Objectives: To assess if baseline iron deficiency, with or without anemia, is associated with incident infections over 3 years among community-dwelling older adults. Design: Prospective secondary analysis of DO-HEALTH, a 3-year randomized, double-blind controlled trial. Setting And Participants: 2157 community-dwelling adults age 70+ from 5 European countries with good cognitive function and mobility and no major health events in the 5 years prior to enrollment Measurements: Incident infections, their severity and type were recorded every 3 months throughout the 3-year follow-up. Iron deficiency was defined as soluble transferrin receptor (sTfR) levels > 28.1 nmol/l and anemia as hemoglobin levels < 120 g/l for women and 130 g/l for men. We applied negative binomial mixed effects regression models with random effects for countries, and controlling for treatment allocation, age, sex, body mass index, polypharmacy, number of comorbidities, smoking status, living situation, alcohol intake, frailty status, and physical activity levels. A pre-defined stratified analysis was performed to explore if the associations between iron deficiency and infections were consistent by baseline anemia status. Results: In total, 2141 participants were included in the analyses (mean age 74.9 years, 61.5% of women, 26.8% with iron deficiency). Across all participants, baseline iron deficiency was not associated with incident overall infections, but was associated with a 63% greater rate of incident severe infections requiring hospitalization (incidence rate ratio [IRR] 1.63, 95% Confidence Interval [CI] 1.11–2.41, p=0.01). This association was more pronounced among the 2000 participants who did not have anemia at baseline (IRR=1.80, 95% CI 1.20–2.69, p=0.005). Conclusion: Based on this prospective study among generally healthy European community-dwelling older adults, iron deficiency was not associated with the incidence rate of overall infections but may increase the incidence of severe infections. Intervention studies are needed to prove the causality of this observation.