Guideline to Aid Treatment of Suicidal Behavior

July 1, 2004

Assessing and treating patients with suicidal behavior is not an easy task. Acts of suicide cannot be predicted; the best a psychiatrist can hope for is the ability to identify a patient's risk factors and reduce them. With the publication of a new practice guideline, it is hoped that psychiatrists will be better equipped to deal with this particularly vexing challenge.

Assessing and treating patients with suicidal behavior is not an easy task. Acts of suicide cannot be predicted; the best a psychiatrist can hope for is the ability to identify a patient's risk factors and reduce them.

To that end, psychiatrists now have help in the form of a practice guideline from the American Psychiatric Association. The Practice Guideline for the Assessment and Treatment of Patients With Suicidal Behaviors, the first by the APA to address these behaviors, was published in November 2003 and is being disseminated through professional meetings and journals.

Suicidal ideation occurs in about 5.6% of the U.S. population, with about 0.7% of the population attempting suicide. The incidence of completed suicide is far lower, at 0.01%. "This rarity of suicide, even in groups known to be at higher risk than the general population, contributes to the impossibility of predicting suicide," according to the guideline.

"The challenge in working with a suicidal person is that the psychiatrist has to be familiar with a wide range of treatments," Douglas G. Jacobs, M.D., told Psychiatric Times. Regardless of the disorder being treated, "The psychiatrist has to be wary of when suicide can occur in a particular patient."

Psychiatrists do a good job of assessing and treating suicide, despite the fact that "no study demonstrates that we are able to predict suicide," said Jacobs, who is chairperson of the workgroup that developed the guideline, executive director of Screening for Mental Health and associate clinical professor of psychiatry at Harvard Medical School.

None of the suicide scales used in research are recommended for use in clinical practice by the guideline, in part because they produce so many false positives and false negatives, Jacobs said. However, the questions used in suicide scales are useful for psychiatrists to know because they add to one's armamentarium about the appropriate questions to ask.

Mental illness is a major risk factor, present in 90% to 95% of suicides, with the majority suffering from depression, according to Jacobs. Basically, this means that every patient a psychiatrist treats is at some risk for suicide. "I feel that we do a pretty good job, given that the majority of people in treatment don't commit suicide."

However, psychiatry doesn't have a foolproof mechanism for preventing suicides, and even with good treatment, some people still commit suicide. For instance, 5% of all suicides occur in hospitals, so inpatient care isn't absolutely preventive, he said.

Jacobs' hope is that the guideline will provide psychiatrists with a useful structure for assessment and treatment by offering guidance for a variety of situations, such as inpatient care, partial hospitalization, risk management and even how to approach a family in which there has been a suicide.

The guideline is a combination of clinical consensus and evidence-based medicine and gives psychiatrists the most current overview of the risk factors of suicide. Its tables give the psychiatrist an organized approach in terms of assessment, diagnosis, comorbidity, the role of family and patient histories, and current treatment approaches. It also provides a structured approach for conducting a detailed suicide inquiry.

One of the things that psychiatrists sometimes do not appreciate enough is the importance of asking detailed questions about suicidal thoughts, the existence of a suicide plan and any efforts that the patient has made to actualize that plan, said Jacobs. Sometimes a self-destructive act may not be correlated with lethality, and psychiatrists need to try to focus on the intent behind a specific act.

Psychiatrists also need to be aware of the range of treatments available for reducing suicide risk. Lithium (Eskalith, Lithobid), for example, may not be helpful for treating a manic episode but may still be useful in its ability to reduce the risk of suicide in people with bipolar disorder (BD). The guideline also covers the significance of atypical antipsychotics, electroconvulsive therapy, inpatient care and psychotherapy.

Reducing the Rate

According to Jacobs, one of the hard parts of treating for suicide is that only a few treatments demonstrate evidence for actually reducing the suicide rate--lithium, clozapine (Clozaril) and ECT for short-term reduction. Antidepressants lack an evidence base for preventing suicide, but clinical consensus supports that they are also valuable for reducing suicide in people with depression.

The challenge in evaluating the role of antidepressants in reducing the suicide rate is constructing a study, Jacobs said. A handful of studies of U.S. Food and Drug Administration trial data suggest that there is little difference in the suicide rate between patients who receive an antidepressant and those who receive a placebo. That would seem to address the ethical concern of giving a placebo to a patient who is suffering from potentially life-threatening depression. Even so, clinical trials, as a rule, exclude patients who are suicidal.

A report on the adolescent suicide rate by Columbia University researchers found a 10% drop in the 1990s that suggested a link to the increase in antidepressant use. However, the finding represents an important statistical association, although not direct evidence, that the increase in antidepressant use among teens caused a reduction in suicide rates, said Jacobs.

The workgroup found the lack of evidence base supporting clinical treatments disillusioning, Jacobs said. "That means we are going to have to be creative in doing research, particularly with the antidepressants."

Determining Risk

Suicide scales have been used in research since the early 1980s, and the guideline workgroup reviewed them all. They concluded that the scales, which are used in controlled settings, were not clinically useful because of the complexity of suicide.

According to Jacobs, "None of the scales that are out there are specific enough or sensitive enough for general clinical usage. Suicide assessment still is left to the judgment of the psychiatrist based upon a comprehensive psychiatric examination, a review of current and past treatment, and a detailed inquiry into suicidal thinking and behavior."

The guideline purposefully refrained from including a table that quantified risk factors, Jacobs told PT. "We left it that the overall determination of risk is a clinical judgment, but it's based upon an organized examination."

The group did not feel that scientific evidence supported a specific scale or tool that psychiatrists could use in determining risk level, and the whole point of determining risk is to then inform a treatment plan. For example, a high-risk patient usually requires hospital treatment; however, there are always exceptions, such as if a patient is well known to the psychiatrist and has a supportive family. Likewise, a patient with moderate risk might need to be hospitalized due to clinical circumstances.

The guideline's recommendations include the consideration of various factors, some of which are listed below.

Comorbidity. Psychiatrists need to be aware that comorbidity increases the risk of suicide, particularly when it involves an Axis I disorder with borderline personality disorder or substance abuse, Jacobs said. Having more than one disorder complicates treatment, and it may mean that the psychiatrist has to work a little harder and be more aggressive in terms of follow-up.

Suicidal behavior. The guideline contains a whole section on the significance of suicidal behavior. Even though suicidal behavior in itself is not predictive of suicide, it is a significant risk factor. Follow-up is critical for a person who has made a suicide attempt once they leave the hospital.

ECT. The guideline's task force determined that ECT is useful in short-term reduction of suicide risk, but it does not result in long-term reduction. Psychiatrists who administer ECT have to consider follow-up as an essential component of treatment.

Anxiety. The guideline emphasizes the role of anxiety in suicide and the appropriate pharmacotherapy for treating it.

Psychotherapy. The group knew from other practice guidelines that studies providing evidence for psychodynamic psychotherapy were lacking but that clinical consensus supported its utility, particularly when working with the chronically suicidal patient. The psychotherapeutic relationship may be the most significant aspect of reducing risk.

In terms of cognitive-behavioral therapies (CBTs) and dialectical behavioral therapies, the evidence is there, but not as strong as people have thought, Jacobs said. As with other therapies, the evidence suggests a reduction in suicidal behavior but not necessarily in suicide. "The guidelines, I think, uncover the reality that we have a lot of work to do in terms of demonstrating that our treatments can reduce suicide."

This is particularly true for dialectical behavioral therapy and CBT, Jacobs said. "Our overall clinical consensus is that our treatments are effective, but it is the combination of psychotherapy and having a psychotherapeutic relationship, regardless of orientation, and pharmacotherapy."

The Process

The APA's Practice Guidelines series began in 1993 and initially focused on specific disorders, such as major depressive disorder, schizophrenia and BD. Then came the idea to focus not only on disorders but behaviors such as suicide. That becomes a little more complicated because suicidal behavior occurs across disorders, Jacobs said. The suicidal behavior guideline refers to the other guidelines, such as depression guidelines on ECT use and selecting dosages for antidepressants.

Putting together the suicidal behavior practice guideline took about 2.5 years, Jacobs said. The workgroup began the process in January 2001 and finished during the summer of 2003. The group screened about 18,000 articles, ultimately selecting 800 articles that went through the review process.

The group created evidence tables for each article, using a format that summarized the article's findings and limitations, Jacobs said. Each article was then placed into a category based on strength of evidence.

The task force tried to cover every clinical situation, and the end result is an intense document, Jacobs said. "We've covered a good portion of the landscape here."

The first half of the practice guideline was published in November 2003, and Jacobs presented it at the APA annual meeting earlier this year. The full document can be found on the APA Web site at <www.psych.org/psych_pract/treatg/pg/prac_guide.cfm>.

Jacobs' hope is that psychiatrists will look at the document in bits and pieces and pay particular attention to the tables that cover risk and protective factors and the questions to ask about suicidal thinking.

Jacobs is also interested in knowing how useful the guideline is. "In one sense, the guidelines have been published but the work is not finished," he told PT. The task now is to get the information out there and see what questions people have and how useful they find it, he added.