Recommendations for proficiency testing criteria for hemoglobin A1c based on the Shanghai Center for Clinical Laboratory's study.

Objectives The US Centers for Medicare & Medicaid Services proposed in 2019 that glycated hemoglobin A1c (HbA1c) be a CLIA'88 regulated analyte. People who commented expressed concerns that the proposed acceptance limit (AL, HbA1c in NGSP unit) ±10% for proficiency testing (PT) would be unable to maintain already improved analytical performance and guarantee the clinical utility of HbA1c testing. Assessing impact of various ALs on PT performance is needed to provide scientific evidence for adopting an appropriate AL. Methods Ten patient EDTA-whole blood specimens were distributed to 318 and 336 laboratories in the 2018 and 2019 PT events organized by Shanghai Center for Clinical Laboratory (SCCL). HbA1c concentrations were measured by participants using various methodologies commonly used in the USA and China. Targets were determined using secondary reference measurement procedures (SRM) at SCCL. "Failed Results" were those outside the SRM-defined target ± AL (5% through 10%). Laboratories with Failed Results ≥2 out of five samples per PT event obtained Event Unsatisfactory Status. Results HbA1c target values ranged 33.3 mmol/mol (5.2 NGSP%) -102.2 mmol/mol (11.5 NGSP%) for 2018 event, and 33.3 mmol/mol (5.2 NGSP%) -84.7 mmol/mol (9.9 NGSP%) for 2019 event. Overall Laboratory Event Unsatisfactory Rates were 11.3-12.2%, 4.8-5.3%, 0.9-3.1%, 0.6-2.2%, 0.6-1.4% and 0.6-1.4%, at AL of ±5, ±6, ±7, ±8, ±9 and ±10%, respectively. Conclusions The AL (in NGSP unit) of ±6% or ±7% for PT evaluation of HbA1c results would be appropriate, with satisfactory event scores for about 95% of participant laboratories in a PT event.