| Description |
Deciding whether to admit a patient to the hospital is often a difficult and coslty decision. The huge cost of medical care in the United States and its rate of increase have focused attention on all areas of high expenditure. The hospitalization decision is one of these, and various studies have suggested that many admissions are discretionary. One disagnostic group for which this is true is pneumonia. Guidelines aree being developed to help in the admission decision, but these are often diagnosis-specific. Therefore the diagnosis must be accurate for the guidelines to be valid. There is evidence that discharge diagnoses contain significant errors, and it seems reasonable to hypothesize that if the diagnosis is incorrect, the admission decision could also be. This study took a random sample of Utah Medicare patients who were discharged with the principal diagnosis of pneumonia, and each case was reviewed by an expert physician to assess the appropriateness of admission. Each case was also entered into a computerized decision support system (DSS), Iliad, to obtain a probability of pneumonia. These two variables were then compared. A significant association was found, most noticeable in the rural"" subpopulation. Comparing this approach with a well-known tool used to review need for admission, a modified Appropriateness Evaluation Protocol (AEP), Iliad was found to be more specific but less sensitive. This suggests that the use of both tools concurrently could improve discrimination for appropriate admission of ""pneumonia"" cases. Limitations of this study and future directions are discussed."" |
