Stephanie Balters ^1* Marc Schlichting ² Thomas O. Walton ³ Mykel J. Kochenderfer ² Debra Kaysen ¹

• ¹ Department of Psychiatry and Behavioral Sciences, School of Medicine, Stanford University, Stanford, California, United States
• ² Department of Aeronautics and Astronautics, School of Engineering, Stanford University, Stanford, California, United States
• ³ Department of Psychiatry and Behavioral Sciences, School of Medicine, University of Washington, Seattle, Washington, United States

Sexual victimization (SV) is common among college women, with approximately half of those who have experienced SV meeting criteria for posttraumatic stress disorder (PTSD) within a year. Both SV and PTSD are associated with alcohol misuse among college women, often explained by the self-medication hypothesis. Existing literature focuses on overall PTSD severity rather than potential day-to-day fluctuations in specific symptoms, which might play a crucial role in understanding alcohol misuse risk. Studies also examine only same-day or next-day associations between PTSD and drinking, neglecting the potential for longer-term changes. This study explores the short-term longitudinal stability and time-lagged predictive dynamics of PTSD symptoms, affect, and drinking behavior among 174 female college heavy episodic drinkers over four weeks. Participants were categorized into three groups: those with a history of SV and PTSD (n=77), women with SV but without PTSD (n=59), and women without prior trauma history (n=38) to be able to examine differences by trauma exposure, and PTSD. We compared the longitudinal stability of PTSD symptom networks, affect (arousal, positive affect, and negative affect), and drinking behavior across groups. Support vector regression determined which PTSD symptom networks and affect best predict drinking behavior at specific time lags within a 0-7 day range. The PTSD group showed higher longitudinal stability for PTSD symptom networks (adjusted ps<.049) and arousal (adjusted ps<.048), but lower stability for negative affect (adjusted p=.013) and drinking behavior, including alcohol cravings (adjusted p=.019) and consumption (adjusted ps=.012), compared to the comparison groups. This suggests individuals with PTSD have more stable symptoms and arousal levels but greater fluctuations in negative affect and alcohol-related behaviors. Secondary analysis revealed PTSD symptom networks optimally predicted alcohol cravings with a three-day time lag (r=.88, p<.001) and consumption with a four-day time lag (r=.82, p<.001). These findings challenge assumptions regarding immediate effects of PTSD and affect on drinking behavior and underscore the need for therapeutic approaches that consider longer-range effects. Future research should expand on these findings by incorporating longerrange assessments and exploring a broader range of symptom interactions.