Using an anesthesia information management system as a cost containment tool: Description and validation

Background: Medical informatics provide a new way to evaluate the practice of medicine. Anesthesia automated record keepers have introduced anesthesiologists to computerized medical records. To derive useful information from the stored data requires programming that is not currently commercially available. The authors describe how they custom-programmed an automated record keeper's database to perform cost calculations, how they validated the programming, and how they used the data in a successful pharmaceutical cost-containment program. Methods: The Arkive(R) (San Diego, CA) automated record keeper database was programmed at Duke University Medical Center as an independent noncommercial project to calculate costs according to standard formulae and to follow adherence to Duke University Department of Anesthesiology's prescribing guidelines for anesthetic drugs. Validation of that programming (including analysis of discarded drugs) was accomplished by comparing database calculated costs with actual pharmacy distribution of drugs during a 1-month period. Results: Validation data demonstrated a 99% accuracy rate for total costs of the drugs studied (atracurium, vecuronium, rocuronium, propofol, midazolam, fentanyl, and isoflurane). The study drugs represented approximately 67% of all drug costs for the period studied. Conclusions: Programming of an anesthesia automated record keeper's database yields essential information for management of an anesthetic practice. Accurate economic evaluation of anesthetic drug use is now possible. In the future, as definitive identification of best anesthetic practices that yield optimal patient outcomes and higher measures of patient satisfaction is pursued, large numbers of patients should be studied. This is only possible through database analysis and complete computerization of the perioperative medical record.