The COVID epidemic has shown very high mortality among older people, especially among poly-morbid and dependent subjects. In addition to the classic risk factors of age, dependence and associated co-morbidities, community life exposes to specific increased risks in the event of this easily inter human transmissible viral epidemic. In France, according to the Direction of research, studies, evaluation and statistics (DREES) data (Ehpa study, 2015) more than 600,000 elderly people currently live in nursing homes (NH). Since March 28, a national guidance for monitoring the COVID epidemic in NH has just been set up. In France, 14 178 of the 29 319 COVID deaths (48.35%) by June 10th 2020 occurred among NHs residents. Work to consolidate these data is underway, suggesting a much heavier balance sheet. Faced to this threat, in addition to practical recommendations (barrier protection gestures), strict instructions were also announced to all NH to keep their residents safe from COVID : restricting all visitors, all volunteers and nonessential personnel, and more recently, confining residents in their room in case of incident case of COVID in the NH. Organizational factors of NH such as the prevention strategies deployed before and during the epidemic (pneumococcal vaccination, restricting group activities), as well as NH internal resources (equipment, nursing staff) and health resources in the NH environment (hospital partnerships, support devices, telemedicine) lead to heterogeneous situations and could influence the death rates of residents. On the other hand, social isolation can also precipitate the decline of fragile residents. Beyond the immediate and directly risks linked to COVID-19, the present hypothesize that the organizational measures (guidance and recommendations) put in place can have, during and at a distance from the outbreak, beneficial effects but also deleterious effects depending on the severity of the outbreak of a geographic area. More precisely, the hypothesis is that strong and well-followed recommendations at the time of the epidemic were associated with a reduction in the risk of total death in particular of deaths related to COVID in the zones most affected by the epidemic but also that strong and well-followed recommendations were associated with an increased risk of total death, in particular of deaths unrelated to COVID in the areas least affected by the epidemic.

The PIANO COVID-19 study is an observational, ambidirectional multicenter cohort study. The study will compare the death rates (COVID and non-COVID) in NHs according to the application of the general and specific recommendations of the region by the NH teams, taking into account the COVID infection rate in the NH department, the transfer capacities to hospitals, the medicalization of NH and the general characteristics of residents of NH. These parameters having been fluctuating over time, the analysis will also take into account the period considered.

3 sources of data will be used :

- Questionnaire: The NH team leaders will be asked to report variable related to the implementation of the recommendation/guidance by the nursing home staff, the general characteristics of the residents, and characteristics of the NH, and the monthly rate of death from January to December 2020.

A score will be built in regard to the implementation of the recommendation/guidance from the nursing home staff.

- Data related to the recommendations/guidance of the 13 regions of France (general guidance such as barriers measures and specificity guidance), the COVID infection rate in each NH department, the transfer capacities to hospitals and emergency department, will be collected by the PIANO COVID-19 research team.

For this purpose, referent teams for the COVID-19 outbreak of the each Regional Health Agency (RHA) of the 13 regions of France are contacted in order to recover the recommendation documents having been sent to the NHs.

• Official national data related to the COVID infection rate in the NH administrative region, and transfer capacities to hospitals, will be collected from open access website of the "Santé Publique France" website or other official websites currently providing these data. These data are updated on a daily basis for all administrative region of France and have been registered from the beginning of the outbreak.
• Qualitative study: To improve the understanding of the effects of the recommendations/guidance on resident's health, we will carry out a qualitative study with the NH staff of 20 public and private NH (half clusters NH for COVID-19 and half NH saved from COVID-19). Onsite semi-structured interviews will be conducted with the NH staff to better understand the subjective assessment of healthcare professional on the impact of the recommendations/guidance on the residents' health status. The main questions to which the interviews with the nurses and the nursing assistants will answer, will relate to the functional decline, the appetite and the nutritional status, the mood, the evolution of the cognitive capacities, their psycho-behavioral disturbances and their quality of life during the epidemic period. This qualitative survey will be carried out among NHs having been strongly impacted by the epidemic (NHs cluster with more than 20 deaths) and NHs having not at all been affected by the epidemic (no case of COVID-19 among residents or health care professional). The qualitative study will involve NHs located in very variable geographical regions in regards to the outbreak.

Inclusion Criteria:

• Within each NH, all NH residents living in the NH the 1st of January 2020 will be included.

Exclusion Criteria:

• NHs or Long Term Care Facilities (LTCFs) that refuse to participate.
• NHs that are not voluntary or not located in the regions and NHs that do not have a coordinating doctor .
• Within each NH, all NH residents not living in the NH or LTCFs the 1st of January 2020 will be not be included.

NH volunteers located in different regions more or less affected by the epidemic and NH coordinating physician/directors of NHs via the SFGG website, directors of large private groups (Korian, Orpea, Edenis, DOMUS-vi ..) or public groups (APHP), regional (FMC Gériatres 31) and national (FFAMCO) NH coordinating physician associations (FFAMCO, MCOOR), the referents geriatricians of large cities that have been particularly exposed to the epidemic (Mulhouse, Strasbourg, Paris) or particularly least exposed (Bordeaux, Brest).