Theoretical Framework:  A conceptual model for this study was developed from previous research on failure to rescue. 

 Subjects (N=10,187 hospitalizations): Inclusion criteria: 1) ³ 60 years old; 2) admitted to a Midwest tertiary hospital; 3) at risk for falls.  Mean age was 74, most were Caucasian, retired, and admitted from home.

 Methods: Data for this study was abstracted from 9 electronic data repositories at one Midwestern tertiary hosptial.  Failure to rescue was defined as a death that occurred after a complication found in the medical record abstract (MRA).  Complications were coded using the International Classification of Diseases, 9^th Revision (Clinical Modification) (ICD-9-CM), located in the MRA.  A predictive model was built using logistic regression.

 Results:   Males were 56% more likely to experience failure to rescue than females. Both surgical (e.g. laminectomy, peripheral vascular bypass, gastrointestinal OR procedures) and non-surgical (e.g. blood transfusion, conversion of cardiac rhythm) medical treatments were positively associated with failure to rescue.  Two pharmaceutical treatments were associated with being less likely to experience failure to rescue and four pharmaceutical treatments were associated with being more likely to experience failure to rescue.  Four nursing treatments (i.e. Bathing, Cough Enhancement, Airway Management, and Artificial Airway Management) were associated with failure to rescue.

 Conclusion: Failure to rescue measures the early detection and the appropriate, quick treatment of complications to prevent death.  The findings from this study help identify variables associated with failure to rescue. The results indicate that medical, pharmaceutical, and nursing treatments are predictors of failure to rescue.