Comparing Process- and Outcome-Oriented Approaches to Voluntary Incident Reporting in Two Hospitals
Published In: Joint Commission Journal On quality and Patient Safety, v. 35, no. 3, Mar. 2009, p. 139-144
Posted on RAND.org on March 01, 2009
BACKGROUND: The debate over whether patient safety efforts should focus on adverse events or errors logically extends to voluntary incident reporting in hospitals. Reports emphasizing adverse events take an outcome-oriented approach to improving quality, whereas those emphasizing errors take a process-oriented approach. These approaches were compared in an analysis of 2,228 paper incident reports for 16,575 randomly selected inpatients at an academic hospital and a community hospital in the United States in 2001. METHODS: Measures were developed for process orientation (care varying from the norm) and outcome orientation (physical or nonphysical patient harms, regardless of cause); preventability; and any patient, system, and provider factors contributing to the incidents. RESULTS: Fifty percent of the reports were only process-oriented, 35% only outcome-oriented, and 10% both. Exclusively process-oriented reports were better than exclusively outcome-oriented reports for ascertaining preventability (as determined from 96% versus 25% of reports, respectively), system factors (described in 49% versus 5%), and provider factors (37% versus 4%) but were worse for identifying patient factors (5% versus 63%), all at p < .01. DISCUSSION: Many incident reports contain process information or outcome information but not both. Outcome-oriented reports lack the information needed to assess risk and formulate safety improvements; therefore, follow-up investigations are required. Because process-oriented reports include the necessary information more often, they are more directly useful for improving patient safety. Hospitals should focus voluntary incident reporting systems on capturing process-oriented reports and should train staff to describe contributing factors. This focus should not only improve the quality of the information in the reports but is consistent with efforts to promote a blame-free reporting culture.