Waist-to-Height Ratio Exposed Risk Hidden by Body Mass Index
In a large analysis from the All of Us Research Program, waist-to-height ratio separated cardiometabolic risk far more cleanly than body mass index. The finding matters because BMI classifies total body size, not fat distribution, and central adiposity is the phenotype most tightly linked to insulin resistance, vascular inflammation, and atherothrombosis.
The cohort included 308,600 adults. Investigators excluded participants with the corresponding prevalent disease separately for eight cardiometabolic outcomes, then used cause-specific Cox models adjusted for demographic, socioeconomic, smoking, and BMI-related factors. In those models, a 0.05 higher WHtR tracked with higher hazard across every outcome studied, with adjusted hazard ratios ranging from 1.07 to 1.31.
Among 77,684 adults with normal BMI, 24,301 had WHtR of 0.50 or higher. That single cutoff identified a substantial subgroup whose BMI alone would have suggested low or ordinary risk.
Central Adiposity Carried the Strongest Signal for Diabetes and Hypertension
The sharpest separation emerged for diabetes. In BMI-stratified interaction models, the hazard ratio per 0.05 higher WHtR was 1.52 in normal BMI adults, 1.42 in overweight adults, and 1.22 in adults with obesity, with P for interaction below 0.001. Hypertension also behaved differently across BMI strata, again with P for interaction below 0.001. The pattern fits basic pathophysiology: visceral fat does not merely add weight, it drives hepatic fat accumulation, worsens glycemic control, and promotes sympathetic activation and vascular stiffness long before scale-based obesity becomes obvious.
A 0.50 WHtR threshold separated risk even within the normal-BMI group. Compared with WHtR below 0.50, WHtR at or above 0.50 was associated with diabetes, myocardial infarction, and ischemic stroke in that subgroup. The corresponding hazard ratios were 1.74 for diabetes, 1.49 for myocardial infarction, and 1.52 for ischemic stroke.
Central obesity is not a cosmetic phenotype. It is a metabolic one.
Normal-Weight Adults Were Not Low-Risk When Waist Measures Were High
The analysis directly challenges the reflex to equate normal BMI with cardiometabolic safety. Among adults whose BMI fell in the normal range, WHtR captured excess risk that BMI classification alone would have missed. In practical terms, a normal-weight patient with abdominal adiposity may sit in a falsely reassuring category while carrying the biology of insulin resistance and vascular injury.
The landmark analyses strengthen that interpretation. When investigators excluded the first year of follow-up, the associations with all eight outcomes persisted in the normal-BMI group. That reduces the chance that early, undiagnosed disease was driving the waist-to-height relationship. Flexible age adjustment preserved the principal findings as well, suggesting the signal was not an artifact of a particular age specification.
Why the metric works
WHtR functions as a crude surrogate for central fat burden relative to body size. The waist component reflects abdominal adiposity, while the height denominator reduces confounding from stature and body habitus. In adults with the same BMI, a higher waist-to-height ratio usually means more visceral fat and less favorable fat partitioning. That biology explains why the association was strongest for diabetes and hypertension and remained visible for myocardial infarction and ischemic stroke.
Discrimination Improved When WHtR Was Added to the Model
Risk prediction also improved when WHtR was modeled with nonlinearity. Adding nonlinear WHtR improved cross-validated discrimination, with the largest gain for diabetes, where the C-index increased by 0.0134. The numeric increment may appear modest, but in population risk modeling even small gains can matter when a measure is cheap, immediate, and available at the bedside or in primary care.
The result suggests WHtR is not simply another correlated anthropometric variable. It carries information that BMI does not capture, especially in people whose weight falls below the obesity threshold. For clinicians, that means waist measurement deserves a more central place in routine cardiometabolic assessment rather than serving as an optional add-on.
What the Findings Mean for Clinical Assessment
These results support a dual-screening approach: BMI for overall body size, and WHtR for fat distribution. A patient can have a normal BMI and still carry substantial abdominal adiposity, with risk for diabetes, blood pressure elevation, coronary events, and stroke. That discordance is exactly where BMI tends to fail.
The study is also notable for its scale. More than 300,000 adults from a contemporary U.S. research cohort provide enough depth to examine heterogeneity across BMI strata and across multiple outcomes. The consistency across diabetes, hypertension, myocardial infarction, and stroke argues against a chance finding isolated to one endpoint.
Still, the work remains observational and depends on baseline anthropometry rather than serial body-composition tracking. The preprint format also means the analysis has not yet undergone peer review. Residual confounding is possible, and waist-to-height ratio, while clinically practical, is still a surrogate rather than a direct measure of visceral fat.
Even with those limitations, the clinical message is concrete: BMI alone can miss cardiometabolic danger, and WHtR recovers part of that lost signal, especially in adults who look weight-normal on paper but are metabolically at risk.