In the 30-year period from 1956 to 1985, 471 Rochester, MN residents had an initial operation for peptic ulcer disease, 438 of whom were followed for at least 30 days (median 14.8 years per subject). In this population-based cohort, risk was elevated for all of the fracture sites traditionally associated with osteoporosis, including the proximal femur (standardized incidence ratio [SIR] 2.5, 95% CI 1.9-3.3), vertebra (SIR 4.7, 95% CI 3.8-5.7), and distal forearm (SIR 2.2, 95% CI 1.5-3.1). Fracture risk rose with age and was greater among women than men, but there was no influence on overall fracture risk of ulcer type or nature of the operation. In multivariate analyses, the independent predictors of vertebral fractures were age (hazard ratio [HR] per 10-year increase 1.8, 95% CI 1.6-2.0), use of corticosteroids (HR 2.3, 95% CI 1.01-5.2), thyroid replacement (HR 2.5, 95% CI 1.4-4.6), chronic anticoagulation (HR 2.3, 95% CI 1.1-4.6), and the presence of one or more conditions associated with secondary osteoporosis (HR 1.6, 95% CI 1.2-2.1). Gastrectomy with Billroth II reconstruction appeared to be relatively protective (HR 0.5, 95% CI 0.3-0.9), but such patients still had an increased risk of vertebral fractures compared with community residents generally (SIR 3.6, 95% CI 2.4-5.4). The independent predictors of hip fracture risk in this cohort were age (HR 2.7, 95% CI 2.1-3.5) and use of corticosteroids (HR 5.8, 95% CI 2.2-15.3) or anticonvulsants (HR 4.6, 95% CI 1.8-12.0), while higher body mass index was protective (HR 0.9, 95% CI 0.8-0.96). The independent predictors of distal forearm fractures were female gender (HR 4.7, 95% CI 2.2-10.1) and chronic anticoagulant use (HR 2.8, 95% CI 1.1-7.3). Thus, while the risk of osteoporotic fractures was significantly increased among patients operated for peptic ulcers, this appeared to be due more to specific characteristics of the cohort than to adverse effects of particular surgical procedures.