For whom does it work best? Moderators and predictors of outcome of mobile-based approach bias modification for problematic internet use in young adults

Objective: Problematic usage of the internet (PUI) is a growing concern in young adults. Mobile-based approach bias modification (ApBM) has proven efficacious; however, individual differences in intervention response remain poorly understood. This exploratory analysis examined baseline participant characteristics as potential moderators and predictors of outcome following mobile-based ApBM (PROTECTapp) to reduce PUI in university students. Methods: Based on data of 92 university students (Mage = 22.00 years, 69.6% female) with elevated PUI who were randomized to PROTECTapp (n = 45) or a waitlist control group (n = 47), we examined age, baseline PUI severity, motivation to change, comorbid internalizing symptoms, procrastination, and positive intervention expectancies as potential moderators and predictors of PUI (Compulsive Internet Use Scale, CIUS) and adherence at post-intervention. Results: Baseline PUI severity significantly moderated intervention efficacy (b = -0.42, SE = 0.20, p = .040), indicating stronger effects among individuals with higher initial symptom severity. No significant moderating effects emerged for other potential moderators. In a combined cross-over sample, higher baseline PUI severity (aOR = 1.37, p < .001) and younger age (aOR = 0.67, p = .011) predicted greater likelihood of reliable improvement, while adherence was higher in younger participants (b = -2.43, SE = 1.01, p = .019) and those with higher internalizing symptoms (b = 2.68, SE = 1.24, p = .033). Conclusion: Mobile-based ApBM appears particularly effective for young adults with more severe PUI, supporting targeted rather than universal prevention approaches. Other baseline factors did not influence efficacy, underscoring broad applicability across diverse student populations.