Abstract Objective: To assess the role of percutaneous endoscopic gastrostomy (PEG) insertion, and its timing, on ALS survival, and to study prognostic factors of survival before and after PEG placement in a population-based setting. Methods: In this observational population-based, registry study, we enrolled patients with newly- diagnosed ALS, according to the El Escorial revised criteria, who were resident in the Emilia Romagna Region, and who developed severe dysphagia needing enteral nutritional support. The primary outcome measure was tracheostomy-free survival after PEG recommendation. Results: There were 210 patients needing PEG, out of an incident cohort of 545 patients from the Emilia Romagna Registry for ALS, who were diagnosed between 2009 and 2013. One hundred and ninety-three patients were included in the study, and 17 were excluded because they were already tracheostomized at the time of PEG placement. Of the 193 patients included in the study, 152 underwent PEG, whereas 41 did not undergo the procedure. Patients who did not undergo PEG, among the eligible ones, had the same tracheostomy-free survival from onset as patients who did (25 vs. 32 months, p¼0.21). Tracheostomy-free survival from PEG recommendation was greater in patients who underwent PEG placement than in patients who did not (6 vs. 2 months, p¼0.008). Median tracheostomy-free survival from PEG insertion was eight months (95% CI5–12); 30 days after PEG placement, survival was 89.60%. At Cox multivariable analysis, the hazard of death or tracheostomy after PEG insertion was significantly influenced by the difference between BMI at the time of the PEG procedure and BMI at diagnosis (HR 1.05, 95% CI 1.02–1.08; p¼0.002). The hazard of death or tracheostomy was not affected by the timing of PEG insertion. Conclusions: The present study, although it has some limitations, suggests a gain of tracheostomy-free survival from the time of PEG recommendation for patients who undergo PEG placement, and, among patients who undergo PEG, a greater survival if PEG is inserted before a significant weight loss occurs, and if nutritional support avoids further weight loss. Should this association between prevention of weight loss and better clinical outcome be confirmed by further studies, it would have important implications for disease management.