Issues in Measuring Documentation and Coding Change Presented to: American Hospital Association Federation of American Hospitals Association of American Medical Colleges

In the FY 2011 Inpatient Prospective Payment System (IPPS) Proposed Rule, the Centers for Medicare & Medicaid Services (CMS) described its methodology for measuring the documentation and coding changes the agency believed resulted from the shift to Medicare Severity Diagnosis Related Groups (MS-DRGs). The Moran Company (TMC) was asked to examine CMS' approach and determine if it is an appropriate method to isolate documentation and coding from the other factors that could affect the " Case Mix Index (CMI) " —a measurement of average patient severity. Based on our review of the methodology, examination of the logic, and our own analyses, our conclusion is that the CMS methodology does not adequately isolate documentation and coding from other factors when calculating changes in CMI. CMS' methodology cannot distinguish among the different causes of a change in average patient severity. Brief discussion The CMI is measured by calculating the average weight for cases across all Fee-For-Service (FFS) discharges in a fiscal year. With consistent Diagnosis Related Groups (DRGs), the CMI from different years can be compared to see changes in apparent severity. CMS developed a methodology to compare two years of data to measure the impact of coding and documentation on CMI due to the implementation of MS-DRGs. CMS' methodology attributes the entire change in CMI to changes in documentation and coding, which, based on our analyses, we believe to be incorrect. We believe that CMS' methodology does not adequately account for other factors or issues which may lead to changes in CMI from one year to the next. These other factors include historical trends reflecting changes in the beneficiaries' needs and corresponding treatment. Our analyses included: 1. Examination of trends in CMI over a 10 year period. After converting the data to a common grouping system, we found that there were increases in CMI prior to the implementation of MS-DRGs. This raises questions as to why CMS did not test or attempt to account for this historical trend of increasing CMI for reasons other than documentation and coding in response to the introduction of MS-DRGs. 2. Examination of " Within vs. Across DRGs ". DRGs can be collapsed down into what are known as " Base DRGs " which measure underlying conditions, and then can often be split by severity level. CMS' analysis, which we also replicated, found that the change in CMI is due primarily to changes in measured levels …