This study adapted and validated the European Portuguese version of the Fear of Missing Out scale (FoMOs-P) and determined an “optimal” cutoff score for higher education students with high social media engagement. Five hundred Portuguese students (M = 22.37, SD = 5.68) completed FoMOs-P and measures of sleep quality, rumination, and social media. Following previous conflicting results, Confirmatory Factor Analysis (CFA) was conducted to compare two competing factor structures. Model invariance regarding sex was tested using multi-group CFA. FoMOs-P’ internal consistency, convergent and discriminant validity were analyzed, and an “optimal” cutoff point was explored through Receiver Operating Characteristic (ROC) analysis. When comparing CFA models, the two-related-factor model fit best (Δχ2 (1) = 495.01, p < .001), demonstrating a good fit (χ2/df = 2.96; CFI = .98; TLI = .97; RMSEA [90 CI] = .063 [.049;.077]). This model showed metric invariance but not scalar invariance across sex. There was adequate reliability for both factors and total scale (ω > .7). Convergent and discriminant validity were supported. ROC analysis revealed an AUC = .742 (p < .001) and an “optimal” cutoff point of 20 (70.1% sensitivity, 68.4% specificity) to detect probable disruptive social media engagement. FoMOs-P was shown to be a reliable and valid instrument, supporting its use among higher education students. FoMOs-P appears to be composed of two factors [(i.e., internalizing and externalizing Fear of Missing Out (FoMO)] and has moderate accuracy to screen students with high social media usage.