Implications of and lessons learned from external assurance of eight influenza diagnostics in China.

This study evaluated the ability of laboratories in the Chinese mainland to conduct molecular detection of seasonal A(H1N1), A(H1N1)pdm09, A(H3N2), A(H5N1), A(H7N9), A(H9N2), B(Victoria), and B(Yamagata). Based on a genetically engineered system of virus-like particles (VLPs), the National Center for Clinical Laboratories of China (NCCLs) developed an external quality assessment (EQA) panel. The panel was distributed to 35 laboratories in mainland China to investigate the proficiency of the 16 assays for influenza molecular detection. Using genetic engineering technology, VLPs encapsulating the 37 target genes of 8 influenza viruses were generated. After verification and quantification, 26 influenza virus surrogates with different concentrations were prepared for EQA. Among the 35 participating laboratories, 319 datasets were returned to the NCCLs. Overall, 95.6% (305/319) of datasets correctly reported all 30 samples, while 2.2% (7/319) of datasets with more than one incorrect result were considered as "improvable". A total of 16 misdiagnosed and 18 undiagnosed results were reported. The data analyzed in this study showed good reproducibility in China, but improvements are needed to decrease misdiagnosed and undiagnosed cases, particularly for the A(H9N2) NA gene. Moreover, VLPs are a good alternative specimen type for assay training and proficiency testing purposes.