In France, a study has reported that about 0.2% of patients visiting the ED died in the ED. A large survey of 145 EDs in 3 French speaking countries has reported that a median of 2 patients dies each week in each ED and its observation unit.

After the death of a loved one, prolonged grief disorder (PGD) is a bereavement-specific syndrome that is defined as intense, prolonged yearning and preoccupation with thoughts of the deceased. PGD prevalence after the loss of a relative was 10% (95%CI 7-14) in the general population.Post-traumatic stress disorder (PTSD) is a mental health condition that is triggered by experiencing a terrifying event. Symptoms may include flashbacks, nightmares, and severe anxiety. The reported lifetime PTSD prevalence is 7% among adults in the general population. Admission and death in the hospital can be a traumatic and stressful experience for relatives, and is associated with an up to 50% risk of PGD and PTSD at 6 months.

In the setting of ICU, several studies have reported that implementation of simple human interventions (information supports including written information on end-of-life care and pro-active communication strategy with systematic interviews with relatives), was associated with a lower risk of PTSD at 3 months (45% vs. 69%) and PGD at 6 months (21% vs 57%). Furthermore, it has been reported that offering the possibility of relatives to be present during nursing and medical care may be beneficial.

In the out of hospital setting, offering the possibility for relative to be present during resuscitation was also associated with a reduction of PTSD at 3 months (15% vs 26%), which was confirmed at 1 year.

The ED setting differs from the one of ICU mainly because exposition time to the dying process is shorter and healthcare workers are less used to manage end of life.

Whether these strategies are beneficial for patients dying in the ED, where dying patients are older with more end- stage chronic diseases and shorter length of stay, remains unknown. The hypothesize of the study is that a multifaceted intervention, including pro-active communication strategy, visual supports, and offering the possibility to be present during nursing and medical care would decrease the risk of PGD in relatives of patients dying in the ED.

Inclusion Criteria:

- Relative of a patient, that is dying in the ED, defined by an anticipated life expectancy of less than 72H with previous do not resuscitate order or treatment withholding.

If several relative are present, participation to the trial will be offered in priority to the patient's trustworthy person, the relative most involved in the patient's care, the partner, a parent or child, another family member.

• Age >= 18 years
• Agrees to be followed-up by phone interview at 3 and 6 months
• Informed oral consent
• Relative with health insurance (except AME).

Exclusion Criteria:

• Relative who do not understand, read, or speak French
• Relatives of organ donor patients (these relatives benefit from specific support and communication by the transplant coordination teams)
• Inability to follow up
• Participation in another intervention trial
• Relative under legal protection (tutorship or curatorship)
• Relative deprived of their liberty by a judicial or administrative decision
• Relative physically unable to give his/her oral consent