To clarify whether and how psychologic factors might contribute to peptic ulcer, in this study we addressed the relations between psychologic characteristics and known biologic risk factors among ulcer patients. In 75 patients with recent-onset, symptomatic duodenal ulcer, an index of three potential psychologic risk factors (stressful life events, abnormal Minnesota Multiphasic Personality Inventory, mood disturbance) was examined in relation to historical risk factors (sex, age, seasonality, family history, smoking, alcohol use, coffee consumption, nonsteroidal antiinflammatory drug use), and to blood type, serum pepsinogen I, and Helicobacter pylori antibody titers. The more risk factors in a patient's history, the less likely he or she was to have psychopathology or stress (r = -0.45, p = 0.0007). The mean number of conventional risk factors was 2.7 in patients with all three psychologic risk factors and 4.4 in patients with none; conversely, the mean number of psychologic factors was 0.9 among patients with five or more biologic risk factors and 2.1 among patients with fewer than three risk factors. This negative association was strongest among patients with no previous ulcer history. The psychologic factors also tended to vary inversely with H. pylori antibody titers but not with blood type O or pepsinogen. Duodenal ulcer patients who are atypical in terms of their conventional risk factors are likely to be emotionally fragile, under stress, or both, especially at the time of their first ulcer symptoms. A clinician diagnosing an ulcer in an individual who does not match the usual patient profile should be on the lookout for psychologic factors.