Prevalence of Insomnia, Nightmares and PTSD Symptoms: Development and Testing of a Multi-Component Cognitive Behavioural Treatment in Australian, Canadian, and American Wildfire Survivors

Wildfires are now recognised as a global crisis, posing significant threats to the economy, the environment, wildlife, physical and mental health, and human wellbeing. Sleep difficulties and trauma symptoms following the fires are highly prevalent, yet they are often neglected and inadequately treated. If left untreated, sleep and trauma symptoms can perpetuate and become unresponsive to treatment causing more complex health conditions. Therefore, the first aim of the thesis was to establish prevalence rates of sleep and trauma symptoms in an international sample of wildfire survivors. A total of 126 (23 males, 102 females, and 1 nonbinary individual, Mage = 52 years, SD = 14.4) wildfire survivors from Australia, Canada, and the United States of America (USA) took part in an online survey. Participants completed a demographic questionnaire and self-report measures including the Insomnia Severity Index (ISI), PTSD Checklist for DSM-5 (PCL5), and Disturbing Dream and Nightmare Severity Index (DDNSI). Nearly half (49.2%) of the sample reported clinical insomnia, 28.7% reported nightmares, and 77.8% reported PTSD symptoms. In response to these alarming prevalence rates, this thesis addressed its second objective by designing and evaluating the feasibility of a digital, self-paced intervention for the treatment of sleep and trauma symptoms. Sleep Best-i, a digital and self-paced intervention, comprising of cognitive behavioural therapy for insomnia (CBTI), and exposure, relaxation, and rescripting therapy (ERRT) was evaluated in a fourweek clinical trial, targeting wildfire survivors from Australia, Canada, and the USA presenting with sleep difficulties and/or trauma symptoms. To qualify for the study, wildfire survivors had to meet at least one of the following criteria: a score of ≥ 8 on the Insomnia Severity Index (ISI), and/or a score of ≥ 3 on the Nightmare Disorder Index (NDI), and/or a score of ≥ 31 on the PTSD Checklist – Civilian Version (PCL-5). Thirty WILDFIRES AND SLEEP DISTURBANCES iii wildfire survivors were allocated to either the treatment group (n = 16) or the waitlist control group (n = 14) in a sequential manner. Participants' ages ranged from 18 to 79 years, with a mean age of 52.5 years (SD = 16.26). The study sample consisted of 19 females (63.3%) and 11 males (36.7%). They completed self-report secondary outcome measures, including the Generalized Anxiety Disorder Questionnaire (GAD-7), the Patient Health Questionnaire (PHQ-9), and the Pittsburgh Sleep Quality Index (PSQI), via the HealthZone digital platform. Study participants were informed about the purpose, allocation, and the structure of the study. Assessments were conducted at baseline, posttreatment, and 3-months follow-up, with the waitlist group providing an additional assessment at pre-treatment, after 4 weeks of waiting, and prior to crossing over to treatment. The study employed Intention-To-Treat (ITT), and Per Protocol (PP) analyses. Mixed-effects linear regression models and difference-in-difference analyses were utilised to assess the intervention's effect. The ITT analysis revealed significant improvements over time (main effect of time) with a 1.64 point reduction (p = .001) on the NDI and 10.64 point reduction (p = .009) on the PCL-5 at post-intervention. No significant changes were observed in insomnia symptoms. On the secondary measures, there was an interaction effect of condition x time with a 2.22 point reduction (p <.001) on the PSQI and a main effect of time with a 6.48 point reduction (p <.001) on the PHQ9. No significant changes were detected on the GAD-7. The PP analysis yielded comparable results on both the primary and the secondary measures. The findings of the pilot trial demonstrated a reduction in nightmares and trauma symptoms. Future research studies should aim at evaluating Sleep Best-i in a more definitive trial with a larger sample size.