0472 Dynamic Features of the Treatment Process Predict Different Outcomes for Patients Undergoing Cognitive Behavioral Therapy for Insomnia

Abstract

Abstract Introduction Individuals who undergo Cognitive Behavioral Therapy for Insomnia (CBT-I) show different trajectories during treatment and different outcomes after treatment. However, the factors contributing to this interindividual variability have not been adequately identified. Daily sleep logs might help us understand some of these factors. Methods Patients with insomnia (N=34), as confirmed by clinical evaluation and Insomnia Severity Index score >10, completed 4 weeks of group CBT-I conducted in Mandarin via telehealth. Participants completed daily sleep logs and self-reported sleep quality, sleep effort and anxiety before and after treatment. Predictor variables included daily sleep efficiency and daily sleep anxiety. Response variables included pre-post changes in sleep quality, sleep effort and anxiety. Multilevel structural equation modeling was performed to examine the relationship between trajectories of change in predictors and response variables. Results Dynamic features of sleep log data significantly predicted individual differences in response variables. The growth rate of daily sleep efficiency during treatment significantly predicted reduced sleep effort (p=.012), decreased anxiety (p=.046), and improved sleep quality (p<.01) after treatment. Conversely, the growth rate of daily sleep anxiety during treatment significantly predicted increases in sleep effort (p<.01). In addition, daily sleep anxiety had a significant negative effect on daily sleep efficiency (p<.001), and the weaker this negative effect, the greater the reduction in sleep effort (p<.001) and anxiety (p<.01) after treatment. Conclusion Individuals exhibit different trajectories during treatment, which are related to different outcomes after treatment. Those who experience less wakefulness at night across treatment become less anxious about sleep and exert less effort to manage sleep, leading to better sleep quality. However, those who experience increasing anxiety across treatment exert more effort to control sleep. Finally, individuals whose anxiety is not as closely associated with sleep efficiency show greater reductions in anxiety and sleep effort post-treatment. Thus, different trajectories and treatment outcomes appear to distinguish between those individuals who are less anxious about sleep during CBT-I, and thus more willing to implement recommendations, which leads to better outcomes, and those individuals who are more anxious about sleep during treatment, and thus less willing to implement recommendations, which leads to worse outcomes. Support (If Any)