UK Biobank Cohort Links Hemoglobin 1-3 g/dL Above WHO Thresholds to Lowest Mortality

The University of Glasgow analysis examined hemoglobin measured at baseline against WHO anemia cutoffs and tracked deaths through national registries. Investigators observed a clear U-shaped relationship: risk rose both below and substantially above the 1-3 g/dL window above threshold, after adjustment for age, sex, comorbidities and socioeconomic factors. Observational data cannot distinguish whether borderline-low hemoglobin is causal or merely a marker of subclinical disease; reverse causation and residual confounding remain plausible. Original coverage omitted that no randomized hemoglobin-target trial has tested these exact ranges, leaving clinicians without actionable intervention thresholds. The findings align with prior large cohorts in JAMA and Lancet that also reported U-shaped curves yet produced null results when erythropoiesis-stimulating agents were tested in CKD and heart-failure populations. Next steps require prospective validation in contemporary cohorts with serial hemoglobin measures and Mendelian randomization to assess causality before guideline panels consider threshold revision.