Abstract
Iron is central to multiple biological pathways, and treatment of non-anaemic absolute iron deficiency (NAID) is beneficial in certain conditions. However, it is unknown if NAID is associated with increased mortality in older adults. A nationally representative sample of 4451 older adults from the English Longitudinal Study of Ageing was used. NAID was defined as serum ferritin < 30 μg/l and haemoglobin ≥ 120 g/l (women) or ≥ 130 g/l (men). Cumulative mortality was estimated by Kaplan-Meier method. Unadjusted and adjusted hazard ratios (HRs) of mortality were calculated using Cox proportional hazards regression models. Baseline NAID prevalence was 8·8% (95% confidence interval [CI] 8·0-9·7%); 10·9% (95% CI 9·7-12·3%) for women and 6·35% for men (95% CI 5·3-7·5%). The HR for mortality for individuals with NAID compared with non-anaemic individuals without iron deficiency over the 14-year follow-up was 1·58 (95% CI 1·29-1·93). This association was independent of all identified demographic, health-related and biological covariates, and robust in multiple sensitivity analyses. In older adults in England, NAID is common and associated with an increased mortality rate compared to non-anaemic individuals with normal serum ferritin. The association is principally driven by an excess mortality in women.