COVID-19: optimizing surveillance in long-term care facilities

Press release
|

Throughout the world, long-term care facilities have been high-risk sites for virus spread since the emergence of the COVID-19 epidemic. In this context, the capacity for timely surveillance and control of case numbers is essential in such facilities. Scientists at the Institut Pasteur have joined forces with the University of Versailles-Saint-Quentin-en-Yvelines, Inserm, and the National Conservatory of Arts and Trades to analyze various methods for surveillance and testing of residents at these facilities. The results of this study reveal the importance of tailoring surveillance strategies to facilities' testing capacities. A "testing cascade" strategy is more effective for facilities with adequate capacity. In contrast, a group testing strategy is more effective for those with low capacity. The results were published in BMC Medicine on December 8, 2020.

Faced with the COVID-19 pandemic, hospitals are on the front line of tackling the disease and are high-risk sites for virus spread. In hospitals providing follow-up and long-term care, which are not supposed to admit patients with COVID-19, the virus can also spread actively if such sites are accessed by individuals such as patients and staff who have been infected with the virus in the community or in other care facilities.

It is crucial for these facilities to detect any resulting nosocomial outbreaks (i.e. originating in healthcare facilities) as early as possible in order to implement adequate control measures to prevent widespread transmission of the virus within facilities and infection of already frail patients.

However, it is difficult to detect the virus quickly due to several factors: a large proportion of infected individuals present with few or no symptoms, and most of these care settings have limited resources for testing patients and staff.

Different strategies based on testing capacities

To help implement effective surveillance strategies for COVID-19 cases in long-term care facilities, Institut Pasteur scientists collaborated with the University of Versailles-Saint-Quentin-en-Yvelines, Inserm, and the National Conservatory of Arts and Trades to conduct a modeling study based on a detailed description of contacts within one such facility. 

This involved scientists analyzing the effectiveness of a wide range of surveillance strategies. "The results reveal that strategies initially recommended and implemented since March in long-term care facilities can be tailored to individual facilities’ testing capacities," explains Lulla Opatowski, a scientist in the Epidemiology and Modeling of Bacterial Escape to Antimicrobials Unit at the Institut Pasteur and joint last author of the study. These strategies entail conducting RT-PCR tests only on individuals presenting with COVID-19 symptoms.

Based on facilities' local testing capacities, the scientists subsequently identified test usage strategies enabling more effective surveillance. Where adequate testing resources are in place (5 or more tests per 100 beds per day), "testing cascade" methods enable significant and rapid detection of virus spread, thus preventing large-scale nosocomial outbreaks. This "cascade" strategy entails prioritizing testing of individuals at high risk of infection, namely those with symptoms or patients recently admitted to hospital.

In contrast, where resources are limited (less than 2 tests available per 100 beds per day), the study results reveal the value of a strategy based on group testing.  This strategy, which is already implemented in several countries, involves combining samples from several individuals1 and analyzing them with a single test. Although somewhat less sensitive, these group tests enable detection time to be reduced, thus cutting the number of secondary infections in hospitals, with relatively few resources required to achieve this.

"Overall, our results reveal the challenges of monitoring COVID-19 outbreaks in long-term care settings. Although other testing technologies are emerging, RT-PCR tests are still the most widely adopted solution in these care settings and remain the gold standard strategy in COVID-19 surveillance systems," concludes Lulla Opatowski.

This is one of the first modeling studies addressing COVID-19 surveillance in long-term care settings. By increasing testing capacities and consequently updating surveillance procedures, it may be possible to facilitate early detection of disease outbreaks in care facilities.

 


Source

Optimizing COVID-19 surveillance in long-term care facilities: a modelling study, BMC Medicine, december 8th, 2020

David RM Smith*1,2,3, Audrey Duval*1, Koen B Pouwels4,5, Didier Guillemot1,2,6, Jérôme Fernandes7, Bich-Tram Huynh1, Laura Temime§3,8, Lulla Opatowski§1,2.

1. Institut Pasteur, Epidemiology and Modelling of Antibiotic Evasion (EMAE), Paris, France
2. Université Paris-Saclay, université de Versailles Saint-Quentin-en-Yvelines, Inserm, CESP, Anti-infective evasion and pharmacoepidemiology team, Montigny-Le-Bretonneux, France
3. Modélisation, épidémiologie et surveillance des risques sanitaires (MESuRS), Conservatoire national des arts et métiers, Paris, France
4. Health Economics Research Centre, Nuffield Department of Population Health, University of Oxford, Oxford, United Kingdom
5. The National Institute for Health Research (NIHR) Health Protection Research Unit in Healthcare Associated Infections and Antimicrobial Resistance, University of Oxford, Oxford, UK
6. AP-HP, Paris Saclay, Public Health, Medical Information, Clinical Research, Le Kremlin-Bicêtre, France
7. Clinique de soins de suite et réadaptation, Choisy-Le-Roi, France
8. PACRI unit, Institut Pasteur, Conservatoire national des arts et métiers, Paris, France

 

§ : Contributed equally

* : Contributed equally

AURÉLIE PERTHUISON
Head of press office

ANNE BURLET-PARENDEL

Press officer

MYRIAM REBEYROTTE
Press officer

NATHALIE FEUILLET
Press office assistant

 

MARGAUX PUECH PAYS D'ALISSAC
Press officer

 

 

SEND THE PRESS OFFICE AN EMAIL

 

Subscribe to our newsletter

Back to top