Back

Comparative Analysis of Detection Efficacy Across Different Nucleic Acid Testing Modalities in Blood Centers

Xuelong, G.; Mingming, Q.; Xiaohua, Y.; Liwei, Z.; Yanbin, W.

2025-09-08 hematology
10.1101/2025.09.05.25334984 medRxiv
Show abstract

Background and ObjectivesBlood center Nucleic Acid Testing (NAT) laboratories primarily utilize Individual testing(ID-NAT) and minipool testing(MP-NAT) modes. ID-NAT offers higher sensitivity, while MP-NAT is more cost-effective. While direct comparisons of the two modes exist, the efficacy of a parallel implementation combining both modes has been less explored. Materials and MethodsThis laboratory employs both testing systems in parallel. From March to August 2024, the ID-NAT system was suspended for equipment upgrades. NAT indicators during this suspension period were retrospectively analyzed and compared with data from the preceding seven years to assess impacts on economic benefits and positive yield. ResultsDuring the suspension period, 96,941 specimens were tested. The NAT yield rate was 0.0536%, NAT resolution reactive rate was 0.8624%, NAT resolution efficiency rate was 49.06%, MP-NAT reagent utilization rate was 82.35%, NAT invalid rate was 1.5228%, and NAT invalid pool rate was 0.0328%. Comparisons with the same period in 2023 and other months in 2024 showed no statistically significant differences (p > 0.05). Testing costs decreased by approximately 28.57%. When compared to historical data from the same period, the NAT yield rate,and MP-NAT yield rate, showed statistically significant differences (p < 0.05). ConclusionDifferences exist in NAT yield detection between the parallel ID-NAT/MP-NAT mode and the single MP-NAT mode, requiring cautious interpretation. Economically, the parallel mode demonstrated superior performance.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.