Norovirus transmission mitigation strategies during simulated produce harvest and packing

Abstract

In the agricultural setting, core global food safety elements, such as hand hygiene and worker furlough, should reduce the risk of norovirus contamination on fresh produce. However, the effect of these practices has not been characterized. Using a quantitative microbial risk model, we evaluated the individual and combined effect of farm-based hand hygiene and worker furlough practices on the maximum risk of norovirus infection from three produce commodities (open leaf lettuce, vine tomatoes, and raspberries). Specifically, we tested two scenarios where a harvester's and packer's norovirus infection status was: 1) assumed positive; or 2) assigned based on community norovirus prevalence estimates. In the first scenario with a norovirus-positive harvester and packer, none of the individual interventions modeled reduced produce contamination to below the norovirus infectious dose. However, combined interventions, particularly high handwashing compliance (100%) and efficacy (6 log₁₀ virus removal achieved using soap and water for 30 s), reduced produce contamination to <1-82 residual virus. Translating produce contamination to maximum consumer infection risk, 100% handwashing with a 5 log₁₀ virus removal was necessary to achieve an infection risk below the threshold of 0.032 infections per consumption event. When community-based norovirus prevalence estimates were applied to the harvester and packer, the single interventions of 100% handwashing with 3 log₁₀ virus removal (average 0.02 infection risk per consumption event) or furlough of the packer (average 0.03 infection risk per consumption event) reduced maximum infection risk to below the 0.032 threshold for all commodities. Bundled interventions (worker furlough, 100% glove compliance, and 100% handwashing with 1-log₁₀ virus reduction) resulted in a maximum risk of 0.02 per consumption event across all commodities. These results advance the evidence-base for global produce safety standards as effective norovirus contamination and risk mitigation strategies.

Keywords: Foodborne illness; Hand hygiene; Produce microbial safety; Quantitative microbial risk model.

Figure 1.. Full conceptual model.

Conceptual norovirus risk model applied to farms and packing facilities. This model depicts the three discrete modeling phases used in this study, as indicated by the large boxes representing Production-Harvest, Packing-Processing, and Consumer Risk, following fresh produce consumption. Produce commodities pass through the model starting in the Production-Harvest phase with a single direct hand contact by a harvester, followed by transfer to produce bins, contact with other produce (comingling), and transfer to a conveyor belt. From the conveyor belt, produce transitions into the Packing-Processing phase with a single direct hand contact by the packer during transfer from the conveyor belt to a final produce box. While the lettuce and tomato commodities followed this sequence of events, raspberries were contacted only by the harvester followed by placement into a clamshell. Virus decay occurs during a holding period prior to produce distribution and consumption. In the Consumer Risk phase, norovirus infection risk characterization integrates exposure assessment using US EPA commodity-specific consumption rates with a fractional Poisson norovirus dose response model. In this model, maximum infection risk represents the risk from consumption of the final packed product associated with the first tactile event, rather than risk associated with diffuse contamination across an entire produce lot. Boxes represent discrete steps and shaded boxes represent produce within the Production-Harvest and Packing-Processing phases along the fresh produce supply chain. Circles in the Consumer Risk phase represent a transition from norovirus contamination events to risk assessment. Arrows represent norovirus transmission between compartments influenced by transfer efficiency and frequency of contacts. Single and bidirectional arrows indicate contamination pathways, with norovirus assumed to transfer from highest to lowest concentration.

Figure 2.

Impact of combined hand hygiene interventions (handwashing compliance, handwashing efficacy, and glove use compliance) on norovirus concentration in two fresh produce commodities (lettuce, raspberry). For lettuce, contamination represents a single focal contamination event resulting from direct tactile contact with two norovirus-positive food handler. Raspberries are contacted only by the norovirus-infected harvester and placed in a clamshell with no repackaging or contact with the norovirus-infected packer. The horizontal reference line represents the lower range of the estimated norovirus infectious dose of 18 virus particles (Teunis et al., 2008). (A) Joint effects of handwashing compliance and handwashing efficacy (in the absence of glove use) on norovirus contamination (estimated total virus particles) of the first raspberry and lettuce commodities handled by the food worker(s). (B) Joint effects of glove usage compliance and handwashing compliance (with handwashing efficacy set at 2 log₁₀ virus removal) on norovirus contamination (estimated total virus particles) of the first lettuce and raspberry commodities handled by the food worker(s). For all figures, error bars represent 95% credible intervals.

Figure 3.

Contribution of hand hygiene interventions (handwashing compliance, handwashing efficacy, and glove use compliance) interventions, using the first model scenario with two norovirus-infected workers, on maximum consumer norovirus infection risk, defined as the risk of infection to the one individual consuming the daily serving amount of the final packed product. For raspberries, only the infected harvester contacted the fruit, therefore hand hygiene interventions were targeted to the harvester only. Complete glove use was defined as 100% glove use compliance for both the harvester and packer. Complete handwashing was defined as 100% handwashing compliance for both the harvester and packer.

Figure 4.

Contribution of hand hygiene interventions and furlough of a symptomatically infected worker, using the community-based norovirus prevalence model scenario, on maximum consumer norovirus infection risk, defined as the risk of infection to the one individual consuming the final packed product associated with the first tactile event. In contrast to the first model scenario with two norovirus-infected workers, this second model scenario adjusts for a community-based norovirus prevalence estimate (4%) of asymptomatic norovirus infection among the harvester and packer workers. The same viral shedding titer distribution was applied to this scenario as was used in the first scenario with the two norovirus-infected workers. Single worker furlough was defined as the exclusion of the symptomatically infected packer with replacement by a healthy worker for the duration of the simulation. As this model assumes there are only two workers, the removal of one norovirus-infected worker represents partial (50%) worker furlough and does not fully eliminate the potential source of norovirus infection nor infection risk. Complete glove use was defined as 100% glove use compliance for both the harvester and packer. Complete handwashing was defined as 100% handwashing compliance for both the harvester and packer. For raspberries, only the harvester was in contact with the produce, resulting in no impact on consumer risk estimates by furlough of the packer.