1. SUICIDE: A WORKSHOP ABOUT: The Notes contain hyperlinks that are highlighted and underlined. These permit the viewer to see printable handouts and to enlarge some slides.
The slides contain references to the scientific literature. In addition, there are three handouts. Six Interviewing Strategies Relevant to Suicide Risk Assessment is a very condensed summary of interviewing techniques that are more fully described in Dr. Shawn Shea's book, The Practical Art of Suicide Assessment: A Guide for Mental Health and Substance Abuse Counselors. Shea's excellent book was published in 1999; it is available from John Wiley & Sons.
This workshop also makes available to you two 8 X 5 cards. The green colored card is a shorthand Guide to Suicide Assessment; the yellow colored card pertains to general risk management and documentation guidelines . Both are intended to be carried in your pocket for immediate reference. Selected card content will be covered in subsequent slides.

2. ACKNOWLEDGEMENTS: The number of individuals that have contributed to this presentation are numerous; many must go unmentioned due to space limitations.
The most current review of the scientific literature and of current recommended practices are available from the American Psychiatric Association. Dr. Douglas Jacobs chaired the APA's Work Group on Suicidal Behaviors; Practice Guideline for the Assessment and Treatment of Patients with Suicidal Behaviors is the product of this workgroup. This detailed review and practice guide was published in November 2003. The University of Michigan Depression Center is most grateful to the American Psychiatric Association for making available a draft version.

3. DISCLAIMER: There are no special insights into the way the world works. Surely, suicide assessment is complex and the scientific literature is extensive. There are one-week long courses that teach what will be presented here in one hour. This presentation has been forced to be selective about what is immediately important and of immediate value. It would be unfortunate for anyone to leave this presentation with conclusions about a personal mistake that has resulted in an adverse event. Conclusions about any individual patient require a complete data base and should stem from a comprehensive peer-review of all the facts. If anything has been learned about suicide outcomes it is: Don't jump to conclusions!

4. SEVERAL THINGS GO WRONG: Click here to view an enlargement of this slides.

Every suicide is different; every suicide has its own story.
This slide describes one convenient way of organizing the chapters of every story. For example, Fred is an 80 year old white male; his father committed suicide (i.e., biological factors). At the present time, Fred has major depression (i.e., predisposing factor); Fred starts drinking (i.e., proximal factor). Fred's wife is caught having an affair with Fred's best friend; Fred has access to a firearm (i.e., immediate triggers) -- one story.
Jane is the high school senior with bipolar disorder who has told everyone she is going to Princeton. When she doesn't get accepted, Jane feels publicly humiliated; she impulsively jumps from a parking structure -- another story.
There are countless such stories.
The goal of suicide assessment is to "read" all the chapters and intervene before the last, and final, chapter is written.

5. 4% HAVE SUICIDE IDEATION: A tiny minority of persons that are at risk for suicide actually die from suicide.
For this reason, it is common to think that predicting suicide is much like trying to find a "needle in a haystack."
NOT TRUE. Predicting suicide is not a needle in a haystack problem.
For a particular individual with a particular story, there are periods of increased risk. These periods can be identified, and when an individual is in a period of increased risk, a specific suicide evaluation needs to be administered to estimate the magnitude of risk. Interventions are then tailored to the degree of risk.

6. SET REALISTIC GOALS: While statistical prediction is impossible for large populations of patients, an astute clinician can increase the odds of identifying a single individual who is at high risk.
Prevention of suicide requires knowing which collection of features are associated with the highest risk and focusing interventions so as to reduce the weight of modifiable risk factors.

7. PROTOCOL FOR SUICIDE RISK: The standard protocol for reducing suicide risk follows the logic presented in the previous slides.
Use a 1-10 severity scale to help quantify risk.
For example, "Mrs. Doe, I have just told you that you have many characteristics that place you at high risk for suicide. If zero represents no risk and ten represents extremely high risk for killing yourself, what number between zero and 10 estimates your risk in the immediate future."

8. PROTOCOL FOR SUICIDE RISK, cont.: Obviously, people who are at highest risk need to be in a protected environment.
The rationale supporting the decision to send home any patient at some risk for suicide requires detailed documentation. Such documentation is the single best strategy to manage practitioner and institutional risk.

9. HIGH RISK ALARMS: Some risk factors are more important than others. Near lethal attempt, available firearm, and the presence of a suicide note sound the alarms of highest risk. There is no more dangerous combination.
For the older adult: 1 in 4 attempts are completed suicides; older individuals use more denial and have greater resolve and planning.

10. SUICIDE MORTALITY-2002: Click here to view an enlargement of this slide.

Beginning at age 65, white male suicide rates increase linearly with age.
Black male suicide rates peak at about 18/100,000 between ages 35-45.
Suicide is uncommon before the teenage years. Suicide rates for all college-age individuals have been climbing since 1950.

11. SUICIDE DEATHS AND DIAGNOSIS: Having a major psychiatric syndrome predicts the vast bulk of suicides.
Therefore, a detailed inquiry about suicide should be a very common feature of every mental health professional's clinical practice. Some forms of mental illness are particularly associated with suicide risk.

12. MORTALITY RATIO/RATES: Suicide risk for each major mental illness has been calculated in the form of a Standardized Mortality Ratio. The further the ratio deviates from one, the greater the risk.

13. SUICIDE IN MAJOR PSYCHIATRIC SYNDROMES: Affective disorder is most predictive of suicide; this sort of table also permits you to approximate multifactor risk.
For example, a 70 year old white male with depression and panic attacks and a recent increase in alcohol use is at very high risk for suicide.
The eating disorder ratio is based on patients in treatment; if less severe eating disorders are counted, eating disorder would rank much lower.

14. DEPRESSION: ANNUAL SUICIDE RATES: From the previous data the following information comes as no surprise. Note the magnification of risk if there is a familial history. For this reason, every interview must include questions about a family history of suicide.

15. THE PERSON AT THE HIGHEST RISK: While it is uncommon to find anyone with major depression advertising their suicide like this woman, some features of major depression are more predictive of suicide than are others.

16. SCIENTIFIC LITERATURE: Every mental health professional has been taught some facts about suicide that are misleading or just plain wrong! The major problem with the vast bulk of studies about risk factors is that they are retrospective studies and tend to mix patients from three groups: completed suicides, suicide attempts, and suicidal ideation. The major difficulty with retrospective studies is that they cannot generate a true control group.

17. MAJOR DEPRESSION: STRONG PREDICTORS: For this reason, Jan Fawcett's study is considered a landmark accomplishment. Fawcett's results come from patients who were involved in the NIMH Collaborative Program on the Psychobiology of Depression. Fawcett followed 954 subjects between 1977-1981. Over this period, there were 25 suicides and comparisons were made to the 929 patients who remained alive. The bulk of the suicides were in the first year --- early in the course of treatment. The following descriptors help to further define selected crucial variables:
Hopelessness: The belief there is no chance of recovery.
Mood cycling: Dark and darker as opposed to good days and bad days.
Turmoil: Agitated quality to the patient's behaviors (e.g., hand wringing, pacing, etc.).
Unusual thinking: Catastrophic interpretation of events leading to an exaggerated sense of public humiliation and shame is one example.
Fewer episodes: No experience getting better and, therefore, no belief improvement is possible.

18. MAJOR DEPRESSION: WEAK PREDICTORS: These data have to be applied cautiously. For example, the loss of a job and public humiliation may place a particular patient at very high risk; other individuals may be less vulnerable. Remember, each suicide is a different story.

19. MAJOR DEPRESSION: POOR PREDICTORS: Suicide ideation is a poor predictor because strong intent is associated with hidden information; people with high intent don't tell you they are thinking about suicide. For this reason, a clinician needs strategies to uncover hidden information. That prior suicide attempt was found to be a poor statistical predictor requires some explanation. Most suicide attempts are non-lethal which makes this attribute a poor predictor. However, suicide is predicted by the lethality of the present and any past attempt. The more lethal the attempt the greater the risk.

20. PSYCHIATRIC INPATIENTS: Katie Busch, Jan Fawcett and Douglas Jacobs collaborated on another landmark study that allows us to consider near-term predictors for suicide while the patient resides on a psychiatric inpatient unit. The authors were able to collect a nationwide convenience sample of 76 patients who committed suicide as inpatients or immediately after discharge. The records were stripped of all identification data and assigned code numbers; thereafter, standard rating scales were employed to characterize each patient who died. These data force every inpatient unit to reconsider the wisdom of 15 minute checks.

21. PSYCHIATRIC INPATIENTS: HIGH RISK: This slide's list of attributes is quite similar to the list of strong predictors of suicide for depressed outpatients. Obvious targets for admission psychopharmacology are anxiety symptoms and insomnia. Included are references to other studies that support Busch/Fawcett/Jacob's conclusions.

22. PSYCHIATRIC INPATIENTS: HIGH RISK, cont.: Note what is not on this list: suicidal ideation. Suicidal ideation was obviously present, but each patient considered was able hide this essential fact. Psychiatrists must learn interview and examination strategies that uncover hidden information.

23. CHRONIC SUICIDALITY: Questions about suicidality are not usually the first aspect of the clinical interview. Begin by asking questions leading to an understanding of the patient's immediate psychosocial problem. How does this patient experience his or her world? Patient's who are chronically suicidal are put off by immediate questions about suicide risk because this is all too frequently the focus of previous interviews. An alliance is built on the foundation of understanding the patient's experience of his or her situation and related emotions.
Lambert gives us a language and a conceptual framework for considering chronic suicidality. The most clinically useful understanding requires some knowledge of the patient's baseline.
Contingent threats are presented in a manipulative or overly dramatized manner as are behaviors seeking some sort of secondary gain.

24. CHRONIC SUICIDALITY: Those patients chronically at risk, and especially those patient that use non-contingent threats, are at increased risk with the attributes listed on this slide. Often, intense feelings associated with these attributes have been reactivated by some recent stressor.
It is important to inquire about the lethality of prior attempts because the lethality of prior attempts is a window on the lethality of future attempts.

25. CHRONIC SUICIDALITY, cont.: Mental health professionals get used to carrying an enormous amount of risk, and we are always in danger of a certain amount of "compassion fatigue." This slide's list of attributes may characterize someone who is chronically suicidal and who says "everything is fine." Such inconsistency should trigger suspicions that everything is not fine and that the patient is hiding information.

26. >75% DENY RISK: This slide tells us why skilled interviewing and examination are critical.

27. SUICIDE RISK ASSESSMENT:

28. GUIDE TO SUICIDE RISK ASSESSMENT: Some of the assessment tools listed on this pocket-handout card will now be considered. At the outset, homework was mentioned. One homework assignment is to study this card and to practice the suggested strategies. click here for pocket-handout

29. RISK ASSESSEMENT SCALES: The scale scores have to match the clinical data. If a patient presents with a serious depression and has a score of zero, this requires an explanation. A very high score sends another message -- at the very least the patient wants to be taken seriously. One advantage of many risk assessment scales is their comprehensiveness. Most scales cover all the basic risk factors and most use a variety of assessment strategies.

30. CHRONOLOGIZE: Again, every suicide is a story. Patients have the most difficulty telling the first part of the story.
For example, after an attempt, patients would rather talk about what has happened since the attempt. Clinically, more is to be gained by a detailed understanding of all the events and behaviors leading up to the attempt.
In general, questions of suicide come after the development of a shared empathic understanding of the person's world.
Look for self-denigration, catastrophic thinking, cognitive distortions and so forth. Shawn Shea's methods are clinical common sense; however, there has been no research done to "prove" that Shea's methods are better than a more open-ended approach. It is important that you develop your own wording so that each strategy fits you.
The handout, Six Interviewing Strategies Relevant to Suicide Risk Assessment provides an overview of Shea's key methods and strategies together with clinical examples.

31. CORROBORATION & SUICIDE RISK: Breach of confidentiality is not done in a knee-jerk fashion; there is discussion and every attempt is made to give the patient choices (e.g., Would you like me to call your father or your mother?).

32. NORMALIZE & OVERESTIMATE: "Symptom amplification" is Shea's word for "overestimation."

33. CHALLENGE & PROHIBIT: This slide describes one strategy that really seems to work for many clinicians. The examiner uses methods that allow the patient to persuade the examiner that the patient is at no immediate risk (i.e., be the Devil's Advocate.).
An example of the Devil's Advocate Strategy: "It would be normal for someone in your situation to consider suicide seriously. Why aren't you suicidal?" This is a particularly powerful clinical tool.
The idea that asking about suicide encourages suicide is false.

34. IMPULSIVENESS & WEAPONS: Most successful suicides are accomplished with a firearm. Every suicide assessment must contain an inquiry about firearms.

35. PAST ATTEMPTS: What is the severity and lethality of suicidal behaviors during the patient's life history? Another attempt may be much more serious.

36. DISPOSITION AND DISCHARGE PLANNING:

37. RISK ASSESSMENT GOALS: Suicide assessment is often an iterative processs in an attempt to achieve the goals presented on this slide.
Several iterations are done on a psychiatry inpatient unit. In reality, clinicians are never done -- just more and more convinced that they have a reasonable handle on suicide risk.
When in doubt, obtain consultation; consultation is a powerful risk management strategy! Use it.

38. INVOLUNTARY HOSPITALIZATION: There is a mistaken notion that a patient has to actually reveal his/her suicide intent or ideation to qualify for civil commitment. Some courts may be uncomfortable with this slide's means for civil commitment; however, a judge is placed in the position of deciding the merits of the clinician's professional opinion.

39. THE NO-SUICIDE CONTRACT: Contracting for safety cannot be used if the patient is intoxicated or incompetent.

40. DISCHARGE CONSIDERATIONS: A common clinical error is made when the clinicians expect too much of their patient.
Not wanting to disappoint the clinician, patient's go along and may even encourage plans that demand too many new and untried behaviors.
"Availability" means being in eye site of the patient -- not, for example, doing office work in another room. "Availability" means being a good companion.

41. TREATMENT CONSIDERATIONS:

42. COGNITIVE BEHAVIORAL THERAPY: This standard has implications for every outpatient and inpatient program. Every clinical program that takes care of suicidal patients has to ask: Are personnel trained to provide CBT? Is CBT administered? When? How often? By whom?

43. PHYSICIAN-ADMINISTERED TREATMENTS: These studies are of suicidal behavior. The multi-site clozapine study of 980 patients with schizophrenia or schizoaffective disorder is based on a comparison with olanzapine; the study was conducted over an 18 month period. Suicide attempts: clozapine 8%, olanzapine 14%. Completed suicides: 5 vs 3 = NS. The hazards ratio analysis gives clozapine a 26% advantage over olanzapine.

44. PHYSICIAN-ADMINISTERED TREATMENTS: The scientific literature does not allow the conclusion that SSRIs have a statistically significant advantage in treating suicidal behaviors. If completed suicides are the outcome measure, a sample of 20,000 would be needed to prove SSRI superiority.

45. DOCUMENTATION AND RISK MANAGEMENT: More homework: Understanding and applying the contents of this pocket-handout card.

46. DOCUMENTATION GUIDELINES: Please read and study the examples of actual clinical documentation printed on the yellow card. The first rule of risk management is to carefully document your work and your thinking. If the outcome is adverse, there needs to be a clear rational as to why certain risks were taken; what were the expected therapeutic benefits, and why did the expected benefits outweigh the potential for risk and adverse events.

47. ILLUSTRATION: Every suicide has its own story. When risk is highest, that is the time patients are most secretive about their true intentions. The clinician's job is to identify situations when patients are most likely to be developing a suicide story and to use specific techniques to get the patient to reveal to them the contents of every chapter. Of course, once the story is known, appropriate interventions must be administered to prevent the patient from ever writing the last chapter.

48. APPENDIX SLIDES: Click here to view the slides. The following slides were excluded in order to confine the verbal presentation to one hour. The scientific literature pertaining to suicide risk assessment is extensive. The subsequent slides are examples of some of the most important information to be found in this literature.

49. CREDITS: Click here to view the credit slides.