Objective <p>This study examines the longitudinal influence of quarantine and the COVID-19 surge after quarantine was released on behavioral and mental problems among the Chinese university students.</p> Methods <p>A longitudinal observation design was utilized. There were 10 waves of surveys including the pre-quarantine period, the quarantine period, and the quarantine releasing period. The non-parametric linear mixed-effects model and generalized estimating equations were used to examine the association between the dependent and independent variables.</p> Results <p>Two-hundred and two (88.21%) participants completed 10 waves of the survey. The COVID-19 surge was positively associated with perceived severity for COVID-19 infection (β: 0.2162, <i>p</i> &lt; 0.01), the quarantine period was negatively associated with perceived risk (β: -0.3632, <i>p</i> &lt; 0.01). The quarantine was negatively associated with both behavior belief (β: -0.6164, <i>p</i> &lt; 0.01) and outcome belief for lockdown (β: -0.0976, <i>p</i> &lt; 0.01). The COVID-19 surge was only positively associated with behavior belief for the lockdown (β: 0.1073, <i>p</i> &lt; 0.01). Both the COVID-19 surge and the quarantine periods were positively associated with mental disorders, and the standard β values were 0.2611 and 0.3846. The quarantine also had an influence on short sleep duration (β: 0.2681, <i>p</i> &lt; 0.05).</p> Conclusion <p>This study yielded new information about the influence of the COVID-19 surge, the quarantine period and the period after quarantine was released on the behavioral and mental problems among Chinese university students. Policy changes and health education are essential for minimizing the adverse health effects of these responses. This may have important implications for policies and disease prevention strategies targeted at controlling COVID-19.</p>