Open AccessTesting of four-sample pools offers resource optimization without compromising diagnostic performance of real time reverse transcriptase-PCR assay for COVID-19
Author(s) -
Anirudh Singh,
Ram Kumar Nema,
Ankur Joshi,
Prem Shankar,
Sudheer Gupta,
Abhishek Yadav,
Shashwati Nema,
Bijina J. Mathew,
Arti Shrivas,
Chitra Patankar,
Arun Raghuwanshi,
Ritu Pandey,
Ranu Tripathi,
Kudsia Ansari,
Kuldeep Singh,
Jogender Yadav,
Debasis Biswas,
Sarman Singh
Publication year - 2021
Publication title -
plos one
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.99
H-Index - 332
ISSN - 1932-6203
DOI - 10.1371/journal.pone.0251891
Subject(s) - concordance , turnaround time , pooling , medicine , sample (material) , gold standard (test) , sample size determination , covid-19 , biology , statistics , computer science , mathematics , disease , chemistry , chromatography , artificial intelligence , operating system , infectious disease (medical specialty)
Quick identification and isolation of SARS-CoV-2 infected individuals is central to managing the COVID-19 pandemic. Real time reverse transcriptase PCR (rRT-PCR) is the gold standard for COVID-19 diagnosis. However, this resource-intensive and relatively lengthy technique is not ideally suited for mass testing. While pooled testing offers substantial savings in cost and time, the size of the optimum pool that offers complete concordance with results of individualized testing remains elusive. To determine the optimum pool size, we first evaluated the utility of pool testing using simulated 5-sample pools with varying proportions of positive and negative samples. We observed that 5-sample pool testing resulted in false negativity rate of 5% when the pools contained one positive sample. We then examined the diagnostic performance of 4-sample pools in the operational setting of a diagnostic laboratory using 500 consecutive samples in 125 pools. With background prevalence of 2.4%, this 4-sample pool testing showed 100% concordance with individualized testing and resulted in 66% and 59% reduction in resource and turnaround time, respectively. Since the negative predictive value of a diagnostic test varies inversely with prevalence, we re-tested the 4-sample pooling strategy using a fresh batch of 500 samples in 125 pools when the prevalence rose to 12.7% and recorded 100% concordance and reduction in cost and turnaround time by 36% and 30%, respectively. These observations led us to conclude that 4-sample pool testing offers the optimal blend of resource optimization and diagnostic performance across difference disease prevalence settings.