Attempts to reduce high utilisation of mental health inpatient care by targeting the critical time of hospital discharge are rare. In this study, we test the effect of a needs-oriented discharge planning intervention on number and duration of psychiatric inpatient treatment episodes (primary), as well as on outpatient service use, needs, psychopathology, depression and quality of life (secondary). Four hundred and ninety-one adults with a defined high utilisation of mental health care gave informed consent to participate in a multicentre RCT carried out at five psychiatric hospitals in Germany (Düsseldorf, Greifswald, Regensburg, Ravensburg and Günzburg). Subjects allocated to the intervention group were offered a manualised needs-led discharge planning and monitoring intervention with two intertwined sessions administered at hospital discharge and 3 months thereafter. Outcomes were assessed at four measurement points during a period of 18 months following discharge. Intention-to-treat analyses showed no effect of the intervention on primary or secondary outcomes. Process evaluation pending, the intervention cannot be recommended for implementation in routine care. Other approaches, e.g. team-based community care, might be more beneficial for people with persistent and severe mental illness.