Author ORCID Identifier
Year of Publication
Doctor of Philosophy (PhD)
Arts and Sciences
Dr. Shannon E. Sauer-Zavala
AUD and high-risk drinking are significant public health concerns, and prevailing diagnostic models fail to account for the within-person variability in AUD presentation and the role of functional mechanisms that maintain drinking outcomes. The ETOH Framework highlights the importance of three primary superdomains that maintain and characterize AUD – reward salience, cognitive dyscontrol, and negative emotionality. The current study was the first attempt to investigate the role of the primary ETOH mechanisms, along with positive and negative affect, on alcohol use outcomes, using a novel approach to model individual-level networks of relations over time. This study gathered twice-daily diary data from a sample of alcohol-using adults and measured alcohol use outcomes (i.e., drinking, drinking urges, and intensity of urges), ETOH mechanisms, positive affect, and negative affect. The aims of the study were to (1) characterize the current sample and analyze baseline nomothetic relations among personality, ETOH mechanisms, and drinking outcomes, (2) assess the feasibility of longitudinal personalized network models of mechanisms, affect, and daily alcohol outcomes, and (3) investigate the presence of subgroups of individuals based around core ETOH mechanisms. It was hypothesized that (1) baseline nomothetic relations would replicate established conventions and characterize the sample, (2) individual network models would demonstrate that ETOH mechanisms and affect predict alcohol use outcomes contemporaneously and over time, and (3) subgroups would be identified that converge around the three primary ETOH mechanisms. Participants were 40 adults aged 21-62 years (Mage = 37.53, 70% male, 90% white) who endorsed at least moderate alcohol use and who provided twice-daily diary data over thirty days measuring alcohol use outcomes, ETOH mechanisms, and positive and negative affect. Results showed mixed support for these hypotheses: while there were no group-level associations found in the network analyses, several individual-level networks demonstrated unique contemporaneous and lagged relations between mechanisms, affect, and alcohol use outcomes. Furthermore, while subgroups were identified, they did not exhibit significant associations between alcohol use outcomes and functional mechanisms. Overall, this study highlighted the feasibility of within-person networks that demonstrate the variability of mechanisms and affect that predict alcohol use outcomes over time in a sample of alcohol use participants. Future research should consider alternative measures of mechanisms and alcohol use outcomes, utilize additional longitudinal research designs, and expand sample selection.
Digital Object Identifier (DOI)
This study was supported by a Lipman Foundation research grant in 2022.
Semcho, Stephen, "Personalized Longitudinal Network Models of Alcohol Use: A Mechanistic Approach" (2023). Theses and Dissertations--Psychology. 241.