Does the Emergency Department Need a Dedicated Stat Laboratory?:Continuous Quality Improvement as a Management Tool for the Clinical Laboratory

Open Access
Publisher Website
Google Scholar
Abstract
Using the principles of continuous quality improvement (CQI), the authors conducted a study in response to a request from the Department of Emergency Medicine, Los Angeles County, and University of Southern California Medical Center, Los Angeles, for a dedicated stat laboratory in the emergency department. The stat orders to test serum electrolyte, glucose, blood urea nitrogen, creatinine, amylase, and lipase levels, prothrombin time, and complete blood count (CBC) were evaluated. The study was done in two phases. First, a baseline on stat laboratory test turnaround time was established, and problems were identified. Then, the authors reexamined the turnaround after problems were addressed and a new laboratory information system was installed. In the first phase, median within-laboratory turnaround for chemistry tests was 61 minutes from the time the specimens arrived in the laboratory and 70 minutes for CBCs from the time of accessioning. Delay in physician review of the results (45 minutes) was the longest component of overall turnaround. The second phase of the study found that the median within-laboratory turnaround had improved to 36 minutes for chemical and 55 minute for hematologic tests. However, other preanalytic factors outside the control of the laboratory, such as collecting blood and sending the specimens to the laboratory, and postanalytic delays in physician acknowledgment of the results remained the major components of the perceived turnaround delays. In conclusion, stat laboratory service for the emergency department improved with CQI. The study suggested that resources required to establish a dedicated stat laboratory in the emergency department would be more beneficial if directed toward reducing the preanalytic delays. Further, CQI has great potential as a management tool for the clinical laboratory.

This publication has 0 references indexed in Scilit: