Hemoglobin glycation index tracks faster cardiometabolic disease accumulation in CHARLS
A higher hemoglobin glycation index (HGI) was linked to a faster move from no cardiometabolic disease to single disease, from no disease to multimorbidity, and from single disease to multimorbidity in a large prospective cohort of middle-aged and older adults in China. The pattern matters because cardiometabolic risk rarely arrives as a single event; it accumulates in stages, and the biologic signal embedded in HGI appears to identify people who move through those stages more quickly.
The analysis drew on 8,075 participants in the China Health and Retirement Longitudinal Study, with baseline HGI and follow-up data spanning 2011 to 2018. Four states framed the progression: no cardiometabolic disease, single cardiometabolic disease, cardiometabolic multimorbidity defined as at least two of diabetes, heart disease, and stroke, and death. Instead of collapsing outcomes into one endpoint, the investigators used continuous-time multistate Markov models to estimate where risk concentrated along the disease pathway. That approach captures a clinical reality often lost in standard cohort analyses: progression is directional, and the transition from one stage to the next may be biologically distinct from the risk of death itself.
HGI was calculated as observed HbA1c minus fasting plasma glucose-predicted HbA1c, which makes it a measure of interindividual glycation variation beyond fasting glucose alone. In practical terms, two people can share a similar fasting glucose and still differ in the degree to which hemoglobin is glycosylated. A persistently higher HGI may reflect more intense tissue-level glycation, altered red blood cell biology, oxidative stress, or a tendency toward downstream advanced glycation end product signaling. Those mechanisms are plausible contributors to vascular injury, endothelial dysfunction, and the slow stacking of diabetes, coronary disease, and stroke over time.
Each 1-SD increase in HGI was associated with a 21.5% higher transition intensity from no cardiometabolic disease to single disease, with a transition intensity ratio of 1.215 and a 95% confidence interval of 1.150 to 1.283. The association was even stronger for the move from no disease to multimorbidity, where the transition intensity ratio reached 1.596 (95% CI, 1.273 to 2.001). Among participants who already had one cardiometabolic condition, higher HGI was also associated with progression to multimorbidity, with a transition intensity ratio of 1.094 (95% CI, 1.019 to 1.174).
One result stood out sharply.
HGI was also associated with the transition from cardiometabolic multimorbidity to death in the primary analysis.
That finding suggests the biomarker may not only identify accelerated disease accumulation but also capture a later-stage vulnerability once multi-organ vascular injury is established. Glycation-related biology is a credible explanation. Advanced glycation end products interact with the receptor for advanced glycation end products, amplifying inflammatory signaling, oxidative stress, and vascular stiffness. In a patient already carrying diabetes, heart disease, or stroke, that biology can plausibly worsen plaque instability, microvascular injury, and metabolic fragility, each of which may tilt the balance toward fatal outcomes.
The most consistent signal, however, was not mortality. It was disease accumulation. Associations for the transitions from no disease to single disease and from single disease to multimorbidity remained generally stable across sensitivity analyses, and the link between HGI and cardiometabolic disease accumulation persisted after additional adjustment for fasting plasma glucose. That persistence is important because it implies HGI is not merely a surrogate for higher glucose exposure. Instead, it may capture an additional layer of glycemic biology not fully reflected by fasting glucose alone.
From a bedside perspective, the result points to a possible risk-stratification role for routine laboratory data. HbA1c is already widely available, and HGI can be derived by comparing it with fasting glucose. A patient with apparently modest fasting glucose but a disproportionately elevated HbA1c may carry hidden risk that becomes visible only when long-term glycation is considered. In a prevention clinic, that could sharpen attention toward blood pressure control, lipid management, weight reduction, smoking cessation, and closer surveillance for early vascular disease. The study does not prove that acting on HGI will change outcomes, but it does suggest that the biomarker may help identify people who deserve more intensive cardiometabolic follow-up.
The cohort also reinforces the value of looking beyond a single composite endpoint. Diabetes, heart disease, and stroke do not arrive in a uniform sequence, and the transition model used here makes that sequencing explicit. A higher HGI seemed to matter most when disease burden was building, rather than only after multimorbidity had fully declared itself. That pattern fits the biology of glycation, which tends to accumulate gradually and may accelerate vascular damage long before a terminal event occurs.
Several guardrails still apply. The study was observational, so causality cannot be inferred. HGI was measured at baseline, and changes over time in glycemic status, medication use, or other risk factors were not the focus of the summary data provided. The cohort consisted of middle-aged and older adults from CHARLS, which supports population relevance but also limits direct generalization to other ethnic or age groups without further validation. Even so, the signal was consistent enough across analyses to suggest that HGI deserves attention as more than a statistical curiosity.
For clinicians, the central message is straightforward: a higher HGI marked faster cardiometabolic progression across multiple stages, with the most robust associations appearing where disease burden was accumulating. Mortality linkage appeared in the primary analysis, but the study’s clearest contribution lies in showing that abnormal glycation biology may be visible before multimorbidity becomes established, not only after it has already arrived.