Improving Remote Breathalyzer Procedures Used by Clinicians and Researchers to Remotely Monitor Alcohol Use

Devices that can continuously and/or remotely monitor alcohol use have created new opportunities for both research and intervention, initially with transdermal alcohol monitors (TAMs) that continuously and remotely detect alcohol excreted through the skin. TAMs were first designed and used in the criminal justice system, but their disadvantages have limited widespread adoption (e.g., availability, cost, size, interference with physical activities, and stigma) (Allessi et al., 2017; Caluzzi et al., 2019). Thus, other alcohol monitoring technologies (and procedures) to detect and characterize alcohol consumption are being developed and used. One example is remote breathalyzers (e.g., SCRAM Remote Breath®), a potential alternative to TAMs. They are small, can be used daily and more privately, use Automated Facial Intelligence™ (AFI™) technology to verify identity, and have safeguards to prevent circumvention (SCRAM, 2020). Heavy drinking may go undetected when using prescribed remote breathalyzer monitoring protocols. Remote breathalyzers are being used more frequently in legal and clinical research settings (e.g., Alessi & Petry, 2013; Koffarnus et al., 2018; Moody et al., 2018; Nehlin et al., 2018; Recovery Healthcare Corporation, personal communication, June 22, 2021; Skipper et al., 2014). The underlying assumption is that breathalyzers objectively detect alcohol consumption. In fact, among social drinkers (n = 12), remote breath alcohol concentration (BrAC) monitoring protocols had a 98.8% rate of correspondence with self-reported drinking (Skipper et al., 2014), where there were no drinking-associated contingencies/consequences. Manufacturer monitoring protocols have not been validated among individuals who do have contingencies/ consequences associated with detected drinking. Remote breath sampling protocols may provide ample opportunity for individuals to drink heavily, especially among populations where detected drinking is associated with penalties. Importantly, when contingencies are present, individuals may adopt or adjust their drinking patterns to "beat" the system. For instance, procedures used in the judicial system (which correspond with manufacturer recommendations), individuals are prompted to submit 4 breath samples/day, beginning at a start time chosen by the person being monitored and then every 5 hours thereafter (e.g., 7am, 12pm, 5pm, 10pm). Each person's schedule for sampling is the same every day and breath samples must be submitted within an hour after prompting. Therefore, these schedules have "on- and off-cycle" times, with the on-cycle schedule covering only ~15 hours of each day. The off-cycle schedule, from the submission of the last sample on one day to the first sample the next day, allows time for sleep, but during this time (up to 10 hours, including an allowance for the sampling window) drinking could occur. People could drink right after their last breath sample (e.g., 10pm) and still submit a negative BrAC (<.02 g/dL) the next day (required by 8am). BrAC ≥.02 g/dL are considered positive for alcohol use (NHTSA, 2018). Based on our prior alcohol self-administration study, at varying levels of intoxication, both men and women could be below .02 g/dL within ~6 hours (Figure 1; Hill-Kapturczak et al., 2015), showing considerable latitude for engaging in drinking and not violating monitoring criteria. Thus, it is unknown if remote BrAC monitors can yield reliable and valid measures of drinking events in those who: (a) may find it difficult to control drinking, and (b) are motivated to avoid detection of drinking due to consequences.