Wearable Alcohol Sensors Facilitate Monitoring of Drinking Intensity and Risk in Young Adult Heavy Drinkers

Newswise — Sensors that measure alcohol concentration through the skin can provide valid measures of drinking intensity and predict alcohol consequences among young adult drinkers, according to a study in Alcoholism: Clinical and Experimental Research. Youth alcohol misuse is linked to many short- and long-term negative consequences, and finding ways to predict and prevent these consequences is an important goal for research and public health. In research studies, prediction of short-term consequences is typically based on the number of drinks that an individual reports consuming in a given drinking episode. Self-reports can be retrospective, in which the person reports on their consumption the previous day, or episodic, in which the individual provides real-time reports via smart devices. However, self-reported data have some limitations: number of drinks is an imperfect proxy for biologic alcohol concentrations (and therefore for alcohol-related risk), and self-report accuracy diminishes the more a person drinks. Directly measuring biological alcohol concentrations could be an effective complement to self-reports. Measuring blood and breath alcohol concentration is too invasive or burdensome for field use (i.e. use in non-laboratory settings), but wearable transdermal alcohol concentration (TAC) sensors offer a viable and unobtrusive option by passively and continuously measuring perspired ethanol.

During and after a drinking episode, the TAC rises to a peak before diminishing; the precise trajectory can be plotted as a graph and different features assessed, including the peak TAC (maximum intoxication level), rise rate (speed of intoxication), fall rate (rate of ethanol elimination), duration (number of hours of measurable TAC), and total area under the plotted datapoints (AUC, reflecting the cumulative biologic exposure to alcohol). These features could facilitate prediction of consequences, aiding prevention efforts. In the new study, researchers from The Pennsylvania State University derived and tested these five features from TAC sensor data as indicators of self-reported drinking and predictors of alcohol consequences in over 200 heavy drinking young adults (aged 21 to 29).

Participants wore an ankle TAC sensor for six consecutive days, and carried a mobile phone with a custom app for completion of two types of drinking surveys: daily retrospective scheduled surveys, and episodic surveys initiated by participant when drinking begins. In the morning retrospective survey, participants also reported on whether they had experienced any of 13 negative consequences (such as a hangover, being sick, or getting into a fight) as a result of the previous day’s drinking. Across all participants, 1274 days of data were obtained, including 554 self-reported drinking days. Statistical analysis showed that all five TAC features correlated strongly with the number of self-reported drinks from the retrospective morning reports and moderately with those from the episodic reports. TAC features – particularly higher peak TAC and larger AUCs – were also independently predictive of alcohol-related consequences, even after adjusting for self-reported drinks. However, sensor data failed to capture ~15% of self-reported drinking days, mainly those of lower-intensity drinking.

This was the first study to test an array of TAC features as predictors of alcohol consequences in a large sample. From the research perspective, the findings support the use of TAC sensors in future studies of young adult drinking in everyday settings, and highlight the specific TAC features that help predict consequences. The findings may also facilitate development of novel self-report measures that reflect these features. Importantly, TAC features could also have utility in individualized alcohol prevention and treatment protocols – for example, by emphasizing peak TAC levels and implementing real-time interventions to warn against high peak levels and their associated risks.

Transdermal sensor features correlate with ecological momentary assessment drinking reports and predict alcohol-related consequences in young adults’ natural settings. M. A. Russell, R. J. Turrisi, J. M. Smyth (pages xxx).

ACER-21-4794.R3