BackgroundNurses have a very stressful and demanding job, which probably affects their health and well-being. This has resulted in a search for an optimal nurse schedule. Self-scheduling is one method to enhance flexibility and give more control to the employee. Literature on self-scheduling is scarce and mainly focused on the implementation process. So far, little is known about the long-term effects of self-scheduling. ObjectiveThe comparison of effects of self-scheduling versus fixed scheduling on three dimensions of organizational justice (distributive, procedural, interactional justice) and three work attitudes (job satisfaction, affective commitment, global empowerment). This total effect is decomposed into direct and indirect effects via hypothesised mediators (leader-member exchange and organizational justice). DesignA cross-sectional multisite-matched study. SettingA questionnaire was distributed to nurses and nurse assistants of twelve nursing homes in Flanders (Belgium). Six matched pairs of nursing homes were formed, each containing a nursing home that uses self-scheduling (for at least one year) and one that uses fixed scheduling. ParticipantsThe study sample included 308 nurses and nurse assistants, of which 160 surveys were from nursing homes using fixed scheduling, versus 148 surveys from nursing homes using self-scheduling. MethodsUsing a double robust fixed effects regression model adjusted for the propensity score, the total effect of self-scheduling compared to fixed scheduling was analysed on the proposed outcomes. Subsequently, a (multiple) mediation analysis was performed. ResultsIf all people would have used self-scheduling instead of fixed scheduling, on average a significant decrease in the mean procedural justice score (estimated mean difference = -0.20, 95% CI -0.36 to -0.05) would be found. There is a significant negative direct effect (estimated mean difference = -0.12, 95% CI -0.21 to -0.03) and indirect effect via the mediator leader-member-exchange (estimated mean difference = -0.08, 95% CI -0.15 to -0.01) on procedural justice. In addition, significant negative indirect effects of self-scheduling were found via the mediator leader-member exchange on all the investigated outcomes. ConclusionDrawing up the work schedule is perceived as less fair in the case of self-scheduling. Furthermore, self-scheduling has a negative indirect effect on several work attitudes as a result of the changed relationship between the head nurse and the employees. The increase in task variety and decision-making autonomy for the employees and the altered leadership role of the head nurse can have a major impact on the dynamics of the entire team.