Mortality predictions on admission as a context for organizing care activities.

摘要: BACKGROUND Favorable health outcomes are more likely to occur when the clinical team recognizes patients at risk and intervenes in consort. Prediction rules can identify high-risk subsets, but availability of multiple for various conditions present implementation assimilation challenges. METHODS A prediction rule 30-day mortality beginning hospitalization was derived a retrospective cohort adult inpatients from community hospital Midwestern United States 2008 2009, using laboratory values, past medical history, diagnoses on admission. It validated 2010 data same different hospital. The calculated then used predict unplanned transfers intensive care units, resuscitation attempts cardiopulmonary arrests, condition not admission (complications), unit utilization, palliative status, in-hospital death, rehospitalizations within 30 days, 180-day mortality. RESULTS The predictions derivation validation datasets had areas under receiver operating characteristic curve 0.88. turn strong predictor mortality; modest arrests; weaker other events interest. CONCLUSIONS The probability provides systems with an array prognostic information that may provide common reference point organizing activities many professionals involved patient. Journal Hospital Medicine 2013;8:229–235 © 2012 Society