Author: Sasha Tanner
After consuming any amount of alcohol, it is difficult for individuals to gauge their specific level of intoxication. There may be behavioral signs of alcohol impairment present, but their actual blood or breath-alcohol content (BAC/BrAC) remains unknown. Moreover, self-awareness of impairment can decrease when the individual consumed alcohol. These behaviors reveal how different people respond to alcohol consumption, i.e., tolerance.
Alcohol tolerance varies from person to person. The legal BAC or BrAC limit to drive while impaired (0.08 g/dl) yields different cognitive and behavioral effects on a range of people, depending on how much they consume alcohol and the level of tolerance they have built up. Kaczor and colleagues (2026) acknowledge this discrepancy between intoxication levels and behavioral variation across people in their study called Detecting Ethanol Intoxication and Impairment Using Wearable Biosensors.
Summarizing the Research
As a response, they featured an ethanol detection biosensor that could be worn on the wrist. The specific brand, EmbracePlus, was a Food and Drug Administration (FDA)-approved, research grade watch designed by Empatica, a company that manufactures medical technology.
This watch recorded physiological and behavioral information of the person wearing it, such as accelerometry data, transdermal BAC, and heart rate with the following objectives in mind:
- Test the feasibility of these biosensors to accurately detect biometric data about alcohol intoxication levels.
- Over time, this data could possibly create algorithms that the device can store to detect intoxication unique to the individual, thus increasing accuracy.
The purpose of the study was to determine how accurately the biosensor can detect intoxication through a series of tests. First, the researchers gathered 28 participants aged 21 to 64 years old to test the feasibility of this device, where 43% were female and 57% were male. After completing a demographic survey, participants were given the biosensor to wear on their non-dominant wrists to collect data over a course of 6 hours.
Within this timeframe, participants were asked to perform the Standardized Field Sobriety Test (SFST) for police academy instructors to determine impairment. They served different types of alcoholic beverages, such as wine, beer, and liquor, during the experiment. Then, the research team collected BrAC data, recording participants as either impaired (0.08 BAC or greater) or not impaired (below 0.08 BrAC).
Next, the data was processed and organized for analysis. The researchers found different percentages in impaired versus unimpaired participants between the SFST and BrAC results:
- BrAC Test for Impairment: 19 (68%) out of the 28 participants failed the SFST, proving impairment.
- Standardized Field Sobriety Test: 24 (86%) out of the 28 participants had a BrAC of 0.08 or above, qualifying them as impaired.
This reveals that while 24 participants were physiologically impaired based on BrAC, 5 of these participants did not test as impaired for the SFST, signaling differences in tolerance levels.
Among these participants, biometric data from the watch was analyzed for accuracy, sensitivity, and specificity levels between the BrAC measurements and SFST results.
- Compared to the BrAC measurements, the watch yielded 80% accuracy, 78% sensitivity, and 81% specificity to impairment based on 66 physiological features, such as heart rate, sex, and body temperature.
- As for the SFST, the watch yielded 74% accuracy, 85% specificity, and 52% sensitivity to impairment.
Considerations & Policy Recommendations
Although improvements should be made to increase the accuracy of this device, there are a few considerations worth addressing if it is released publicly:
- Effects on Behaviors & Decisions: The behavioral and decision-based effects of wearing the watch are critical to assess if this product is a deterrent of impaired driving. For example, people can monitor their BrAC and behavioral impairment levels to determine whether it is safe to drive, thus suggesting that this device can deter impaired driving behaviors.
- Cannabis & Other Drugs: Since this device only measures alcohol impairment, more research and technological development on cannabis and other drugs are helpful to reduce impaired driving.
- Ethics of Technological Innovation: This device has the potential to be incorporated into personal, smart watch models. However, monitoring people’s physiological content can impose privacy concerns if there is no guarantee to protect such personal information, keeping people from wanting to buy the product to begin with. This concern tapers into how the product can be sold and assessed for consumer satisfaction.
- Marketing & Consumer Insights: A survey that collects data on potential interest in buying the device could reveal whether people would buy the product—and why. This information is especially useful to determine how this device could be used, answering the latter considerations about effects on behaviors and decisions. It can also reveal why people wouldn’t buy it, possibly due to privacy and ethical reasons.
As discussed in the previous blog, technology can be our greatest tool if we are cautious about how it is used. It is critical to touch on the potential outcomes of commercializing this watch while addressing how it could benefit overall traffic safety. For more information on the study or the device, take a look at the following sources!
References
- Empatica. (n.d.). About Empatica: We create groundbreaking technology that is friendly, caring and more human. https://www.empatica.com/company/
- Empatica. (n.d.). EmbracePlus: The world’s most advanced smartwatch for continuous health monitoring. https://www.empatica.com/embraceplus/
- Kaczor, E. E., M. Golleru, D. C. Carter, O. B. Painter, K. J. Kelleran, J. J. Lynch, B. M. Clemency, L. A. Cavuoto, & P. R. Chai. (2026). Detecting Ethanol Intoxication and Impairment Using Wearable Biosensors. Proceedings of the 59th Annual Hawaii International Conference on System Sciences. Annual Hawaii International Conference on System Sciences, Department of Health and Human Services, 3532–3542.
