OCR Text |
Show Using Administrative Data to Identify Adverse Events Paul Hougland, M.D. The Utah/Missouri Patient Safety Consortium is conducting a three-year AHRQ funded grant "Patient Safety Improvement Using State Reporting Systems ". This project is examining the use of ICD-9-CM discharge diagnosis codes as a means of assessing patient safety in acute care hospitals. While there are limitations in using administrative data for this purpose, utilizing a system already in place at all acute care hospitals and that captures all inpatients is clearly attractive. Review of charts from all acute care hospitals will examine the utility of these codes in patient safety assessment and will address incidence of adverse events in Utah. In 2001 the Utah Department of Health attempted for the first time to use discharge data submitted by acute care hospitals in Utah to examine patient safety. A set of ICD-9-CM codes that were thought to be likely associated with adverse events was split into three main categories: misadventures (adverse events associated with medical errors), procedure complications, and adverse drug events. Over the five-year period from 1995-1999, a total of 4248 discharges (.42% of all discharges) suffered a misadventure of medical/surgical care. Over 90 percent of these were due to a cut, puncture or perforation during medical care; 60,000 discharges had a complication of a medical or surgical procedure, while 25,000 discharges had an adverse drug event (Utah Department of Health, 2001). There clearly are limitations in using discharge data in this manner, including the limited capacity to differentiate events that occurred prior to admission from those that occurred during hospitalization, and the inability to capture near misses. These limitations, however, must be weighed against the advantages of incorporating a system that is already in place at most healthcare facilities to evaluate patient safety on a statewide basis. In terms of evaluating patient safety on a statewide basis, leveraging a system that most healthcare facilities already use is clearly an extremely attractive option. Groups in Utah and Missouri applied for and received funding to study "Patient Safety Improvement Using State Reporting Systems." This three-year grant started in 2001 and is part of a federal initiative to reduce medical errors and improve patient safety. The project's focus on using discharge diagnosis codes as an indirect patient safety indicator takes its cues from a recommendation in the Institute of Medicine (IOM) report "To Err is Human" that calls for "collection of standardized information by state governments about adverse events that result in death or serious harm." (Kohn, Corrigan & Donaldson, 2000). While the grant in Utah funds the examination of a wide range of adverse events (surgical complications, nosocomial infections, medical complications, etc), the primary focus is on adverse drug events. Analysis of chart review data from all of the acute care hospitals in Utah seeks to answer two primary questions: • What is the utility of using the selected ICD-9-CM codes as markers for adverse events? • What is the occurrence of adverse events in Utah? Previous studies have shown that adverse events are generally underreported and that with better recognition and documentation of these events the reported rates actually can increase initially. Of interest is an increase in the percentage of Utah inpatient discharges reported to have an adverse drug event, from 3.1% for the first six months of 2001 to 3.2% to the second six months of 2001, 3.4% for the first six months of 2002 and 3.5% for the second six months of 2002. The increases from 2001 to 2002 are statistically significant. In addition to a series of educational sessions focusing on patient safety, quarterly reports detailing these potential adverse event codes are provided to hospitals. These reports include both state-level information and confidential hospital specific information. It is hoped that these measures will increase awareness of adverse events in the hospital and lead to better capture of information in the ICD-9-CM system. ACKNOWLEDGEMENT This project is supported by grant number U18 HS11885 from the Agency for Healthcare Research and Quality. REFERENCES Kohn LT, Corrigan JM, Donaldson MS. Eds. (2000). To err is human:Building a safer health system. Washington, DC: National Academy Press. Utah Department of Health. (2001). Adverse Events Related to Medical Care, Utah: 1995-99. ABOUT THE AUTHOR Paul Hougland, MD, Utah Department of Health, Physician Program Manager, Patient Safety Project Manager. Utah's Health: An Annual Review Volume DC 132 |