Sleep Chatbot Intervention Can Improve Mental Health and Sleep Quality in Young Black/African American Adults

Researcher(s)

  • Stanley Micek, Psychology, University of Delaware

Faculty Mentor(s)

  • Xiaopeng Ji, School of Nursing, University of Delaware

Abstract

Introduction: Young adults often experience mental health issues and poor sleep health, with young Black/African American (BAA) Adults often being at higher risk.  Improving sleep has been shown to improve mental health. Chatbots, artificial intelligence programs designed to simulate human conversation, are potentially accessible ways for sleep promotion programs. This feasibility study aimed to investigate the effect of sleep chatbot intervention on sleep quality, mental health, and executive function in young BAA adults. 

 

Methods: In this pretest-posttest study, 15 BAA college students (19.46±1.30 y.o., 11 females) completed the 4-week sleep intervention. The sleep chatbot mobile app automatically delivered weekly sleep coaching modules, tracked daily sleep, and provided weekly progress reviews and goal setting. Participants completed the Pittsburgh Sleep Quality Index (PSQI), Patient Health Questionnaire (PHQ-9) that assessed depressive symptoms, Generalized Anxiety Disorder (GAD-7), and Behavior Rating Inventory of Executive Function (BRIEF) at baseline and the end of intervention. The student’s t tests were used to compare the sleep, mental health, and EF scores before and after intervention.

 

Results: Compared with baseline scores, participants had significantly lower scores on PHQ-9 (7.53±3.38 vs.4.66±2.98,  P=0.02) and PSQI (8.46±2.41 vs. 5.80±2.56, P=0.01), suggesting lessened depressive symptoms (and improved sleep quality. There were marginal improvements in anxiety and executive function, but they were not statistically significant. 

Conclusion: Sleep chatbot intervention can reduce depressive symptoms and improve sleep quality, but not anxiety and executive function,  in young BAA adults. Future randomized controlled trials with a large sample size are needed to replicate study findings.