Use of Rapid Response Teams in Psychiatry: Variables that Impact Safety

The addition of rapid response teams (RRTs) has emerged as a treatment option for medically deteriorating psychiatric inpatients. The goal of this study was to identify risk factors for use of RRTs at the time of initial psychiatric evaluation and develop measures to predict use of this emergent medical intervention. We have not located any studies that have previously addressed this issue.

We studied the use of RRTs and code teams (CTs) on inpatient general and specialty psychiatric services in a large urban academic psychiatric hospital. RRT or CT calls were made for 60 patients on these units between January, 2010 and December, 2012. Our retrospective chart review study compared the cases of these 60 patients with 120 matched controls. Using χ and the Fisher exact tests, we identified variables that differed significantly between the case and control groups. Next, we used a forward stepwise regression model to create prediction tools to risk stratify patients at the time of initial assessment.

Bivariate analyses identified 14 independent statistically significant patient variables. We created 2 risk prediction tools: (1) an "exhaustive" tool (which used 12 of the total 38 factors we considered) based on a stepwise regression model that yielded an area under the receiver operating curve (AUC) of 0.91, and (2) a simplified tool referred to by the acronym "SCHEME" with 6 factors and an AUC of 0.76.

We propose the acronym SCHEME to refer to a tool that can be used to quickly and easily assess medical risk in prospective psychiatric inpatients at the time of admission, which can help reduce the use of RRTs and CTs.