Evaluation of two methods for quality improvement in intensive care: Facilitated incident monitoring and retrospective medical chart review*

OBJECTIVE: Quality assurance techniques applied within the healthcare industry have been widely used and are intended to improve patient outcomes. Two methods that have been utilized are incident reporting and medical chart review (MCR). The objectives for this study were to evaluate facilitated incident monitoring (FIM) and MCR in the intensive care setting. DESIGN: Cross-sectional comparison of prospective FIM and retrospective MCR. SETTING: Tertiary, 12-bed, closed intensive care unit (ICU) in Australia providing adult and pediatric intensive care to surgical, medical, trauma, and retrieval patients. PATIENTS: Patients present or admitted to the ICU during the 2-month study period. MEASUREMENT AND MAIN RESULTS: During the study period, there were 176 admissions involving 164 patients. A total of 100 FIM reports, of which 70 related to care provided by the ICU team, identified 221 incidents. There were 30 FIM reports that described adverse events, of which only one related to ICU team care. Potential of harm was estimated to be minimal in 49% and significant in 51%; 84% of incidents were considered preventable. Important contextual information was provided, including evidence for the importance of system factors. MCR identified 132 adverse events involving 48% of charts, and 47 related to ICU team care. Common adverse events included nosocomial infections, aspiration, neurologic compromise, respiratory arrest, delayed diagnosis, and treatment. Twenty percent of adverse events were considered preventable, and in 41%, there was evidence of system causation. CONCLUSION: FIM provided more contextual information about incidents and identified a larger number and higher proportion of preventable problems than MCR, but FIM identified few iatrogenic infections, problems with pain management, or problems leading to ICU admission. FIM is easily incorporated into the clinical routine. This study suggests that incident monitoring may be more useful for identifying quality problems, and it could be supplemented by selective audits and focused MCR to detect problems not reported well by FIM.