What is it about?
Heavy drinking does not look the same for everyone, even from one day to the next. In this study, we used wearable alcohol sensors to track real-time alcohol levels in young adults over several days. We found that some people show consistently high levels of intoxication, while others shift between heavier and lighter drinking depending on what happened the day before. Importantly, certain aspects of drinking—such as how long intoxication lasts—vary more between people than others and may be especially important for identifying risk. These findings suggest that prevention and intervention efforts should focus not only on how much people drink, but also on patterns that make heavy drinking persist over time.
Featured Image
Photo by engin akyurt on Unsplash
Why is it important?
Most research on heavy drinking relies on self-reports or averages taken over long periods, which can miss how drinking actually unfolds from one day to the next. This study is unique because it combines continuous data from wearable alcohol sensors with a new way of identifying both stable drinking tendencies and short-term changes in intoxication. By showing that some young adults are consistently intoxicated while others get “stuck” in heavy drinking through day-to-day carryover effects, this work reveals patterns that would be invisible in traditional analyses. These insights are especially timely as wearable health technologies become more common and prevention efforts increasingly aim to deliver personalized, real-time interventions rather than one-size-fits-all approaches.
Perspectives
For me, this paper represents an important step toward making dynamic models feel more concrete and relevant to real people’s lives. Much of my work focuses on statistical methods that can seem abstract, but this project reminded me why those methods matter—because they help us see patterns of risk that unfold in everyday time. Working with continuous data from wearable sensors pushed me to think carefully about how theory, measurement, and modeling need to align if we want results that are both rigorous and useful. I hope this article encourages others to look beyond averages and snapshots and to think more deeply about how behavior persists, changes, and sometimes gets “stuck” in ways that have real consequences for health.
Ivan Jacob Agaloos Pesigan
The Pennsylvania State University
Read the Original
This page is a summary of: Common and unique latent transition analysis (CULTA) as a way to examine the trait–state dynamics of alcohol intoxication., Psychology of Addictive Behaviors, December 2025, American Psychological Association (APA),
DOI: 10.1037/adb0001106.
You can read the full text:
Contributors
The following have contributed to this page
