What’s Behind the High Numbers

Both preventable and sometimes unpredictable, suicides in hospitals represent a significant challenge for healthcare professionals, who often experience feelings of failure and guilt following these events. But what do we really know about the frequency and occurrence of such incidents? Pierre-Emmanuel Michels, psychiatrist at the University Hospitals of Angers and Laval, France, addressed these questions during a session at Congrès de l’Encéphale, a psychiatry conference held recently in Paris, France.

No Sector Is Immune

“All healthcare sectors are affected by suicides and suicide attempts, including hospitals, medico-social institutions, and the community. No sector is immune to this risk, but psychiatric services report the highest number of cases,” the French High Authority for Health (HAS) noted in a report on the issue.

Because suicide is intentional, some believe prevention is nearly impossible, he said. However, it is widely classified as preventable or at least partially preventable, the report stated.

Healthcare institutions are urged to implement structured prevention programmes, allocate resources for training, and better understand the contexts of suicides to enhance prevention efforts.

Risk in Psychiatric Hospitals

Epidemiological studies from the United States and Sweden found that the suicide rate in psychiatric hospitals is 50-72 times higher than that in the general population.

“The identified risk factors include a family history of suicide, previous suicide attempts or self-harm, despair, guilt, and depression,” noted Michels. The studies also found an extremely high suicide rate immediately after discharge, particularly within the first week post-discharge. “From a medico-legal perspective, a death occurring within 10 days of discharge is considered a hospital death,” he added.

Findings indicate a high suicide risk at the start of hospitalisation (77%), emphasising the need for caution when granting early discharge and implementing random patient monitoring.

“This highlights the potential risk of granting permission at the beginning of a hospital stay and the need to establish a random rotation when monitoring patients in their rooms. The other conclusion of these papers is that no patient in psychiatric hospitals can be considered at low risk of suicide,” Michels emphasised.

Start and End of Hospitalisation

A retrospective cohort study conducted in the United States during 2014-2015, examining suicide deaths in hospitals, found that 73% of them occurred in psychiatric units. Hanging was the most common method (70%) in all settings.

“The risk is significantly higher among individuals who express suicidal thoughts,” Michels explained. Specifically, the odds are 2.35 times greater for those with mood disorders (such as depression or bipolar disorder) and 8 times higher for non-psychiatric patients.

Two systematic literature reviews conducted 3 years apart found the highest suicide risk in the weeks following hospital discharge, particularly in the first few weeks. “These findings highlight the need for close post-discharge monitoring,” he stressed. The reviews confirmed a prolonged risk, highlighting the need for long-term follow-up.

Emergency Departments

Research on suicide in emergency departments is limited. A recent qualitative study in the United States analysing 184 suicide attempts among 118 patients found significant variation in patient profiles, with most not expressing suicidal intent at admission.

Common factors included psychosis, intoxication, acute distress, and manipulative behaviours.

“The study advocates for a ‘zero suicide’ plan, recognizing that while eliminating risk entirely is impossible, efforts should be made to minimise it as much as possible,” Michels said.

Hospital Security

“In France, no new data have emerged since a 2010 review, which found that 5% of suicide attempts occur in hospitals — 3% in psychiatric settings and 2% in general hospitals. As previously mentioned, approximately 70%-75% of hospital suicide attempts occur in psychiatric units,” Michels recalled.

The same study highlighted that the highest risk periods were the first week of hospitalisation and the 2 weeks following discharge. “It already recommended securing hospital environments, both psychiatric and general”, he noted.

In a study conducted as part of a master’s course at HAS, Michels analysed the French national data on serious care-related adverse events. He found that 20% of these incidents involved suicides or suicide attempts. Over 4 years, only 3917 reports were identified, far fewer than the estimated 150,000-200,000 serious care-related adverse events.

From 795 reported cases of suicide or suicide attempts between March 2017 and June 2021, Michels identified five main contributing factors:

1. Inadequate security — unsafe rooms or furniture
2. Communication failures — within and between hospitals and with families
3. Staff-related issues include insufficient training and excessive workload
4. Service organisation — lack of clear suicide risk assessment protocols, limited psychiatric referral access, and staff shortages
5. Patient-related factors — refusal of care, social isolation

Impact on Psychiatrists

Recognising that patient suicide is deeply challenging for mental health professionals, Michels concluded his presentation by discussing a survey on its prevalence and impact on French psychiatrists.

A total of 764 psychiatrists completed the questionnaire. Nearly 9 in 10 (87.3%) had encountered patient suicide during their career, with 13.7% reporting posttraumatic stress disorder symptoms, including guilt, sadness, and shock symptoms.

Notably, 37.1% of respondents reported receiving no support after a patient’s suicide. Additionally, 15.1% of affected psychiatrists considered changing careers, with self-perceived responsibility for suicide being the strongest predictor of its negative impact.

This story was translated from Medscape’s French edition using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.