Smartphone Sensor Data May Be Able to Detect Marijuana Intoxication: Study

A smartphone sensor, much on the lines of the GPS system, may help determine if someone is intoxicated after consuming marijuana, a new study has demonstrated. The intoxication resulting from the consumption of cannabis is associated with slowed response time, effect on work or school performance, and even influencing the driving behaviour often resulting in accidents and fatalities. 

The research, by Rutgers Institute for Health, Health Care Policy and Aging Research, was published in the Drug and Alcohol Dependence journal, and it assesses the feasibility of using smartphone sensor data to identify cannabis intoxication. Remarkably, the study, which made use of a combination of time features and sensor data, was found to be 90 percent accurate. This is significant, for the existing methods including blood, urine, and saliva tests to detect cannabis intoxication have limitations, experts say.  

Here’s how they did it

Researchers evaluated daily data collected from young adults using cannabis at least twice every week. As part of the study, they analysed the phone surveys, self-initiated reports of cannabis use, and continuous phone sensor data. 

According to the analysis, while the time of day and the day of the week had 60 percent accuracy in detecting self-reporting of cannabis intoxication, the combination of time features and smartphone sensor data could detect marijuana intoxication with 90 percent accuracy. 

Tammy Chung, professor of psychiatry, said that the phone’s sensors may help us detect when a person is experiencing cannabis intoxication and deliver a brief intervention when and where it might have the most impact to reduce harm.

According to a statement, researchers used low burden methods — tracking time of day and day of the week and analysing phone sensor data — to detect intoxication in daily life. The study demonstrated that the feasibility of using phone sensors to detect subjective intoxication from cannabis consumption was strong.