A comparison of nursing and medical diagnoses in predicting hospital outcomes

The main premise of the Nursing Minimum Data Set (NMDS) is that nursing data should be included in the hospital discharge abstract. Yet to date, little empirical evidence has been published to measure the efficacy or usefulness of these nursing data elements. We report the results of a comparison between a daily collection of nursing assessments using nursing diagnoses (NDX) to the Diagnostic Related Group (DRG) and the All Payer Refined DRG (APR-DRG) in their ability to predict three common outcome variables: hospital days, ICU day, and total charges. A secondary data analysis was performed from a large existing data set of four years patient data from a Midwest University hospital.

NDX is significantly associated with hospital length of stay, ICU length of stay, and total charges. NDX also improves explanatory power when added to models with DRG or APR-DRG. This suggests that nursing data compliments existing data and is not redundant with the DRG or APR-DRG. The findings also suggest that NDX explains a different portion of the variance of the three outcome variables in this series. The results of this study support the argument that nursing data should be included in the hospital discharge abstract.