Baseline HbA1c/HDL-C ratio, anthropometric product indices, and the RAND-consolidated self-reported stroke endpoint in the health and retirement study: a prospective cohort analysis
Year: 2026
Type: article
Study design: Observational Study
Abstract: Background Glucose and lipid abnormalities are important risk factors for stroke, but single biomarkers may not capture their combined metabolic context. The hemoglobin A1c to high-density lipoprotein cholesterol ratio (HbA1c/HDL-C ratio, HHR) has been associated with incident stroke. Whether exploratory products of HHR and anthropometric measures provide additional information remains uncertain. Objective To examine the associations of baseline HHR and exploratory HHR-anthropometric product indices with the incident RAND-consolidated self-reported stroke endpoint, which includes responses coded as possible stroke or transient ischemic attack (TIA), among middle-aged and older adults. Methods We analyzed HRS baseline wave 10 (2010) and follow-up waves 11–15, including delayed interviews after 2020. Baseline HHR and anthropometric measures were assessed once in 2010. At each wave, the primary RAND-consolidated endpoint was positive when at least one of RwSTROK, RwSTROKE, or RwSTROKS equaled 1, missing when all three fields were missing, and negative otherwise. Raw RwSTROK codes 2 and 3 did not independently override this rule. Events were assigned to the first positive interview date, and non-events were censored at their last negative assessment. Cox, design-weighted, strict raw-code self-reported-stroke, observation-window competing-risk, administrative-horizon competing-risk, and longitudinal observation-IPCW analyses were conducted. Results The analysis included 5,989 participants and 399 incident endpoint reports over 47,619 person-years. The adjusted HR was 1.10 per 1-SD higher HHR (95% CI, 1.01–1.19; p = 0.022). Forty-nine events (12.3%) were first identified at a wave with a possible stroke/TIA response. The strict raw-code self-reported-stroke analysis included 364 events (HR, 1.11; 95% CI, 1.02–1.20; p = 0.015). The observation-window Fine-Gray estimate was 1.10 (95% CI, 1.02–1.18; p = 0.017), and the longitudinal observation-IPCW estimate was 1.09 (95% CI, 1.01–1.18; p = 0.024). Conclusion Higher baseline HHR was associated with a modestly higher hazard of the RAND-consolidated self-reported stroke endpoint. The association was similar after excluding possible stroke/TIA-coded events and in analyses that respected observed stroke-assessment windows. Product-index analyses were exploratory and do not establish synergy, incremental predictive value, or causality.
Source: Frontiers in Neurology
Authors: Jing Fan(Conceptualization, Data curation, Formal analysis, Methodology, Writing – original draft), Xiangyu Zhang(Formal analysis, Funding acquisition, Writing – original draft), Lin Wang(Methodology, Software, Visualization, Writing – original draft), Huanan Li(Funding acquisition, Investigation, Resources, Software, Validation, Writing – original draft), Xiaoxiao Liu(Investigation, Writing – review & editing), Chen Li(Validation, Writing – review & editing), Zekun Wang(Investigation, Writing – review & editing), Xianggang Meng(Conceptualization, Supervision, Writing – review & editing)
Language: English
Subfield: Endocrinology, Diabetes and Metabolism
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $3,801
OpenAlex ID: W7235960545
