Using Machine Learning to Develop Just-in-Time Adaptive Interventions for Smoking Cessation

The University of Texas System (UT)

Status: Completed

Conditions: Smoking Cessation

Treatments: Behavioral: Adaptive Treatment; Behavioral: interviewing-based counseling; Drug: Nicotine Patch; Behavioral: Android Wear smartwatch

Study type: Interventional

Funder types: Other; NIH

Identifiers

NCT04839198

R00DA046564 (U.S. NIH Grant/Contract)

HSC-SPH-20-1159

Details and patient eligibility

About

The purpose of this study is to evaluate the feasibility and preliminary effectiveness of delivering a personalized, just-in-time adaptive intervention driven by machine learning prediction of smoking lapse risk in real time.

Enrollment: 60 patients

Sex: All

Ages: 18+ years old

Volunteers: No Healthy Volunteers

Inclusion criteria

• a score greater than or equal to 4 on the Rapid Estimate of Adult Literacy in Medicine Short Form (REALM-SF),12
• willingness to quit smoking 14 days after the baseline visit
• no contraindications to using Nicotine replacement therapy (NRT).
• If participants would like to use their own phone to complete the EMAs, they must additionally have an Android smartphone (Android 5.2 or higher), and be willing to install the InsightTM mHealth app on their phone.

Exclusion criteria

• currently smoking less than 5 cigarettes per day

Trial design

Primary purpose: Treatment

Allocation: Randomized

Interventional model: Parallel Assignment

Masking: None (Open label)

---

60 participants in 2 patient groups

Adaptive Treatment plus usual care

Experimental group

Treatment:

Behavioral: Android Wear smartwatch

Drug: Nicotine Patch

Behavioral: interviewing-based counseling

Behavioral: Adaptive Treatment

Usual care

Active Comparator group

Treatment:

Behavioral: Android Wear smartwatch

Drug: Nicotine Patch

Behavioral: interviewing-based counseling