Written by: Supriyo Rana
TW: Mentions of sexual assault, rape
Alcohol-related incidents on college campuses are known to lead to various issues, including sexual assault, academic problems, mental health issues, and even death. About 1,825 college students between the ages of 18 and 24 die from alcohol-related unintentional injuries and about 97,000 students report experiencing alcohol-related sexual assault or rape each year. About 1 in 4 college students report academic consequences from drinking, including missing class, falling behind, doing poorly on exams or papers, and decreasing grades. Other consequences include suicide, injuries, unsafe sex, driving under the influence, and legal trouble.
Alcohol leads to these issues at virtually every college in the country, and Vanderbilt and other elite schools are no exception. Sometimes, college kids have good intentions but the judgment impairments caused by alcohol can lead someone to do things they may later regret.
Fortunately, scientists are working on a new way to determine intoxication: by looking at gait, or the way you walk. A smartphone’s onboard accelerometer can detect changes in gait. Scientists from the University of Pittsburgh, led by Brian Sulfoletto, just published research showing that smartphone motion data can be used to detect if a subject is intoxicated, with an average accuracy of 93 percent. This technology could be extremely valuable in reducing the number of alcohol-related incidents and injuries that occur on Vanderbilt’s campus by alerting students to when they are too drunk to be driving or doing other activities that require great coordination.
In the study, 22 participants had an hour to finish a drink containing vodka, lime juice, and syrup. The researchers strapped smartphones to the lower backs of the subjects and had them walk 10 steps forward, turn around, and walk 10 steps back to determine their baseline gaits. The participants then repeated this sequence once every hour for 7 hours while the smartphone logged motion data.
Because each participant had a unique sober gait and unique intoxicated gait, Suffoletto, and his colleagues used individualized mathematical models to compare each person against himself or herself. He hopes to eventually determine how to achieve similar accuracies when the phone is held in a person’s hand or is in their pocket.
Even though this research is in its early stages, it builds on a growing body of evidence showing that a device’s motion data could be an accurate measure of intoxication. “The work is well done and extends some of our group’s work on the same problem,” says Worcester Polytechnic Institute computer scientist Emmanuel Agu.
Previous research shows that people don’t realize they are impaired by alcohol consumption up to 50% of the time. But, gait sensing can be used to alert users to when they are impaired regardless of their opinion, which is why it has the potential to be a super useful tool in society and especially on college campuses, where large volumes of alcohol are likely to be accessible.
The intoxicated brain may lie to a person, but smartphone gait data won’t.
The authors of the study said their findings could develop a warning to alert users that they are over the drunk-driving limit. This feature could be implemented here at Vanderbilt, although it is worth acknowledging that students who are over the drunk-driving limit may simply disregard the alert. However, there are many methods to make this work: alerts could be sent to a trusted, sober friend’s phone or pairing the technology with the user’s car could prevent them from unlocking their car if they are over the legal limit. These are all logistical challenged that need to be considered, but they are important to explore and discuss.
Additionally, by monitoring gait, health experts might be able to better predict when a person’s problematic drinking might occur. Too many people who start drinking alcohol casually may develop an addiction, but patterns generated from gait-sensing data may help health specialists determine how to best help their patients.
Since data related to smartphone use and sensors, including accelerometers, are widely collected, this information could be harvested by third parties to try to determine whether a smartphone user is drunk, says Suffoletto, meaning that anyone who wanted to process and analyze this data in real-time could make inferences about changes in gait. What is done with this data once it is collected is still not entirely clear, but if this technology does become accessible to college students, Vanderbilt should consider how they can potentially use this data to make sure that students are safe and healthy.
The availability and future of this technology are still not entirely clear. Likely, this technology will eventually become available to phone manufacturers like Apple and Android, but the types of data they will be collecting are hard to predict at this point. Of course, applications with access to your phone’s accelerometer may run into some consumer privacy issues which will need to be addressed before the technology will be ready for use by the general public.
Although the technology is still in its early stages, it is important for the Vanderbilt community to think about how we are going to implement this advancing technology to make Vanderbilt a safer learning and living environment. Promoting this technology to students and making them aware that it exists and can literally be the difference between life and death.