Abstract Background Due to the poor outcomes reported in older patients with hematological malignancies (HM) and pulmonary infections, there is a need to identify patients at high risk of mortality. Objective: This study aims to find the association between frailty and mortality in patients above 65 years of age with HM hospitalized with pulmonary infections. Methods A cross-sectional, retrospective study was conducted on the clinical data of 64 HM older patients with pulmonary infection hospitalized in a tertiary university hospital in Bucaramanga, Colombia, between January 2015 and December 2020. Patients were assessed using Clinical Frailty Scale (CFS) and were divided into three groups: CFS 1–4 fit; CFS 5–6 frail and CSF 7–9 severely frail. The association between frailty and in-hospital mortality was the primary outcome. All statistical analyses were performed using Stata/IC 14.0. Results The number of participants classified as fit, frail, and severely frail was 21 (32.8%), 22 (34.3%), and 21 (32.8%), respectively. 48.4% were women, and the mean age was 75.4 ± 6.85 years. A total of 45 patients (70.3%) died during hospitalization. We found a significantly higher in-hospital mortality rate among frail (23.8% vs. 90.9%; p < .001) and severely frail patients (23.8% vs. 95.2%; p < .001), when compared to fit patients. After multivariate analysis, high CFS score was an independent risk factor for mortality, OR was 6.8 (1.40–32.97, p = .001) for frail and 14.4 (1.76–117.32 p = .013) for severely frail compared to fit patients, after adjustment for sex, type of HM and the presence of comorbidities. Table 1Baseline characteristics of study participants.Image 1Kaplan-Meier survival estimate for mortality for patients ≥65 years with hematological malignancies and pulmonary infection. Conclusion Clinical Frailty Scale (CFS) could be used as a potentially useful tool in predicting mortality of pulmonary infections in elderly patients with HM. These results could suggest that the use of this score extends beyond evaluating the degree of frailty and could predict adverse outcomes and help decision making in complex clinical scenarios. Disclosures All Authors: No reported disclosures.