The association of adverse childhood experiences (ACEs) with chronic disease prevalence among middle-aged and elderly people has become a research hot spot, but the association between ACEs and self-rated health in older people is still unclear, and relevant studies in Chinese population are especially rare.

To explore self-rated health in Chinese elderly people, and its association with ACEs.

The study was conducted in October 2021. Data about ACEs and self-rated health were acquired by applying from the 2014 and 2018 waves of CHARLS conducted by Peking University, involving 7 579 older people (≥ 60 years old) . The ACEs were stratified into three groups by the times of ACEs events (10 kinds in total) : high-risk (4 or more) , low-risk (1-3) , and reference (zero) . The self-rated health was divided into three levels: good, fair and poor. Ordinal logistic regression was used to analyze the correlation of self-rated health with the number and types of ACEs events.

Among the participants, the prevalence of good, fair and poor self-rated health was 22.06% (1 672/7 579) , 59.03% (4 474/7 579) , and 18.91% (1 433/7 579) , respectively, and the prevalence of experiencing ≥4, 1-3 and no ACEs events was 25.36% (1 922/7 579) , 69.71% (5 283/7 579) , and 4.93% (374/7 579) , respectively. The self-rated health differed significantly by age, gender, place of living, and types of ACEs (parental divorce, parental violence, parental depression, parental disability, living in an unsafe neighborhood, childhood loneliness, childhood famishment, bullying) (P<0.05) . The self-rated health also differed significantly across the three groups stratified by the number of ACEs events (P<0.05) . Univariate ordinal logistic regression analysis indicated that the probability of a worse self-rated health increased by a factor of 1.358〔95%CI (1.110, 1.663) 〕in those with 1-3 ACEs events, and increased by a factor of 2.151 in those with 4 or more ACEs events compared with those with no ACEs events (P<0.05) . The ordinal Logistic regression analysis indicated that results showed that parental divorce, parental violence, parental depression, parental disability, living in an unsafe neighborhood, childhood loneliness, childhood famishment, and bullying were associated a worse self-rated health (P<0.05) . The statistical results remained unchanged after adjusting for age, gender and place of living.

The self-rated health status in Chinese older people was generally fair. To improve the self-rated health to promote healthy aging in this population, interventions could be provided for them targeting reducing the negative influence caused by parental divorce, parental violence, parental depression and disability, living in an unsafe neighborhood, childhood loneliness, and childhood famishment.

儿童期不良经历（ACEs）对中老年慢性病的影响已成为研究热点，但ACEs对老年时期自评健康的影响尚不明确，尤其是以我国人群为样本的研究鲜有报道。

探索我国老年人群的自评健康状况，研究ACEs对老年时期自评健康的影响。

于2021年10月申请获取北京大学发布的中国健康与养老追踪调查（CHARLS）2014年和2018年中≥60岁受试者（n=7 579）的ACEs与自评健康数据。提取的ACEs共10种，按照事件发生数量分为高危组（≥4种）、低危组（1~3种）、零ACE组（0种），自评健康状况分为好、中、差3级。采用有序Logistic回归模型分析ACEs分组和ACEs种类对自评健康的影响。

7 579例研究对象中，分别有1 672例（22.06%）、4 474例（59.03%）、1 433例（18.91%）自评健康状况为好、一般、差；1 922例（25.36%）经历过≥4种ACEs（高危组），5 283例（69.71%）经历过1~3种ACEs（低危组），374例（4.93%）未经历过ACEs（零ACE组）。不同年龄、性别、居住地点、ACEs种类（父母离婚、父母暴力、父母抑郁、父母残疾、成长环境危险、童年孤独、童年饥饿、霸凌）、ACEs分组的研究对象自评健康状况比较，差异有统计学意义（P<0.05）。单因素有序Logistic回归结果显示，低危ACEs组导致更差的自评健康等级是零ACEs组的1.358倍〔95%CI（1.110，1.663）〕，高危ACEs组导致更差的自评健康等级是零ACEs组的2.151倍〔95%CI（1.735，2.667）〕（P<0.05）。有序Logistic回归结果显示，父母离婚、父母暴力、父母抑郁、父母残疾、成长环境危险、童年孤独、童年饥饿、霸凌是导致自评健康更差的因素（P<0.05）；校正年龄、性别、居住地点因素后的统计结果保持不变。

我国老年人自评健康状况整体尚可，未来可通过对经历过父母离婚、父母暴力、父母抑郁和残疾、不安全的成长环境、童年孤独、童年饥饿、霸凌者进行干预，以改善老年期自评健康结果，促进健康老龄化。