Evaluating the Impact of Accessible and Scalable Digital Interventions to Address Co-Occurring Insomnia and Depression in Breast Cancer Survivors Using an Adaptive Design

Project Summary/Abstract Insomnia and depression are highly prevalent among breast cancer survivors, and the co-occurrence of these conditions significantly increases the risk of premature death. There is robust evidence that cognitive- behavioral therapy, which is highly scalable when delivered digitally, is effective for reducing symptoms of insomnia and depression. SHUTi and IntelliCare are automated evidence-based digital interventions that have individually been shown to address symptoms of insomnia and depression, respectively. However, their impact on survivors with co-occurring symptoms of insomnia and depression have not been thoroughly evaluated. Furthermore, it is not currently understood how multiple digital interventions could work together sequentially to address these reinforcing conditions. This is important because providing one of these interventions alone may not address the diverse ways insomnia and depression can manifest or their interrelatedness; while some survivors will respond to a single intervention, others will require additional treatment to gain relief from their co-occurring symptoms. Members of our research team previously found that in a study of breast cancer survivors experiencing clinical levels of insomnia, SHUTi had a large effect on sleep quality. Conversely, in a study of breast cancer survivors experiencing clinical levels of depression, IntelliCare had a strong effect on depression. The positive results from these trials lead us to the next logical step of testing the efficacy of SHUTi and IntelliCare against a robust control in a randomized study with long-term follow-up, as well as developing adaptive treatments that optimally sequence these interventions to cater to the diverse needs of breast cancer survivors. Using an innovative experimental design, called the Sequential, Multiple Assignment, Randomized Trial (SMART), 747 breast cancer survivors with elevated symptoms of insomnia and depression will be randomized to initially receive SHUTi, IntelliCare, or active control. We will test the impact of SHUTi and IntelliCare on survivors’ response for their co-occurring symptoms of insomnia and depression after 9 weeks (Aim 1). Responders are defined as those reporting clinically meaningful improvements in symptoms of both conditions, while non-responders are those who report improvements in only one or neither condition. After 9 weeks, non-responders in the SHUTi condition will be re-randomized to either receive IntelliCare or not; similarly, non-responders in the IntelliCare condition will be re-randomized to either receive SHUTi or not (Aim 2). This will allow us to evaluate whether added intervention improves response for survivors whose co- occurring symptoms do not respond to a single intervention. Finally, after the trial we will match subgroups of survivors with the treatment model that is most efficacious for them, based on their characteristics and treatment preferences (Aim 3). This will be the first proposal to inform how to treat co-occurring insomnia and depression among breast cancer survivors by understanding how to optimally sequence highly accessible and scalable digital interventions that have immediate potential for public impact for serious health conditions.