The sleep of 15 adult subjects who reported heightened generalized anxiety in the absence of other psychiatric syndromes and a 15-adult contrast group were studied by means of nocturnal polysomnography. Analysis of polysomnography variables revealed a significant discriminant function that accounted for 79% of the variance between groups, indicating that high-anxiety/worry subjects took longer to fall asleep, had a smaller percentage of deep (slow-wave) sleep, and more frequent transitions into light sleep [stage 1 nonrapid eye movement (NREM)]. Additional analyses indicated that high-anxiety/worry subjects had a greater percentage of light sleep, more early microarousals, a lower rapid eye movement (REM) density relative to low-anxiety subjects. These subjects also showed more electrodermal storming when slow-wave sleep and REM sleep variables were covaried. Results indicated disrupted sleep depth and continuity similar to that documented in clinical anxiety disorder patients and distinct from that of depressed patients. These results indicate that generalized anxiety and worry in otherwise healthy individuals may act to produce a clinically significant sleep disturbance in the absence of other psychiatric symptoms.