目的 揭示衰弱长期演变规律及与多病共存的关联，为制订老年健康管理策略提供依据。方法 基于中国健康与退休纵向调查2011—2018年的数据，运用群组轨迹模型识别衰弱轨迹，采用多因素Logistic回归及交叉滞后模型分析其与多病共存的关联。结果 纳入3696名参与者，识别出持续低值组、轻度渐进上升组、中度快速上升组、高位前期激增组和高位末期激增组5条轨迹。以持续低值组为参照，其余各组的多病共存风险均升高，其中，中度快速上升组的风险升高最为显著[OR=21.94，95%CI(16.40~29.72)，P＜0.001]，衰弱指数在随访期内从20.5增至30.2。交叉滞后模型证实衰弱与多病共存存在双向关联。结论 老年人衰弱轨迹呈现异质性且与多病共存双向关联。建议将轨迹动态监测纳入常规老年综合评估体系，并实施基于轨迹特征的分层干预。

Objective To reveal the long-term evolution law of frailty and its association with multimorbidity,and to provide evidence for formulating elderly health management strategies.Methods Based on the data of China Health and Retirement Longitudinal Study from 2011 to 2018,group-based trajectory modeling was used to identify frailty trajectories,and multivariate logistic regression and cross-lagged modeling were adopted to analyze the association between frailty trajectories and multimorbidity.Results A total of 3696 participants were included,and 5 frailty trajectories were identified:the frailty index of the persistent low-value group,the mild progressive increase group,the moderate rapid increase group,the high-level early surge group from,and the high-level late surge group from. With the persistent low-value group as the reference,the risk of multimorbidity was increased in all the other groups,among which the moderate rapid increase group had the most significant increase [OR=21.94,95%CI(16.40-29.72),P＜0.001]. The frailty index increased from 20.5 to 30.2 during the follow-up period.Conclusions The frailty trajectories of elderly people are heterogeneous and have a bidirectional association with multimorbidity.It is suggested that dynamic trajectory monitoring be incorporated into the routine comprehensive geriatric assessment system,and stratified interventions based on trajectory characteristics be implemented to delay the process of disability.

R473.59

国家自然科学基金（72074168）