Detecting Drinking Episodes in Young Adults Using Smartphone-based Sensors

Abstract

Alcohol use in young adults is common, with high rates of morbidity and mortality largely due to periodic, heavy drinking episodes (HDEs). Behavioral interventions delivered through electronic communication modalities (e.g., text messaging) can reduce the frequency of HDEs in young adults, but effects are small. One way to amplify these effects is to deliver support materials proximal to drinking occasions, but this requires knowledge of when they will occur. Mobile phones have built-in sensors that can potentially be useful in monitoring behavioral patterns associated with the initiation of drinking occasions. The objective of our work is to explore the detection of daily-life behavioral markers using mobile phone sensors and their utility in identifying drinking occasions. We utilized data from 30 young adults aged 21-28 with past hazardous drinking and collected mobile phone sensor data and daily Experience Sampling Method (ESM) of drinking for 28 consecutive days. We built a machine learning-based model that is 96.6% accurate at identifying non-drinking, drinking and heavy drinking episodes. We highlight the most important features for detecting drinking episodes and identify the amount of historical data needed for accurate detection. Our results suggest that mobile phone sensors can be used for automated, continuous monitoring of at-risk populations to detect drinking episodes and support the delivery of timely interventions. CCS Concepts: • Human-centered computing ➝ Human computer interaction (HCI) ➝ Empirical studies in HCI

Assessment of Alcohol Use in the Natural Environment

“…An emerging line of inquiry has attempted to use machine learning algorithms to passively detect alcohol use based upon sensor data collected from mobile technologies. Investigators have used measures of gait and posture derived from inertial sensors in smartphones and watches (Aiello and Agu, ; Arnold et al., ; Gharani et al., ; McAfee et al., ; Suffoletto et al., ) and other smartphone data such as communications, typing, and detected physical activity to make predictions concerning the alcohol consumption of the user (Bae et al., , ). In these studies, the stream of sensor data is segmented and each segment is labeled according to indices of alcohol use collected in the laboratory (e.g., real or simulated BAC) or in the field (e.g., number of drinks or estimated BAC based on EMA reports).…”

Toward the prevention of alcohol use disorders: Overdrinking (unintentional binge drinking) in a community sample

“… Bae et al (2017) developed a machine learning model that was able to identify three different types of behavior—non-drinking, drinking, and heavy drinking—96.6 percent of the time outside of a laboratory in a sample of 21- to 28-year-olds using passively collected smartphone data. Their collection of individual historical data such as the average time between keystrokes and travel activity helped improved their predictions.…”

#Drink Or #Drunk

“…limited recall [24,43]. The advent of ubiquitous sensors and smartphones aided researchers to collect larger amounts of data including insitu responses via SMS on features phone [44], wearable sensor data [9,11,38], and hybrid data (including sensor data, and human-generated data like photos, captions, and location) [16,56,57,62]. However, these methods of data collection have the disadvantage of being intrusive because participants are asked to intentionally report their alcohol intake.…”