Context: Suicide risk assessment in the emergency department is a challenging task for psychiatrists and is further complicated when patients are admitted involuntarily.
Objectives: To evaluate the quality of suicide risk assessments in the psychiatric emergency setting by reviewing physician documentation of process indicators.
Methods: A retrospective review of medical records for patients who were admitted involuntarily to the Kern Medical Center Psychiatric Emergency Service in Bakersfield, Calif. All patients were deemed a “danger to self” as defined by California Law and were admitted for evaluation in June 2006. Medical records were reviewed for 19 process indicators, which were identified from risk factors and treatment guidelines described in the literature. Documentation that a process indicator was not met by a patient was included in the data. Patients were then separated into two study groups: those who were admitted to the inpatient psychiatric unit, and those who were released. Data were analyzed using t tests for continuous variables and χ2 tests for categorical variables.
Results: The medical records of 145 patients were reviewed. None of the suicide risk assessments documented all 19 process indicators. The three most commonly documented process indicators were access to firearms (75.9%), recent stressful life events (75.2%), and “contract for safety” (74.5%). According to medical records, patients admitted to the inpatient unit were more likely than patients released to home care to have been assessed for command auditory hallucinations (P=.02) or prior psychiatric diagnoses (P=.001). Discharged patients were more likely to have been assessed for a family history of suicide (P=.001) or symptoms of major depressive disorder (P=.02).
Conclusions: Many important risk factors for suicide were not documented in emergency department assessments, suggesting that overall quality of psychiatric risk assessments was not optimal. This lack of documentation has important implications from a treatment and medicolegal perspective.