时间标签：数据日期 2026-11-05 · 生成时间 2026-11-05 05:30（Asia/Shanghai）｜ 检索窗口 PubMed edat 2026/11/02 – 2026/11/06 ｜ 情报发生日期 2026-11-05
研究问题
Second, most studies have been conducted within a single country, limiting assessment of whether findings are consistent across different healthcare systems, reproductive history documentation practices, and social contexts.；Because parity was missing for approximately 70.4% of CHARLS participants, parity-adjusted analyses were restricted to ELSA and HRS person-wave data Fig. 1Retention of participants for the individual-level main analysis across cohorts We next compared CMM burden by pregnancy loss history to examine whether exposed and unexposed women showed consistent crude differences before model fitting.；The parity-adjusted person-wave analysis included two cohorts (ELSA and HRS), and therefore HKSJ intervals are not reported Fig. 4Summary of the main analysis and key sensitivity analyses To examine whether the overall pattern depended on the individual-level last-observation design, we repeated the analysis using all person-wave observations as a sensitivity analysis.
方法
Large cohort studies, including the Nurses’ Health Study II, the China Kadoorie Biobank, and the Women’s Health Initiative, have suggested that pregnancy loss may be associated with subsequent CVD risk [7–9].；Third, few studies have used harmonized individual participant data to examine the same research question across multiple ageing cohorts.；We therefore used data from China, England, and the United States to estimate country-specific associations and then evaluate the extent to which these estimates could be meaningfully summarized.Cross-cohort comparisons are challenging because reproductive history variables and cardiometabolic outcome definitions are not always fully comparable.
核心结论
These findings indicated a cohort-specific signal led by CHARLS rather than a uniformly replicated association across countries, motivating the subsequent random-effects meta-analysis.；These findings suggested that the average cross-cohort association was not robust across analytic specifications.；These findings support a possible cohort-specific association but do not establish temporality or causality.A key implication is that the pooled estimate should be read as a comparative summary rather than as a single universal effect.
局限
However, most existing studies have focused on single cardiovascular endpoints, such as coronary heart disease or stroke, rather than CMM [7–11].Potential pathways linking pregnancy loss with later cardiometabolic risk include shared vascular, inflammatory, and metabolic risk factors, as well as autoimmune mechanisms such as antiphospholipid antibodies in recurrent fetal loss [12–15].；Thus, HRS estimates should be interpreted in light of wider confidence intervals and possible selection due to sample reduction.；This pattern was consistent with the direction of the CHARLS main analysis, but because it was based on repeated person-wave data and could not be replicated in ELSA or HRS, it should be interpreted as supplementary evidence within CHARLS.
对本项目的价值
中等价值：属「血管 / 心血管老化」方向；血管 / 心血管老化栏目可用作本项目文献库与综述素材。 ｜ 价值分 2.6/5（★★★☆☆）
全文来源
PubMed PMC JATS（正文 39,290 字符，题名指纹 0.93，已过 PubMed PMCID 官方关联 + 题名指纹 + 正文长度归属校验）
原文题录
- 中文题名：妊娠丢失与心血管代谢多病共存：CHARLS、ELSA 与 HRS 两阶段个体参与者数据分析
- 英文标题：Pregnancy loss and prevalent cardiometabolic multimorbidity: a two-stage individual participant data analysis of CHARLS, ELSA, and HRS.
- 期刊与标识：BMC women's health · PMID 42972542 · DOI 10.1186/s12905-026-04767-7 · PMC PMC13729411