Abstract Background and Aims Iron deficiency (ID) is a poor prognostic factor in chronic diseases such as chronic kidney disease (CKD). Indeed, independently of anemia, ID is associated with more rapid clinical worsening of CKD and is a risk factor for mortality in this condition. In older patients, CKD is common and often associated with other comorbidities. Method The CARENFER PA study is a national multicenter study designed to evaluate ID in older patients (>75 years) newly hospitalized in a geriatric unit. The primary endpoint was the prevalence of ID, defined as serum ferritin < 100 µg/L and/or transferrin saturation coefficient (TSAT) <20%. Absolute ID was defined as serum ferritin < 100 µg/L and functional ID as serum ferritin ≥ 100 µg/L and TSAT < 20%. In this analysis, we explored ID prevalence according to the degree of CKD. The Short Physical Performance Battery (SPPB) test was used to identify older patients at high risk of adverse events (e.g., disability, falls, hospitalization, death). Results 888 patients (mean age, 85.2 years; women, 63.5%) from 16 French centers were analyzed. Stages of CKD were: stage I, 3.8%; II, 53.2%; IIIa, 22.3%; IIIb, 13.3%; IV, 6.0% and V, 1.4%. ID prevalence in the entire cohort was 57.6% (495/859). When present, ID was absolute in 56.2% (278/495) of cases and functional in 43.8% (217/495). ID prevalence was 54.3% (265/488) in patients with stages I-II, and 61.9% (227/367) in patients with stages III-V (p = 0.0271). In the subgroup of anemic patients, ID prevalence was 58.9% (112/190) and 65.9% (147/223), respectively (p = 0.14). We assessed, according to the stages of CKD, various factors likely to affect or be associated with ID. Thus, in patients with at least three comorbidities, ID prevalence was 56.9% (74/130) in patients with stages I-II and 64.8% (142/219) with stages III-V (p = 0.14); in patients with heart failure, ID was present in 63.2% (36/57) and 60.3% (76/126), respectively (p = 0.71). If patients received treatments with possible effect on ID/anemia (e.g., anticoagulant, proton pump inhibitors, platelet antiaggregant), ID prevalence was 58.4% (192/329) for stages I-II and 62.7% (188/300) for stages III-V (p = 0.27). In patients with CRP ≥ 12 mg/L, ID prevalence was 73.8% (96/130) and 72.2% (117/162), respectively (p = 0.76). In univariate analysis of the entire cohort, degree of CKD (stages III-V vs. I-II), at least three comorbidities, CRP ≥ 12 mg/L, anemia and low serum albumin were significantly associated with ID. In multivariate analysis, CRP ≥ 12 mg was the only independent predictor of ID (odds ratio, 2.78; 95% CI, 1.92–4.08; p < 0.0001). In patients with stages I-II, physical performances were poor (SPPB score 0-6) in 44.5% (77/173) of patients without ID and 57.5% (119/207) with ID; in patients with stages III-V, SPPB scores were low in 55.6% (60/108) of patients without ID and 63.5% (106/167) with ID. Conclusion ID prevalence is high (over half of patients) in older patients admitted to a geriatric unit—particularly those with CKD—and is associated with inflammation and a low SPPB score. These new data underline the importance of more systematic screening for ID in older patients. Future randomized controlled trials should evaluate the efficacy of iron supplementation on the outcome of older patients with ID.