The overall aim of this thesis is to describe and explore the concept of caring in intensive psychiatry. An initial inventory was made of nursing care activities in a PICU, based on an analysis of critical incidents. This inventory resulted in four categories: supporting, protecting and use of the structured environment (Study I). Caring in intensive psychiatry was also studied through ethnographic fieldwork that that led to the conceptualization of the PICU staff as projecting a culture of stability. Within this culture, the overall goal was to prevent, maintain and restore stability as turbulence occurred. Cultural knowing, as expressed through nursing care, was further described in terms of providing surveillance, soothing, being present, trading information, maintaining security, and what has been termed reducing (Study II). A focused approach was applied to study the staff’s different approaches to observing patients in relation to the practice of surveillance in psychiatric nursing care. PICU staff moved flexibly between a latent and a manifest approach to surveillance (Study III). Having conceptualized the culture as one of stability, a concept analysis was conducted upon the concept of stability. The analysis revealed that stability is by no means a static condition; it fluctuates and can be distorted. Intervening with nursing care when turbulence occurs, can involve both the use of active and passive stability systems (Study IV). Further, I argue that caring in intensive psychiatry can be accurately described as the projection of rhythm and movements. Nursing care in terms of movements creates fluctuations in stability as it entails a rhythm of caring in intensive psychiatry. In conclusion, physical boundaries and incorporated control along with tactful sensibility involve rhythm and movements within limited structures and closeness in care. This thesis contributes to articulating advanced nursing practice within intensive psychiatry