The prevention of recurrent suicidal about people who have made a suicide attempt is a major strand in the prevention of suicide. It is estimated that 10-15 % of people who made a suicide attempt die by suicide. Recidivism rate of suicide increases even faster than the subject is close to the index suicide attempt. A one month recurrence rate is 5 %, 12-25 % at one year. Most people who made a suicide attempt receive ambulatory monitoring. On this population, there is a low adherence to care.

The main objective of the study is to test the effectiveness of a prevention program of recurrent suicidal acts for people who made a suicide attempt.

The secondary objectives of this study are the assessment of adherence to care; the identification of sub - populations benefiting most from this program; the evaluation of the possible generalization level of the program (eligibles persons rate) and its feasibility level.

This is a randomized, controlled and prospective trial comparing an experimental group (usual treatment + interventions) to a control group (usual treatment only). All participants are evaluated after one year of monitoring. The search duration is 2 years (one inclusion year and one year of follow-up). It is planned to include 330 patients, 165 patients in each group.

The program (experimental group) includes three interventions:

• A series of three telephone calls ( the second week , one month and three months after discharge from hospital )
• A systematic telephone contact with the referring physician
• A telephone helpline for people who made a suicide attempt and referring physicians.

Inclusion Criteria:

• Adults ≥ 18 years.
• Patients admitted to the emergency department for a suicide attempt.
• Patients referred to outpatient care.
• Patients who have given their written consent.

Exclusion Criteria:

• Patients not affiliated to social security.
• Patients do not have the faculties needed to be evaluated (cognitive or delusional disorder).
• Patients hospitalized for longer than 72 hours.
• Patients can not be recalled by phone (no phone, homeless, incarcerated)
• Patients who do not speak French.