Suicide Risk 30 Days After a Potentially Traumatic Event in Patients Treated by a Medico-psychological Emergency Unit

The national french survey "Mental health in general population", showed that the risk of suicide was significantly increased during exposure to a potentially traumatic event (PTE), with an increasing gradient depending on the psychological impact. This included not only people suffering from post-traumatic stress disorder but also those confronted with PTE, presenting or having presented phenomena of reviviscence or at least an element of psychopathological impact. This risk could concern 30.2% of the population over the entire lifetime. Internationally, recent reviews of the literature tend to show that victims of interpersonal violence are more at risk of suicide (OR 1.99 95% CI: 1.73-2.28). Nevertheless, the literature remains heterogeneous, the definitions vague and the temporal proximity of the violence little taken into account. It is therefore a large-scale phenomenon that remains little explored.

To increase our knowledge in this area, the investigators will focus on patients treated by Emergency Medical Psychological Cells (CUMP), that is victims of disasters, accidents involving a large number of victims or events that may have significant psychological repercussions due to the circumstances surrounding them.

Estimate the prevalence of suicide risk 30 days after exposure to a potentially traumatic event, in people treated by a CUMP

The secondary objective aim to identify:

1. The main psychiatric disorders at 30 days (depressive disorders, bipolar disorders, panic disorder, agoraphobia, social phobia, generalized anxiety, post-traumatic stress disorder, obsessive-compulsive disorder, disorder linked to the consumption of alcohol or other substances, psychotic disorders, eating disorders)
2. The number of death by suicide 1 year after the trauma
3. The incidence of suicide attempts 1 year after the trauma
4. The incidence of suicidal ideation 1 year of trauma Prognostic factors of mortality by suicide, suicide attempts or suicidal ideation will be sought.

A number of 215 individuals would give a precision of ± 5% to a point estimate equal to 15% of the prevalence of suicide risk at 30 days. Assuming a 10% refusal rate to participate in the follow-up visit, this number should be increased to 240 patients.

This number of 215 patients would give 80% power to a logistic regression model of the binary primary endpoint (Y, risk of suicide at 30 days) on a binary independent variable (X) in the presence of a binary covariate (Z). to detect, with an alpha risk of 0.05, an odds ratio equal to 2.50. This number is calculated under the assumption of a baseline risk for the primary endpoint Y of 0.15, the independence of observations, a prevalence of the independent variable X of 50%, a prevalence the covariate Z of 50% and an odds ratio of the primary endpoint Y associated with the covariate Z of 1.50. This number is calculated for a two-tailed Wald test.

The investigators believe that their study could have a real impact on CUMP practices and suicide risk prevention. In fact, the CUMP represents an original device and the context of the emergency and support groups may overlook the risk of suicide for some patients. In addition, there is currently no systematic monitoring planned within the CUMPs. It is therefore an innovative and particularly topical subject that could change clinical practices but also the organization of the healthcare offer. Indeed, several treatment centers for psychotrauma are currently being deployed at the national level (including a center in Lyon, within the Hospices Civils de Lyon (HCL)) and an important work of coordination of treatment structures is in progress.