EXPLORE RISK ADJUSTMENT USING CLAIMS DATA FOR SURGICAL OUTCOME MEASURES

The Deficit Reduction Act (DRA) of 2005 requires the Secretary to expand the domains of Hospital Quality of Care Measures for public reporting on Hospital Compare (HC) to include Outcomes/Efficiency in addition to process of care mandated by the Medicare Prescription Drug, Improvement, and Modernization Act (MMA) of 2003. In compliance with these requirements, the Centers for Medicare & Medicaid Services (CMS) developed and posted six 30-Day Mortality and Readmission Measures for Acute Myocardial Infarction (AMI), Heart Failure (HF) and Pneumonia on HC incrementally since 2007. However, these measures are claims-based and limited to measuring outcomes of medical conditions. To provide the public with a broader spectrum of quality measures, CMS plans to expand the domains of the measures posted on HC to include surgical outcomes, preferably outcomes based on clinical data such as the registry data. The American College of Surgeons (ACS) has developed five risk-adjusted Surgical Outcome Measures using the clinical data collected by the NSQIP. The five measures are listed below. 1. Risk Adjusted Lower Extremity Bypass (LEB) Surgery Outcomes 2. Risk Adjusted Colorectal Surgery Outcome Measure 3. Risk Adjusted Case Mix Adjusted Elderly Surgery Outcomes Measure 4. Risk Adjusted Surgical Site Infection Outcome Measure 5. Risk Adjusted Urinary Tract Infection Outcome Measure One of these ACS measures, Lower Extremity Bypass Surgery Complications, was recently endorsed by the National Quality Forum (NQF) and the other four have been submitted to the NQF and are currently undergoing evaluation. Since the risk adjustment methodology of the four measures under NQF evaluation is similar to that of the endorsed LEB measure, they are likely to be endorsed. CMS is considering adopting these measures for public reporting on HC since they met CMS's program need for outcome measures that are based on clinical data. Currently hospitals are collecting and submitting to CMS the clinical data elements needed to calculate 44 chart-based process-of-care measures for posting on HC in order to obtain their Annual Payment Update. The clinical data elements necessary to calculate these ACS measures are not part of the data elements submitted to CMS by hospitals for these 44 measures. Adopting these ACS measures for reporting on HC would clearly add additional data collection burden to hospitals. Therefore prior to adopting these measures, CMS would benefit from the information regarding the degree the data collection burden will increase for hospitals. Specifically the information that CMS is seeking is whether some (or all, if feasible) of the clinical data elements necessary to calculate the ACS measures can be replaced with existing claim data. The purpose is to minimize data collection burden for hospitals without compromising the integrity of the measure methodology. This Purchase Order is to request the ACS to conduct analyses to explore ways to elucidate whether the risk adjustment embedded in the ACS measure methodology can be performed using a hybrid database (i.e., a mix of chart and claims data) or a database with claims data only. The findings of this investigation is important to inform CMS decision making on public reporting of the ACS surgical outcome measures on HC.