Patient safety measures in burn care: do National reporting systems accurately reflect quality of burn care?

Recently, much attention has been placed on quality of care metrics and patient safety. Groups such as the University Health-System Consortium (UHC) collect and review patient safety data, monitor healthcare facilities, and often report data using mortality and complication rates as outcomes. The purpose of this study was to analyze the UHC database to determine if it differentiates quality of care across burn centers. We reviewed UHC clinical database (CDB) fields and available data from 2006 to 2008 for the burn product line. Based on the September 2008 American Burn Association (ABA) list of verified burn centers, we categorized centers as American Burn Association-verified burn centers, self-identified burn centers, and other centers that are not burn units but admit some burn patients. We compared total burn admissions, risk pool, complication rates, and mortality rates. Overall mortality was compared between the UHC and National Burn Repository. The UHC CDB provides fields for number of admissions, % intensive care unit admission, risk pool, length of stay, complication profiles, and mortality index. The overall numbers of burn patients in the database for the study period included 17,740 patients admitted to verified burn centers (mean 631 admissions/burn center/yr or per 2 years), 10,834 for self-identified burn centers (mean 437 admissions/burn center/yr or per 2 years), and 1,487 for other centers (mean 11.5 admissions/burn center/yr or per 2 years). Reported complication rates for verified burn centers (21.6%), self-identified burn centers (21.3%), and others (20%) were similar. Mortality rates were highest for self-identified burn centers (3.06%), less for verified centers (2.88%), and lowest for other centers (0.74%). However, these outcomes data may be misleading, because the risk pool criteria do not include burn-specific risk factors, and the inability to adjust for injury severity prevents rigorous comparison across centers. Databases such as the UHC CDB provide a potential to benchmark quality of care. However, reporting quality data for trauma and burns requires stringent understanding of injury data collection. Although quality measures are important for improving patient safety and establishing benchmarks for complication and mortality rates, caution must be taken when applying them to specific product lines.