The first rule of life is to reveal nothing, to be exceptionally cautious in what you say, in whatever company you may find yourself.1
Unfortunately, many patients, when they find themselves in the company of a psychiatrist or other mental health professional, may adopt the above dictum, especially when asked about suicide. Those patients in the most angst, and with the strongest intention of relieving that angst by suicide, are the exact same patients most likely to follow the above dictum. Clinicians are left with a disturbing paradox. Those patients most likely to die by suicide are those patients least likely to relate their intention to do so.
Whether the patient has a desire to share this critical information is often directly related to the interviewing skills of the clinician involved. Our challenge is akin to a surgeon's challenge in the sense that with every new patient we meet, there exists the possibility that our skill-set will be a determining factor in a life-and-death situation. But as interviewers we face a challenge that is not present for our surgical colleagues. Although a surgical patient's body may present anatomical challenges unique to that body, they do not do so intentionally. They do not lie, but potentially dangerous patients do.
Some patients with serious suicidal intent may relay their actual intent in stages during an interview. When evaluating such patients, one would expect that after being asked about suicide, a nuanced interpersonal dance may unfold in which the patient shares some information, reads how the clinician responds, shares some more information, re-evaluates “where this session is going,” and so on. Indeed, some patients with serious suicidal intent who are trying to decide how much to share, may intentionally withhold their method of choice (such as a gun) for fear of its removal until they arrive at a decision during the interview that they are not certain they want to die. If they have previously thought of, or implemented, several suicidal methods, an interviewer may not hear of the method of choice until several abandoned methods are revealed first.
Arguably, the single most important take-home point from this article is this—many patients who strongly intend to kill themselves, when first asked how they plan to do it, may frequently choose not to share their method of choice even if well-engaged and asked directly. Not understanding this principle can lead to unnecessary tragedy. Therefore, the question is, “are there interviewing techniques and strategies that might decrease the likelihood of such untoward outcomes?”
First, as per Figure 1, the interviewer explores the patient's presenting suicidal “events” (defined in the CASE Approach as including the patient's suicidal feelings, death wishes, ideations, plans, intent, and actions) reported by the patient as being present during the past 48 hours or several weeks (Region #1 - Presenting Events). Particularly dangerous patients will often not relay their method of choice here for fear that their method of choice, such as a gun, will be confiscated. Second, the clinician explores the patient's suicidal ideation/actions that have been experienced during the previous 2 months preceding the presenting event (Region #2 - Recent Events). It is here, in the Region of Recent Events, that clinicians will often uncover an immediately or imminently dangerous patient's method of choice, time spent contemplating the method of choice, and actions taken on the method of choice. Consequently this article focuses solely upon the interviewing techniques used during the exploration of the patient's Recent Events. Interested readers can find the most recently updated description of the complete CASE Approach (including streaming video demonstrations) in the book Psychiatric Interviewing: The Art of Understanding.8 After finishing the Region of Recent Events, the clinician subsequently completes the CASE Approach by proceeding to selectively explore the patient's past suicide attempts (Region #3 - Past Events) and the patient's thoughts and intentions concerning suicide that arise during the interview itself (Region #4 - Immediate Events).
Microstructure: Validity Techniques and the Importance of Their Sequence
Over 15 behaviorally operationalized interviewing techniques, known as “validity techniques,” have been delineated for uncovering particularly sensitive or taboo topics ranging from substance abuse and suicide to domestic violence and incest.9 Of these, the following five are of immediate use in uncovering the method of choice in patients who are predisposed to withholding such information: the behavioral incident, gentle assumption, denial of the specific, catch-all question, and symptom amplification.
Behavioral incidents. Originally described by Pascal,10 behavioral incidents are questions that ask for specific facts, behavioral details, or trains of thought (called fact-finding behavioral incidents) such as, “How many pills did you take?” and “Have you ever gotten the gun out while you were having thoughts of killing yourself?” or that simply ask the patient what happened sequentially (called sequencing behavioral incidents), such as, “What did you do next?” By using a series of behavioral incidents (an interviewing strategy known as making a “verbal video” in the CASE Approach and as a behavioral analysis in cognitive-behavioral therapy) the interviewer can help a patient enhance validity by re-creating, step by step, the unfolding of a potentially taboo topic such as a suicide plan or attempt.
Gentle assumption. Originally delineated by Pomeroy et al.11 for use in eliciting a valid sex history, gentle assumption is used when an interviewer suspects that a patient may be hesitant to share a taboo behavior such as substance abuse or suicide. With gentle assumption, the clinician assumes that the potentially incriminating behavior is occurring and frames the question accordingly using a gentle tone of voice. Thus, with regard to suicide, one would ask, “What other ways have you thought of killing yourself?” instead of “Have you thought of other ways of killing yourself?” Gentle assumption is used as a technique for uncovering methods of suicide that patients have not mentioned when first asked about how they would kill themselves.
Denial of the specific. Many patients, who are hesitant to share their method choice, may do so after the strategic employment of a gentle assumption, but not all. Some particularly dangerous patients will respond with a “no.” In these patients, a clinician can use a validity technique called “denial of the specific.”12 As its name suggests, with denial of the specific, the interviewer asks the patient about any specific common method of suicide that the interviewer is suspicious might be the patient's method of choice and is being purposefully withheld. Examples of denials of the specific are as follows: “Have you been having any thoughts of shooting yourself?” or “Have you been having any thoughts of hanging yourself?”
The take-away interviewing principle is simple: if a patient responds to a gentle assumption with a negative, which the clinician doubts, consider using a denial of the specific or a series of denials of the specific. This interview strategy, called “a Gentle Push” (gentle assumption, patient responds with a not-believable negative response, use of a denial of the specific), is powerful, elegant in its simplicity, easily taught, and easily tested for competence.
Let's see it at work with a clinician interviewing a veteran with depression back from his third deployment in Afghanistan. The topic of suicide has been raised. The veteran admitted to an overdose of aspirin 1 month ago. We are picking-up the interview with the very last sentence of the verbal video the clinician has made regarding the overdose. The clinician has doubts that overdosing is the patient's method of choice:
Vet: . . . I wouldn't use pills in the future . . . It's never a done deal.
Clinician: What other ways have you thought of killing yourself? (gentle assumption)
Vet: (pausing, looks away, sighing heavily) Nothing really. (sighs again) Everything is going to be okay. I'm not thinking of anything right now, just getting by day by day.
Clinician: Have you been having any thoughts of shooting yourself, you know, even fleeting in nature? (denial of the specific)
Vet: (pauses, looks away) Maybe.
Clinician: Do you still have a gun at home? (behavioral incident)
Vet: (glances at clinician) I suppose so.
Clinician: Have you ever gotten the gun out while you were thinking about shooting yourself? (behavioral incident)
Clinician: When was the last time? (behavioral incident)
Vet. (pauses, then looks at the clinician) I guess about two weeks ago . . . .
The clinician is “in.” She has just uncovered the vet's method of choice. A potentially life-ending lethal means (firearm) can now be removed from the household. When using denial of the specific, clinicians should avoid jamming various methods into a single question such as, “Have you been having thoughts of shooting yourself, overdosing, jumping, or hanging yourself?” A simple yes or no is likely to be inaccurate since patients will often only respond to the last item. A separate denial of the specific should be used for each suspected method.
Catch-all question. Some patients may persist in withholding their method of choice even after the use of one or more denials of the specific. The catch-all question allows an interviewer to unobtrusively see “if something has been missed” by literally asking if such is the case.13 The catch-all question looks like this, “We've been talking about a lot of different ways that you've been thinking of killing yourself, are there any other ways you've thought of that we haven't talked about?” Even a slight hesitancy here can be the signal that a method of choice is still under the table.
Symptom amplification. In response to its culturally taboo nature (as well as fears about what will happen if one accurately shares the amount of suicidal preoccupation), suicidal thinking and behaviors are often minimized, even with patients having only moderate amounts of such thinking. Symptom amplification bypasses this defense mechanism without risking the disengagement that could result by confronting it.14 It sets the upper limits of the quantity at such a high level that, even after the patient downplays the amount, it remains clear that a significant problem is present.
With regard to suicidal ideation, a typical symptom amplification might sound like this, “On your very worst days, how much time do you spend thinking about killing yourself (slight pause): 10 hours a day, 14 hours a day, 18 hours a day?” A patient may minimize saying, “Oh, not that much, probably at least 6 or 8 hours on a really bad day.” Despite the patient's defense of minimization, the resulting number reflects a great deal of time spent on suicidal thought. To effectively use symptom amplification, the interviewer must always suggest an ascending series of numbers, set high.
Importance of Sequence
When trying to uncover the presence of a hidden method of choice, the sequencing of the validity techniques cannot be overemphasized. The techniques generally work best when used in conjunction with each other and in the sequenced order described in the CASE Approach (Figure 2). This interviewing principle is particularly true with patients with an intense need to relieve their angst via suicide.
Exploring region of recent events. “—”, denies ideation; BI, behavioral incident; CAQ, catch-all question; DS, denial of the specific; GA, gentle assumption; S, suicidal plan; SA, symptom amplification. Reprinted with permission from Shea.8
It is also critical to remember that, long before one begins employing the CASE Approach in any given interview, the preceding minutes of that interview must have been done with skill and empathy. Engagement, both before the CASE Approach is performed and while it is being performed, is the single most important prerequisite for successfully uncovering suicidal ideation and intent.
One can see from Figure 2 that, in the Region of Recent Events, the basic sequence follows a simple pattern. After thoroughly exploring the first method relayed by the patient in the region of Presenting Events, the clinician uses a gentle assumption to try to elicit a second method, “What other ways have you thought of killing yourself?”
If a second method is uncovered, sequencing and fact-finding behavioral incidents are used to create another verbal video reflecting the extent of action taken. The interviewer continues this use of gentle assumptions, with the creation of a follow-up verbal video for each newly uncovered method, until the patient denies any other methods in response to the gentle assumption, “What other ways have you thought of killing yourself?”
In the situation where the patient denies all other methods but the interviewer has reason to believe that the patient may be withholding his or her method of choice or that considerable suicidal thought remains unexplored, the interviewer should ask about any suspected methods, using denials of the specific, particularly if the method is common to the patient's culture. A verbal video is made, using behavioral incidents to delineate the extent of action and intent of any subsequently uncovered new methods.
After the use of denials of the specific, the clinician can use the catch-all question if deemed to be useful (“We've been talking about a lot of different ways that you've been thinking of killing yourself. Are there any other ways that you've thought of that we haven't talked about yet?”). Some patients may inadvertently pause after being asked the catch-all question, revealing through nonverbal leakage that other plans may have been considered. A simple comment, said gently, can be surprisingly powerful at such moments as with, “Mr. Thompson, it looks like you may have thought of some other ways. I know it can be hard to talk about suicide, but I really want to help you. Try to share with me your thoughts even if they were fleeting in nature.” The resulting information is sometimes lifesaving.
After establishing which methods the patient is considering and how much action has been taken on each method, the interviewer finds out about the frequency, duration, and intensity of the suicidal ideation with a symptom amplification such as: “On your very worst days, how much time do you spend thinking about killing yourself, 10 hours a day, 14 hours a day, 18 hours a day?”
The above strategy is easy to learn and simple to remember. It also flows naturally in a conversational fashion. Patients are often pleased and surprised that they can speak freely about potentially shame-inducing material. They also recognize that the interviewer has explored such material with many others, and that he or she appears to be comfortable doing so.
Let us imagine a prototypical patient, named Sean, constructed from several emergency department patients that I interviewed when I was the medical director of the Diagnostic and Evaluation Center at Western Psychiatric Institute and Clinic at the University of Pittsburgh, where the CASE Approach was first developed. Sean is a freshman in college. He has been referred from the university counseling center. He comes from a divorced family in which his mother did not want him. He and his father adored each other and were essentially best friends. Unfortunately, his father died unexpectedly from a heart attack about 1 year ago.
When not at school Sean lives with a loving uncle, but he has become progressively depressed over time. Originally excited about college (he wants to go to medical school to make his dad proud), he learned before Thanksgiving break that he is failing calculus and freshman chemistry. He just returned to school after Thanksgiving break (where he did not share his academic problems with his uncle). It is now mid-December, and final examinations are looming.
The clinician engages Sean well. About 20 minutes into the interview, the clinician begins exploring the topic of suicide. Using the CASE Approach, the interviewer deftly explores the region of the Presenting Events. The clinician uncovers that several weeks ago, Sean had taken 25 pills of aspirin and several pills of diphenhydramine and subsequently threw up. Sean indicated that, “I pretty much wanted to die that night, but I couldn't keep the s**t down.”
The clinician effectively uncovered this material, during his exploration of the Region of the Presenting Event of the CASE Approach, by creating a verbal video of Sean's actions using fact-finding and sequencing behavioral incidents. We are picking-up the conversation at the tail end of the clinician-guided verbal video regarding Sean's overdose attempt. Sean is commenting that, after the overdose, he simply fell asleep.
Student: Nope, I just went to sleep.
Clinician: Sean, how about, let's say, over the past 2 months, have you been having any other thoughts of overdosing? (fact-finding behavioral incident, the clinician is gracefully moving into the region of Recent Suicide Events with a bridging question).
Student: Sort of.
Clinician: When was that? (fact-finding behavioral incident)
Student: I don't know, maybe about 5 weeks ago. It was right before Thanksgiving break (sighs) and I just knew it wasn't going to be like going home when my Dad was alive. Don't get me wrong, my Uncle's a good enough guy, in fact, he's been awesome, but it's not the same.
Clinician: Of course not (said gently). Dads, especially one as great as your dad sounds like he was, are not easy to replace.
Student: (looks up at clinician and nods in agreement)
Clinician: Did you get any pills out that night? (fact-finding behavioral incident)
Student: Nope, I didn't get that far (pauses) just thinking.
Clinician: What other ways have you thought about killing yourself? (gentle assumption)
Student: Oh, I don't know. I suppose I briefly thought about hanging myself, but that's pretty lame and doesn't always work.
Clinician: Have you ever gotten a rope out or something else to hang yourself? (fact-finding behavioral incident)
Student: Nope, I never seriously considered it. A kid in my high school did it. Doesn't appeal to me. Never did like just hanging around (smiles, musters a bit of humor)
Clinician: (smiles in response) What other ways have you thought about killing yourself? (gentle assumption)
Student: That's about it. Nothing else really.
Note that, at this juncture, Sean has denied any other methods when presented with a gentle assumption (“What other ways have you thought about killing yourself?”). It is the exact spot where a clinician, if suspicious that a method of choice is being withheld, can opt to use the strategy of a “Gentle Push” by employing a series of denials of the specific:
Clinician: Have you been having any thoughts, even in a fleeting way, of things like driving your car off the road? (denial of the specific)
Student: Don't own one. (long pause). I took my Uncle's car out one night though.
Clinician: Thinking of killing yourself? (fact-finding behavioral incident)
Student: I guess so.
Clinician: When was that? (fact-finding behavioral incident)
Student: Thanksgiving break.
Clinician: What happened? (fact-finding behavioral incident)
Student: Went down to a road where my Dad used to take me for a spin in his beat-up old Triumph. He loved that thing. He thought it was a Ferrari. (smiles faintly) Anyway, I was gonna run into a tree and I did, but I'd jammed on the breaks in time. Broke out a f***ing headlamp. My Uncle was really pissed. What an ass. (pauses) Not him, me.
Clinician: What do you think stopped you? (fact-finding behavioral incident)
Student: The tree. (looks serious for a moment, then cracks a smile)
Clinician: (smiles, acknowledging Sean's humor) What actually stopped you? (fact-finding behavioral incident)
Student: I guess I didn't have the guts. Gotta get me some more guts.
Clinician: (pauses, truly sensing the depth of Sean's angst) Hard to put into words.
Clinician: The pain.
Student: Yeah. (pauses) Yeah, it is. Wish it would stop.
Clinician: Have you been thinking of any other ways, like shooting yourself? (denial of the specific)
Student: Don't own a gun.
Clinician: You didn't own a car either. (this garners a smile from Sean) Does your Uncle have any guns?
Clinician: I've got a feeling he might. (said gently)
Student: I suppose so.
Clinician: Ever get one out while you were thinking of killing yourself? (fact-finding behavioral incident) (Sean looks down) It looks like you might have given that some thought. (said gently)
Student: Yeah. (pauses) I got one out once. It's the only sure thing.
Clinician: When's the last time you did that? Thanksgiving? (fact-finding behavioral incident)
Student: (said very quietly, almost mumbled) Yeah (pauses and squirms a bit in his chair) over break.
Clinician: Help me to understand a little better, Sean. It sounds like a really rough time. Where does your Uncle keep his gun . . .
Clinician: Oh, guns. Is he a hunter? (fact-finding behavioral incident)
Student: Yeah, big time. Just like my Dad.
Clinician: Sort of walk me through what happened next. (sequencing behavioral incident)
Student: Well, I went to my Uncle's gun cabinet. He keeps it unlocked. He's not real bright about this sort of thing, you know, considering how depressed I am. I love the guy, but he's not what I'd call psychologically minded, if you know what I mean.
Clinician: (nods) So what happened next? (sequencing behavioral incident)
Student: I took out one of his handguns.
Clinician: Did you load it? (fact-finding behavioral incident)
Student: Oh yeah. (nods his head yes) Yeah, I loaded it. Been thinking about what I was gonna do a lot in my head for weeks.
Clinician: What did you do then? (sequencing behavioral incident)
Student: I put it in my mouth, read somewhere on the web that that's the best way to make it a sure thing.
Clinician: Did you click the safety off? (fact-finding behavioral incident)
Student: Yeah. (looks down, lips start quivering)
Clinician: (said very gently) You really really miss him don't you?
Student: Oh God. (bursts into tears) I miss him so much. I just want to be with him for Christmas. That's all I want. That's all I f***ing want.
Clinician: It's alright to cry (hands Sean a box of tissues).
Clinician: (waits a moment for Sean to collect himself) How long do you think you were holding the gun that night? (fact-finding behavioral incident)
Student: I don't know. (pauses) Seemed like forever. Probably about a half hour or so. Long time.
Clinician: Where's the gun now? (fact-finding behavioral incident)
Student: Don't know for sure, but I put it back in the cabinet.
Clinician: Sean, on your very worst days, like over Thanksgiving, how much time are you spending thinking about killing yourself, 70% of your waking hours, 80%, 90%? (symptom amplification)
Student: You want to know the truth?… I can't get it out of my f***ing head.
Sean's originally stated intent, uncovered during the clinician's exploration of the Region of Presenting Events, was much lower and dangerously misleading when compared to his true level of intent, which was uncovered in the exploration of the Region of Recent Events. In addition, the removal of the guns from his uncle's house before Sean returns home for Christmas may prove to be lifesaving.
The interviewer's ongoing “read” on how dangerous the patient is serves as the basis for how extensively to ask questions during the Region of Recent Events. If a patient has low-risk factors, denies thoughts of suicide during the exploration of presenting events, and reports only one fleeting thought of shooting himself (no gun at home) during the first part of the Recent Events, a clinician most likely would not use denials of the specific, the catch-all question, or symptom amplification. There would be no reason to use such questions, and their use might appear puzzling to the patient. The CASE Approach is always flexibly implemented, matching the best techniques to the unique needs of the patient and the clinical presentation.