This cross-sectional study examined the associations between waking movement behaviour composition and depressive symptoms, self-rated health (SRH) and eating-disorder (ED)-related symptomatology (SCOFF) in university students using compositional data analysis (CoDA) and isotemporal substitution modelling. Data were collected through an anonymous web-based survey among students enrolled at the University of Milan, Italy. The analytical sample included 2636 participants. International Physical Activity Questionnaire (IPAQ) was used to measure daily time spent in moderate-to-vigorous physical activity (MVPA), light physical activity (LPA), and sedentary behaviour (SB) and analysed as a three-part composition. Outcomes included depressive symptoms (Patient Health Questionnaire-9 [PHQ-9]), SRH and SCOFF score. In adjusted compositional regression models, a higher relative proportion of MVPA was associated with lower depressive symptoms ( β  = -0.39, 95% confidence interval [CI]: -0.53 to -0.24) and higher SRH ( β  = 0.11, 95% CI: 0.08-0.13), whereas a higher relative proportion of SB was associated with higher depressive symptoms ( β  = 0.42, 95% CI: 0.21-0.64) and lower SRH ( β  = -0.11, 95% CI: -0.14 to -0.07). LPA was not significantly associated with outcomes. Isotemporal substitution analyses showed that reallocating 15 min from SB to MVPA was associated with lower PHQ-9 scores (-0.114, 95% CI: -0.156 to -0.072), higher SRH (0.031, 95% CI: 0.024-0.038) and SCOFF score ( β  = 0.04, 95% CI: 0.00-0.07). These findings suggest that, within the daily movement composition, increasing a greater relative allocation of time to MVPA at the expense of sedentary time is associated with modestly more favourable mental and perceived health in university students.