Predicting Anxiety Trajectories from Individual Differences in Sleep–Related Distress Alleviation

Importance

Anxiety of fluctuating severity is common in many psychiatric disorders. Few studies addressed factors determining individual differences in trajectories of the recovery phase, while their identification could inspire treatment innovation. Given the role of REM sleep in overnight alleviation of emotional distress, we here investigate whether individual differences in this overnight regulatory process matter for anxiety recovery.

Objective

To investigate whether individual differences in overnight alleviation of distress by REM sleep predict anxiety recovery rate.

Design

People tend to volunteer for intervention trials when fluctuating symptoms peak. This results in a significant recovery even in waitlist or control conditions. Leveraging this opportunity to recruit people prior to the recovery phase, in this cohort study, we utilized data from people who volunteered for optional sleep EEG and overnight distress assessment prior to their participation in an intervention trial (2021-2025). Anxiety severity was assessed at baseline and two months later.

Setting

Home-based assessment in the Netherlands.

Participants

Adults with insomnia alongside cross-threshold symptom severities of generalized anxiety disorder, social anxiety disorder, panic disorder, posttraumatic stress disorder, or borderline personality disorder (N = 223, 157 female [70.4%]; mean [SD] age, 45.7 [14.5] years; clinical diagnoses confirmed in 165 [74.0%]).

Exposures

Cognitive behavioral therapy for insomnia (CBT-I) or waitlist control.

Main Outcomes and Measures

Predicting 2-month anxiety improvement by individual differences in the strength of the effect of REM sleep on overnight distress alleviation at baseline.

Results

Within-subject mixed model analysis showed stronger overnight distress alleviation across nights with longer REM sleep ( b = -0.011; 95% CI, -0.016 to -0.007; P < .001). Individual differences in the strength of REM–related distress alleviation predicted anxiety improvement after two months ( b = −0.521; 95% CI, −0.854 to −0.188; P = .002), irrespective of treatment or waitlist control (interaction b = 0.011; 95% CI, −0.656 to 0.678; P = .97).

Conclusions and Relevance

Individual differences in the degree to which REM sleep drives overnight alleviation of distress predict the trajectory of anxiety recovery in people with clinically relevant psychiatric complaints. These findings suggest REM-related emotion regulation as a mechanism linking sleep physiology to anxiety recovery.

Key Points