Objective: In the Netherlands, seclusion of patients with a psychiatric disorder is a last-resort measure to be used only in the event of (imminent) severe danger or harm. Although aggressive behavior is often involved, seclusions not preceded by aggression also seem to occur. We sought insight into the non-aggressive reasons underlying seclusion and investigated the factors associated with it. Method: We included all patients admitted to a Dutch psychiatric hospital in 2008 and 2009. Seclusions had been registered on Argus-forms, and aggression incidents had been registered on the Staff Observation Aggression Scale-Revised (SOAS-R), inspectorate forms and/or patient files. Determinants of seclusion with vs. without prior aggression were analyzed using logistic regression. Reasons for seclusion without prior aggression were evaluated qualitatively and grouped into main themes. Results: Of 1,106 admitted patients, 184 (17%) were secluded at some time during admission. Twenty-one (11.4%) were excluded because information on their seclusion was lacking. In 23 cases (14%), neither SOAS-R, inspectorate forms nor individual patient files indicated any aggression. Univariable and multivariable regression both showed seclusion without preceding aggression to be negatively associated with daytime and the first day of hospitalization. In other words, seclusion related to aggression occurred more on the first day, and during daytime, while seclusion for non-aggressive reasons occurred relatively more after the first day, and during nighttime. Our qualitative findings showed two main themes of non-aggressive reasons for seclusion: "disruptive behavior" and "beneficial to patient." Conclusion: Awareness of the different reasons for seclusion may improve interventions on reducing its use. Thorough examination of different sources showed that few seclusions had not been preceded by aggression. The use of seclusion would be considerably reduced through interventions that prevent aggression or handle aggression incidents in other ways than seclusion. However, attention should also be paid to the remaining reasons for seclusion, such as handling disruptive behavior and focusing on the beneficial effects of reduced stimuli. Future research on interventions to reduce the use of seclusion should not only aim to reduce seclusion but should also establish whether seclusions preceded by aggression decrease different from seclusions that are not preceded by aggression.