University of Verona Researchers Challenge BMI Accuracy with Groundbreaking Body Composition Analysis
The University of Verona's Department of Neurosciences, Biomedicine and Movement Sciences has led a pivotal study highlighting significant flaws in the Body Mass Index (BMI), a metric widely used to classify weight status. Published in the journal Nutrients on June 29, 2025, and presented at the European Congress on Obesity (ECO) 2026 in Istanbul, the research reveals that BMI misclassifies over one-third of adults when compared to precise body fat measurements obtained through dual-energy X-ray absorptiometry (DXA), the gold standard for assessing body composition.
Led by Professor Chiara Milanese, alongside collaborators from Beirut Arab University and the University of Modena and Reggio Emilia, the study involved 1,351 white Caucasian adults aged 18 to 98 from northern Italy. Participants underwent comprehensive DXA scans to measure body fat percentage (BF%), providing a direct contrast to BMI classifications defined by the World Health Organization (WHO): underweight (<18.5 kg/m²), normal weight (18.5–24.9 kg/m²), overweight (25–29.9 kg/m²), and obesity (≥30 kg/m²).
The Mechanics of BMI: A Simple Formula with Complex Shortcomings
Developed in the 19th century by Belgian mathematician Adolphe Quetelet, BMI calculates weight in kilograms divided by height in meters squared. It's inexpensive, quick, and requires no specialized equipment, making it a staple in clinical practice, public health policy, and epidemiological research across Europe. However, BMI does not differentiate between fat mass, muscle mass, bone density, or fat distribution—key factors in metabolic health risks.
In the Verona study, BMI indicated a 41% overweight/obesity prevalence, aligning with Veneto region data. Yet DXA reassessments showed only 37% (23.4% overweight, 13.2% obese), exposing discrepancies. Professor Marwan El Ghoch noted, "BMI inflates the prevalence of abnormal weight categories, potentially leading to misguided interventions and resource allocation."
Detailed Findings: Misclassification Rates Across BMI Categories
The study's results were stark. Overall, 32.5% of participants were misclassified. Breakdown by BMI category:
- Underweight (1.4%, n=19): 68.4% misclassified, mostly as normal weight by BF%.
- Normal weight (58.3%, n=787): 78.1% correctly classified; 21.9% misclassified (9.7% underweight, 11.4% overweight, 0.8% obese).
- Overweight (26.2%, n=354): 53.4% misclassified, with 75% reclassified as normal and 25% as obese.
- Obesity (14.1%, n=191): 34% misclassified as overweight by BF%.
Misclassification was higher in females (underweight 60%, overweight 54.8%) and young adults (18–39 years: 42.1%). Weighted kappa agreement was a mere 0.126, indicating weak concordance.
Why BMI Falls Short: Muscle, Age, Sex, and Ethnicity Factors
BMI's inability to account for body composition leads to errors. Athletes or those with high muscle mass (e.g., resistance-trained individuals) often register as overweight/obese despite low BF%. Elderly adults face sarcopenia (age-related muscle loss), inflating BF% at lower BMIs. Women typically have higher BF% than men at equivalent BMIs due to physiological differences. Ethnicity also plays a role; while the Verona cohort was Caucasian, studies show Asians have higher BF% at lower BMIs, prompting adjusted WHO thresholds (e.g., overweight ≥23 kg/m²).
European research echoes this. A UK Biobank analysis found BMI overestimates obesity in muscular individuals, while a Spanish study using bioelectrical impedance analysis (BIA) reported 20–30% discordance rates. Professor Milanese emphasized, "Different people are identified by BMI versus DXA, even if overall prevalences align."
University of Verona's Expertise in Body Composition Research
Housed within the University of Verona, a leading Italian institution ranked among Europe's top universities for sports sciences and biomedicine, the Department of Neurosciences, Biomedicine and Movement Sciences excels in obesity and body composition studies. Prior work includes revising BMI cutoffs for male athletes (BMI ≥28.5 for overweight) and analyzing multicomponent training effects on cancer survivors' body composition. This interdisciplinary hub integrates physiology, genetics, and clinical nutrition, contributing to national guidelines and EU-funded projects on metabolic health.
The department's DXA facilities enable precise, non-invasive assessments, positioning Verona as a European leader. Collaborations with Modena and Beirut underscore higher education's role in transnational research tackling obesity epidemics.
European Context: Obesity Trends and BMI's Policy Influence
Europe faces rising obesity, with WHO estimating 23% adult prevalence (BMI ≥30), varying from 15% in France to 28% in the UK. Italy reports ~10% obesity but 40% overweight. BMI drives policies like the EU's Health at a Glance reports and national screening programs. Yet, misclassification risks overpathologizing healthy individuals (e.g., muscular laborers) or missing metabolically obese normal-weight (MONW) cases.
Recent studies from Dutch and German universities advocate BF%-adjusted classifications. A pan-European review found BMI-BF% discordance up to 40% in middle-aged cohorts, urging integration into primary care.Read the full Verona study here.
Promising Alternatives: From DXA to Practical Surrogates
DXA provides segmental BF% with <2% error but is costly for routine use. Alternatives include:
- Bioelectrical Impedance Analysis (BIA): Portable, estimates BF% via electrical conductivity; validated in European cohorts with 3–5% accuracy.
- Waist Circumference (WC) or Waist-to-Height Ratio (WHtR ≥0.5): Targets visceral fat, strong cardiovascular predictors; endorsed by EASO (European Association for the Study of Obesity).
- Air Displacement Plethysmography (BodPod): Accurate whole-body density measure.
- Body Adiposity Index (BAI): Hip/height-based, ethnicity-independent.
Scandinavian universities promote WHtR in apps and wearables for population screening. The Verona team recommends combining BMI with surrogates for those outside normal range.
| Method | Accuracy | Accessibility | Cost |
|---|---|---|---|
| DXA | Gold standard (<2% error) | Clinical/lab | High |
| BIA | 3–5% error | Home/clinic | Low |
| WHtR | Good for visceral fat | Universal | Free |
Health Implications and Stakeholder Perspectives
Misclassification affects clinical decisions, insurance, and stigma. Overdiagnosed individuals may face unnecessary diets or medications; underdiagnosed MONW patients risk undetected cardiometabolic issues. In Europe, where obesity costs €70 billion annually, accurate tools optimize interventions.
Stakeholders like the European Society of Clinical Nutrition and Metabolism (ESPEN) advocate multiparametric assessments. Italian health authorities are reviewing guidelines post-Verona findings. Patient advocates stress reducing weight bias in higher education health curricula.ScienceDaily coverage.
Future Directions: Research and Policy Evolution in European Academia
University of Verona plans longitudinal DXA studies tracking BF% changes. EU Horizon projects fund AI-enhanced BIA for telemedicine. Other institutions like KU Leuven (Belgium) and Lund University (Sweden) explore ethnicity-specific models. Policymakers eye BF%-inclusive national surveys.
For academics, this underscores body composition's centrality in curricula—from physiology to public health. Emerging wearables promise real-time BF% via photoplethysmography, bridging research and practice.
Photo by Benjamin Lim on Unsplash
Actionable Insights for Clinicians, Students, and Individuals
Clinicians: Pair BMI with WHtR (>0.55 risky) or BIA. Students: Prioritize resistance training to preserve muscle. Individuals: Track waist/height; consult DXA if BMI extreme. European universities offer affordable BIA screenings via student health services.
This Verona-led breakthrough signals a paradigm shift, enhancing precision in Europe's fight against obesity.
