Critical care nurses' decision making: Sedation assessment and management in intensive care

Aims. This study was designed to examine the decision making processes that nurses use when assessing and managing sedation for a critically ill patient, specifically the attributes and concepts used to determine sedation needs and the influence of a sedation guideline on the decision making processes. Background. Sedation management forms an integral component of the care of critical care patients. Despite this, there is little understanding of how nurses make decisions regarding assessment and management of intensive care patients' sedation requirements. Appropriate nursing assessment and management of sedation therapy is essential to quality patient care. Design. Observational study. Methods. Nurses providing sedation management for a critically ill patient were observed and asked to think aloud during two separate occasions for two hours of care. Follow-up interviews were conducted to collect data from five expert critical care nurses pre- and postimplementation of a sedation guideline. Data from all sources were integrated, with data analysis identifying the type and number of attributes and concepts used to form decisions. Results. Attributes and concepts most frequently used related to sedation and sedatives, anxiety and agitation, pain and comfort and neurological status. On average each participant raised 48 attributes related to sedation assessment and management in the preintervention phase and 57 attributes postintervention. These attributes related to assessment (pre, 58%; post, 65%), physiology (pre, 10%; post, 9%) and treatment (pre, 31%; post, 26%) aspects of care. Conclusions. Decision making in this setting is highly complex, incorporating a wide range of attributes that concentrate primarily on assessment aspects of care. Relevance to clinical practice. Clinical guidelines should provide support for strategies known to positively influence practice. Further, the education of nurses to use such guidelines optimally must take into account the highly