1047 FOSI Mediates Nightmares and Insomnia in Women Who Have Experienced Sexual Trauma
Sophie Nilsson, Garrett Baber, Elijah Nichols, Anna K Quesada, N. Hamilton
SLEEPJ May 1, 2026 DOI: 10.1093/sleep/zsag091.1046 (opens in new tab)
Study at a glance
AI-extracted from the abstract| Characteristics | Cross-sectional study Peer reviewed |
|---|---|
| Sample size | 497 |
| Population | College-aged women (mean age 18.9 years) reporting a history of sexual trauma |
| Measures | Life Event Checklist (LEC), Post Traumatic Stress Disorder Checklist-5, Civilian edition (PCL-C), Insomnia Severity Questionnaire (ISI), Nightmare Distress Inventory (NDI), Fear of Sleep Inventory (FOSI) |
| Topics | Dreaming |
| Key findings | Fear of sleep, and most prominently fear of nightmares, mediated the association between nightmares and insomnia in this sample, consistent with the model proposed by Werner and colleagues. Fear of nightmares accounted for 35.6% of the total effect, vulnerability 30.4%, vigilance 21.1%, and fear of darkness 12.2%. The authors caution that the cross-sectional design does not rule out other causal trajectories. |
Abstract
Trauma exposure is a known risk for sleep disruption from both nightmares and insomnia. Informed by recent research proposing specific mechanisms linking trauma to sleep disruption and between nightmares and insomnia, (Werner et al. 2021), we tested a model in which the relationship between nightmares and insomnia was mediated by fear of sleep. Study participants were college-aged women (N=497; Mage=18.9, SD=2.0) who reported a history of sexual trauma. Participants completed a one-time battery of questionnaires: Life Event Checklist (LEC), the Post Traumatic Stress Disorder Checklist-5, Civilian edition (PCL-C), the Insomnia Severity Questionnaire (ISI),the Nightmare Distress Inventory (NDI), and the Fear of Sleep Inventory (FOSI), using factor analytically derived subscales. The average PCL-C score was 47.6 (SD=15.1) with approximately 60% of the participants scoring at or above44, indicating high prevalence of PTSD symptoms in this sample. The total effect of ISI score on NDI was significant (t(496) = 10.76, p < 0.001), accounting for 19% of the total variance in the ISI. The results include: FOSI vigilance mediated the association between ISI and NDI (indirect effect = 0.13, SE = 0.03, p < 0.001, standardized βIndirect = 0.09), accounting for 21.1% of the total effect. FOSI Fear of Darkness mediated the association between NDI and ISI (indirect effect = 0.08, SE = 0.02, p = 0.002, Standardized βIndirect = 0.05), accounting for 12.2% of the total effect. FOSI Fear of Nightmares mediated the association between NDI and ISI (indirect effect = 0.22, SE = 0.04, p < 0.001, Standardized βIndirect = 0.15) accounting for 35.6% of the total effect. FOSI Vulnerability mediated the association between ISI and NDI indirect (effect = 0.19, SE = 0.04, p < 0.001, Standardized βIndirect = 0.13), accounting for 30.4% of the total effect. Consistent with the model proposed by Werner and colleagues, Fear of Sleep, and most prominently, Fear of Nightmares, appears to link nightmares to insomnia in this sample of young adult women reporting a history of sexual trauma. Although we have modeled nightmares upstream of Fear of Sleep, these are cross-sectional data and thus do not rule out other causal trajectories. Support (if any)