Pooled and person-specific machine learning models for predicting future alcohol consumption, craving, and wanting to drink: A demonstration of parallel utility.

Abstract

Background and Aims: The specific factors driving alcohol consumption, craving, and wanting to drink, are likely different for different people. The present study sought to apply statistical classification methods to idiographic time series data in order to identify person-specific predictors of future drinking-relevant behavior, affect, and cognitions in a college student sample. Design: Participants were sent 8 mobile phone surveys per day for 15 days. Each survey assessed the number of drinks consumed since the previous survey, as well as positive affect, negative affect, alcohol craving, drinking expectancies, perceived alcohol consumption norms, impulsivity, and social and situational context. Each individual’s data were split into training and testing sets, so that trained models could be validated using person-specific out-of-sample data. Elastic net regularization was used to select a subset of a set of 40 variables to be used to predict either alcohol consumption, craving, or wanting to drink, forward in time. Setting: A west-coast university. Participants: Thirty-three university students who had consumed alcohol in their lifetime. Measurements: Mobile phone surveys. Findings: Averaging across participants, accurate out-of-sample predictions of future drinking were made 76% of the time. For craving, the mean out-of-sample R 2 value was .27. For wanting to drink, the mean out-of-sample R 2 value was .27. Conclusion: Using a person-specific constellation of psychosocial and temporal variables, it may be possible to accurately predict drinking behavior, affect, and cognitions before they occur.

Emotion Dynamics as Predictors of Drinking Behavior: Idiographic and Pooled Approaches

“…Thus, the sample average provided better prediction accuracy for these individuals than a personalized model. This finding is consistent with the parent study for the current data [2]. In addition, the models of craving and wanting to drink for P001, P010, craving for P006, P008, P023, wanting to drink for P018, and the models for drank/did not drink for P002 and P005 lack emotion variables altogether, and instead retained only time variables as predictors.…”

“…This study demonstrated the possibility that certain and separate emotions could predict craving alcohol, and unlike drank/did not drink or wanting to drink outcomes, craving on average was predominated by negatively valenced emotions. As shown in other analyses of these data, the present analysis showed that certain days of the week, Thursday through Sunday, were common predictors of craving, and were even more common predictors in the pooled model for whether a person drank [2]. While the pooled predictors may provide some insight into which theories of alcohol use best apply to college students in general, our results indicate that the relationship between emotions and alcohol use vary per person.…”

“…All study procedures were approved by the University of California, Berkeley Committee for the Protection of Human Subjects. The present study was a reanalysis of Soyster et al [2].…”

“…Given that the present sample was relatively homogeneous (see Method), there is a possibility that common influences may provide greater predictive utility. Prior research with the current sample has demonstrated predictive utility in pooled prediction modeling [2].…”

“…Given that alcohol consumption is a behavior enacted within the course of individual lives, contexts, and demands, the specific factors driving alcohol consumption are likely idiosyncratic. No universally accepted theory exists to explain the observed variability in alcohol use both within-and between people, and despite research efforts and attention to alcohol use, how these factors apply to or influence each person remains unknown [1,2]. Precise methods are likely needed to investigate the cognitive, affective, behavioral, psychological, and social mechanisms driving an individual to drink.…”