Introduction:
Acute gastroenteritis outbreaks caused by noroviruses are common public health incidents on cruise ships¹. Understanding the main drivers of sustained outbreaks and the dynamics of transmission onboard is crucial for evaluating the effectiveness of preventive interventions such as the isolation of infected individuals and to provide real-time forecasts to support outbreak control.

Methods:
We analyzed anonymized line-lists of cases diagnosed in 7 norovirus outbreaks on cruises, totaling 365 cases (range: 20-121). We probabilistically reconstructed likely transmission chains for the largest outbreak using Bayesian approaches^2,3 based on the time of symptoms and diagnosis of each case and on the time spent on- and off-board while infectious. We developed different statistical models, based on renewal equation approaches⁴, to forecast future case counts in real-time. Our statistical models rely on observed case counts and consider delay distributions between symptoms and diagnosis/case isolation (“diagnostic delays”). The forecasting ability of the models was compared on data from the seven outbreaks.

Results:
Reconstructed transmission chains suggest that 47-63% of secondary cases are caused by a 10% of infected individuals with longer diagnostic delays (mean 73 hours vs. 37 for the remaining 90%). The isolation of diagnosed individuals in their cabins immediately upon symptoms reduced their transmissibility by 94%. Integrating diagnostic delays and isolation effectiveness in the forecasts improved the accuracy compared to a baseline model based on the observed incidence of cases. Reducing diagnostic delays markedly improved both outbreak control and forecasting accuracy.

Conclusions:
Diagnostic delays play a key role in norovirus outbreaks on cruise ships. Targeted information campaigns to passengers, encouraging immediate health assistance-seeking upon early gastrointestinal symptoms may promote a significantly improved control of outbreaks.