Relation Of Blood Pressure Categories Defined By The New ACC/AHA Guidelines With Cardiovascular Mortality And Cost-effectiveness Analysis

Part 1 The Relations of Blood Pressure Categories Defined by the New ACC/AHA Guidelines with All-Cause and Cardiovascular Mortality in Mainland ChinaObjective: The recent American College of Cardiology/American Heart Association(ACC/AHA)guidelines for high blood pressure(BP)lowered the diagnostic threshold of hypertension from systolic/diastolic blood pressure(SBP/DBP)?140/90 mm Hg to ?130/80 mm Hg.The potential impact of the new guidelines on Chinese adults remains unclear.Methods: We evaluated the new BP categories in relation to all-cause and cardiovascular disease(CVD)mortality among 154,407 Chinese adults,who are participants of three prospective cohorts: the Shanghai Women's Health Study,the Shanghai Men's Health Study and the Dongfeng-Tongji cohort.All cohorts measured BPs at baseline or follow-up visits and BP was categorized as follows: normal BP(SBP<120 mm Hg and DBP <80 mm Hg),elevated BP(SBP 120-129 mm Hg and DBP <80mm Hg),stage 1 hypertension(SBP 130-139 mm Hg or DBP 80-89 mm Hg)and stage2 hypertension(SBP/DBP ?140/90 mm Hg).Death events were tracked via linkages to medical insurance system or vital statistics registries.Cox proportional hazards regression models were used to estimate the hazard ratios(HRs)and 95% confidenceintervals(CIs)with adjustment for demographic characteristics,lifestyles and comorbidities.Results: During a total follow-up of 1,718,089 person-years,14,692 deaths were documented including 5,086 CVD deaths [1,277 ischaemic heart disease(IHD)deaths and 2,509 cerebrovascular disease(CBVD)deaths].Compared to normal BP,newly defined stage 1 hypertension was associated with increased CVD mortality [HR(95%CI)=1.40(1.16-1.69)for CVD,1.36(1.12-1.65)for IHD,and 1.53(1.18-2.00)for CBVD],but not with all-cause mortality(1.04,0.89-1.21).Stage 2 hypertension showed significant associations with CVD and all-cause mortality(HRs ranged from 1.26 to2.49),while elevated BP showed null associations.The associations between hypertension(both stage 1 and 2)and CVD mortality were stronger in adults younger than 65 years and adults without pre-existing CVD compared with their counterparts(P for interaction <0.05).When analysed as a continuous variable,each 10 mm Hg increase in SBP was associated with an increased risk of all-cause mortality(1.07,1.02-1.12)and CVD mortality(1.15,1.09-1.21),and the risk estimate was 1.09(1.00-1.18)and 1.25(1.10-1.42),respectively,for each 10 mm Hg increment in DBP.Conclusions: The newly defined stage 1 hypertension is associated with an increased risk of CVD mortality in the Chinese population,particularly among younger adults and those without CVD history.Part 2 Cost-effectiveness analysis of implementation of the new ACC/AHA guideline based on the Dongfeng-Tongji cohortObjective: To evaluate whether the implementation of the new ACC/AHA guideline for stage 1 hypertensive population without CVD history and with a 10-year atherosclerotic cardiovascular disease(ASCVD)risk ?10% was cost-effective,compared to implementation of the current Chinese guideline.Methods: A Markov state-transition model was constructed to simulate costs and effectiveness of implementation of the new ACC/AHA guideline and the current Chinese guideline for stage 1 hypertensive population without CVD history and with a10-year ASCVD risk ?10%.Data on the transition probability were obtained from Dongfeng-Tongji cohort.Utility and costs for specific status were obtained from published literatures.Robustness and uncertainty were evaluated with a series of sensitivity analyses.Results: Based on a time horizon of 20 years,the cost of the current Chineseguideline and the new ACC/AHA was 314 million yuan and 359 million yuan respectively,and the respective quality adjusted life years(QALYs)were 162 thousand and 151 thousand.For each QALYs obtained,the respective cost were 1938.76 Yuan and 2380.43 Yuan.Compared with the Chinese guideline,the incremental cost-effectiveness ratio of the new ACC/AHA guideline was-4042.87 yuan/QALY.The most influential factors were cost for stroke after one year of occurrence,cost for stroke within one year of occurrence,all-cause mortality based on the new ACC/AHA guideline,cost of antihypertensive therapy,and cost for coronary heart disease(CHD)after one year of occurrence.Conclusions: Based on the Dongfeng-Tongji cohort,we found that implementation of the new ACC/AHA guideline was not cost-effective in the long run,whereas the current Chinese guideline could cost less and gain more utilities.But the results should be validated in other Chinese cohorts.Part 3 Establishment of a risk prediction model for CVD mortality based on the Dongfeng-Tongji cohortObjective: In this study,a prediction model of CVD mortality was established based on the Dongfeng-Tongji Cohort.The performance of the model with or without laboratory indicators was evaluated and validation was conducted in the same population.Methods: A total of 21,596 people were included in the analysis after deleting participants with missing information on blood pressure,waist circumference,blood glucose,lipid profile and those with missing or invalid follow-up time.Gender-specific models were formulated with a randomly selected 60% of the population(derivation population)and validated in the remaining 40% populations(validation population).Predictors included non-laboratory indicators [social demographic indicators(age,marital status,education level,etc.),lifestyle(smoking,alcohol,physical activity,etc.),health status [menopausal status(only for women),medical history and physical examination indicators(body mass index,waist circumference and blood pressure)] and laboratory tests(fasting blood glucose,total cholesterol,high-density lipoprotein cholesterol,low-density lipoprotein cholesterol and triglyceride).The death outcome was ascertained by linkage to medical insurance system or related electronic medical record registries.The Cox proportional hazards regression model was used to estimatethe risk of CVD death in the subjects.The performance of the model was evaluated with discrimination and calibration and risk assessment scores were established.Results: Both the laboratory-based and non-laboratory-based models overestimated the actual risk of CVD death(EOR > 1)to varying degrees(4%-8%).But the laboratory-based model had better performance.Validation of laboratory-based model showed that the model overestimated the risk of CVD death in the male population by 14% compared to the Kaplan-Meier adjusted CVD mortality,and overestimated by 31% in the female population.Conclusions: Compared with the non-laboratory-based model,the laboratory-based model could predict CVD mortality more accurately,but further research is needed to establish models that could be generalized in other population.