A study by Rutgers Institute for Health, Health Care Policy and Ageing has shown that the sensors within smartphones can be useful when trying to determine whether or not somebody is under the influence of cannabis. The study produced preliminary evidence that the same sensors used for GPS can predict if someone is high, with a very high accuracy rate.
The purpose of the study was to evaluate the feasibility of using smartphone sensor data to identify episodes of cannabis intoxication in the natural environment. Researchers were able to predict 60 out of 451 reported episodes of cannabis use by using only the time feature. They were able to see patterns in the day of the week, and the times of the day when cannabis use was most likely to happen. This accuracy rate increased to 90% when time feature data and data from the phone’s accelerometer were analysed together. Accelerometers in mobile phones are used to detect the orientation of the phone by measuring the linear acceleration of movement.
The study was published in the journal Drug and Alcohol Dependence and involved 57 young adults aged between 18 and 25 in Pennsylvania, USA. The participants reported to the researchers when they used cannabis, which was at least twice a week, the team then used the data from the phones of the participants to work out patterns in their behaviour and link to cannabis use.
“Using the sensors in a person’s phone, we might be able to detect when a person might be experiencing cannabis intoxication and deliver a brief intervention when and where it might have the most impact to reduce cannabis-related harm,” said corresponding author, Tammy Chung, professor of psychiatry and director of the Center for Population Behavioral Health at the Rutgers Institute for Health, Health Care Policy and Aging Research.
Cannabis use may produce slow reaction times in some people, which could have dangerous consequences for people in the workplace. Existing methods for detecting cannabis intoxication, such as blood or saliva tests have their limitations because they cannot measure a person’s impairment accurately, thus not being able to make an accurate assumption that somebody can safely perform the task at hand.
When the data collected by the team was analysed they found that the feasibility of using phone sensors to detect subjective intoxication from cannabis is strong. However, further research is needed to evaluate the performance of the algorithm in differentiating between intoxicated reports versus non-intoxicated reports.
