Cost-benefit analysis of two quality control approaches for infectious disease testing

The traditional methods for monitoring quality control (QC) results of using the mean and standard deviation of 20 external QC results to establish acceptance criteria have been widely accepted in medical pathology. The use of Westgard rules to monitor assay performance has been used by clinical chemists for decades and is now applied to infectious disease testing. However, previous reports have indicated this approach creates frequent 'false rejections' leading to a waste of time and resources. This study evaluated the true cost of false rejections in a single laboratory by mapping the QC activities over a 5-month period. From January 2023 to May 2023 inclusive, a laboratory logged all QC results that failed Westgard rules using Bio-Rad Unity software. All activities arising from those QC failures were logged and the cost calculated in Australian dollars. The costs included the cost of reagents, external quality control (EQC), kit control and calibrators, and the staff time. As each laboratory has different pricing structures, certain assumptions were made. The laboratory used the traditional methods for patient result release. Over the same period, the EQC results were submitted to EDCNet software and subjected to National Serology Reference Laboratory-developed QConnect Limits. At the end of the period, the outcomes of the traditional approach to EQC monitoring were compared with QConnect Limits. The senior scientist identified which QC rejections were 'true rejections'.Over the 5-month period, there were a total of 70 flags raised across 15 of the 23 test kits used on two different test platforms. All but one of the 70 episodes resulted in repeat testing of the EQC sample. On 17 occasions, the QC sample was tested more than three times. Recalibration of the assay occurred 12 times, and the results were referred to the senior scientist six times. The acceptance criteria for the EQC were re-set seven times. After ensuing investigations, only one of the 70 flags was deemed to be a 'real error' by the senior scientist. This was also detected by QConnect Limits. The cost of EQC flag investigations over the 5-month period was calculated to be just under $12,000 for 5 months. Extrapolating this to a 12-month period, the additional cost to the laboratory for EQC follow-up was estimated to be approximately $28,700 or about $2,400 per month. Delays in the release of patient results occurred in 42 of 70 episodes. In total, over the 5-month period, the accumulated delay in patients' reports was 68 h ​or 816 min per month. The use of traditional methods for the monitoring EQC is not fit for purpose of infectious disease testing. This is because frequent, normal reagent lot-to-lot variation is expected in serology tests. These changes cause frequent 'false rejections' if the acceptance criteria are too tight, as is the case when using 20 data points to establish limits. The consequences of false rejections are the waste of resources and staff time, additional, unnecessary costs to the laboratory, and delays in the release of patient results.