Using Smartphone Survey and GPS Data to Inform Smoking Cessation Intervention Delivery: Case Study

Area of Science:

• Digital Health
• Public Health
• Geospatial Analysis

Background:

• Young adults face challenges quitting smoking due to limited access to tailored interventions.
• Smartphone-based interventions offer a promising approach for delivering timely smoking cessation support.
• Geofencing, using spatial buffers around high-risk locations, is an emerging strategy for personalized interventions.

Purpose of the Study:

• To explore a method for generating person-specific geofences around high-risk smoking areas.
• To evaluate different geofence construction methods using location data and surveys.
• To inform the design of automated, just-in-time adaptive interventions for smoking cessation.

Main Methods:

• Utilized ecological momentary assessment data from young adult smokers (2016-2017).
• Combined self-reported smoking events with passively tracked GPS location data.
• Constructed person-specific geofences around smoking locations using kernel density estimates and compared different spatial zone types (census blocks, fishnet grids).

Main Results:

• Geofences captured over 50% of smoking events for most cases within 3-hour intervals.
• 1000 ft² fishnet grid geofences demonstrated higher capture rates of smoking events compared to census blocks.
• Geofences successfully identified high-risk smoking situations based on time and place.

Conclusions:

• The geofence construction method effectively identifies high-risk smoking situations by time and place.
• Individually tailored geofences show potential for enhancing smoking cessation interventions.
• Future studies will utilize fishnet grid geofences to guide the delivery of intervention messages via smartphone apps.