中国中老年人群慢性病轨迹与代谢相关指标的关联研究——基于CHARLS纵向研究

Study on the Association between Chronic Disease Progression Trajectories and Metabolism-Related Indicators in Middle-Aged and Elderly Chinese Populations—A Longitudinal Study Based on CHARLS

摘要: 目的：探究中老年人群共病核心慢病及其进展轨迹与代谢相关指标的关联。方法：本研究基于中国健康与养老追踪调查(CHARLS) 2011年的基线数据及2013、2015、2018、2020年追踪调查数据，共纳入8224名45岁及以上中老年人。采用Apriori算法筛选共病中的核心慢性病并挖掘共病关联规则，并运用时间序列聚类分析识别共病核心慢病进展轨迹。最后，通过多分类Logistic回归模型，在调整年龄、性别、文化程度、婚姻状况、吸烟状况、饮酒频率、社交频率等因素后，分析代谢相关指标与共病核心慢病及其进展轨迹的关联。结果：通过Apriori算法筛选出6种共病核心慢病(包括关节炎或风湿病、高血压、消化系统疾病、血脂异常、心脏病、慢性肺部疾病)，并识别出以关节炎或风湿病和高血压为中心的两大共病集群。基于时间序列聚类分析，将中老年人群中共病核心慢病进展轨迹划分为三类：“共病低负担组”(n = 3855，基线核心慢病数：0.59)、“共病中负担组”(n = 3046，基线核心慢病数：1.91)和“共病高负担组”(n = 1323，基线核心慢病数：3.48)。多分类Logistic回归结果显示，与Q1组相比，TyG Q2组(OR = 1.212, 95% CI: 1.079~1.363)与BMI Q3组(OR = 1.136, 95% CI: 1.008~1.281)均与关节炎或风湿病风险升高存在相关。所有代谢指标的Q2、Q3分组均与高血压风险升高有关，其中以BMI Q3组关联最强(OR = 5.210, 95% CI: 4.509~6.020)。以“共病低负担组”为对照，“共病中负担组”和“共病高负担组”的风险随代谢相关指标分位数上升而增加，其中与BMI Q3 (共病中负担OR = 1.736，95% CI：1.530~1.969；高负担OR = 4.251，95% CI：3.570~5.062)和cMetS Q3 (共病中负担OR = 1.892，95% CI：1.675~2.138；高负担OR = 3.555，95% CI：3.005~4.207)的关联最为显著。结论：中国中老年人慢性病共病以关节炎或风湿病和高血压为两大核心共病集群，建议将体重管理作为中老年人慢病防控的核心策略。同时，中老年人共病发展轨迹存在异质性，应基于异质性的轨迹实施精准干预，代谢相关指标可为精准化防控策略提供指标依据。

Abstract: Objective: To explore the association between core chronic comorbidities, their progression trajectories, and metabolism-related indicators in middle-aged and elderly populations. Methods: This study was based on the baseline data (2011) and follow-up survey data (2013, 2015, 2018, 2020) from the China Health and Retirement Longitudinal Study (CHARLS), with a total of 8224 middle-aged and elderly individuals aged 45 years and above enrolled. The Apriori algorithm was used to screen core chronic diseases among comorbidities and mine comorbidity association rules, while time series cluster analysis was applied to identify the progression trajectories of core chronic comorbidities. Finally, a multinomial Logistic regression model was used to analyze the association between metabolism-related indicators and core chronic comorbidities as well as their progression trajectories, after adjusting for factors such as age, gender, educational level, marital status, smoking status, drinking frequency, and social interaction frequency. Results: Six core chronic comorbidities were identified via the Apriori algorithm, including arthritis or rheumatism, hypertension, digestive system diseases, dyslipidemia, heart disease, and chronic lung disease. Two major comorbidity clusters centered on arthritis or rheumatism and hypertension were detected. Based on time series cluster analysis, the progression trajectories of core chronic comorbidities in middle-aged and elderly populations were classified into three categories: “low comorbidity burden group” (n = 3855; mean number of core chronic diseases at baseline: 0.59), “moderate comorbidity burden group” (n = 3046; mean number of core chronic diseases at baseline: 1.91), and “high comorbidity burden group” (n = 1323; mean number of core chronic diseases at baseline: 3.48). Multinomial Logistic regression results showed that compared with the Q1 group, the TyG Q2 group (OR = 1.212, 95% CI: 1.079~1.363) and BMI Q3 group (OR = 1.136, 95% CI: 1.008~1.281) were both associated with an increased risk of arthritis or rheumatism. The Q2 and Q3 groups of all metabolic indicators were related to an elevated risk of hypertension, among which the BMI Q3 group showed the strongest association (OR = 5.210, 95% CI: 4.509~6.020). Taking the “low comorbidity burden group” as the reference, the risks of the “moderate comorbidity burden group” and “high comorbidity burden group” increased with the elevation of percentiles of metabolism-related indicators. The most significant associations were observed with BMI Q3 (moderate burden: OR = 1.736, 95% CI: 1.530~1.969; high burden: OR = 4.251, 95% CI: 3.570~5.062) and cMetS Q3 (moderate burden: OR = 1.892, 95% CI: 1.675~2.138; high burden: OR = 3.555, 95% CI: 3.005~4.207). Conclusion: Obesity plays a central role in the occurrence and development of comorbidities, and weight management is recommended as a core strategy for chronic disease prevention and control in middle-aged and elderly populations. Heterogeneity exists in the progression trajectories of comorbidities among middle-aged and elderly individuals, and precise interventions should be implemented based on these heterogeneous trajectories. Metabolism-related indicators can provide an index basis for refined prevention and control strategies.