1218 The Effect of a Sleep- and Physical Activity-Related Texting-Based Conversational Agent on Perceived Cognitive Function in Breast Cancer Survivors

Abstract Introduction Cancer-related cognitive impairment is a common symptom among breast cancer (BC) survivors that may be improved with sleep and physical activity. Cecebot, a 6-week texting-based conversational agent intervention, integrates cognitive behavioral therapy for insomnia and physical activity components. This study examines the effect of the Cecebot intervention on subjective cognitive function in BC survivors. Methods This 12-week crossover randomized controlled trial enrolled female BC survivors experiencing insomnia symptoms based on the Patient Reported Outcomes Measurement Information System Sleep Disturbance 8b scale (≥55 T-score). Data were collected at baseline, 6 and 12 weeks. The Insomnia Severity Index (ISI) was used to measure insomnia symptoms, and subjective cognitive function was assessed using the Perceived Cognitive Impairment subscale of the Functional Assessment of Cancer Therapy-Cognitive Function. The intervention group received Cecebot from Week 1–Week 6 and the control group from Week 6–Week 12. Linear mixed-effects models were used to analyze data. Results 60 participants (mean age 53.2±11.1) with mild-to-moderate insomnia symptoms (mean Insomnia Severity Index=13.6±4.35) were enrolled (n=32 intervention, n=28 control). There was no significant baseline between-group difference in cognitive function (intervention mean=48.5, control mean=48.3; p=0.96). The intervention group showed a significant improvement in subjective cognitive function at Week 6 compared to baseline, with a mean increase of 6.47 points in cognitive function (95% confidence interval [CI]: 3.24, 9.69; p< 0.01), while the waitlist control group reported an decrease of 1.08 points. No significant between-group difference was observed (p=0.12). At Week 12, the improvements in the intervention group were sustained, and there was no between-group difference (p=0.63) due to a significant improvement in the waitlist control group from Week 6 to Week 12 (+5.55 points, 95% CI: 1.87, 9.22; p< 0.01). Conclusion This study suggests that a texting-based conversational agent sleep intervention may support improvements in subjective cognitive function in BC survivors, warranting a double-blind confirmatory study with a larger sample. Future studies may explore objective measures of cognitive function and investigate whether sleep or physical activity improvements mediate cognitive function improvements to better understand how Cecebot affects cognitive function. Support (if any) SWOG/Hope Foundation Impact Award and the UW School of Nursing Center for Innovation in Sleep Self-Management