The AQI: Present and Future
Richard P. Dutton, MD, MBA
Executive Director, Anesthesia Quality Institute, Park Ridge, IL
The Anesthesia Quality Institute was chartered in 2009, and it began collecting case data in the National Anesthesia Clinical Outcome Registry (NACOR) on January 1, 2010. NACOR was designed to harness the power of the Information Age by aggregating and analyzing large quantities of data. Unlike traditional registries that depend on a trained abstractor to examine medical records and pull out the facts of interest, NACOR accumulates data by direct reporting from the electronic health records (EHR) that are in use every day, including administrative systems such as the ABC billing software and clinical support systems such as ePreop. As anesthesia practices become increasingly digital — driven by the “meaningful use” requirements of the federal government discussed elsewhere in this issue of the Communiqué — even larger quantities and types of data will be available. The barrier is no longer the creation of digital records; it is now the enormous challenge of putting this mass of information to work. Here’s how the AQI is moving forward:
More than 150 anesthesia groups now participate in NACOR, representing 38 states and a broad range of practice types and styles. About 100 of these groups are submitting clinical data, providing information on every case, every day. NACOR now includes information from 4,500 anesthesiologists and 3,500 nurse anesthetists, working in 1100 facilities. Medium sized community hospitals (100-500 beds) represent the largest group of facilities and contribute the greatest number of cases, but NACOR also includes large and small community hospitals, university medical centers, specialty hospitals, attached and freestanding ambulatory surgery centers, pain clinics and office-based practices (Figure 1). As of April 1, NACOR includes more than 4,000,000 cases. NACOR has doubled in size in the past four months, and will double again by the end of 2012. Perhaps even more important, NACOR is growing in depth as well as breadth. More than 85% of cases submitted now include some kind of process or outcome measure and 20% include detailed clinical information from an Anesthesia Information Management System. The ability to link patient risk factors and intraoperative anesthetic decisions to long-term outcomes is the key analytic step that will power improvements in care. In a nutshell, this is the mission of the AQI.
While the AQI has many stakeholders, and receives numerous requests for data and analysis, the most important reporting we do is back to our participating practices. Each group contributing data to NACOR can log in to the AQI Reporting Server at any time to see both their information and how it stacks up against national benchmarks. The “data map” for a typical practice report is shown in Table 1 on page 12. Each element is presented as an individual graph and table, which the practice can slice and dice using drop-down boxes and filters to produce custom displays, down to the level of individual facilities,surgical case types, and providers. At the highest level, data from NACOR allows the practice to understand its own performance and track it over time. At a more granular level, access to this data allows practice managers and researchers to find information specific to whatever administrative, business, quality or scientific questions are most relevant to them. This might include the average duration for a particular kind of case, the difference in postoperative nausea and vomiting between inpatients and outpatients, or the number of cases done per provider per month at a new facility.
A second AQI registry, launched in October 2011, is the Anesthesia Incident Reporting System, or AIRS. This is a national collection of unusual occurrences, near misses, and serious adverse events, collected confidentially direct from the bedside. Any anesthesia provider can contribute to AIRS, simply by accessing the website at www.aqiairs.org. Reporting to AIRS is legally protected from discovery by the AQI’s status as a federally-designated Patient Safety Organization. Aggregated data from AIRS is used to identify emerging risks to patient safety at an earlier point in time than traditional mechanisms such as FDA surveillance or the Anesthesia Closed Claims project. Aggregation of rare events nationwide will provide early warning of new risks, and a mechanism for addressing them. De-identified cases from AIRS provide teaching material for a regular column in the ASA NEWSLETTER, a national-level morbidity and mortality conference that allows the many to learn from the experience of the few.
The newest AQI project is our partnership with ASA to create interactive modules for the Maintenance of Certification in Anesthesiology (MOCA) Practice Performance Assessment and Improvement (PPAI) requirements. Every participant in MOCA now has to make an assessment of their own practice – including clinical data from real patients – demonstrating a quality management problem that they have measured, addressed with new policies or practice, and then re-measured. The ASA modules will provide an easy-to-use format for doing this, focusing on the clinical issues that face us every day, such as monitoring in patients with sleep apnea or perioperative management of blood sugar. The AQI is providing a registry to collect and protect the clinical data entered by MOCA participants, and a reporting feature that will make it easy to document comparisons to national benchmarks and improvement in outcomes over time. For anesthesiologists in practices that participate in NACOR, the AQI will go a step further. We will soon be piloting a voluntary program that helps providers identify the cases they need to collect for their MOCA-PPAI project. In the long run this system will even auto-populate much of the required data.
Helping anesthesiologists in pain management practice keep track of their long-term outcomes is an important future goal of the AQI. This is a substantially different challenge than collecting data from perioperative care, which consists of easily defined events. Instead, pain management is a dynamic process that may extend over many years. Fewer standard definitions exist for good and bad outcomes, and electronic health records in this area are less sophisticated than in the operating room. Working with experts from the American Society for Regional Anesthesia and Pain Management (ASRA) we have laid out a template for data and definitions. The challenge now is to build those into existing EHRs in such a way that the data can be periodically transferred to the AQI without creating an undue burden of reporting for individual clinicians.
Comparative Effectiveness Research (CER), defined as assessment of medications and techniques in a real-world practice setting, is increasingly recognized as an important approach to expanding scientific knowledge. This is because the artificial conditions of a highly-controlled prospective randomized clinical trial (RCT) may not adequately represent the real patients who will be exposed to a new product, creating the potential for unanticipated consequences once the product is approved for use. Further, the cost of performing a good RCT makes including large numbers of patients prohibitively expensive. In a very safe specialty such as anesthesiology, traditional scientific approaches will not have the statistical power to identify rare complications, or to show differences in safety between different approaches. In CER very large sets of data — such as those collected in NACOR — are analyzed to identify outcome differences resulting from common practice decisions in which there is natural variability in approach (e.g. the choice between regional and general anesthesia for lower-extremity orthopedic procedures). Other research topics of interest might be the rate of anaphylaxis to anesthetic drugs, the variation in surgical time associated with resident training, or the impact of quality management feedback on the occurrence of postoperative nausea and vomiting. The development of sophisticated statistical methodology for risk adjustment and propensity scoring has made it possible to advance scientific knowledge through the retrospective study of large clinical data sets. The AQI is positioned to facilitate this kind of research, and is actively seeking collaborators for future projects.
The AQI is growing rapidly in both size and scope, and is collecting more clinical practice information ever day. The challenge of the future is to turn this data into improvements in our practice that lead to better outcomes for our patients. We want your data, we want your ideas for reporting and research, and most of all we want your participation. You can learn more about the AQI at our website: www.aqihq.org.
Richard P. Dutton, MD, MBA is Visiting Professor of Anesthesiology, University of Maryland School of Medicine and AQI Executive Director. To contact Dr. Dutton or the AQI, visit www.aqihq.