The Burden of Add-On Orders in a Laboratory With Manual Processing. (11th September 2019)

• Record Type: Journal Article
• Authors: Barouqa, Mohammad; Fecher, Roger; Wolgast, Lucia R; Forest, Stefanie
• Is Part Of: American journal of clinical pathology. Volume 152(2019)Supplement 1
• Page Start: S16
• Page End: S16
• Publication Date: 2019-09-11
• Subjects: Diagnosis, Laboratory -- Periodicals; Pathology -- Periodicals; 616.07
• DOI: 10.1093/ajcp/aqz112.031
• Languages: English
• ISSNs: 0002-9173

Abstract

Abstract: In order to avoid additional blood draws for their patients, clinicians frequently order add-on tests on existing specimens in the laboratory. Manually processing these add-ons is problematic, utilizes tremendous resources, and raises concerns regarding specimen integrity. The objective of this study was to analyze add-on ordering patterns and assess the time and resources required to complete these orders. In this retrospective study conducted at a large, multisite, academic medical center, a report was generated from the laboratory information system (LIS) to identify all the add-on orders that were placed with details about the type of add-on test, which specimen it was added to, and location from where an add-on order was placed for a 2-month period (August 5 to October 4, 2018). The workflow was observed and financial cost was calculated. The laboratory received 5, 658 add-on orders during the study period. By laboratory protocol, 859 tests were cancelled, leaving 4, 799 tests to be processed. Add-on orders were most common for liver tests (7.48%), creatine kinase (6.35%), troponin (6.31%), vancomycin level (5.93%), thyroid-stimulating hormone (4.91%), magnesium (4.81%), and vitamin B12 (4.33%). The add-on orders were mainly generated for inpatient (74.07%) followed by emergency (17.79%) and outpatient departments (8.12%). The add-on request is placed by the clinical provider in the hospital information system (HIS) as a generic "Add-on Order" test, with free text to specify the test and specimen. A clerk in the laboratory uses the LIS-generated work list to electronically order the requested tests to the original specimen. Subsequently, a clerk must manually locate the original specimen and deliver it to the performing lab to run the add-on test. The median turnaround time from the provider placing the generic add-on order in the HIS to the lab placing the add-on in the LIS is 119 minutes. The median time for the provider to place an add-on after the initial order is 462 minutes. The average time needed to monitor the add-on work list, electronically place the add-on to the original order, and retrieve the sample for one test for a skilled senior statistical clerk is 7 minutes. The average number of add-on requests received daily is 71. Therefore, the daily time to process add-ons is 497 minutes (8.2 hours). Based on the hourly cost and fringe in the laboratory ($28.92/hour), these add-on tests cost $237/day in labor ($86, 000/year). Our study demonstrates the significant cost and labor burden of add-on tests in a laboratory with manual processing. The laboratory is considering transitioning to direct provider ordering of add-ons to existing specimens using laboratory-defined rules in the HIS and moving to total laboratory automation with robotic specimen archival and retrieval to reduce the manual efforts, which would streamline the add-on workflow.