Examining the impact of pre-treatment life stressor exposure and sleep reactivity on chronic insomnia symptoms and response to cognitive behavioural therapy for insomnia

Purpose This study assessed the relationships between baseline sleep reactivity and life stressor exposure with insomnia symptoms before and after cognitive behavioural therapy for insomnia (CBTi). Methods 62 participants with chronic insomnia were randomized to either receive 8 weekly sessions of group-CBTi or to a wait-list control group. At baseline, participants reported sleep reactivity scores and underwent a structured life stress interview (LSI) assessing chronic and episodic stressors in various psychosocial domains. Symptoms were assessed multimodally from baseline to 1-year follow-up including subjective measures of sleep disturbances (Insomnia Severity Index [ISI]; Pittsburgh Sleep Quality Index [PSQI]; daily sleep diaries), depression (Beck Depression Inventory-II [BDI-II]) and polysomnographic (PSG) sleep assessments. Our main analysis investigated baseline symptom associations with LSI and sleep reactivity scores, and then predicted CBTi response at post-treatment (treatment versus wait-list) and 1-year following CBTi (pooled) using the baseline stress-related variables. Results Higher sleep reactivity was associated with higher ISI and PSQI scores, and lower PSG-derived sleep efficiency at baseline. Although higher global chronic stressor severity was associated with higher BDI-II scores at baseline, chronic and episodic stressors were not related to baseline sleep or insomnia symptoms. Sleep reactivity, chronic stressor severity, and episodic stressor exposure at baseline did not predict differences in CBTi response at post-treatment or 1-year follow-up. Conclusions Although higher sleep reactivity was associated with greater baseline insomnia symptoms, no strong associations were found between pre-treatment life stressors or sleep reactivity and CBTi responses. These findings suggest that CBTi may remain effective in stressful contexts, although replication employing larger, more clinically representative samples is needed. The present manuscript is a secondary analysis of a project registered as a clinical trial (ISRCTN13983243 - doi.org/10.1186/ISRCTN13983243).