OBJECTIVE: The effect of an intervention on neuropsychiatric symptoms (NPS), particularly agitation and aggression, and psychotropic drug use (PDU) in institutionalized people with young-onset dementia (YOD) was evaluated. METHODS: A randomized controlled trial was conducted using a stepped wedge design. Thirteen YOD special care units were randomly assigned to three groups, which received the intervention at different time points. Four assessments took place every 6 months during a period of 18 months. Two hundred seventy-four people with YOD who resided in YOD special care units participated, of whom 131 took part in all assessments. The intervention consisted of an educational program combined with a care program, which structured the multidisciplinary process of managing NPS. The care program included the following five steps: evaluation of psychotropic drug prescription, detection, analysis, treatment, and evaluation of treatment of NPS. The Cohen-Mansfield Agitation Inventory and the Neuropsychiatric Inventory-Nursing Home version were used to assess NPS. Data on PDU were retrieved from residents' medical files. Multilevel models were used to evaluate the effect of the intervention, which accounted for clustering of measurements in clients within units. RESULTS: No significant differences were found in agitation, aggression, other NPS, or PDU after crossing over to the intervention condition. CONCLUSION: We found no evidence that the intervention for management of NPS in nursing home residents with YOD was more effective in reducing agitation, aggression, other NPS, or PDU compared with care as usual.