The effects of age on the prevalence of sleep apnea in the general population remain unclear, because previous studies have focused on specific populations. The effects of age on the severity of apnea are unknown. This study was based on a two-stage general random sample of men (aged 20 to 100 yr), consisting of a telephone survey (n = 4,364) and a sleep laboratory evaluation of a survey subsample (n = 741). Obstructive sleep apnea (OSA), based on both sleep laboratory and clinical criteria (apnea/hypopnea index [AHI] > or = 10 and the presence of daytime symptoms) was found in 3.3% of the sample, with its maximum prevalence in the middle age group (45 to 64 yr). Also, based solely on laboratory criteria, the prevalence of OSA (obstructive AHI > or = 20) showed an age distribution similar to that of OSA diagnosed by laboratory and clinical criteria. The prevalence of any type of sleep apnea (central and obstructive) increased monotonically with age. However, central apnea appeared to account for this monotonic relationship with age. Severity of sleep apnea, as indicated by both number of events and minimum oxygen saturation, decreased with age when any sleep apnea criteria were used and when controlling for body mass index (BMI). The study shows that the prevalence of sleep apnea tends to increase with age but that the clinical significance (severity) of apnea decreases. On the basis of these findings, the sleep laboratory criteria used for diagnosis of sleep apnea should be adjusted for age.