To evaluate the morbidity associated with obstructve sleep apnea syndrome (OSAS), we undertook a seven-year follow-up study of 198 OSAS patients seen between 1972 and 1980. The patients had been submitted to tracheostomy (71 patients) or had received a weight-loss recommendation (127 patients). Despite a lower mean apnea index (AI) (43 vs 69) and a lower mean body mass index (BMI) (31 vs 34 kg/m2) at entry, excessive daytime sleepiness (EDS) and vascular morbidity were significantly higher in the conservatively treated group. The relative risk (odds ratio) of finding EDS in the conservatively treated group, after adjustment for BMI at seven-year follow-up, was 3.7 (95 percent confidence interval [CI] = 2.6-5.3). The relative risk of developing new vascular problems in the same population, estimated by Cox models, was 2.3 (95 percent CI = 1.5-3.6). The effect of tracheostomy, independent of age, BMI, and AI at entry, was highly significant. At entry, 56 percent of the population already had a vascular problem, particularly hypertension, thus emphasizing the need for earlier treatment of the sleep-related abnormal breathing.