In March of 2019, the DADSS Program received OMB approval to run a series of controlled, in-vehicle tests involving sober drivers and drinking passengers to determine how the sensors respond to real-world driving conditions. These tests began in Massachusetts in late June, in partnership with McLean Hospital, a Harvard Medical School affiliate as well as at KEA Technologies, Inc. The data collected from this research will inform the development and calibration of the next generation of breath sensors, as well as future vehicle integration.
Once the recruited research participants pass the initial screening, the drivers receive their keys and one of five pre-set driving routes. The passengers are dosed by trained staff, based on their height and weight, to reach a target BAC. Over the course of the route, passengers provide regular breath samples into an approved reference device, the passenger door sensor, and the front passenger sensor (by the glove box). The sensors — installed on both the driver and the passenger sides of the vehicle — are sniffing for and recording alcohol levels the entire time, helping the team determine if the sensors can accurately distinguish between driver and passenger breath. At the end of the drive, the passenger is kept onsite until their BAC measures below 0.05% and then is sent home with a rideshare service.
DADSS and its research partners collect and track all of the breath samples from each ride to observe the accuracy and precision of the sensor readings, as well as the ease with which the driver’s breath sample was captured. The team also tracks the different variables — from the person (such as age, sex, ethnicity, and distance from the sensor); to the vehicle (such as windows up or down; air conditioning on or off; and cabin temperature and humidity); to the environment (such as outside temperature and weather conditions, location, and elevation). This data is supplemented by video footage of each trip, along with post-drive interviews with both passengers and drivers.