Impact of COVID-19 Nonpharmaceutical Interventions on the Extent of Norovirus Contamination in Oyster Production Areas in Ireland during Winter 2020 to 2021.

ABSTRACT: A significant decrease in norovirus prevalence and concentration was observed in oyster production areas in Ireland during winter 2020 to 2021. Oyster production areas impacted by human wastewater discharges that had been undergoing norovirus surveillance since 2018 were investigated. Samples collected in the winter seasons of 2018 to 2019 and 2019 to 2020, prior to when the COVID-19 pandemic interventions were applied, showed a prevalence of 94.3 and 96.6%, respectively, and geometric mean concentrations of 533 and 323 genome copies per g, respectively. These values decreased significantly during the winter of 2020 to 2021 (prevalence of 63.2% and geometric concentration of below the limit of quantification), coinciding with the control measures to mitigate the transmission of severe acute respiratory syndrome coronavirus 2 of the genus Betacoronavirus. Divergence between norovirus GI and GII prevalence and concentrations was observed over the 3-year monitoring period. Norovirus GII was the dominant genogroup detected in winter 2020 to 2021, with over half of samples positive, although concentrations detected were significantly lower than prepandemic winters, with a geometric mean concentration of below the limit of quantification.

Estimating the distribution of norovirus in individual oysters.

Norovirus in oysters is a significant food safety risk. A recent ISO detection method allows for reliable and repeatable estimates of norovirus concentrations in pooled samples, but there is insufficient data to estimate a distribution of copies per animal from this. The spread of norovirus accumulated across individual oysters is useful for risk assessment models. Six sets of thirty individual Crassostrea gigas oysters were tested for norovirus concentration levels by reverse-transcription quantitative PCR (RT-qPCR): three from a commercial harvest site, and three post-depuration. Five sets had norovirus GII means above the limit of quantification (LOQ), and one below the LOQ, but above the limit of detection. No norovirus GI was detected in pooled tests, and individual oysters were not tested for norovirus GI. Depuration was shown to reduce the mean concentration of GII copies, but not to affect the shape of the distribution around the mean. Deconvoluting the uncertainty of the method, the coefficient of variation was stationary (0.45 ±â€¯0.2). The best fit distribution was either a lognormal distribution or a gamma. Multiplying these distributions by the weight of oyster digestive tissues gave an estimate for the count mean. This was used as the parameter λ in three compound Poisson distributions: Poisson-lognormal, Poisson-gamma, and Poisson-K. No model was found to fit better than the others, with advantages for each. All three could be used in future risk assessments. Preliminary validation of sampling uncertainty using repeated testing data from a previous study suggests that these results have predictive power.