Research published in the November issue of Anesthesiology describes development of a new Risk Stratification Index (RSI) that allows important clinical outcomes such as length-of-stay and mortality for surgical patients to be accurately compared among hospitals using only publically available billing data. The new risk stratification system is more accurate than existing outcomes measurements such as the Charlson Comorbidity Index (CCI) and the investigators have made it freely available.
"Standardized risk stratification metrics coupled with outcome and cost measures are central to the success of the existing healthcare model," said the lead investigator Dr. Daniel I. Sessler, M.D., Professor and Chair of the Department of Outcomes Research at the Cleveland Clinic. "Clinicians and hospitals use major outcomes to evaluate performance and guide improvement efforts. Patients and insurers also use outcomes to select hospitals. However, outcomes from various hospitals can only reasonably be compared after adjusting for patients’ baseline risk and the risk associated with different operations. Our Risk Stratification Index permits outcomes such as duration of hospitalization and mortality to be compared fairly across institutions."
To develop the Risk Stratification Index, Dr. Sessler and his colleagues analyzed more than 35 million Medicare records dating from 2001to 2006. From these records, the group developed highly predictive risk-adjustment models for length-of-stay, and for in-hospital, 30-day, and 1-year mortality. The validity of the system was then confirmed by applying the system to the Cleveland Clinic.
"An important aspect of our Risk Stratification models is they are entirely objective, reproducible, and transparent; they do not include any ‘adjustments’ or subjective ‘fixes,’" said Dr. Sessler. "The system thus provides a transparent and fair basis for comparing outcomes among hospitals."
The anesthesia community is excited at the prospect of the new RSI. In a recent review, Alexander A. Hannenberg, M.D. and Norman A. Cohen, M.D., commented that, "the risk stratification methodology presented by Sessler and colleagues is an enormous contribution to the quality and uniformity of hospital outcome reporting," but they note some potential shortcomings of the system, particularly the use of Medicare claims (billing) database.
"Applying RSI methodology to current physician payment administrative claims data will be unlikely to generate as robust a formula for predicting mortality, morbidity or other endpoints as the authors demonstrated here with inpatient data. However, the reorganization of Medicare contracting into combined Part A and B Medicare Administrative Contractors has the potential to link patient-specific quality, cost, facility, and provider data in a way that could allow a modified RSI to include quality and cost inputs," said Dr. Hannenberg and Dr. Cohen.
In a second review of the RSI, reviewer Fredrick K. Orkin, M.D. concurred with Dr. Hannenberg and Dr. Cohen, "The devil is in the details, principally data quality." Dr. Orkin recognized that "faced with this trade-off between data quality and volume of cases, Sessler et al. opted for administrative data because their goal was a risk-adjustment methodology ‘broadly applicable’ (generalizeable) to all hospitals."
Dr. Orkin noted another important consideration for the RSI when he said, "While Sessler et al. propose using their methodology for public reporting of hospital-level outcomes, the notion of report cards is problematic: Consumers pay more attention to ratings when buying a toaster than selecting hospitals, possibly due to restriction imposed by their health insurance plans."
Dr. Sessler and his team have committed to releasing their Risk Stratification Index into the public domain, including future updates and potential extensions to other clinical endpoints and other types of administrative data. The goal is for any entity to be able to risk stratify their patients using only standard billing records and basic patient demographics — thus permitting important outcomes such as mortality to be fairly compared among hospitals.