Body mass index, or BMI, is one of the most used indirect measures for assessing body fat and health risks. It’s a simple calculation using height and weight and is considered the standard body composition measurement endorsed by many health organizations. However, it does have its limitations. It doesn’t differentiate between muscle and fat. Some fall into the overweight category if they are muscular despite being very fit. Furthermore, people with a normal BMI can still have metabolic issues, while people with higher BMIs might be metabolically healthy.
Waist circumference (WC) is another body composition measurement performed in the office. The concern with WC and even the waist-to-hip ratio is that the measurements can be imprecise.
Other, more direct measures of body composition might predict risk better than BMI and include dual-energy X-ray absorptiometry (DXA) scans and bioelectrical impedance analysis (BIA). BIA measures electrical properties of body tissues to estimate body composition. DXA scans make accurate assessments; however, their use is not as practical in the clinical setting due to cost, complexity, and accessibility. Whereas BIA is inexpensive, quick, and easy to use.
The question is whether body fat percentage (BF%) is a better measure of mortality risk than BMI.
The Study
Researchers assessed BMI versus BF% for 15-year mortality risk and published their findings in the Annals of Family Medicine. They used 1999-2004 National Health and Nutrition Examination Survey (NHANES) data, which included 2,821 males and 1,431 females aged 20 to 49 years. This group is representative of the US population, which is important for applying these results in everyday practice. NHANES data was linked to the National Death Index through 2019, providing a 15-year mortality follow-up.
BMI, waist circumference (WC), and BF% derived from BIA were compared, and healthy measures were defined as follows:
| Men | Women | |
| Healthy BMI | 18.5-24.9 kg/m2 | |
| Healthy BF% | <27% | <44% |
| Healthy WC | </=40 inches | </=35 inches |
Both adjusted and unadjusted Cox proportional hazard models were used to examine 15-year all-cause, heart disease, and cancer mortality based on the three body composition measures, while adjusting for important factors like age, race, and poverty. BF% and WC have statistically significant relationships with all-cause and heart disease mortality in both adjusted and unadjusted models, but not with cancer mortality. Whereas BMI shows a statistically significant association with heart disease mortality in the unadjusted model only, meaning it is a weak predictor when other risk factors are considered.
BF% and WC may be better predictors of 15-year all-cause and cardiovascular mortality than BMI.
Conclusions
BF% is a more direct measure of body composition compared to BMI and may be more strongly linked to mortality in young adults. This may inform future guidelines for risk stratification. Study authors noted that the focus was on mortality, but future studies looking at morbidity outcomes like myocardial infarction may help to understand BF% as a risk factor in disease development. The study showed that WC is also a more reliable risk measurement than BMI.
The authors state that measuring BF% directly via BIA is accurate, reproducible, and may be more practical in clinical practice. There are several BIA models, and it is unclear which model is more reliable. BIA machines range from hand-held devices to upright machines to electrodes placed while lying down. Standardized practices will help to better understand how to extract the most accurate data from these devices.
