Strategies for Understanding and Assessing Suicide Risk in Psychotherapy

by: The American Psychotherapy Association

This 1-credit continuing education opportunity is co-sponsored by the American College of Forensic Examiners International (ACFEI) and the American Psychotherapy Association. ACFEI maintains responsibility for all continuing education accreditations. This article is approved by the following for 1 continuing education credit:

APA provides this continuing education credit for Diplomates.

The American College of Forensic Examiners International is an NBCC Approved Continuing Education Provider (ACEP) and may offer NBCC approved clock hours for events that meet NBCC requirements. The ACEP solely is responsible for all aspects of the program. Provider #5812.

The American College of Forensic Examiners International is approved by the American Psychological Association to sponsor continuing education for psychologists. ACFEI maintains responsibility for this program and its content.

The American College of Forensic Examiners International is an approved provider of the California Board of Behavioral Sciences, approval PCE 1896. Course meets the qualifications for 1 hour of continuing education credit for MFTs and/or LCSWs as required by the California Board of Behavioral Sciences.

This organization, American College of Forensic Examiners International Approval Number 1052, is approved as a provider for continuing education by the Association of Social Work Boards 400 South Ridge Parkway, Suite B, Culpeper, VA 22701. www.aswb.org. ASWB Approval Period: 9/13/2004 to 9/13/2007. Social workers should contact their regulatory board to determine course approval. Social workers will receive 1 continuing education clock hours in participating in this course.

Suicide is one of the few topics that almost uniformly triggers anxiety and apprehension in clinicians, both novice students and seasoned practitioners (Rudd, 2006). Moreover, the actual assessment and subsequent treatment plans for suicidal clients are perhaps the most challenging clinical endeavors mental health practitioners may face during their careers. Literature shows that this is often the case because one concrete outcome of negligence in this area is a client fatality and resultant liability for the clinician (Jobes, 2006; Jobes & Drozd, 2004; Packman, Marlitt, Bongar, & O’Connor-Pennuto, 2004; Peruzzi & Bongar, 1994). Perhaps this explains one reason why psychotherapists seem to focus on collecting data surrounding lethality and risk factors instead of exploring the narrative story of the suicidal client (Rogers & Soyka, 2004). As Schwartz and Rogers (2004) explain, psychotherapists should realize that although they will be unable to successfully prevent all instances of suicide due to the unpredictability of human nature, clinicians can lessen the number of completed suicides by being able to better identify at-risk populations and common themes of suicidality. However, clinicians should also remember not to omit a thorough exploration of the individual meanings of suicidality for a particular client.

Approximately 71% of psychotherapists report managing at least one client who has attempted suicide, with 28% reporting having had at least one client die by suicide (Rogers, Gueulette, Abbey-Hines, Carney, & Werth, 2001). For psychotherapists, the psychological impact of losing a client by suicide is similar to the stress and trauma that would be experienced in the death of a loved one (Chemtob, Hamada, Bauer, Torigoe, & Kinney, 1988). Therefore, updated information surrounding suicide risk factors, myths, assessment strategies, treatment options, and additional resources are crucial when working with this particularly challenging population.

Information on Suicide Risk Factors

Suicide takes the lives of over 30,000 Americans every year according to the Centers for Disease Control and Prevention’s (CDC) Fatal Injury Report, making it the eighth leading cause of death for males and the 19th leading cause of death for females (Centers for Disease Control [CDC], 2006). Overall, in the United States, suicides outnumber homicides 3:2 and take the lives of twice as many persons as HIV/AIDS (CDC, 2006). Over the last century, researchers have tried to produce a set of “risk factors,” which, when identified, would label a client as someone who might take his or her own life (Maris, Berman, & Silverman, 2000). The underlying premise is that if there were a way to predict suicidal behavior, lives would be saved (Maris et al., 2000). Unfortunately, research has not yielded a specific flow chart that all clinicians can follow when working with a suicidal individual. In fact, according to research by Plutchik (1995), 41 factors correlate with the risk of completed suicide. Although no one person could either remember or assess all potential risk factors, below are some of the most common ones discussed in the literature.

First, suicide among young people between the ages of 15 and 24 ranks as the third leading cause of death (National Center for Health Statistics, 2002). This represents 7.9 deaths per 100,000 persons, with a male-to-female ratio of 3:1. Between the ages of 20–24, suicide claims the lives of 12 persons per 100,000, with a male-to-female ratio of 7:1 (National Institute of Mental Health, 2001). Within the last few decades, teen suicides have steadily been on the rise. In fact, in a study conducted of high school age students, as many as 15% have made at least one suicide attempt (King, 1997), with teen girls being particularly vulnerable (Lewinsohn, Rohde, Seeley, & Baldwin, 2001). Teens who suffer from depression and substance abuse are at a higher risk, and both of these factors are on the rise (Gould & Kramer, 2001). Perhaps teens are more at risk due to their lack of financial and social resources, lack of emotional self-control, poorer problem-solving capacity, and lack of mobility (Reynolds & Mazza, 1994).

Currently, the age group considered most at risk contains white males who are over the age of 65 (CDC, 2006). Suicide among the elderly represents 14.6 deaths per 100,000 persons, a highly at-risk age group that is often under-assessed by mental health professionals. Particularly distressing is that 75% of the elderly use a gun of some sort, leaving a significantly reduced margin for failed attempts (Frierson & Melikian, 2002). It has been speculated that at this age, the elderly are struggling with physical and mental depreciation, as well as with the loss of friends and family members, leading to a mild or moderate depression that they may never have experienced before. Therefore, their coping strategies may be inadequate (CDC, 2006; Maris et al., 2000).

Although age is considered one important risk factor to evaluate, gender also provides information regarding the plausibility of a client attempting and/or completing suicide. As stated above, suicide is the eighth leading cause of death for males and the 19th leading cause of death for females (CDC, 2006). Subsequently, there are four male-completed suicides for every one female-completed suicide, but there are three female-attempted suicides for every one male attempt (CDC, 2006). Simply stated, more men complete suicide, while more women attempt it.

In addition to age and gender, people suffering from a mental illness (e.g., DSM diagnosis) are another at-risk group, accounting for an estimated 95% of all completed suicides (Shea, 2002). One of the most reliable predictors of suicidality is current, severe, depressive symptoms. In fact, the risk of suicide in clients with Major Depressive Disorder is approximately 20 times that of the general population (American Association of Suicidality [AAS], 2005). Research shows that seven out of every 100 men and one out of every 100 women who have had clinical depression at some point in their lifetime will go on to complete suicide (AAS, 2005). Although depression is a primary risk factor, a diagnosis of schizophrenia, bipolar disorder, and severe borderline personality disorder are also considered to put an individual at risk for completing suicide (Maris et al., 2000; Schwartz & Rogers, 2004; Shea, 2002).

It is important to monitor clients with mental illness under a psychiatrist’s care, as well as those currently in psychotherapy. Between 50% to 67% of individuals completing suicide had seen a doctor less than one month prior, between 10%–40% saw a doctor in the week preceding death, and in over half of suicides via overdose, the prescription had been either written, or refilled a week prior to the overdose (U.S. Preventive Services Task Force, 1996). Therefore, clients taking psychotropic medications should be monitored closely. Also, clients who are actively engaging in substance use and abuse are more likely to complete suicide due to the exacerbation of other environmental problems, as well as lowered inhibition when making a suicide attempt (Maris et al., 2000; Westefeld et al., 2000). Moreover, clients who are dependent on substances often have a number of supplementary risk factors for suicide (i.e., depression, engagement in high risk/self-injurious behaviors, or financial problems), which should be assessed by psychotherapists (Jobes, 2006; Shea, 2002).

Clients who are coping with chronic illness or chronic pain may be unable to imagine the possibility of change or progress in their struggle and may look to suicide as a way of absolving themselves of being “stuck” (Reeves, Bowl, Wheeler, & Guthrie, 2004). Moreover, these clients may be psychologically overwhelmed to the point that they can no longer cope with their current suffering, nor find a means of relief from it (Jobes, 2006; Shneidman, 1993; Schwartz & Rogers, 2004). Finally, the three most critical at-risk factors for suicide assessment are the number and severity of previous attempts, a family history of suicide, and current suicidal ideation (Jobes, 2006; Maris et al., 2000; Peruzzi & Bongar, 1994; Rogers & Soyka, 2004). Consequently, research by Packman, Marlitt, Bongar, and O’Connor-Pennuto (2004) found that multiple attempters possessed a greater overall baseline risk, indicating that suicide attempts increase the overall vulnerability for future suicide completion. Moreover, a familial pattern of suicidal behavior is considered an amplifier of risk through genetic and temperament influences and possible behavioral modeling (Packman et al., 2004).

Myths About Suicide and the Psychotherapy Relationship

There are several myths surrounding suicide that may inadvertently influence a clinician’s ability to accurately assess a client’s lethality. Most importantly, individuals in general (and some clinicians, as well) often believe that discussing suicide may directly lead to increased suicide risk. As Schwartz and Singer (2005) point out, clients kill themselves because they decide to, not because it was discussed in a psychotherapy session. In fact, there are data suggesting that psychotherapists rarely explore with their clients past experiences with suicidal thoughts or attempts (Rogers & Soyka, 2004). This clinical pattern may serve the purpose of helping clinicians to “feel better” while unwittingly contaminating the suicide assessment process (Schwartz & Singer, 2005).

Other common misperceptions about suicide are that suicide is an “irrational” act, or that suicidal behaviors are always “impulsive” acts, that children and elderly may be at risk but do not actually complete suicide, and that people who commit suicide usually do not actively seek help beforehand (Peruzzi & Bongar, 1994; Schwartz & Rogers, 2004; Schwartz & Singer, 2005; Wingate, Joiner, Walker, Rudd, & Jobes, 2004). However, a review of 71 completed suicides showed that more than half of the victims communicated their suicidal ideation within 3 months before the fatal attempt (Isometas et al., 1994). A final myth that should be noted is that people whose suicide attempts have failed really were not seriously contemplating suicide. That is, these clients were only looking for sympathy or attention (Segal, 2000). Unfortunately, research has shown that 40% of all suicide victims (i.e., those who completed suicide) made previous attempts or threats, and as the number of attempts increases, so does the likelihood that a future attempt will be fatal (Goldstein, Black, Nasrallah, & Winokur, 1991). In fact, all of the myths described above have been disputed both by clinical reports, as well as empirical research findings. Even though these suicide-related myths abound in American popular culture, it is crucial that clinicians do not succumb to their damaging influence. For various reasons—discomfort with the suicide assessment process, fears of client vulnerability and suicidality, clinician countertransference (perhaps one’s friend or relative attempted or completed suicide)—psychotherapists are at risk of not hearing clients’ calls for help. Clinicians should be aware of the myths outlined above because by increasing their understanding of what is, and is not, linked to suicidality, psychotherapists can remain open and objective during the assessment process (Schwartz & Rogers, 2004).

It is vital that psychotherapists listen intently to what clients mean behind what they say, objectively and empathically, in order to fully engage clients in a thorough suicide assessment (Schwartz & Singer, 2005). Jobes (2006, p. 7) observes that given what we currently know about people who commit suicide, there are “three essential truisms” for clinicians to note:

1) Most suicidal people do not want to end to their biological existence; rather, they want an end to their psychological pain and suffering.

2) Most suicidal people tell others (including mental health professionals) that they are thinking about suicide as a compelling option for coping with their pain.

3) Most suicidal people have psychological problems, social problems, and poor methods for coping with pain—all things that mental health professionals are usually well trained to tackle.

Strategies for Suicide Assessment

Despite the fact that several useful surveys and questionnaires are available to help clinicians evaluate suicide risk, a face-to-face clinician/client interview is thought to be both preferential and necessary to the assessment process (Reeves, Bowl, Wheeler, & Guthrie, 2004). Whether this interview is done from a crisis intervention framework, a cognitive framework, an existential-constructionist framework, or a collaborative framework, a face-to-face thorough assessment remains the only valid method for determining risk (O’Connor, Warby, Raphael, and Vassallo, 2004). The psychotherapy relationship therefore becomes the pivotal pathway for clinicians to access clients’ lethality. In order to accomplish this task, it is the responsibility of the psychotherapist to maintain an awareness of current information on suicide risk assessment practices (Westfeld et al., 2000). In this regard, the crisis interview method utilizing Shea’s validity techniques (2002), the Collaborative Assessment and Management of Suicidality (CAMS) model (Jobes, 2006), and the Aeschi Group’s Guidelines for Clinicians will be examined below.

One of the first things a clinician must be willing to participate in is a self-inventory for the identification of biases regarding suicide as an act. This self-reflection can determine whether an intervention will be a success or a failure (Shea, 2002). Self-exploration is not a static awareness, but on ongoing process (Shea, 2002). Attitudes can range vastly from “suicide is wrong” to “suicide has intrinsic positive worth” (Shea, 2002). Suicide is a difficult topic for discussion, even for the experienced therapist. It is for this reason that the therapist should be aware and keep track of his or her values and ongoing emotional experiences. Counter-transference is one phenomenon the psychotherapist should be continually checking in with, as this can create a power struggle between client and therapist. For example, Maris, Berman, and Silverman posit that suicidal clients can actually be “help-rejecting” as well as engaging in a wide variety of “interpersonally alienating behaviors” (p. 513), which may create negative counter-transference.

There are many schools of thought on how to assess a suicidal individual. One such assessment is the crisis interview wherein the psychotherapist directly asks questions regarding suicidality (e.g., ideation, intent, plan, means of completion). During this process, a helpful hint is to use very specific and concrete wording such as “kill yourself” or “commit suicide” versus general “softer” words such as “stop the pain” (Shea, 2002). The client needs to know that the psychotherapist can handle their thoughts surrounding taking their own life, as many clients do not have anyone else with whom to discuss these confusing thoughts.

Shea (2002) offers several other points to keep in mind when assessing a client’s lethality. First, the slightest hesitancy in a client’s response may suggest that he or she has thought about suicidal ideation (even if they deny it). Next, answers such as “no, not really” when clients are questioned about suicidal ideation usually means there have been at least some suicidal thinking. Clinicians should also try to be as present with the client as possible to pick up on any non-verbal cues he or she may be sending. For this reason, it may be beneficial for clinicians not to take notes (or to do so sparingly) during the suicide assessment, so they may be 100% available to the client during the process. Clinicians should routinely check themselves during the interview, asking “What am I feeling right now?” and “Is there any part of me that doesn’t want to hear the truth right now?” These simple preparations can help guide the techniques the clinician will use when eliciting suicidal intent.

In The Practical Art of Suicide Assessment, Shea (2002) discusses six validity techniques that clinicians can utilize to explore sensitive material with a client. These can be used with a variety of sensitive topics, such as domestic violence, substance abuse, antisocial behavior, sexual abuse, and suicide. The first validity technique, the behavioral incident, is when the clinician asks about concrete behavioral facts. Questions like, “Exactly how many pills did you take?” provide the facts of the incident. The next technique is shame attenuation, which relates to the therapists’ ability to inquire about information without making the client feel shame or guilt. Instead of asking the client, “Do you have a bad temper and tend to pick fights?” the clinician could ask, “Do you find people tend to pick fights with you when you are out trying to have a good time?” Or, “Some people have told me that when they get angry they tend to pick fights, has that happened with you?”

The next technique is designed to help increase the chances the client will be open with sensitive information. Gentle assumption is a technique that proposes that the behavior is already happening. Instead of asking, “Do you drink?” The therapist can ask, “How much do you drink?” In the case of potential suicidality, if the client is severely depressed the clinician may ask, “During the past two weeks how difficult has it been to not think about taking your own life?” This technique helps clients bypass the psychological hurdle of admitting to problem behaviors in the first place.

The technique symptom amplification uses the client’s natural tendency to minimize or downplay quantitative information about problem behaviors. By setting the upper limits of the quantity higher than average during questioning, the client has “room to move” while being more truthful about the actual number. For example, rather than asking, “Have you had thoughts of suicide during the past week?” the therapist could ask, “How many times has the thought of suicide entered your mind during the past week, fifteen or twenty?” This allows the client to ease his or her natural defense mechanisms and avoid confrontation. The question may be particularly effective after a gentle assumption (see above) has already exposed suicidality.

The technique denial of the specific involves asking the client specific questions versus generic or global questions. The rationale is that it is easier to deny a generic question than a specific one. If trying to assess the use of drugs a clinician might ask, “Have you ever tried cocaine?” or, “Have you ever smoked crack?” or, “Have you ever used crystal meth?” or, “Have you ever dropped acid?” rather than, “Do you use illegal drugs?” Regarding suicidality, when assessing a plan after suicidal ideation and/or intent has been revealed, the clinician may ask, “Have you thought about overdosing on your medication?” and, “Have you thought about taking your life by hanging?” and, “Have you considered using a gun to take your life?”

The last validity technique Shea (2002) offers is normalization. By normalizing their problem behavior, the client may not feel as embarrassed or anxious when discussing it. For example, regarding depressive symptoms, the therapist may ask, “Sometimes when people are depressed they will have a decrease in their sex drive . . . has this happened to you?” When assessing suicidality, a therapist might ask, “Many times when people are sad and ‘in the dumps’ as you have described yourself, they say the thought of wanting to die comes into their minds . . . has this thought surfaced for you?” Letting people know they aren’t the only ones to experience the behavior allows them to feel less anxious about it and free to share it with the interviewer.

When completing an assessment of a potentially suicidal client, the clinician must be aware of the most important information needed from the client: mainly, the client’s current level of suicidal ideation, suicidal intentions, whether a plan for action has been considered, and what access the client has for the means of completion (O’Connor et al., 2004; Packman et al., 2004; Schwartz & Rogers, 2004; Shea, 2002; Wingate et al., 2004). As the amount of information from these four areas increases, so does the probability that the client may be truly at risk. For example, if suicidal ideation is present, the clinician should evaluate how often these thoughts are occurring, how long the thoughts have been present, whether or not the thoughts have become more intense over time, and how difficult is it for the client to keep from acting upon these thoughts (Schwartz & Rogers, 2004). Another clinically important area would be to determine whether or not the client has a specific plan to harm him or herself. If a plan exists, the clinician would need to determine how well developed the plan is and whether the client has the means accessible to complete the plan. Not only will this exploration of ideation help to determine the lethality of the client, but it will also provide direct suggestions for setting up a safety plan.

Lethality is a function not only of risk factors, but also of whether or not protective factors are present (Maris et al., 2000). Below are some general guidelines provided by Schwartz and Rogers (2004) that may be helpful in determining the lethality of a client who acknowledges suicidal ideation:

* Low lethality—suicidal ideation is present but intent is denied, client does not have a concrete plan, and has never attempted suicide in the past.

* Moderate lethality—more than one general risk factor for suicide is present, suicidal ideation and intent are present but a clear plan is denied, and the client is motivated to improve his/her psychological state if possible.

* High lethality—several general risk factors for suicide are present, client has verbalized suicidal ideation and intent, has a coherent plan to harm him or herself, and reports access to resources needed to complete the plan.

* Very high lethality—client verbalizes suicidal ideation and intent, he or she has communicated a well thought out plan with immediate access to resources needed to complete the plan, demonstrates cognitive rigidity and hopelessness for the future, denies any available social support, and has made previous suicide attempts in the past.

Although suicide involves a complex range of behaviors, thoughts, and affective states, the evaluation of concrete suicide markers (i.e., ideation, intent, planning, and means) may increase a clinician’s success in predicting a client’s overall lethality (Schwartz & Rogers, 2004; Shea, 2002). However, as O’Connor et al. (2004) state, it is important to realize that “every clinician lives with the knowledge that even with our best efforts and exemplary care, there will still be some suicide deaths” (p. 359).

Another assessment approach that has been gaining popularity is an inclusive or “team-building” approach called the Collaborative Assessment and Management of Suicidality (CAMS), created from the research of David Jobes and associates. The main focus and uniqueness of this assessment model is that it targets the client’s subjective suicidality as the central clinical problem, independent of objective diagnosis (Jobes, 2006). In addition, by utilizing the Suicide Status Form (SSF), both the clinician and client develop a shared understanding of the client’s suicidality by rating the client’s current psychological pain, press (stress), perturbation (agitation), hopelessness, and poor self-regard (self-hate) (Jobes, 2006; Jobes & Drozd, 2004). With the CAMS model, even the traditional face-to-face seating is changed once suicide is mentioned. The clinician asks for permission to sit side-by-side the client while filling out the SSF in order to facilitate a more collaborative feeling (Jobes, 2006; Jobes & Drozd, 2004).

In addition to ranking risk-related characteristics, the CAMS model also helps the client to identify reasons for living as well as reasons for dying. In doing this, the clinician receives a glimpse at some of the protective factors that have kept the client from taking his or her life up to this point. As Jobes & Drozd (2004) profess, it is our job as clinicians to help suicidal individuals find alternative ways of coping with the unbearable pain and stress in their lives in order to alleviate suicide as a viable option. Finding a common ground and being able to agree on mutual goals increases collaboration immensely (Ellis, 2004).

The CAMS model incorporates its own documentation throughout each of the stages. In this model, the SSF has 4 sections:

Section A: This initial section is completed collaboratively in order to extract a true understanding of the meaning the world has for the client currently.

Section B: This section is completed by the clinician who asks specific questions regarding plan, preparation, rehearsal, history of suicidality, and so on.

Section C: This section is completed collaboratively and explicitly states what the outpatient treatment plan will be.

Section D: This section is completed by the clinician post-session and includes a mental status exam, preliminary diagnosis, and the client’s overall suicide risk level. Also, this section provides a place for the clinicians to jot down any additional information not otherwise covered in sections A-D.

There is a place at the bottom of each section for the client and clinician’s signature and date. This aspect of the model also reinforces collaboration during the assessment process, because the information collected is reviewed and agreed to by both clinician and client. This same process would be completed each session until there were three consecutive sessions without suicidal ideations (Jobes, 2006). As Jobes (2006) states, “I truly believe that through collaboration all things are possible, not the least of which is coaxing a life to be meaningfully lived back from the jaws of suicidal death” (p. 137)

By Michelle E. Toth, MA; Robert C. Schwartz, PhD; and Sandy T. Kurka, MA