A reanalysis of NHANES data spanning 1999–2020 (N = 23,363 adults) finds that replacing BMI with DXA-derived appendicular lean mass index (ALMI) and weight-loss history in malnutrition screening preserves diagnostic yield — 17.9% vs. 18.9% prevalence — while eliminating a structural flaw called 'circularity bias.' Critically, low BMI alone was the sole indispensable diagnostic driver in just 233 individuals (5.2% of malnutrition cases), suggesting BMI contributes minimally to identifying true malnutrition in contemporary U.S. cohorts.
This finding lands at a critical moment in nutrition science. As obesity rates climb past 40% in U.S. adults, clinicians routinely overlook sarcopenic malnutrition — significant muscle wasting hidden beneath excess adipose tissue, a condition increasingly linked to poor surgical outcomes, accelerated aging, and reduced healthspan. The current GLIM consensus, while a landmark standardization effort, was designed partly for global populations where low BMI still signals undernutrition clearly. In high-obesity Western populations, BMI actively obscures what DXA body composition data can reveal.
Limitations matter here: NHANES is cross-sectional and observational, precluding causal inference. DXA access remains limited outside research settings, reducing immediate clinical scalability. This is also a preprint posted on medRxiv and has not yet undergone peer review, so methodology and conclusions may shift. If validated, however, a weight-independent malnutrition standard could reshape screening protocols across hospitals, elder care facilities, and epidemiological surveillance — a genuinely paradigm-shifting proposal for clinical nutrition.
