Email this article to a Colleague Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles via Google Scholar Articles by Moran, M. Search for Related Content

Research Designs Need Not Exclude Suicide Risk

In routine clinical trials of new medications, patients with a history of suicidal behavior are typically excluded, so that the specific effect of drugs on suicidality is not known.

Research on interventions for suicidal behavior is dogged by ethical problems: how to design a study that rigorously tests the effectiveness of treatment, but that adequately protects subjects at high risk of killing themselves.

Yet it is possible to design a randomized, controlled trial that minimizes the risk of morbidity and mortality for suicidal patients while providing valuable data on effectiveness of interventions, according to Maria A. Oquendo, M.D., and colleagues, in a report in the August American Journal of Psychiatry.

Oquendo described her own federally funded intervention study on suicidal patients with bipolar disorder to highlight six strategies for designing methodologically and ethically rigorous research. Among those strategies:

• Use of surrogate outcome measures to protect subjects from actual suicidal acts.
  
• No use of placebo control.
  
• Implementation of psychosocial interventions to monitor for exacerbation or increased risk.
  
• Close monitoring of patients.
  
• Retention of subjects in the study protocol in the event of hospitalization.
  
• Early detection of large effect sizes.
  

Oquendo told Psychiatric News that because of the complications around protection of patients, there has been a dearth of research specifically on interventions for suicidal behavior. And in routine clinical trials of new medications, patients with a history of suicidal behavior are typically excluded, so that the specific effect of drugs on suicidality is all but unknown.

"The issue is that there are not a lot of data about what works well for suicidal patients, because of the nature of study designs," Oquendo said. "There have been meta-analyses using databases from the Food and Drug Administration, and these have been helpful. But generally the number of suicidal patients is low because of the exclusion criteria."

Oquendo is a clinical professor of psychiatry at Columbia University College of Physicians and Surgeons.

Maximizing Equal Selection

Oquendo and colleagues described a randomized, controlled trial comparing the effects of lithium and valproate on suicidal behavior in patients with bipolar disorder. In that study, a treating psychiatrist who was blinded to the study medications was advised about study-drug dosing by an unblinded nontreating psychiatrist on the basis of blood levels.

Subjects with all bipolar disorder subtypes were enrolled as either outpatients or inpatients, and patients with comorbid substance abuse, personality disorders, or psychosis were also included.

However, all patients had to have had a previous suicide attempt and had to be currently in a mixed or depressed state.

These strategies maximized equal selection of subjects and generalizability to the bipolar population, while increasing the statistical power of the study (as well as its beneficence) by ensuring that subjects were those at highest risk for suicide, Oquendo said.

Potential subjects who expressed an interest in the study were evaluated by two clinicians for capacity to provide informed consent. Patients deemed to be imminently suicidal were hospitalized and had an independent assessment of capacity by the inpatient clinical team. The informed-consent process includes a discussion about clinical monitoring, availability of hospitalization as an intervention, and the staff's mandate to act in the face of acute suicidal risk.

In addition, the study was designed to address phases of illness the patients may be experiencing. In an acute phase, patients may have received, in addition to one of the study medications, an antidepressant or an antipsychotic. In the second, or continuation, phase beginning when the subject was euthymic for two weeks, patients continued to take the study medication and antidepressant for six months, or the study medication and the antipsychotic for two months.

In the third phase, when the antidepressant or antipsychotic was tapered off, the subject received only the maintenance study medication. It was in this phase that the study hypothesis could be tested.

Finally, a flexible treatment algorithm allowed for "rescue medications" to treat subjects with emerging mood episodes.

Strategies Reduce Morbidity, Mortality

Within this protocol, the six strategies were employed to obtain pertinent data on a high-risk population while maximally reducing morbidity and mortality.

• Surrogate outcome measures: The outcome measure is time of survival until hospitalization for suicidal ideation, a suicide attempt, or suicide completion or use of rescue procedures. If patients report one or more "planning items," such as writing a suicide note, researchers intervene with hospitalization, a change in medication, or increased monitoring.
  
• No placebo control: "The efficacy of lithium and valproate is well established, and it was deemed unethical to maintain high-risk subjects without medication," the authors stated.
  
• Psychosocial monitoring: The study design includes use of "family-focused" therapy to monitor for exacerbation or increased risk. "This is basically a psychoeducational therapy designed to create an early alarm system and give the family tools to live with these patients, who are sometimes difficult to live with," Oquendo said.
  
• Close monitoring: After family-focused therapy, subjects continue to have semiweekly telephone appointments with a psychologist and monthly in-person appointments with a psychiatrist.
  
• Subject retention when hospitalized: Patients admitted to the hospital continue to take study medications.
  
• Early detection of large effect sizes: An independent data-safety monitoring board established guidelines for interim analysis. If an early large effect of the study drug is seen, the study is terminated preventing unnecessary risk for the subjects.
  

The study, "Prospective Study of Clinical Predictors of Suicidal Acts After a Major Depressive Episode in Patients With Major Depressive Disorder or Bipolar Disorder," is posted online at <http://ajp.psychiatryonline.org/cgi/content/full/161/8/1433?>.

Am J Psychiatry 2004 161 1433[Abstract/Free Full Text]

Email this article to a Colleague Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles via Google Scholar Articles by Moran, M. Search for Related Content

Get information about faster international access.

Privacy Policy

Copyright © 2004 American Psychiatric Association. All rights reserved.

Home | Search | Current Issue | Past Issues | Subscribe | All APPI Journals | Help | Contact Us