Adherence to Choosing Wisely Campaign Guidelines at Three Health Systems

Introduction/Objective With rising healthcare costs in the United States, there has been a push for lab stewardship to improve the quality of patient care while reducing costs. To optimize the use of clinical laboratory testing, the ASCP working with other medical specialty organizations, developed the Choosing Wisely Campaign to promulgate evidence-based guidelines to optimize clinical laboratory testing. Methods/Case Report We examined adherence to three Choosing Wisely guidelines over a four-year period (2017- 2020), through queries of internal cost accounting databases to return aggregate volumes as well as variable and total costs at three large academic health systems. We measured concurrent orders for: 1) erythrocyte sedimentation rate (ESR) with C-reactive protein (CRP), 2) serum/plasma amylase with lipase, and 3) free thyroxine (FT4) and/or total triiodothyronine (TT3) with thyroid stimulating hormone (TSH) when the TSH is within the reference range (using an frequency estimate of 85% based on other studies). We also examined another guideline for concurrent orders for serum aldolase with creatine kinase (CK). We also quantified aggregate variable costs for the non-recommended test in each Choosing Wisely guideline (amylase, ESR, FT4 and/or TT3), and for serum aldolase when ordered with CK. Results (if a Case Study enter NA) Over the four-year period, there were 322,853 unnecessary tests based on these four guidelines (120,587 ESR and CRP, 30,444 amylase and lipase, 164,818 FT4 and/or TT3 with TSH, and 7,004 aldolase). Overall, unnecessary testing decreased between 2017 and 2020 for amylase with lipase, remained essentially unchanged for aldolase, and increased for the other two test guideline scenarios. The largest changes were concurrent orders for amylase and lipase at one health system (38% decrease), and orders for TT3 with a normal TSH result at another health system (324% increase). The four-year variable cost of these unnecessary tests was $1,215,309 ($303,827 mean annual cost), resulting in potential annual variable cost savings of $101,276 for each health system for the four guidelines we examined. Variable costs for unnecessary testing increased by 16.5% ($45,571) over the four-year period. Conclusion Guideline-based unnecessary testing remains as a target to improve laboratory diagnostic testing. There is potential to realize significant achievable cost savings if guidelines are implemented and maintained.

Control Charting Genomic Data

Background Control charting is routine in the quality assurance of traditional clinical laboratory testing. Genomic tests are not typically managed by control charting. We examined control charting to monitor the performance of a clinical next-generation sequencing (NGS) assay. Methods We retrospectively examined 3 years of control material (NA12878) data from clinical genomic epilepsy testing. Levey-Jennings plots were used to visualize changes in control material depth of sequencing coverage in genomic regions of an epilepsy genomic panel. Changes in depth of coverage were correlated with changes in the manufactured lot of capture probe reagent. Depth of coverage was also correlated between quality control material and clinical samples. Results Fifty-seven sequencing runs of NA12878 were analyzed for 1811 genomic regions targeting 108 genes. Manufactured probe lot changes were associated with significant changes in the average coverage of 537 genomic regions and the lowest coverage of 173 regions (using a critical cut-off of P < 5.52 x 10−6). Genomic regions with the highest sensitivity to lot-to-lot variation by average sequencing depth of coverage were not the same regions with the highest sensitivity by lowest sequencing depth of coverage. Levey-Jennings plots displayed differences in genomic depth of coverage across capture probe reagent lot changes. There was moderate correlation between the changes in depth of sequencing across lot changes for control material and clinical cases (r2 = 0.45). Conclusions Genomic control charting can be used routinely by clinical laboratories to monitor assay performance and ensure the quality of testing.