Abstract: Objective  To analyze the trends in the burden of stroke deaths attributable to high LDL-C in China from 1990 to 2021, explore the effects of age, period, and cohort factors, and predict the future changes in the burden of death.
Methods  Using the data from the global burden of disease database 2021 on stroke deaths attributable to high LDL-C in Chinese adults aged≥20 years from 1990 to 2021, the Joinpoint regression model was used to analyze the trends of age-standardized death rates over time. Through the age-period-cohort model, the age, period, and cohort effects of stroke deaths attributable to high LDL-C in China were estimated. The autoregressive integrated moving average (ARIMA) model was used to predict the burden of stroke deaths attributable to high LDL-C in China from 2022 to 2030.
Results  Compared with 1990, the number and rate of stroke deaths attributable to high LDL-C in China have significantly increased in 2021, while the age-standardized death rates has decreased. In 2021, the total number of deaths for the whole population, males, and females were 300 100, 170 500, and 129 500, respectively, which were 161.87%, 187.04%, and 134.60% higher than those in 1990. The mortality rates in 2021 were 21.08/100 000, 23.42/100 000, and 18.64/100 000, respectively, which were 116.65%, 139.22%, and 92.56% higher than those in 1990. The age-standardized death rates in 2021 were 15.93/100 000, 20.96/100 000, and 12.36/100 000, with estimated annual percentage changes of -0.43%, -0.02%, and -0.88%, respectively (all P<0.05). The results of the Joinpoint regression model showed that from 1990 to 2021, the age-standardized death rates of stroke attributable to high LDL-C in China showed a trend of first increasing and then decreasing in general, and the age-standardized death rates were higher in males than in females. The results of the age-period-cohort model analysis showed that the stroke mortality rates attributable to high LDL-C generally increased with age, with RR ranging from 0.028 to 12.214 in the whole population, 0.027 to 14.661 in males, and 0.032 to 15.258 in females. The period effect showed an increasing trend, with RR ranging from 0.677 to 1.418 in the whole population, 0.623 to 1.532 in males, and 0.750 to 1.298 in females. The cohort effect showed a decreasing trend, with RR ranging from 0.213 to 4.210 in the whole population, 0.221 to 3.637 in males, and 0.170 to 3.978 in females. The results of the ARIMA model predicted that from 2022 to 2030, the number and rate of stroke deaths attributable to high LDL-C in China’s whole population, males, and females will continue to rise. The age-standardized death rates for males will show an increasing trend, but the age-standardized death rates for the whole population and females will remain steady.
Conclusions  From 1990 to 2021, the burden of stroke deaths attributable to high LDL-C has been increasing in China, and the burden of death is heavier in males than in females. Mortality increases with age, grows over periods, and decreases by birth cohorts. The burden of death is expected to remain heavy for the foreseeable future.

Key words: Stroke; Low density lipoprotein cholesterol; Global burden of disease; Joinpoint regression; Age-period-cohort model

摘要： 目的 分析1990—2021年中国归因于高LDL-C的卒中死亡负担变化趋势，探讨年龄、时期、队列因素的影响，并预测未来死亡负担变化。 
方法 基于2021年全球疾病负担数据库中1990—2021年中国≥20岁人群归因于高LDL-C的卒中死亡负担数据，运用Joinpoint回归模型分析年龄标化死亡率随时间的变化趋势，通过年龄-时期-队列模型估算中国归因于高LDL-C的卒中死亡率的年龄、时期和队列效应，使用整合移动平均自回归（autoregressive integrated moving average，ARIMA）模型预测2022—2030年中国归因于高LDL-C的卒中死亡负担。
结果 相较于1990年，2021年中国归因于高LDL-C的卒中死亡数、死亡率均明显增长，年龄标化死亡率有所下降。2021年全人群、男性和女性死亡数分别为30.01万例、17.05万例和12.95万例，较1990年分别升高161.87%、187.04%和134.60%；2021年全人群、男性和女性死亡率分别为21.08/10万、23.42/10万和18.64/10万，较1990年分别升高116.65%、139.22%和92.56%；2021年全人群、男性和女性年龄标化死亡率分别为15.93/10万、20.96/10万和12.36/10万，年度变化百分比估计值分别为-0.43%、-0.02%和-0.88%（均P＜0.05）。Joinpoint回归模型分析结果显示，1990—2021年中国归因于高LDL-C的卒中年龄标化死亡率整体呈先升后降趋势，且男性的年龄标化死亡率高于女性。年龄-时期-队列模型分析结果显示，归因于高LDL-C的卒中死亡率总体上随年龄的增长而增加，全人群、男性、女性的RR变化范围分别为0.028～12.214、0.027～14.661和0.032～15.258；随时期推移呈上升趋势，全人群、男性、女性RR变化范围分别为0.677～1.418、0.623～1.532和0.750～1.298；总体上随着队列变化呈下降趋势，全人群、男性、女性RR变化范围分别为0.213～4.210、0.221～3.637和0.170～3.978。ARIMA模型预测结果显示，2022—2030年中国归因于高LDL-C的全人群、男性和女性卒中死亡数和死亡率将持续升高，男性年龄标化死亡率将呈上升趋势，但全人群和女性的年龄标化死亡率将保持稳定。
结论 1990—2021年，中国归因于高LDL-C的卒中死亡负担呈上升趋势，且男性的死亡负担重于女性。死亡率整体上随年龄的增长而增加，随时期的推移而增长，随出生队列的变化而降低。预计未来一段时间内死亡负担仍将持续加重。

关键词: 卒中; 低密度脂蛋白胆固醇; 全球疾病负担; Joinpoint回归; 年龄-时期-队列模型