TUCSON, ARIZ. – As of this month, the Joint Commission on Accreditation of Healthcare Organizations has made the identification of patients who are at risk for suicide one of its patient safety goals for behavioral health care.
Yet little is known about hospital-based suicide and how to prevent it, according to speakers at a workshop on inpatient suicide at the annual meeting of the Academy of Psychosomatic Medicine.
“So far, there is no guidance whatsoever on how to do that, who should do that, what instruments you should use, and how frequently you should assess people,” Dr. Donald L. Rosenstein said of the new patient safety goal.
Despite its being rare, inpatient suicide is the most common sentinel event reported to JCAHO, according to Dr. Rosenstein, clinical director of the National Institute of Mental Health in Bethesda, Md. He said JCAHO is informed of about 50 inpatient suicides each year, most of them occurring on psychiatric units or within 72 hours of discharge.
The new JCAHO mandate applies to psychiatric hospitals and to patients who are being treated for emotional or behavioral disorders in general hospitals. Suicides also occur on medical and surgical wards–albeit less frequently–and speakers focused on all patients for whom consultation-liaison psychiatrists may be called to assess suicide risk.
Most standard risk assessments do not predict inpatient suicides, according to Dr. J. Michael Bostwick. These are often acute events brought on by anxiety, pain, and delirium, said Dr. Bostwick, a psychiatrist at the Mayo Medical School in Rochester, Minn. Accordingly, the reduction of these risk factors can be more important than psychiatric interventions when an inpatient is at risk.
“The key is, suicidality is not really the point. The agitation and anxiety [are],” he said.
Dr. Bostwick cited seminal work by Dr. Jan A. Fawcett, a psychiatrist at the University of New Mexico, Albuquerque, who found that many–but not all–inpatient suicides do not exhibit chronic risk factors, such as hopelessness, suicide ideation, and prior attempts. Suicidal inpatients are more likely to present with acute risk factors, such as psychic anxiety or panic; severe anhedonia; and recent alcohol abuse, according to research by Dr. Fawcett and his colleagues in the 1980s. More recently, Dr. Fawcett reported that 79% of inpatient suicides followed severe anxiety and agitation (J. Clin. Psychiatry. 2003;64:14–9).
Although a medical illness, such as cancer, is a risk factor for suicide, Dr. Bostwick said it is not a useful predictor because most seriously ill patients are not suicidal.
Several reports cited by Dr. Bostwick found that older male patients are more likely to commit suicide. In a Finnish study that compared 26 general hospital suicides to 1,397 suicides outside of hospitals, the hospital patients were significantly older (mean age 58.7 vs. 44.3 years); they were also more likely to use a violent method (96% vs. 62%) or to jump (35% vs. 2%), to have a major depression (62% vs. 30%), and to be delirious (12% vs. 1%).
Conversely, the hospital patients who committed suicide were significantly less likely to depend on alcohol (12% vs. 33%), to have an Axis II disorder (8% vs. 31%), or to present with a borderline personality disorder (0% vs. 14%) (Gen. Hosp. Psychiatry 2002;24:412–6).
“Jumping is a big deal,” Dr. Bostwick said. Although many hospitals have been redesigned to eliminate opportunities for suicidal patients to jump from high places or throw themselves down stairs, he said, such events still happen.
Dr. Bostwick concluded with an unpublished review of 25 cases evaluated as suicide risks at the Mayo Clinic. The patients came from a wide range of medical and surgical units. Of these, 12 had simply said the “S word” (suicide); 5 had been drunk, 4 had engaged in self-injurious behavior, 2 were delirious, and 2 were end-of-life patients.
Consultation-liaison psychiatrists must recognize the limits of their own ability both to predict suicide and to protect patients who are suicidal, said Dr. James L. Levenson. Among the factors he cited for consideration were countertransference by psychiatrists and nonpsychiatric staff; the complexity of depression and decision making; and the legal and medical implications when a patient with an advanced or terminal illness wants to stop treatment.
Assessment can be difficult even when patients are severely depressed, openly suicidal, and/or have already attempted suicide, advised Dr. Levenson, chairman of consultation-liaison psychiatry at Virginia Commonwealth University Medical Center, Richmond.
For example, one patient may be hospitalized after a near-lethal attempt that was not meant to be lethal, such as an unintended overdose of an over-the-counter painkiller. Another person in less serious condition might have expected an overdose of a prescription drug, such as Valium, to be deadly.