Investigation of a COVID-19 Outbreak on an Aircraft Carrier: An Epidemic Model Based on Real Data

Background: From mid-March to mid-April 2020, the French aircraft carrier Charles de Gaulle suffered a COVID-19 outbreak. An investigation was performed to describe the outbreak, including how the virus was introduced on board. We present the clinical pictures of COVID-19 cases with risk factors for infection and severity, effectiveness of preventive measures, and we discuss the real collective protective rate. Methods: A confirmed case was any service member with a positive SARS-CoV-2 RT-PCR and/or who presented symptoms of anosmia and/or ageusia. We considered the entire crew as a cohort and questioned them about individual, epidemiological, and clinical data. We performed viral genome sequencing and searched for SARS-CoV-2 in the environment. Results: The attack rate was 65% (1085/1767). The sex ratio was 6·9, and median age was 29 years. There were four clinical profiles: asymptomatic (13·0%), non-specific symptomatic (8·1%), specific symptomatic (76·3%), and severe (i.e. requiring oxygen therapy, 2·6%). Active smoking prevented severe COVID-19; age and obesity were risk factors. The instantaneous reproduction rate Rt and viral sequencing supported the hypothesis of several introductions of the virus on board, with an acceleration of the Rt when preventive measures were lifted. Physical distancing prevented infection (ORa, 0·55, 95% CI, 0·40-0·76). In the end, transmission stopped when the proportion of infected personnel was large enough to prevent the virus from circulating (65%, 95% CI, 62-68). Discussion: Asymptomatic and non-specific clinical pictures of COVID-19, combined with a lack of knowledge at that time about the specific symptoms of COVID-19 (anosmia, ageusia), delayed detection of the outbreak. Once it was identified, the lack of an isolation ward made it difficult to manage transmission on board, and the outbreak spread until a collective protective rate was reached. However, physical distancing was effective when applied. Syndromic surveillance and point-of-care biology could enable early detection of such viral emergences or outbreaks. Funding Information: No funding to declare. Declaration of Interests: The authors declare that they have no competing interests. Ethics Approval Statement: This was not an experimental protocol, but an outbreak investigation with routine care provided to infected individuals, so no ethical approval from any named institutional and/or licensing/ethics committee was required. We obtained individual informed consent to analyze data. No administrative authorization was required to access and use medical records. We chose not to provide certain details and to aggregate certain data in order to maintain patient anonymity. All methods were carried out in accordance with relevant guidelines and regulations.