Background. The value of endoscopy in dyspeptic patients is questionable. Aims. To examine the prevalence of significant endoscopic findings (SEFs) and the utility of alarm features and age in predicting SEFs in outpatients with dyspepsia. Methods. A retrospective analysis of outpatient adults who had endoscopy for dyspepsia. Demographic variables, alarm features, and endoscopic findings were recorded. We defined SEFs as peptic ulcer disease, erosive esophagitis, malignancy, stricture, or findings requiring specific therapy. Results. Of 650 patients included in the analysis, 51% had a normal endoscopy. The most common endoscopic abnormality was nonerosive gastritis (29.7%) followed by nonerosive duodenitis (7.2%) and LA-class A esophagitis (5.4%). Only 10.2% had a SEF. Five patients (0.8%) had malignancy. SEFs were more likely present in patients with alarm features (12.6% versus 5.4%, p = 0.004). Age ≥ 55 and presence of any alarm feature were associated with SEFs (aOR 1.8 and 2.3, resp.). Conclusion. Dyspeptic patients have low prevalence of SEF. The presence of any alarm feature and age ≥ 55 are associated with higher risk of SEF. Endoscopy in young patients with no alarm features has a low yield; these patients can be considered for nonendoscopic approach for diagnosis and management.