Objective: To explore the key risk factors influencing in-hospital mortality in elderly patients with sepsis and pulmonary infection, and to develop and validate a risk prediction model for in-hospital mortality based on the Least Absolute Shrinkage and Selection Operator (LASSO)-Logistic regression algorithm, integrating the lactate-to-albumin ratio (LAR) and other indicators, aiming to provide a quantitative tool for early identification of high-risk patients and optimization of intervention strategies. Methods: A retrospective analysis was conducted on 102 elderly patients (age ≥ 65 years) with sepsis and pulmonary infection admitted to the Intensive Care Unit (ICU). Based on in-hospital outcomes (within 28 days), patients were divided into a survival group (n = 65) and a death group (n = 37). Clinical data of all patients were collected. LASSO regression was used to screen predictive variables. The selected variables were incorporated into multivariate Logistic regression analysis to establish a risk prediction model. The Bootstrap method was employed for internal validation. The area under the receiver operating characteristic curve (AUC) and the Hosmer–Lemeshow goodness-of-fit test were used to evaluate the model’s discrimination and calibration. Results: Univariate and LASSO-Logistic regression analyses revealed that Acute Physiology and Chronic Health Evaluation II (APACHE II) score (OR = 1.202, 95% CI: 1.078–1.326), Sequential Organ Failure Assessment (SOFA) score (OR = 1.366, 95% CI: 1.142–1.590), lactate-to-albumin ratio (LAR) (OR = 1.581, 95% CI: 1.242–1.920), and mechanical ventilation (OR = 5.523, 95% CI: 1.892–9.155) were independent risk factors for in-hospital mortality (P 0.05). Decision curve analysis (DCA) showed that the prediction model had good clinical applicability when the probability threshold ranged from 0.3 to 0.7. Conclusion: The prediction model based on APACHE II score, SOFA score, LAR, and mechanical ventilation, constructed via LASSO-Logistic regression, demonstrates good predictive performance for in-hospital mortality risk in elderly ICU patients with sepsis and pulmonary infection. It aids in the early identification of high-risk patients and facilitates timely intervention.