| Specific Data Considerations | |
|
|
| The models encompass over 94% of Medicare admissions and about 96% of Commercial admissions (excluding early deaths, early transfers to another acute facility or leaving against orders). Generally, only DRG/severity combinations having more than 3,000 admissions for Medicare and 1,500 admissions for Commercial are used. However, this represents almost 600 unique Medicare regression models and over 700 unique non-Medicare models. | |
Additional Considerations |
|
|
|
| The interpretation of the hospital efficiency evaluations must consider the following: | |
| • | The evaluations do not address whether the quality of a hospital is good or bad, but rather their relative efficiency levels based on LOS and admission appropriateness. |
| • | Since the Medicare data is processed uniformly using UB92 forms, and most hospitals use grouper software, there should be a great degree of consistency, completeness and accuracy in coding. This data is coded by the Medicare Fiscal Intermediaries based on the claim information submitted by the hospitals. This process is conducted under audit specifications set forth by CMS. The non-Medicare data may not exhibit this same level of consistency (see Medicare Data and State Data sections for further detail). |
| • | These Indices may vary due to statistical fluctuations and consideration should be given to appropriate confidence levels, which are available in the database. |
| • | Comparisons of hospitals based on the percentage of days avoidable due to longer than necessary LOS may be different than if avoidable admissions are also considered. |
| • | These results measure a hospital’s efficiency for the years that the data is reported using a common benchmark. To the extent that hospitals have subsequently implemented programs to improve their efficiency, their current performance levels are likely to have improved. In order to improve its relative position vis-a-vis other institutions, a hospital must improve by more than the comparison hospitals. |
| • | Some hospitals may be constrained by infrastructure and external community resource limitations from performing at more efficient levels. Usually all institutions within an area are similarly constrained. |
| Indiscriminate use of these results for comparing hospitals is not appropriate. These and other factors should be taken into account. | |
| The measures presented in the Hospital Efficiency Index™ database are clearly not the only factors that should be considered in evaluating a hospital’s performance. Quality measures, such as patient satisfaction, physicians’ qualifications, results of clinical reviews (e.g., chart audits), mortality and morbidity rates, post-surgical complications, etc. are essential to determining an overall picture of a hospital’s performance. | |