A significant and rapid increase in the number of COVID-19 cases and deaths was observed in Perpignan beginning on March 13. Cases were concentrated in particular in two neighborhoods in downtown Perpignan predominantly inhabited by a Romani community: the Saint-Jacques and Haut-Vernet neighborhoods. This alert prompted the implementation of targeted COVID-19 surveillance in a residential area defined in collaboration with local healthcare professionals to monitor the progression and characteristics of this outbreak. This surveillance relied on data collection specifically established at COVID-19 centers set up in the Pyrénées-Orientales, three of which were located in the heart of the affected neighborhoods. The trends in various indicators, both outpatient and inpatient, reflect the escalation of the COVID-19 epidemic in Perpignan during the second week of March. The various available indicators stabilized at a high level by the end of March, then declined to reach the level of low viral circulation observed in the region during the first week of May. In the Saint-Jacques and Haut-Vernet neighborhoods, viral circulation was more intense than in other neighborhoods of Perpignan. More than 2% of the population reported clinical symptoms consistent with infection during a visit to a COVID-19 center between February 24 and May 10, 2020. During this period, the relative incidence of Covid-19 consultations by residents of these neighborhoods was three times higher than in the rest of the city of Perpignan. The measures taken to ensure and reinforce the population’s adoption of preventive measures and lockdown restrictions certainly contributed to a rapid decline in viral circulation in these neighborhoods. The characteristics of cases hospitalized in intensive care at the Perpignan hospital center show a higher prevalence of risk factors in Perpignan compared to cases hospitalized in intensive care across the entire Occitanie region. However, these risk factors—and in particular diabetes and obesity—are particularly prevalent among the Romani population residing in the study area. The outbreak within this community has not only led to a higher incidence, as data from COVID-19 centers tend to show, but has also likely resulted in a greater health burden in terms of severity. These data, collected reactively for epidemiological surveillance purposes, lack specificity and completeness regarding the actual health burden caused by the outbreak in this particularly vulnerable community. These initial findings call for further studies to better understand the direct and indirect health impacts of such outbreaks and to identify the determinants that could be addressed to limit the extent of transmission in the event of a new epidemic wave or cluster. The knowledge thus generated could be extended to the prevention of other infections with comparable transmissibility. Regarding the prospective surveillance of COVID-19 in the coming months, these results highlight the value of developing specific surveillance tools for vulnerable populations where there is a risk of increased viral circulation coupled with greater disease severity.