Clinicians who assess suicidal patients in the emergency department setting face the challenge of whether to admit the patient to inpatient or outpatient care, and data on predictors of compulsory admission are limited, wrote Laurent Michaud, MD, of the University of Lausanne, Switzerland, and colleagues.
To better identify predictors of hospitalization after self-harm, thefrom 1,832 patients aged 18 years and older admitted to four emergency departments in Switzerland between December 2016 and November 2019 .
Self-harm (SH) was defined in this study as “all nonfatal intentional acts of self-poisoning or self-injury, irrespective of degree of suicidal intent or other types of motivation,” the researchers noted. The study included 2,142 episodes of self-harm.
The researchers conducted two analyses. They compared episodes followed by any hospitalization and those with outpatient follow-up (1,083 episodes vs. 1,059 episodes) and episodes followed by compulsory hospitalization (357 episodes) with all other episodes followed by either outpatient care or voluntary hospitalization (1,785 episodes).
Overall, women were significantly more likely to be referred to outpatient follow-up compared with men (61.8% vs. 38.1%), and hospitalized patients were significantly older than outpatients (mean age of 41 years vs. 36 years, P < .001 for both).
“Not surprisingly, major psychopathological conditions such as depression, mania, dementia, and schizophrenia were predictive of hospitalization,” the researchers noted.
Other sociodemographic factors associated with hospitalization included living alone, no children, problematic socioeconomic status, and unemployment. Clinical factors associated with hospitalization included physical pain, more lethal suicide attempt method, and clear intent to die.
In a multivariate analysis, independent predictors of any hospitalization included male gender, older age, assessment in the Neuchatel location vs. Lausanne, depression vs. personality disorders, substance use, or anxiety disorder, difficult socioeconomic status, a clear vs. unclear intent to die, and a serious suicide attempt vs. less serious.
Differences in hospitalization based on hospital setting was a striking finding, the researchers wrote in their discussion. These differences may be largely explained by the organization of local mental health services and specific institutional cultures; the workload of staff and availability of beds also may have played a role in decisions to hospitalize, they said.
The findings were limited by several factors including the lack of data on the realization level of a self-harm episode and significant events such as a breakup, the researchers explained. Other limitations included missing data, multiple analyses that could increase the risk of false positives, the reliance on clinical diagnosis rather than formal instruments, and the cross-sectional study design, they said.
However, the results have clinical implications, as the clinical factors identified could be used to target subgroups of suicidal populations and refine treatment strategies, they concluded.
The study was supported by institutional funding and the Swiss Federal Office of Public Health. The researchers had no financial conflicts to disclose.