Leveraging Natural Language Processing to Evaluate Young Adults' User Experiences with a Digital Sleep Intervention...

Area of Science:

• Digital health interventions
• Behavioral science
• Public health

Background:

• Evaluating digital interventions is crucial for adherence and efficacy.
• Traditional evaluation methods have limitations.
• Personalized digital interventions show promise for behavior change.

Approach:

• Convergent mixed-methods approach combining natural language processing (NLP) with traditional evaluation.
• Randomized clinical trial (N=120) comparing a personalized feedback and coaching digital sleep intervention ('Call it a Night' - CIAN) against two control conditions (web-based advice + monitoring, and advice only).
• NLP analysis of user experiences to identify key motivators and sentiments.

Key Points:

• Participants found personalized feedback and coaching (CIAN) most helpful, while advice was generally beneficial across groups.
• Interest extended beyond alcohol use to broader sleep and whole-health factors.
• High adherence, satisfaction, and feasibility were reported across all intervention groups.
• CIAN and advice + monitoring groups reported significantly higher effectiveness than advice only.
• NLP identified sleep improvement as a primary participant motivator, surpassing alcohol reduction.

Conclusions:

• Digital sleep interventions are an acceptable and novel strategy for alcohol risk reduction.
• Improving sleep and overall wellness may be key motivators for young adults.
• NLP offers an efficient method for evaluating user experiences with digital health interventions.