Study Reveals BMI Misclassifies Over One-Third of Adults’ Weight Status

A recent study from Italy raises critical questions about the accuracy of the body mass index (BMI) as a tool for assessing an individual's weight status. Findings set to be presented at the European Congress on Obesity (ECO 2026) reveal that when evaluated through dual-energy X-ray absorptiometry (DXA)—a highly regarded method for body composition analysis—BMI frequently mislabels people as overweight or obese.

Despite ongoing criticism, BMI remains a staple in both clinical practice and public health policies. This reliance is largely due to its simplicity, but Professor Marwan El Ghoch of the University of Modena points out that BMI's failure to directly measure body fat or its distribution significantly undermines its reliability as an indicator of obesity.

Research Methodology

In this study, researchers from the University of Verona and Beirut University focused on a diverse sample of 1,351 adults, aged 18 to 98, with 60% being female participants. These individuals were all referred to the University of Verona's Department of Neurosciences, Biomedicine and Movement Sciences. Notably, all participants identified as White Caucasian, which reflects established variations in BMI outcomes across different ethnicities.

The participants’ weight status was initially classified using standard WHO BMI categories: 19 individuals (1.4%) were deemed underweight, 787 (58.3%) classified as normal weight, 354 (26.2%) as overweight, and 191 (14.1%) as obese. This led to a total of approximately 41% of participants categorized as overweight or obese, aligning with existing data from Italy’s Veneto region. However, it’s pertinent to ask how this specific demographic of White Caucasian individuals might skew the findings, and whether similar inaccuracies would appear across a broader ethnic mix.

Comparative Analysis of BMI and DXA

When comparing BMI classifications to body fat percentages obtained through DXA, significant disparities emerged. Among patients marked as obese via BMI, over a third (34%) were correctly classified as merely overweight when analyzed by DXA. For those classified as overweight by BMI, more than half (53%) were miscategorized; of those, three-quarters fell within the normal weight range, while a quarter qualified as obese. These discrepancies indicate that BMI could potentially misguide clinical decision-making, affecting treatment pathways.

Interestingly, BMI and DXA showed agreement in the normal weight category (BMI 18.5 to 25) roughly 78% of the time. However, 22% rendered differing results, with 9.7% identified as underweight and 11.4% as overweight according to DXA criteria. It raises the question: how reliable is BMI as a sole measure? The errors in classification, especially in identifying those as underweight, underscore the potential dangers of relying solely on BMI without incorporating more precise tools for body composition assessment.

The most substantial inaccuracies arose within the underweight classification. Two-thirds (68.4%) of those reported as underweight based on their BMI should have actually been classified as normal weight according to DXA. This is alarming, especially regarding public health initiatives that might base interventions on flawed data. (And this is the part most people overlook.) If individuals are misclassified, they may not receive the support or medical guidance they need, from dietary recommendations to more serious interventions.

Implications for Public Health Guidelines

DXA analysis revealed an overall prevalence of combined overweight and obesity at about 37% (23.4% overweight and 13.2% obese), compared with a higher rate of 41% determined by BMI. This discrepancy carries significant implications for public health strategies that seek to address weight-related issues within populations.

Professor El Ghoch emphasizes the importance of these findings, noting the significant number of adults being misclassified and thereby potentially mismanaged in health settings due to reliance on BMI alone. Co-author Professor Chiara Milanese reiterates that while both systems may report similar prevalence rates, they differ in identifying specific individuals, indicating that DXA could reveal more accurate categorizations. This inconsistency illustrates a vital need for reconsideration of public health messaging, particularly how weight categories are communicated to the public.

As a result, the researchers advocate for an update to Italy's public health guidelines, suggesting the incorporation of additional assessment tools alongside BMI. Options might include direct body composition measurements, skinfold techniques, or simpler alternatives like the waist-to-height ratio. This reflects a growing consensus that a multi-faceted approach to weight assessment could yield better health outcomes across demographics. The researchers also propose that similar levels of misclassification may occur in other European populations, warranting further investigation across various ethnicities and regions.

Such findings could have profound impacts on how weight is assessed, emphasizing the need for greater accuracy to improve health outcomes. This isn’t just about refining technical measurements; it’s about bridging gaps in healthcare access and support, addressing a range of complications that arise from weight misclassification. If you're working in this space, the shift toward more accurate assessment tools may soon be on the horizon. Policymakers must take these findings into account to enhance health outcomes for populations across Europe and beyond.

Materials provided by European Association for the Study of Obesity. Content may be edited for style and length.