，或150分钟或以上的中等强度和剧烈活动的组合；中间活动定义为每周1-149分钟中等活动、1-74分钟剧烈活动或1-149分钟或以上中等和剧烈活动的组合；不良活动被定义为身体不活跃。理想饮食定义为四种或四种以上，中间饮食定义为两种或三种，差饮食定义为没有或一种以下饮食项目：每天4.5杯或以上的水果和蔬菜，每周3.5杯或100克以上的鱼，每天三种或85克以上的富含纤维的食物，每天的钠摄入量低于1500毫克，以及每周450千卡或更少的含糖饮料。在基线检查时，社会经济地位近似于英国公务员职业等级：一个代表高级（行政）、中级（专业或行政）和低级（文书或支持）等级的三级变量。该指标是社会经济状况的综合指标，与工资、社会地位、工作责任水平和未来养老金有关。英国公务员的行政级别是最高级别；管理员管理不同的政府部门。经验丰富的临床护士测量身高、体重、收缩压和舒张压，并采集血液样本进行血脂和血糖分析。[6]此外，还指示参与者在每5年一次的临床检查中携带所有药物。The SCORE and ASCVD risk scores were constructed using age, sex, total cholesterol, HDL cholesterol, systolic blood pressure, antihypertensive medication, smoking, and diabetes in accordance with the original descriptions (appendix 1 p 3).[2],? [3] Higher scores denote higher risk. The thresholds for low, borderline, and high cardiovascular disease risk were derived from current guidelines: for SCORE, these corresponded to less than 1%, 1% to less than 5%, and 5% or more; and for ASCVD, these corresponded to less than 5%, 5% to less than 7·5%, and 7·5% or more.[2],? [3] For SCORE, we used the algorithm for a low-risk population because our cohort originated in England, which is classified as a low-risk country according to the original SCORE report.[7]根据原始描述（附录1第3页），使用年龄、性别、总胆固醇、高密度脂蛋白胆固醇、收缩压、抗高血压药物、吸烟和糖尿病构建评分和ASCVD风险评分。[2]，[3]评分越高表示风险越高。低、临界和高心血管疾病风险的阈值来自当前指南：对于分数，这些阈值分别对应于小于1%、1%到小于5%和5%或更多；对于ASCVD，这些分别对应于小于5%、5%到小于7.5%和7.5%或更高的分数。[2]，[3]对于分数，我们对低风险人群使用该算法，因为我们的队列起源于英国，根据原始分数报告，英国被归类为低风险国家。[7]WHO International Classification of Diseases (ICD) codes were retrieved from the National Health Service (NHS) Hospital Episode Statistics database records and from mortality registers using individual NHS identification numbers for linkage.[8] The NHS provides nearly complete comprehensive health-care coverage for all individuals legally resident in the UK. Cardiovascular disease diagnoses, as ascertained via the Hospital Episode Statistics database, have shown 70% sensitivity and 96% specificity against standard biomedical examinations.[8] Incident cardiovascular disease was denoted by stroke (ICD-10 codes under I60, I61, I63, and I64; ICD-9 codes 430, 431, 434, and 436), myocardial infarction (ICD-10 codes under I21; ICD-9 codes under 410), heart failure (ICD-10 codes under I50), definite angina (ICD-10 codes under I20; ICD-9 codes under 410, both verified from medical records), peripheral artery disease (ICD-10 codes I70.2, I73.3, I73.9, I74.3-5, E10.5, E11.5, E12.5, E13.5, and E14.5; ICD-9 codes 250.7, 440.2, 440.4, 443.8-9, 444.2, and 444.81), coronary artery bypass graft (self-reported), or percutaneous coronary intervention (self-reported).WHO国际疾病分类（ICD）代码是从国家卫生局（NHS）医院事件统计数据库记录和死亡率登记簿中检索的，使用单个NHS识别号进行链接。[8]NHS为所有合法居住在英国的个人提供几乎全面的医疗保险。通过医院事件统计数据库确定的心血管疾病诊断与标准生物医学检查相比，显示出70%的敏感性和96%的特异性。[8]突发心血管疾病以中风（I60、I61、I63和I64下的ICD-10代码；ICD-9代码430、431、434和436）、心肌梗死表示（I21项下的ICD-10代码；410项下的ICD-9代码）、心力衰竭（I50项下的ICD-10代码）、明确性心绞痛（I20项下的ICD-10代码；410项下的ICD-9代码，均经病历证实）、外周动脉疾病（ICD-10代码I70.2、I73.3、I73.9、I74.3-5、E10.5、E11.5、E12.5、E13.5和E14.5；ICD-9代码250.7、440.2、440.4、443.8-9、444.2和444.81）、冠状动脉旁路移植术（自我报告）或经皮冠状动脉介入术（自我报告）。Participants were followed up until their first cardiovascular disease event, death, or the end of follow-up (Oct 2, 2019), whichever occurred first. To examine whether incorporating change in risk scores improved the predictive performance of the SCORE and ASCVD risk scores, we used Harrell's C index, continuous net reclassification improvement, the Akaike information criterion (AIC), and calibration analysis.[9],? [10],? [11] A smaller AIC value indicates better fit, and a difference of more than 10 units is considered to provide strong support for a better fit.[12] We used an optimism index to quantify overfitting. Optimism in discrimination and calibration was estimated by drawing 200 repeated bootstrap samples (with replacement) from the original data.参与者被随访至首次心血管疾病事件、死亡或随访结束（2019年10月2日），以先发生者为准。为了检验风险评分的变化是否改善了评分和ASCVD风险评分的预测性能，我们使用了哈雷尔C指数、持续净重分类改进、Akaike信息标准（AIC）和校准分析[9]、[10]、[11]较小的AIC值表示更适合，超过10个单位的差异被认为是更好拟合的有力支持。[12]我们使用乐观指数来量化过度拟合。通过从原始数据中提取200个重复的自举样本（替换样本），估计了对鉴别和校准的乐观态度。We examined the associations of SCORE and ASCVD risk categories (low, borderline, and high) with cardiovascular disease-free life-years, defined as the number of years without cardiovascular disease up to the age of 90 years. In further analyses, we calculated the hazard ratios (HRs) and 95% CIs for the association of change in risk scores with cardiovascular disease-free life-years using flexible parametric survival models. Years free of cardiovascular disease were estimated with change in restricted mean survival times using change in risk score between the ages of 40 years and 75 years. All analyses on change in risk scores were adjusted for baseline cardiovascular disease risk, ethnicity, and socioeconomic status. We used flexible parametric survival models to estimate HRs and 95% CIs for the associations between risk scores and cardiovascular disease events. In all parametric analyses, we first built a Cox proportional hazards model and examined the survival curves, Schoenfeld residuals, and log–log plots to detect any violations in proportionality assumption, and the degrees of freedom needed for the restricted cubic spline function used for the baseline hazard rate and for the potential time-dependent effects. The final model was chosen using the AIC.[12]我们研究了评分和ASCVD风险类别（低、临界和高）与无心血管疾病生命年（定义为90岁之前无心血管疾病的年数）的相关性。在进一步的分析中，我们使用灵活的参数生存模型计算了风险评分变化与无心血管疾病生命年的关联的危险比（HRs）和95%CI。使用40岁至75岁之间风险评分的变化，通过限制平均生存时间的变化来估计无心血管疾病的年数。所有风险评分变化分析均根据基线心血管疾病风险、种族和社会经济状况进行调整。我们使用灵活的参数生存模型来估计风险评分与心血管疾病事件之间的HRs和95%CI。在所有参数分析中，我们首先建立了Cox比例风险模型，并检查了生存曲线、Schoenfeld残差和对数-对数图，以检测比例假设中的任何违规行为，以及用于基线危险率和潜在时间相关效应的受限三次样条函数所需的自由度。使用AIC选择最终模型。[12]We tested the robustness of our findings in several sensitivity analyses. To examine whether our findings were attributable to pharmacological interventions, we repeated the main analysis after excluding individuals who were taking or had initiated risk factor-modifying medication (ie, lipid-lowering, antihypertensive, antidiabetic, or anticoagulation medication) between the surveys. To address potential survival bias, we studied the effect of competing risk of death using Fine and Gray models.[13] Owing to missing risk factor data, we used multiple imputation by chained equations to supplement missing values in change analyses (appendix 1 pp 4–10). The sensitivity to outcome definition was examined by including only major cardiovascular diseases (myocardial infarction and stroke) as the outcome.我们在几个敏感性分析中测试了我们发现的稳健性。为了检验我们的发现是否归因于药物干预，我们在调查期间排除正在服用或已经开始服用危险因素修饰药物（即降脂、降压、降糖或抗凝药物）的个体后，重复了主要分析。为了解决潜在的生存偏差，我们使用精细和灰色模型研究了竞争性死亡风险的影响。[13]由于缺少风险因素数据，我们使用链式方程的多重插补来补充变化分析中的缺失值（附录1第4-10页）。通过仅包括主要心血管疾病（心肌梗死和中风）作为结果来检查对结果定义的敏感性。We derived baseline survival and adjusted coefficients for change in risk for each risk score category between the ages of 40 years and 75 years (appendices 2–4) and then estimated changes in cardiovascular disease-free life-years as a function of changes in SCORE and ASCVD risk scores for continuous age. An extension incorporating this information was integrated into the SCORE and ASCVD risk scores and is available as an interactive online tool. This tool provides an estimate of gained or lost life-years when earlier risk factor measurement is taken into account in addition to the updated risk factor measurement (analysis of risk history), and estimated life-years free of cardiovascular disease in the next risk assessment as a function of anticipated risk factor levels at that time (analysis of targeted change) with accompanying information about how lifestyle changes recommended in current American College of Cardiology/American Heart Association and European Society of Cardiology guidelines[14],? [15],? [16] change risk factor levels (appendix 1 p 11).我们推导了40岁至75岁年龄段各风险评分类别的基线生存率和风险变化调整系数（附录2-4），然后根据评分和ASCVD风险评分的变化估计连续年龄段无心血管疾病生命年的变化。纳入该信息的扩展已整合到评分和ASCVD风险评分中，并作为交互式在线工具提供。除了更新的风险因素测量（风险史分析）外，当考虑早期风险因素测量时，该工具还提供了获得或失去的寿命年的估计值，并在下次风险评估中作为当时预期风险因素水平的函数提供了无心血管疾病的估计寿命年（针对性改变的分析）以及当前美国心脏病学会/美国心脏病协会和欧洲心脏病学会指南[14]、[15]、[16]中建议的生活方式改变如何改变风险因素水平的相关信息（附录1第11页）。We used Stata (version 16.1 MP) and R (version 3.6.0) for statistical analyses and developed the online tool with R (version 3.6.0).我们使用Stata（版本16.1 MP）和R（版本3.6.0）进行统计分析，并使用R（版本3.6.0）开发了在线工具。The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.该研究的资助者在研究设计、数据收集、数据分析、数据解释或报告撰写方面没有任何作用。7996 participants aged 40–63 years between Aug 7, 1991, and May 10, 1993, were included in the baseline analyses (figure 2). Both SCORE and ASCVD risk scores increased over time. This increase was coupled with an increase in systolic blood pressure, use of antihypertensive medication, body-mass index, diabetes, HDL cholesterol, and physical activity, and a decrease in smoking and total cholesterol (table 1). In risk score change analyses, after exclusion of 422 individuals with prevalent cardiovascular disease, those who died, and those who did not attend the clinical visits, 1441 (19·0%; 1042 [72·3%] men, 399 [27·7%] women) of 7574 people remaining in the study (5233 [69·1%] men, 2341 [30·9%] women) were diagnosed with incident cardiovascular disease during a mean follow-up of 18·7 years (SD 5·5). Of the diagnoses, 391 (27·1%) were for myocardial infarction, 288 (20·0%) for coronary artery intervention, 284 (19·7%) for stroke, 233 (16·2%) for definite angina, 173 (12·0%) for heart failure, and 72 (5·0%) for peripheral artery disease. The distribution of risk factors by outcome status after each 5-year survey is presented in appendix 1 (pp 12–16).基线分析包括1991年8月7日至1993年5月10日期间年龄在40-63岁之间的7996名参与者（图2）。评分和ASCVD风险评分均随时间增加。这一增加与收缩压增加、抗高血压药物的使用、体重指数、糖尿病、高密度脂蛋白胆固醇和体力活动以及吸烟和总胆固醇的减少相结合（表1）。在风险评分变化分析中，排除422名患有流行性心血管疾病的患者、死亡患者和未参加临床访问的患者后，7574名研究对象中有1441名（19.0%；1042名[72.3%]男性，399名[27.7%]女性）（5233名[69.1%]男性，2341名[30.9%]女性）平均随访18.7年（SD 5.5），诊断为偶发性心血管疾病。诊断中，391例（27.1%）为心肌梗死，288例（20.0%）为冠状动脉介入治疗，284例（19.7%）为卒中，233例（16.2%）为明确的心绞痛，173例（12.0%）为心力衰竭，72例（5.0%）为外周动脉疾病。附录1（第12-16页）列出了每5年调查后按结果状态划分的风险因素分布情况。Study profileFigure 2全文超过5万字符微信字数限制无法显示全文扫描左侧二维码查看全文?The Lancet Digital Health ArticleJuly 01, 2021: 3 (7), e434-e444 https://doi.org/10.1016/S2589-7500(21)00079-0