In late April 1995, an outbreak of a poorly defined respiratory illness related to the ingestion of leaves of Sauropus androgynus was observed in southern Taiwan. To further evaluate the association between S. androgynus and bronchiolitis obliterans syndrome, a hospital-based case-control study was conducted with one case group and three different control groups at Veterans General Hospital-Kaohsiung between April and September 1995. A total of 54 cases (50 females, 4 males), 54 age- and sex-matched neighborhood controls, 54 matched routine physical check-up controls, and 54 matched self-referred patron controls (who had ingested S. androgynus yet without obstructive physiology) were interviewed for clinical symptoms, history of S. androgynus consumption, and potential confounding factors. All 54 cases (100%) ingested S. androgynus compared with only five (9%) neighborhood controls (matched odds ratio (OR) incalculable, p < 0.001) and two (4%) physical check-up controls (matched OR incalculable, p < 0.001). In the univariate analysis of 54 cases and 54 self-referred patron controls, factors associated with an increased risk of bronchiolitis obliterans syndrome were methods of food preparation (uncooked juice vs. stir fried or boiled dishes, matched OR 10.3 (95% confidence interval (CI) 1.3-84.4)); preparer of the S. androgynus-containing food (vendor only vs. patient only or patient plus vendor, matched OR 2.8 (95% CI 1.1-7.1)); total S. androgynus consumption quantity (> 4,500 vs. 413-2,063 g, matched OR 10.0 (95% CI 1.9-53.0)); duration of consumption (> 45 vs. 6-24 days, matched OR 2.1 (95% CI 1.2-3.8)); and midterm interruption (< 2 vs. 2-5 days per week, matched OR 2.6 (95% CI 1.1-6.1)). Additionally, multiple conditional logistic regression analysis of cases and self-referred patron controls revealed that a larger total amount of S. androgynus consumption, preparation of S. androgynus food without cooking, and ingesting S. androgynus food prepared by a vendor were the significant risk factors associated with bronchiolitis obliterans syndrome.