Introduction: Although rapid response systems (RRS) are ubiquitous, a consistent reduction in hospital mortality has remained elusive. A minimum threshold for RRS encounters has been observed before arrest reduction can be expected. It is unclear whether this same threshold is associated with a reduction in mortality. In addition, an optimal rate of RRS encounters beyond which additional benefit is minimal has not been identified. Here we explore the relationship between RRS encounters and hospital mortality, hypothesizing that an optimal RRS encounter rate could be identified to ensure both program effectiveness and efficiency. Methods: This was a prospective, observational study performed in a 500-bed urban hospital. A standard RRS existed for several years prior to the intervention, which included transition to a mobile/proactive team and implementation of a novel educational curriculum based on principles of cognitive psychology. The main objective for this analysis was to explore the relationship between RRS patient encounter volume and Risk-Adjusted Mortality Index (RAMI). Data were collected on a quarterly basis over a 4-year period that spanned the intervention. Polynomial regression was used to model the number of RRS encounters as a predictor of RAMI, with r-squared used to quantify the strength of association and statistical significance assumed for p< 0.05 Results: An 80% reduction in non-ICU arrests, a 20% reduction in ICU arrests, and a reduction in hospital mortality were observed following the intervention. A strong nonlinear association was observed between RRS encounter volume and RAMI (r-squared=0.73, p< 0.0001). An initial reduction in RAMI with increasing RRS volume reached a horizontal asymptote at approximately 150 encounters/1000 discharges. Conclusions: We observed a substantial reduction in hospital mortality following transition to a mobile, proactive RRS trained using a novel curriculum. A steep decrease in RAMI with increasing RRS volume was observed up to approximately 150 encounters/1000 discharges, beyond which no additional improvement in RAMI was observed. These data support an “optimal dose” of RRS encounters, potentially representing a target RRS volume for hospitals for both clinical effectiveness and program efficiency.
Read full abstract