Groundbreaking New SPI Study Shows That Suicidal Behavior Can Be Predicted

SPI's new journal article in The Journal of Nervous and Mental Disease documents a study with veterans at a major VHA Medical Center, in which the ASQ was able to identify correctly one patient at risk for every two who were not. This is notable since it was achieved with patients not selected for high risk for suicide. It also made its predictions for a short time period of three months, while earlier studies dealt with time periods of from 5 to 10 years.

Although past studies using suicide assessment instruments on patients at high risk for suicide identified most of the patients who went on to suicide, they did so at the cost of misidentifying so many cases (false positives) that their positive predictive value was between 1 to 3 percent. This led investigators to agree that suicidal behavior was unpredictable with instruments then available. It also led many experts to conclude that any such instrument would have to have no more than one false positive for every case correctly identified (i.e. a positive predictive value of 50%).

The article being published in March explains that the ASQ did reach the 50% figure, however, when applied to subgroups of patients with high levels of disability in functioning at work, in social settings, in their families, or with patients who were substance abusers.

The results of this study have encouraged the Veterans Administration to work with Suicide Prevention International in implementing use of the ASQ with more veterans, starting with veterans of the Iraq and Afghanistan wars. The ASQ should be of value, however, to both clinicians and researchers in dealing with a variety of patient populations and in a variety of outpatient and inpatient settings.

Past work by the Hendin group with therapists who had lost a patient to suicide demonstrated that intense affective states that were intolerable were the factor most related to acute risk for suicidal behavior. The uncontrollable nature of the affects -- anxiety, rage, desperation, abandonment, loneliness, hopelessness, self-hatred, guilt, and humiliation -- engendered fear on the part of patients that they were fragmenting i.e., "falling apart." Just before death, the suicides averaged more than 3 times the number of intense affects than comparably depressed nonsuicidal patients.

That work permitted the Hendin group to develop the ASQ. Although the success of the ASQ is primarily due to its use of intense affective states as a measure of painful emotional dysfunction, a large part of that success also comes from the ASQ not asking any questions about suicidal behavior. Patients who denied they were contemplating suicide because they were afraid of the consequences, or were unaware that they were emotionally out of control in ways that were self-destructive, nevertheless had ASQ scores that correctly predicted their subsequent suicidal behavior.

Recognizing the intense, overwhelming, intolerable emotional states that leave patients feeling out of control in a crisis period immediately preceding their suicidal behavior enables us to focus our monitoring and treatment efforts appropriately. The ASQ is demonstrating that identifying these emotional states significantly improves our ability to predict when patients are at acute risk for suicidal behavior.

Herbert Hendin, M.D. is an internationally recognized authority on sucide and suicide prevention. In addition studies of suicide in the United States among college students, the elderly, African Americans, and Vietnam veterans he has done studies of suicide in the Scandinavian countries, the Netherlands, and Hungary. He is one of two U.S. experts on suicide prevention selected by WHO to advise and assist WHO in its global suicide prevention activities. The many awards he has received for his work include the prestigious Louis I. Dublin Award of the American Association of Suicidology for "distinguished contributions to our knowledge of suicide."