Prospective vs retrospective data for evaluating emergency care: a research methodology

Emergency department records and patient charts do not provide enough or sufficiently detailed data for audit of quality of care in a high volume emergency department. As a solution, at the Department of Emergency Medicine, University of Southern California School of Medicine, three emergency medical technicians--hospital-based paramedics--were trained as observers of patient process and treatment. In addition to basic identification information, the form completed by observers listed 21 procedural steps and process data such as sequence, time for completion, type of personnel performing, necessary equipment and supplies, and space for comments. Direct observation of patient process was carried out in 442 patients, a total of 3,882 procedures was observed and recorded. The direct observation is perhaps the most accurate method of data collection for auditing purposes because it reflects actual events. This data was used by the Research Peer Review Committee to help rate the quality of patient treatment process.