Suicide: The Unmet Challenge of the Trauma System

Hypothesis  We hypothesized that a significant number of injuries and deaths due to suicide occurred in patients undergoing psychiatric treatment.

Design  We performed a retrospective cohort study of patients who committed suicide and patients with intentional self-inflicted injury.

Setting  San Francisco General Hospital in San Francisco, Calif, and the San Francisco Violent Injury Reporting System.

Patients  We retrospectively reviewed the San Francisco General Hospital records for all attempted and fatal suicides during calendar years 2001 and 2002. Data were merged with suicide data collected by the San Francisco Violent Injury Reporting System.

Results  Two hundred thirty-five suicides occurred between January 1, 2001, and December 31, 2002. One hundred thirty-two patients (56%) who committed suicide had a known mental health disorder at the time of their suicide. One hundred fifteen (87.1%) of those with a known mental health disorder had received psychiatric treatment at some point. Ninety-one patients (68.9%) with a known mental health disorder who committed suicide were receiving psychiatric treatment at the time of suicide. One hundred sixty-five (70%) of those who committed suicide had a traumatic mechanism of death. During the same 2-year period, 3106 trauma patients were admitted to San Francisco General Hospital. Fifty-five (2%) sustained intentional self-inflicted injuries. Ten (18%) of the 55 patients with intentional self-inflicted injury died after arrival at San Francisco General Hospital.

Conclusion  Creation of a feedback mechanism between the trauma and mental health systems has the potential to improve psychiatric care and prevent injury and death.

Suicide is the 11th leading cause of death in the United States.¹ In 2002, there was 1 suicide every 17 minutes in the United States. We hypothesized that a significant number of deaths and injuries due to suicide occurred in patients who were undergoing psychiatric treatment at the time of self-inflicted injury.

We retrospectively reviewed the San Francisco General Hospital (San Francisco, Calif) trauma and emergency department registries for all attempted and fatal suicides during calendar years 2001 and 2002. Data were coded for age, sex, mechanism of injury, Injury Severity Score, and mortality. These data were merged with suicide data from the City and County of San Francisco collected by the San Francisco Violent Injury Reporting System. The San Francisco Violent Injury Reporting System is a pilot site for the National Violent Injury Statistic System through the Harvard Injury Control Research Center, Boston, Mass. The National Violent Injury Statistic System is working to establish ongoing, national data systems on violent injuries. The project has culminated in the establishment of the National Violent Death Reporting System at the Centers for Disease Control and Prevention, Atlanta, Ga.

Two hundred thirty-five suicides occurred in the City and County of San Francisco during the 2-year period between January 1, 2001, and December 31, 2002. One hundred seventy patients (72%) were male and 65 patients (28%) were female. The age range for the men was 13 to 94 years with an average age of 48 years and a median age of 45 years. The women tended to be older with an average and median age of 53 years and an age range of 14 to 90 years. One hundred sixty (68%) were white. Two thirds (146) committed suicide in their place of residence. Sixty-seven (29%) committed suicide by ingesting poison; 64 (27%), by jumping from heights; 45 (19%), by hanging; 42 (18%), by firearms; 6 (3%), by stabbing; and 8 (3%), by drowning (Figure). Overall, 165 patients (70%) had a trauma mechanism as the cause of death. Of all who committed suicide, 35 (14.9%) had a history of alcohol dependence and 55 (23.4%) had a history of substance abuse (Table). One hundred thirty-two (56%) had a known mental health disorder at the time of their suicide. Eighty (60.6%) of the 132 patients with a known mental health disorder had depression; 15 (11.4%), bipolar disorders; and 8 (6.1%), schizophrenia. One hundred fifteen (87.1%) of those with a known mental health disorder had received psychiatric treatment at some point. Ninety-one (68.9%) of those with known mental health disorders were receiving psychiatric treatment at the time of their suicide.

During the same 2-year period, 3106 trauma patients were admitted to San Francisco General Hospital. Fifty-five (2%) sustained intentional self-inflicted injuries. Ten (18%) of the 55 patients with intentional self-inflicted injury died after arriving at the trauma center.

Twenty-six (58%) of the 45 patients who attempted suicide were male and 19 (42%) were female. The age of the patients who attempted suicide ranged from 17 to 77 years with an average age of 38 years. The mean Injury Severity Score of the patients who attempted suicide ranged from 1 to 50 with an average Injury Severity Score of 13. Twenty-four (53%) of the patients who attempted suicide jumped from heights, 20 (45%) stabbed or cut themselves with sharp instruments, and 1 (2%) had a burn.

Seven (70%) of the patients who committed suicide were male and 3 (30%) were female. Their ages ranged from 28 to 83 years with an average age of 52 years. The Injury Severity Score ranged from 25 to 75 with an average Injury Severity Score of 49. Six (60%) of the patients who committed suicide used a firearm, 3 (30%) jumped from a height, and 1 had a fatal burn.

Suicide and attempted suicide are significant causes of trauma-related death and injury^2,3 and are likely to be underestimated because of lack of documentation, the myriad potential mechanisms of injury,⁴ and the difficulty of hospital E codes to consistently capture suicidal behavior.⁵ Although the molecular mechanism of suicide ideation is unknown and the cause is undoubtedly multifactorial, we do know that mental disorders such as depression, schizophrenia, and bipolar disorders are associated with an increased risk of suicide.⁶ In addition, alcohol and drug use⁷ and a history of both self-inflicted injury and assault⁸ are associated with an increased risk of suicide.

Two hundred nine (22%) of the 931 trauma-related deaths in the City and County of San Francisco in fiscal years 2001-2003 were due to suicide.^9,10 The fact that almost a quarter of the trauma-related deaths were due to suicide is astounding and offers an important opportunity for injury prevention. The progression from the “at risk for suicide” status to intentional self-inflicted injury in patients receiving psychiatric care represents therapeutic failure. Yet, we have no formal system for interacting with or providing feedback to our psychiatric colleagues when their patients enter our trauma system because of intentional self-inflicted injury. A formal system for review of suicide deaths will require cooperation between the trauma service, the medical service, the medical examiner, and the psychiatric community. Challenges to creation of a formal review of suicide deaths include the absence of an existing review body in the psychiatric community, concerns regarding the confidentiality of psychiatric records, and interdisciplinary cultural differences regarding the investigation and adjudication of adverse outcomes. These barriers can be overcome by interdisciplinary goodwill and cooperation focused on patient outcome. We are squandering an opportunity to learn more about suicide and identify opportunities to improve care by failing to review these cases as part of our Trauma System Quality Improvement Program.

One hundred sixty-five (70%) of the 235 suicide deaths in the City and County of San Francisco between January 1, 2001, and December 31, 2002, had self-inflicted traumatic as opposed to poisoning injury. Fifty-five patients (2%) of the 3106 patients admitted to the Trauma Service at San Francisco General Hospital during the same period had self-inflicted injury, with a case fatality rate of 18%. One hundred thirty-two (56%) of those who committed suicide had a known mental health disorder at the time of their suicide. One hundred fifteen (87.1%) of those with a known mental health disorder had received psychiatric treatment at some point. Ninety-one (68.9%) of those with a known mental health disorder were receiving psychiatric treatment at the time of their suicide. No quality improvement program exists to link the trauma and psychiatric communities to analyze the causes of suicide. Creation of a feedback mechanism between the trauma and mental health systems has the potential to improve psychiatric care and prevent injury and death.

Correspondence: William P. Schecter, MD, University of California, San Francisco, Department of Surgery, San Francisco General Hospital, 1001 Potrero Ave, 3A33, San Francisco, CA 94110 (bschect@sfghsurg.ucsf.edu).

Accepted for Publication: May 4, 2005.

Previous Presentation: This study was presented as a poster presentation at the Pacific Coast Surgical Association meeting; February 20, 2005; Dana Point, Calif.