Nearly 400,000 adults, six body measurements, and a surprising result for heart attack risk

By Dr. Liji Thomas, MDReviewed by Lauren HardakerOct 13 2026

Researchers tracked nearly 400,000 adults to investigate whether a simple calculation using waist circumference and height could offer a clearer picture of heart attack risk than conventional body measurements.

Study: Body roundness index and other anthropometric indices in relation to incident myocardial infarction: a cohort study from the UK Biobank. Image Credit: peterschreiber.media / Shutterstock

Commonly used anthropometric measures include body mass index (BMI) and waist circumference (WC). However, these measures have limitations as markers of body fat distribution: BMI cannot describe regional adiposity, and WC cannot distinguish visceral from subcutaneous abdominal fat. A recent study published in the Journal of Health, Population and Nutrition examined the body roundness index (BRI) and its association with the risk of a new heart attack or myocardial infarction (MI).

Background

Obesity is a complex medical condition with multiple health risks. The contribution of high BMI to global disability-adjusted life years (DALYs) has increased over the years, from 3.4% in 2010 to 4.9% in 2023.

High BMI rose from seventh to sixth among 88 global health risk factors between 2010 and 2023. Ischemic heart disease (IHD) was the number one cause of age-standardized deaths around the world from 1990 to 2023 except in 2021, when coronavirus disease 2019 (COVID-19) outranked it.

Heart attack is one of the most serious manifestations of IHD. Identifying simple, inexpensive screening tools may help assess MI risk, especially in high-risk individuals. The study did not evaluate whether using BRI could prevent heart attacks.

Abdominal fat distribution is associated with cardiovascular disease (CVD) risk, beyond what either measure alone can indicate.

Newer measures include the BRI. This metric uses height and WC to estimate fat accumulation and body roundness. Yet almost all studies examining the association of BRI with CVD were based on Asian populations, limiting their generalizability, and have yielded inconsistent results.

Few previous studies compared BRI with conventional measures for cardiovascular risk prediction, and only one assessed its predictive value for heart attack. The current study analyzed data from the UK Biobank prospective cohort to assess associations between heart attack incidence and multiple anthropometric indices, comparing their predictive ability for heart attack.

Study characteristics

The analysis included 395,831 participants aged 40-70 years from the UK Biobank cohort. Participants were categorized into quartiles based on various anthropometric indices: BRI, BMI, WC, weight-adjusted waist index (WWI), body surface area (BSA), and A body shape index (ABSI).

The researchers evaluated these for associations with heart attack risk using Cox proportional hazards regression and restricted cubic spline (RCS). To assess their predictive utility, the researchers used area under the receiver operating characteristic (ROC) curve (AUC), net reclassification improvement (NRI), and integrated discrimination improvement (IDI). The regression models accounted for demographic, lifestyle, and clinical risk factors.

The median follow-up was 13.3 years.

Heart attack incidence across groups

The study recorded 14,419 MI events over the follow-up period. Participants who had heart attacks were older, more often male, with a higher deprivation metric, and more likely to have a smoking history. They were also more likely to have a family history of heart disease and to have hypertension, diabetes, and hypercholesterolemia.

In addition, participants who had heart attacks reported more sedentary hours each day, were more likely to have higher anthropometric measures, and had higher serum creatinine and C-reactive protein levels.

All measures except BSA associated with higher heart attack risk

The findings showed a 52% higher adjusted hazard of heart attack among participants in the highest versus the lowest quartile of BRI, with a progressive increase in the association across quartiles. The adjusted hazard of heart attack was 13% higher with each standard deviation (SD) increase in BRI.

Similar trends were observed for the other indices, except for BSA. Each SD increase was associated with a 9% to 15% higher adjusted hazard, depending on the index used. When comparing the lowest versus highest quartiles of WWI and WC, the adjusted hazards were 46% and 40% higher in the highest quartiles, respectively.

Participants in the highest BMI and ABSI quartiles had a 32% higher adjusted hazard than those in the lowest quartiles. All measures except BSA showed significant nonlinear associations with MI risk.

BRI showed slightly better predictive performance than the other five indices, with an AUC of 0.710, compared with 0.697 to 0.708 for the other measures. These models adjusted for age, sex, and race; BRI's advantage over the next-best index, WWI, was just 0.002 AUC points.

The study reported NRI and IDI values above zero for predicting heart attack risk compared with each of the other five indices, supporting a statistical advantage without establishing a clinically meaningful improvement.

The associations remained largely consistent in sex-stratified models. In females, BRI and WWI had comparable predictive value and exceeded that of the other measures, whereas in males, BRI remained the top-performing measure for heart attack prediction.

The authors hypothesize that differences in fat distribution between males and females may partly explain this, with men more likely to accumulate abdominal fat around the viscera. This has long been associated with metabolic abnormalities and may partly account for the findings. Visceral fat was not directly measured.

By comparison, women are likely to accumulate more peripheral subcutaneous fat, which may help explain BRI's smaller predictive advantage among women, but this mechanism was not tested. The comparable predictive performance of WWI and BRI for heart attack risk in females but not males requires further research to elucidate sex-specific mechanisms tying adiposity to CVD risk elevation.

What BRI measures

By combining height and WC, BRI provides an indirect estimate of body roundness and abdominal adiposity. It does not directly measure visceral fat or distinguish it from subcutaneous abdominal fat. Abdominal adiposity is associated with insulin resistance, dyslipidemia, and systemic inflammation. BRI performed slightly better than WC in this cohort.

Still, the authors caution that “whether these performance differences are sufficiently meaningful to warrant the routine adoption of BRI in clinical practice still requires further validation in independent population cohorts.” If validated, it could serve as a simple, inexpensive measure for assessing MI risk in routine healthcare settings. Future studies could test whether including BRI in existing CVD risk scores improves their accuracy; this analysis did not assess that application.

BRI is derived from the waist-to-height ratio and is highly correlated, underscoring their mathematical near-equivalence. The study did not directly compare the two indices for heart attack prediction.

Comparison with other studies

These findings agree broadly with prior studies examining associations between BRI and CVD. Conversely, only one study compared BRI with BMI for its value in heart attack prediction, finding BRI inferior to BMI in a Chinese population. A separate Iranian cohort found no statistically significant association between BRI and MI, but included only 108 MI events.

The current study shows the need for data from more diverse samples, while adding evidence from a predominantly White UK population.

Limitations

The study had several limitations. The anthropometric indices were measured at baseline, so changes during follow-up were not assessed. The UK Biobank includes a self-selected group of volunteers that is not representative of the general UK population, especially subgroups with a higher cardiovascular risk. This could have led to an underestimation of the association between anthropometric measures and heart attack risk, and it limits the study's generalizability.

Approximately 21% of the original cohort was excluded overall, including nearly 19% because of missing anthropometric or covariate data. The excluded sample differed systematically from the included population, which may have introduced selection bias and limited generalizability. The findings remained consistent in sensitivity analyses accounting for these factors.

The predominantly White cohort limits applicability to other ethnic populations. The study identified MI outcomes from hospital diagnostic codes, including codes for prior MI, and does not describe independent clinical adjudication of individual events. The observational design cannot establish causality, and residual confounding might have introduced bias.

Conclusion

“BRI is positively associated with higher incident MI risk, showing statistically significant but modest predictive improvements over other anthropometric indices.” The small advantage does not establish that BRI would improve existing clinical cardiovascular risk scores. Independent studies must validate these findings in diverse populations before BRI can be considered for routine clinical use.