第一部分 我国心血管代谢疾病流行病学现状分析 研究背景：心血管代谢疾病是目前我国主要的公共卫生问题。遵循健康生活方式在心血管代谢疾病发生发展的各个环节均发挥保护作用。了解我国不同分级心血管代谢疾病流行现状，以及健康生活方式状况，对其进行分级管理将有助于实施早期、持续和有针对性的干预措施。我国尚缺乏近期不同分级心血管代谢疾病比例及不同分级心血管代谢疾病人群健康生活方式状况的数据。 研究方法：本研究基于心血管病高危人群早期筛查与综合干预项目（China Patient-centered Evaluative Assessment of Cardiac Events Million Persons Project，China-PEACE MPP），自2015年至2020年，该项目覆盖31个省、自治区和直辖市，包括170个县和116个区。本研究纳入具有完整的诊断心血管代谢疾病信息的35~75岁社区居民。项目采用统一的仪器和标准完成调查对象的心血管健康状况调查、体格检查和快速血糖和血脂检测。心血管代谢疾病状况依据既往发表的经过验证的心血管代谢疾病分级模型分为0~3级，具体定义为：3级：糖尿病、冠心病或脑卒中；2级：空腹血糖受损或符合代谢综合征诊断；1级：包括一个或两个代谢综合征中危险因素（血压升高、甘油三酯升高、高密度脂蛋白胆固醇降低及腹型肥胖，不包括空腹血糖异常）；0级：没有任何以上心血管病及代谢紊乱。健康生活方式包括5种，分别为健康体型（无普通型肥胖和中心型肥胖）、从不吸烟、目前不饮酒、健康饮食和充分体力活动。评估不同分级心血管代谢疾病比例及不同分级心血管代谢疾病人群健康生活方式遵循状况，并采用逻辑回归模型评估与心血管代谢疾病患病和不同分级心血管代谢疾病人群中遵循健康生活方式相关的因素。 研究结果：本研究共纳入1037116例研究对象。经年龄性别标化后，仅11.1%的研究对象无任何心血管代谢疾病，22.8%、48.7%和17.4%的研究对象分别为心血管代谢疾病1级、2级和3级；代谢综合征患病率为37.3%，糖尿病患病率为15.6%。多元混合模型显示，与患心血管代谢疾病（心血管代谢疾病1级及以上）相关的危险因素包括年龄增长、女性、家庭收入低、职业为农民、未工作或退休、居住于农村和西部地区、目前吸烟或曾经吸烟、目前饮酒、不健康饮食、体重指数升高和筛查时休闲时间中高强度体力活动增加。随着心血管代谢疾病分级升高，遵循3种及3种以上健康生活方式的比例逐渐降低（趋势检验P<0.001），该比例在心血管代谢疾病0级、1级、2级和3级人群中分别为75.8%、56.4%、49.6%和46.5%，而相应的遵循所有5种健康生活方式的比例分别为1.7%、0.9%、0.8%和0.8%。在心血管代谢疾病不同分级人群中，年龄增长、女性、较高的教育程度、较高的家庭收入、职业为退休、居住于城市地区以及居住于东部和中部地区均与遵循5种健康生活方式呈正向关联，而职业为农民呈负向关联，此关联在心血管代谢疾病分级高的人群中更强。 研究结论：在一百万研究对象中，符合代谢健康标准的年龄性别标化比例仅11.1%，且不同分级心血管代谢疾病人群遵循所有5种健康生活方式的比例均不足2%。本研究识别了与心血管代谢疾病患病相关的人群特征及心血管代谢疾病各级人群中与遵循5种健康生活方式相关的人群特征。 第二部分 成年早期至中年期体重增加与代谢综合征患病风险的关联分析 研究背景：既往研究表明，成年早期至中年期体重增加与代谢风险升高有关。然而，在不同成年早期体重指数（body mass index，BMI）类别人群中，成年早期至中年期体重增加与代谢综合征患病之间的剂量反应关系尚不清楚。 研究方法：本研究基于China-PEACE MPP项目。本研究纳入自2015年至2020年，该项目筛查时年龄在35~64岁（定义为中年时期）并回忆25岁时（定义为成年早期）的体重，且自25岁至筛查时体重保持稳定或增加的社区居民。项目采用统一的仪器和标准完成调查对象的心血管健康状况调查、体格检查和快速血糖和血脂检测。25岁BMI类别分别为：体重过低（BMI<18.5 kg/m2）、体重正常（18.5 kg/m2≤BMI<24 kg/m2）和超重/肥胖（BMI≥24 kg/m2）。采用多元逻辑回归模型和限制性立方样条函数，在不同BMI类别人群中，分析成年早期至中年期体重增加与中年期代谢综合征患病风险的剂量反应关联。 研究结果：共纳入437849名研究对象，中位年龄为52岁（四分位间距：46，58），62.1%为女性。在多因素逻辑回归模型中，校正年龄、性别、地理区域、城乡、社会经济学、生活方式特征和合并症，成年早期至中年期体重增加量越大，中年期代谢综合征患病风险越高，与体重稳定（体重增加<2.5 kg）相比，适度体重增加（介于≥2.5和<10.0 kg）、显著体重增加（介于≥10.0和<20.0 kg）和极度体重增加（≥20.0 kg）的研究对象代谢综合征患病的比值比（odds ratio，OR）分别为1.6（95%置信区间：1.57~1.64）、3.85（3.77~3.94）和11.05（10.73~11.38）。成年早期至中年期体重增加与代谢综合征患病风险的关联在成年早期体重过低和体重正常的人群中更强。在成年早期体重过低、体重正常和超重/肥胖的研究对象中，体重每增加5kg，代谢综合征患病的OR值分别为2.01（95%置信区间：1.98~2.05）、1.93（1.92~1.94）和1.67（1.64~1.70）。调整当前BMI后，在体重过低和正常体重的研究对象中，代谢综合征的患病风险依旧随着体重增加而显著增加，尽管这一关联较未校正当前BMI减弱。而在超重/肥胖的研究对象中呈现倒U型关联，代谢综合征患病风险随着体重增加而增加，直到体重增加到12kg左右，然后随着体重增加而降低（非线性P<0.001）。 研究结论：从成年早期至中年期，维持体重可以降低代谢综合征的患病风险，尤其是在体重过低和正常体重的人群中。历史体重增加与代谢综合征患病风险的相关性独立于当前BMI。 第三部分 超重/肥胖年轻成年人长期体重减轻与代谢健康的剂量反应关联 研究背景：已有研究证明，长期体重减轻与超重/肥胖的年轻成年人代谢风险降低有关。然而，年轻成年人长期体重减轻与代谢风险的剂量反应关系尚不确定。 研究方法：本研究基于China-PEACE MPP项目。本研究纳入自2015年至2020年，该项目筛查的年龄在35~64岁并回忆了25岁时体重，且25岁时为超重/肥胖（体重指数≥24 kg/m2），自25岁至筛查时体重保持稳定或减轻的社区居民。项目采用统一的仪器和标准完成调查对象的心血管健康状况调查、体格检查和快速血糖和血脂检测。采用多元逻辑回归模型和限制性立方样条函数，评估25岁至筛查时，长期体重减轻与筛查时代谢综合征患病风险之间的剂量反应关联。 研究结果：共有40150名研究对象（女性占66.4%）被纳入研究。在多因素逻辑回归模型中，校正25岁时的BMI、年龄、性别、地理区域、城乡、社会经济学和生活方式特征后，长期体重减轻量越多，代谢综合征的患病风险越低。与<5%的长期体重减轻相比，长期体重减轻为5~9.9%、10~14.9%、15~19.9%和20%以上的研究对象，代谢综合征患病的OR值分别为0.64（95%置信区间：0.60~0.67）、0.42（0.40~0.45）、0.27（0.25~0.29）和0.15（0.13~0.17）。此外，在按照年龄、性别、BMI类别、吸烟状况、城乡和教育水平分类的所有人群亚组中，长期体重减轻量越多，代谢综合征患病风险均越低。 研究结论：超重/肥胖的年轻成年人的长期体重减轻量与之后代谢综合征的患病风险之间存在单调正向的剂量反应关联。本研究结果提示应探索实现长期体重减轻的方法，以改善超重/肥胖年轻成年人的终生代谢健康状况。

Part I Characteristics of cardiometabolic diseases in Chinese adults Background: Cardiometabolic diseases have been the public health concern in China. Healthy lifestyle behaviors are effective ways to both prevent and manage cardiometabolic diseases. Evaluating the characteristics of the population of different cardiometabolic disease stages, as well as healthy lifestyle behaviors distribution, managing them by different stages, could improve early, sustainable and individualized intervention. However, national data about the proportion of different cardiometabolic disease stages and healthy lifestyle behavior distribution across cardiometabolic disease stages is limited. Method: This study was based on China Patient-Centered Evaluative Assessment of Cardiac Events Million Persons Project (China PEACE MPP), which covered 170 rural counties and 116 urban districts of 31 provinces in China from 2015 to 2020. Participants aged 35–75 years who had complete data on the cardiometabolic disease diagnosis were included. Standardised devices and procedures were used to investistigate cardiovascular-related health status, perform the physical examination and test the rapid blood glucose and rapid blood lipid for all participants. According to the prior published validated cardiometabolic disease staging model, cardiometabolic diseases were classified as: stage 3: diabetes, coronary heart disease or stoke; stage 2: impaired fasting glucose or metabolic syndrome; stage 1: meeting one or two of the metabolic syndrome components (elevated blood pressure, elevated triglycerides, low high-density lipoprotein cholesterol, and central obesity, excluding elevated fasting blood glucose); stage 0: no aforementioned cardiovascular diseases or metabolic disorders. There were 5 healthy lifestyle behaviors including healthy body shape (without general obesity and central obesity), never smoking, non-current alcohol drinking, healthy diet and sufficient physical activity. We assessed rates of cardiometabolic disease stages and distribution of healthy lifestyle behaviors across cardiometabolic disease stages. Multivariable logistic regression models were fitted to assess factors related to cardiometabolic diseases and healthy lifestyle behaviors across cardiometabolic disease stages. Results: A total of 1 037 116 participants were included. After standardizing age and sex, 11.1% participants had no cardiometabolic disorders, and the proportions of participants of cardiometabolic disease stage 1, stage 2 and stage 3 were 22.8%, 48.7% and 17.4%, respectively. The standerdised prevalence of metabolic syndrome and diabetes was 37.3% and 15.6%, respectively. Multivariable mixed models indicated that participants who were older, female, farmer, unemployed or retired, current smokers or former smokers and current drinkers, with higher BMI and leisure-time physical activity levels, followed unhealthy diet, had lower household incomes, and lived in rural areas or western China were more likely to be with cardiometabolic diseases. With the increase of the cardiometabolic disease stages, the rates of following 3 or more healthy lifestyle behaviors decreased (P for trend: <0.001), which were 75.8%, 56.4%, 49.6% and 46.5% among participants of cardiometabolic disease stage 0, stage 1, stage 2 and stage 3 respectively, and the corresponding rates of following all 5 healthy lifestyle behaviors were 1.7%, 0.9%, 0.8% and 0.8%. Participants who were older, female, retired, with higher education attainment or higher household incomes, and lived in urban areas or eastern or central China were more likely to folllow healthy lifestyle behaviors of all cardiometabolic disease stages, while participants who were farmer were less likely to folllow healthy lifestyle behaviors, and the associations were stonger at higher cardiometabolic disease stages. Conclusions: Of the 1 million participants, only 11.1% had no cardiometabolic disease disorders. Adherence to healthy lifestyle behaviors among participants of all cardiometabolic disease stages was no more than 2%. Our study recognized factors related to the prevalence of cardiometabolic diseases and healthy lifestyle behaviors across cardiometabolic disease stages. Part II Association between weight gain from young to middle adulthood and metabolic syndrome Background: Weight gain from young to middle adulthood has been shown to be associated with higher metabolic risk. However, the dose-response associations between weight gain from young to middle adulthood and odds of metabolic syndrome, across BMI categories at young adulthood are still unclear. Methods: This study was based on China Patient-Centered Evaluative Assessment of Cardiac Events Million Persons Project (China PEACE MPP). Middle-aged (35–64 years) participants at screeing from 2015 to 2020 who recalled weight at age 25 years, kept stable weight or gained weight from age 25 years to the screening were included. Standardised devices and procedures were used to investistigate cardiovascular-related health status, perform the physical examination and test the rapid blood glucose and rapid blood lipid for all participants. BMI at age 25 years was classified into underweight (BMI<18.5 kg/m2), normal weight (18.5 kg/m2≤BMI<24 kg/m2) and overweight/obesity (BMI≥24 kg/m2). Multivariable adjusted restricted cubic splines and logistic regression models were applied to assess the dose-response association between weight gain from young to middle adulthood and prevalence of metabolic syndrome at middle adulthood across different BMI categories. Results: In total, 437 849 participants were included. The median age of the participants was 52 years (IQR: 46, 58), and 62.1% were women. In multivariable logistic regression model adjusting for age, sex, geographical region, urbanity, socioeconomic characteristics, lifestyle behaviors and comorbidities, larger weight gain from young to middle adulthood was associated with higher odds of metabolic syndrome at middle adulthood, with odds of 1.6 (95% confidence interval: 1.57–1.64), 3.85 (3.77–3.94), and 11.05 (10.73–11.38) across moderate (weight gain ≥2.5 kg and <10 kg), marked (weight gain ≥10 kg and <20 kg), and extreme amount of weight gain (weight gain ≥20 kg) compared with stable weight (weight gain <2.5 kg), respectively. The ORs of metabolic syndrome were 2.01 (1.98–2.05), 1.93 (1.92–1.94), 1.67 (1.64–1.7) per 5-kg weight gain across participants who were underweight, normal-weight and overweight/obese at young adulthood, respectively. After further adjusting for current body mass index (BMI), although the associations were attenuated, larger weight gains still correlated with higher odds of metabolic syndrome among underweight and normal-weight participants. An inverted U-shaped association was observed in overweight/obese participants: odds of metabolic syndrome increased with increasing weight gain until around 12 kg and then decreased (P for nonlinearity: <0.001). Conclusions: Weight maintenance from young to middle adulthood could be effective to mitigate metabolic syndrome burden, especially among underweight and normal-weight people. Historical weight gain confers information about metabolic syndrome prevalende risk independent of attained BMI. Part III Dose-response association between long-term weight loss in overweight/obese young adults and metabolic health Background: Long-term weight loss has been shown to be associated with lower metabolic risk in young adults with overweight/obesity. However, the dose-response association is uncertain. Methods: This study was based on China Patient-Centered Evaluative Assessment of Cardiac Events Million Persons Project (China PEACE MPP). The participants aged 35–64 years at screeing from 2015 to 2020 who recalled overweight/obesity at age 25 years and experienced long-term weight loss or maintained stable weight were included. Standardised devices and procedures were used to investistigate cardiovascular-related health status, perform the physical examination and test the rapid blood glucose and rapid blood lipid for all participants. The dose-response association between long-term weight loss from age 25 to screening and the odds of metabolic syndrome at screening was assessed using multivariable adjusted regression models with restricted cubic splines. Results: A total of 40 150 participants (66.4% women) were included. In multivariable logistic regression model adjusting for BMI at age 25 years, age, sex, geographical region, urbanity, socioeconomic characteristics, and lifestyle behaviors, the increment of long-term weight loss was associated with continuously decreased odds of metabolic syndrome. The odds of metabolic syndrome were 0.64 (0.60–0.67), 0.42 (0.40–0.45), 0.27 (0.25–0.29), and 0.15 (0.13–0.17) for those with long-term weight loss of 5–9.9%, 10–14.9%, 15–19.9%, and over 20% compared with <5% long-term weight loss, respectively. Moreover, the incremental pattern was observed across all population subgroups defined by age, sex, BMI categories, smoking status, urbanity and education level. Conclusions: An incremental association between long-term weight loss from young adulthood and odds of later-life metabolic syndrome was observed. Our findings highlight the effective ways to achieve long-term weight loss to improve lifetime metabolic health for young adults with overweight/obesity.