【摘要】 目的研究高血压患者中基线体质量指数(BMI)与癌症发病风险之间的关联,为高血压患者的体质量管理提供建议,以期降低癌症发病风险。方法选择2007—2015年上海市闵行区电子健康信息系统(EHR)登记的212 394名中国成年高血压患者作为研究对象,采用回顾性队列研究方法,使用变系数Cox回归模型估计基线BMI与癌症发病风险之间的时变关联。结果多因素Cox回归分析结果显示,在高血压患者中低体质量与较高的癌症发病风险相关,但其关联程度随着随访时间而逐渐减弱,随访1年HR=1.65(95%CI:1.42～1.91),随访5年HR=1.30(95%CI:1.19～1.41),随访10年HR=0.96(95%CI:0.85～1.08)。超体质量与较低的癌症发病风险相关,其保护作用随着时间逐渐减弱,随访1年HR=0.86(95%CI:0.80～0.93),随访5年HR=0.94(95%CI:0.90～0.97),随访10年HR=1.04(95%CI:0.99～1.08)。一级肥胖具有相对稳定的保护效应,随访1年HR=0.86(95%CI:0.82～0.90),随访10年HR=0.87(95%CI:0.83～0.90)。结论高血压患者的健康管理工作应以控制体质量在合理范围内为目标,更多关注体质量过轻的患者。更多还原

【Abstract】 Objective To evaluate the time-varying association between baseline BMI in patients with hypertension and cancer risk to provide recommendations for weight management in order to reduce the risk of cancer.Methods We retrospectively selected 212 394 Chinese adults with hypertension from 2007 to 2015 in the Minhang District Electronic Health Information System(EHR) and used time-varying coefficients Cox regression model to estimate the time-varying association between baseline BMI and cancer risk.Results Multivariate Cox regression analysis showed that underweight was associated with a higher risk of cancer in hypertensive patients, and the magnitude of associations decreased from 1.65(95%CI: 1.42-1.91) at 1 year of follow-up to 1.30(95%CI: 1.19-1.41) at 5 years of follow-up, until 0.96(95%CI: 0.85-1.08) at 10 years of follow-up. Overweight was associated with lower risk of cancer and the protective effects of overweight gradually attenuated over time from 0.86(95%CI:0.80-0.93) at 1 year of follow-up to 0.94(95%CI:0.90-0.97) at 5 years of follow-up, until 1.04(95%CI:0.99-1.08) at 10 years of follow-up. Class I obesity had a relatively stable protective effect, with HRs at 1 year and 10 years of follow-up 0.86(95%CI:0.82-0.90), and 0.87(95%CI:0.83-0.90), respectively.Conclusion Management efforts for hypertension may target controlling body weight in a reasonable range for patients, and probably more attention should be given to underweight patients.更多还原