The frequency of sleep-disordered breathing (SDB) events increases dramatically with age, although the clinical significance of this phenomenon is uncertain. We report herein on data from the fifth follow-up observation of a cohort of healthy elderly which we selected, evaluated, and followed up in an effort to address this issue. Baseline observations on this group of 95 normal older persons medically screened for good health included standard nocturnal polysomnograms and daytime assessment of medical, sleep/wake, and psychological variables. At fifth-year follow-up, 42 subjects returned for assessment. Comparison of returning vs nonreturning subjects indicated no significant differences in baseline characteristics between these groups. Division of returning subjects into moderate (apnea-hypopnea index [AHI] > or = 5) vs low (AHI < 5) SDB at baseline resulted in 10 subjects in the moderate group and 32 subjects in the low group. The two groups were comparable with regard to sex, weight, and education, although the moderate SDB group was significantly older (70.4 years) than the low SDB group (64.8 years). Follow-up medical, sleep/wake, and psychological data were contrasted for the two groups using a mixed design multivariate analysis of variance (repeated measures factor-time-6 observations; between-subjects factor SDB level, 2 groups). There were no statistically significant effects of SDB or interactions of SDB and time across this 5-year follow-up. These results confirm observations from the third-year follow-up. We conclude that observation of a moderate level of SDB in an otherwise healthy older person is apparently not a significant risk across a 5-year follow-up period and therefore does not seem to warrant investigation beyond a thorough medical history and physical examination to rule out other disorders.