目的: 在中国鄞州电子健康档案研究(Chinese electronic health records research in Yinzhou, CHERRY)的队列人群中，评价启动降压药物治疗的不同策略预防心血管病的健康收益与干预效率。方法: 采用马尔科夫模型模拟评价的不同策略包括：策略1，对收缩压≥140 mmHg的人群启动降压药物治疗(根据2020年《中国心血管病一级预防指南》)；策略2，对收缩压≥130 mmHg的人群启动降压药物治疗；策略3，对收缩压≥140 mmHg以及130~140 mmHg且心血管病高风险人群启动降压药物治疗(根据2017年美国心脏病学会/美国心脏协会《成年人高血压预防、检测、评估和管理指南》)；策略4，对收缩压≥160 mmHg以及140~160 mmHg且心血管病高风险人群启动降压药物治疗(根据2019年英国国家卫生与临床优化研究所《成年人高血压诊断与管理指南》)。采用2019年世界卫生组织心血管病风险评估模型进行风险分层。马尔科夫模型的循环周期设为1年，模拟10个周期后计算质量调整生命年(quality-adjusted life year, QALY)、心血管病发病数、全因死亡数等结局事件数以评价策略的健康收益，并计算每预防一例心血管病事件或全因死亡的需治疗人数(number needed to treat, NNT)以评价策略的干预效率。马尔科夫模型的参数主要来源于CHERRY队列与公开发表的文献。采用单因素敏感性分析探讨心血管病发病率对结果的影响，采用概率敏感性分析探讨干预措施效应参数的不确定性对结果的影响。结果: 共纳入213 987名35~79岁基线无心血管病史的人群。相比于策略1，单纯下调降压起始值的策略2可预防的心血管病发病数增加666(95% UI: 334~975)例，但每预防一例心血管病发病的NNT增加10(95% UI: 7~20)人；而考虑定量风险评估的策略3可预防的心血管病发病数增加388(95% UI: 194~569)例，且每预防一例心血管病发病的NNT减少6(95% UI: 4~12)人，提示策略3可增加健康收益并具有更高的干预效率。策略4相比于策略1，可预防的心血管病发病数虽然减少193(95% UI: 98~281)例，但每预防一例心血管病事件的NNT减少18(95% UI: 13~37)人，效率更高。单因素敏感性分析及概率敏感性分析结果与主分析结果一致。结论: 在中国发达地区的社区人群中选择降压药物治疗目标人群时，结合心血管病定量风险评估的策略优于单纯将起始值从140 mmHg降至130 mmHg的策略，前者可提升健康收益且兼顾干预效率；不同地区需因地制宜选择降压起始值并结合定量风险评估的策略，以权衡健康收益与干预效率。

Objective: To evaluate the health benefits and intervention efficiency of different strategies of initiating antihypertensive therapy for the primary prevention of cardiovascular diseases in a community-based Chinese population from the Chinese electronic health records research in Yinzhou (CHERRY) study. Methods: A decision-analytic Markov model was used to simulate and compare different antihypertensive initiation strategies, including: Strategy 1, initiation of antihypertensive therapy for Chinese adults with systolic blood pressure (SBP) ≥140 mmHg (2020 Chinese guideline on the primary prevention of cardiovascular diseases); Strategy 2, initiation of antihypertensive therapy for Chinese adults with SBP ≥130 mmHg; Strategy 3, initiation of antihypertensive therapy for Chinese adults with SBP≥140 mmHg, or with SBP between 130 and 140 mmHg and at high risk of cardiovascular diseases (2017 American College of Cardiology/American Heart Association guideline for the prevention, detection, evaluation, and management of high blood pressure in adults); Strategy 4, initiation of antihypertensive therapy for Chinese adults with SBP≥160 mmHg, or with SBP between 140 and 160 mmHg and at high risk of cardiovascular diseases (2019 United Kingdom National Institute for Health and Care Excellence guideline for the hypertension in adults: Diagnosis and management). The high 10-year cardiovascular risk was defined as the predicted risk over 10% based on the 2019 World Health Organization cardiovascular disease risk charts. Different strategies were simulated by the Markov model for ten years (cycles), with parameters mainly from the CHERRY study or published literature. After ten cycles of simulation, the numbers of quality-adjusted life years (QALY), cardiovascular events and all-cause deaths were calculated to evaluate the health benefits of each strategy, and the numbers needed to treat (NNT) for each cardiovascular event or all-cause death could be prevented were calculated to assess the intervention efficiency. One-way sensitivity analysis on the uncertainty of incidence rates of cardiovascular disease and probabilistic sensitivity analysis on the uncertainty of hazard ratios of interventions were conducted. Results: A total of 213 987 Chinese adults aged 35-79 years without cardiovascular diseases were included. Compared with strategy 1, the number of cardiovascular events that could be prevented in strategy 2 increased by 666 (95% UI: 334-975), while the NNT per cardiovascular event prevented increased by 10 (95% UI: 7-20). In contrast to strategy 1, the number of cardiovascular events that could be prevented in strategy 3 increased by 388 (95% UI: 194-569), and the NNT per cardiovascular event prevented decreased by 6 (95% UI: 4-12), suggesting that strategy 3 had better health benefits and intervention efficiency. Compared to strategy 1, although the number of cardiovascular events that could be prevented decreased by 193 (95% UI: 98-281) in strategy 4, the NNT per cardiovascular event prevented decreased by 18 (95% UI: 13-37) with better efficiency. The results were consistent in the sensitivity analyses. Conclusion: When initiating antihypertensive therapy in an economically developed area of China, the strategy combined with cardiovascular risk assessment is more efficient than those purely based on the SBP threshold. The cardiovascular risk assessment strategy with different SBP thresholds is suggested to balance health benefits and intervention efficiency in diverse populations.