Aim: Performance of negative and positive controls serum (NCS and PCS) should be closely monitored for HLA testing exposed to variables including change of reagents, instrument repairs, and technologists. Variation in the performance of NCS indicates assessing cut-off values for flow cytometry crossmatching (FCXM). In addition, understanding the times of specimen arrival and the count of test types performed by each technologist help understand if the staffing and work flow is efficient enough to equally distribute work among the technologists. Here, we incorporated business intelligence (BI) in our laboratory information system (LIS) to achieve these goals.
Methods: The SCC SoftComputer has an optional product marketed as SoftBI. SoftBI is an online data warehouse for data mining and reporting purposes that is populated in real time throughout the day. Reports were designed in Tableau Desktop and delivered to the department via the Tableau Server emails. Two dashboards were developed for this analysis by taking scheduled data extracts to improve report performance. Performance of NCS and PCS is viewed in standard deviation (SD) plot chart. To determine the workload, a count of specimen in the HLA lab by hour of the day and the performing technologist was prepared. Serial samples of listed patients were reviewed to identify increase in cPRA for updating unacceptable on UNOS.
Results: a) T and B cell FCXM cut-offs based on NCS median channel fluorescence (MCF) (Figure 1A,1B) and PCS MCF, and ranges for every quarter were visualized with ease; b) majority of specimen (Mean±SD = 59±9%; Figure 2) for testing arrived 10 a.m. to 1 p.m.; c) unequal distribution of work was realized as the afternoon shift performed thrice the number of HLA identification test and twice the number of FCXM tests as compared to morning and night shifts; and d) serial samples of all listed patients were viewed conveniently to identify increase in cPRA for updating unacceptable on UNOS.
Conclusion: Quarterly FCXM cutoffs were monitored closely, and monitoring specimen arrival times helped in decision making for new hire for 10 a.m. to 6 p.m. shift. Assignment of tests were personalized based on the number of test type performed by each technologist in the past 60 days. Virtual crossmatches could be performed without unexpected updates to unacceptable and successful accepting organ offers.
