Heart Risk Gaps Between Men and Women Flip Depending on Where They Live

Cardiovascular disease remains the leading cause of death worldwide, but the road to a heart attack or stroke rarely begins in the heart itself. Increasingly, cardiologists and epidemiologists describe a slow-burning constellation of dysfunction spanning the blood vessels, kidneys, and metabolic system, a condition the American Heart Association has formally labeled cardiovascular-kidney-metabolic, or CKM, syndrome. Because the earliest phases of this syndrome unfold silently, often years before any clinical event, they represent a precious window for prevention. A new study published in Biology of Sex Differences now asks a deceptively simple question about that window: does being male or female change the odds of progressing from early CKM syndrome to overt cardiovascular disease? The answer, remarkably, appears to depend on where in the world you live.

The research, led by Xinru Peng, Wenli Li, Yujiong Chen and Guowei Li together with colleagues including Lu Qi, Ai Zhao, Zebing Ye, Gregory Y. H. Lip and Miaoguan Peng, took advantage of an unusually powerful design: two parallel analyses of large prospective cohorts, one in the United Kingdom and one in China. The team drew on 352,297 participants from the UK Biobank and 7,610 participants from the China Health and Retirement Longitudinal Study, known as CHARLS. In both datasets, participants were classified into stages of CKM syndrome using definitions aligned with the American Heart Association’s staging framework, which ranges from stage 0, indicating no detectable risk factors, through stages 1 to 3, reflecting progressively more advanced metabolic, kidney, and vascular abnormalities without established clinical cardiovascular disease.

Classifying CKM stage required an extensive set of measurements. The researchers used body mass index, waist circumference, blood pressure, glycosylated hemoglobin, high-density and low-density lipoprotein cholesterol, triglycerides, and estimated glomerular filtration rate, alongside documented diagnoses such as hypertension, diabetes, and chronic kidney disease. In the UK Biobank, participants could be assigned to any of stages 0 through 3; in CHARLS, the available data allowed classification into combined stages 0 to 1, stage 2, or stage 3. The outcome of interest was incident cardiovascular disease during follow-up, encompassing events such as myocardial infarction, coronary heart disease, and stroke ascertained through hospital and mortality records. Multivariable Cox proportional hazards models, adjusted for a wide range of demographic, clinical, socioeconomic, and behavioral covariates, were used to estimate how much risk differed between males and females at each stage.

The headline numbers from the two cohorts point in opposite directions. In the UK Biobank, 32,268 participants, or 9.16 percent of the sample, developed cardiovascular disease during follow-up. That burden fell unequally by sex: 19,421 males, representing 12.37 percent of males in the cohort, experienced an event, compared with 12,847 females, or 6.58 percent of females. After full statistical adjustment, males had a 79 percent higher risk of cardiovascular disease than females, with a hazard ratio of 1.79 and a 95 percent confidence interval spanning 1.75 to 1.84. Strikingly, this pattern of elevated male risk persisted across every CKM stage from 0 through 3, suggesting that the male disadvantage in the UK population is not confined to a particular point along the disease continuum but permeates the entire preclinical spectrum.

CHARLS told the reverse story. Among 7,610 Chinese participants, 1,379 cardiovascular events occurred, an incident rate of 18.12 percent that reflects the older age structure of the cohort and shorter follow-up intervals typical of repeated survey-based studies. Here it was females who fared worse: 800 females, or 20.06 percent, experienced an incident cardiovascular event, compared with 579 males, or 15.99 percent. In fully adjusted models, males had a 26 percent lower risk of cardiovascular disease than females, with a hazard ratio of 0.74 and a 95 percent confidence interval of 0.62 to 0.87. The direction of this association held across CKM stages 0-1, 2, and 3, although the estimate at stage 3 was imprecise and did not reach statistical significance, a limitation the authors attribute to the smaller number of participants in that stratum.

What could produce such a dramatic geographic reversal? The researchers probed this question through mediation-style analyses examining whether socioeconomic status and lifestyle behaviors could account for part of the observed sex differences, and in both cohorts these factors partially explained the associations. In the UK Biobank, the excess male risk was partly attributable to unfavorable patterns in socioeconomic circumstances and health behaviors such as smoking, diet, physical activity, and alcohol consumption. In CHARLS, the elevated risk observed among females was likewise partly attenuated when socioeconomic and lifestyle variables were taken into account. These findings imply that sex differences in cardiovascular risk are not purely biological; they are woven into the social and behavioral fabric of each population, from occupational exposures and income gradients to gendered patterns of smoking, caregiving, and healthcare access.

The methodological scaffolding behind these conclusions was deliberately robust. The team verified the proportional hazards assumption for the sex coefficient, applied inverse probability of censoring weights in CHARLS to mitigate bias from survey attrition, and compared interval-censored models with standard Cox models to accommodate the coarser timing of events between survey waves. Propensity score matching, multiple imputation for missing covariates, E-value analyses to gauge how much unmeasured confounding would be required to overturn the results, and subgroup analyses by socioeconomic status and healthy lifestyle score all supported the robustness of the primary findings. In CHARLS, the researchers even re-derived CKM staging using the China-PAR risk equation as a sensitivity check, further confirming that the observed sex reversal was not an artifact of how the staging algorithm was implemented.

The implications reach well beyond academic curiosity. Clinical risk prediction tools such as PREVENT, SCORE2, and SCORE2-OP, along with China’s own China-PAR equation, already incorporate sex as a risk input, but they are typically calibrated within single populations. The new findings suggest that the magnitude, and even the direction, of the sex coefficient may not transfer cleanly across borders. A risk model trained predominantly on British or European data could systematically miscalculate cardiovascular risk for Chinese women, while models developed in China might understate the danger faced by men in the United Kingdom. For clinicians, this argues for treating sex-specific risk estimates as population-contextual rather than universal constants, and for vigilance in screening patients of the higher-risk sex within each community.

For public health planners, the study underscores the necessity of localized surveillance. Because socioeconomic status and lifestyle behaviors partially mediate the sex differences, interventions that improve economic conditions, reduce smoking, promote healthy diets, and encourage physical activity could narrow the gap in either direction, benefiting the disadvantaged sex in each setting. The authors emphasize that effective cardiovascular prevention may require strategies tailored specifically to males and females within their distinct contexts, a sentiment that aligns with a growing movement in cardiology toward sex-aware precision medicine. The early CKM stages, the study makes clear, are precisely where such tailoring could pay the greatest dividends, before irreversible vascular damage takes hold.

Certain caveats temper the conclusions. Both cohorts are observational, so residual confounding cannot be excluded despite the extensive adjustments. The CHARLS sample is considerably smaller and relies on self-reported and survey-based clinical data, and the stage 3 finding in that cohort was statistically imprecise. CKM staging was operationalized slightly differently in the two studies, reflecting data availability, although the sensitivity analyses suggest this did not drive the divergent results. Nevertheless, the consistency of the sex reversal across stages, models, and sensitivity analyses within each cohort makes the central message hard to dismiss: the relationship between sex and cardiovascular risk in the preclinical CKM spectrum is cohort-specific, shaped by biology and circumstance in equal measure, and any serious effort to curb the global cardiovascular epidemic must confront that complexity head-on.

Subject of Research: Sex differences in cardiovascular disease risk across cardiovascular-kidney-metabolic syndrome stages in two prospective cohorts

Article Title: Sex differences in risk of cardiovascular disease in participants with cardiovascular-kidney-metabolic syndrome stages 0–3: parallel analyses of two prospective cohorts

Article References: Sex differences in risk of cardiovascular disease in participants with cardiovascular-kidney-metabolic syndrome stages 0–3: parallel analyses of two prospective cohorts. (n.d.). https://doi.org/10.1186/s13293-026-00991-w

Image Credits: AI Generated

DOI: 10.1186/s13293-026-00991-w

Keywords: cardiovascular-kidney-metabolic syndrome, cardiovascular disease, sex differences, UK Biobank, CHARLS, CKM stages, risk prediction, public health, cohort study, metabolic syndrome, prevention, epidemiology