Assignment

Styles of Suicide Intervention: Professionals’ Responses and Clients’ Preferences

Jill C. Thomas and Larry M. Leitner Department of Psychology

Miami University, Oxford, OH

As the rates of suicide in America continue to rise, suicide recently has been declared to be a national public health concern. The crisis intervention model, which has dom- inated the treatment of suicidal individuals in America since the 1950s, is currently believed to be the most effective model for suicide intervention. This study examined this belief by taking a more complex look at professionals’ responses to suicidal cli- ents both by investigating the existence of different ways in which professionals in- terpret the crisis intervention model. In one interpretation (the “fight” response), the professional takes power and agency away from the client and does what is perceived to be “best” for the client. An alternate interpretation, the “ideal” response, allows for a respectful engagement with the client. Professionals also can act contrary to the model (i.e., the “flight” response). In addition, based on the humanistic notion that clients are the experts of their own experience and that their voices are a very valu- able part of evaluating the treatment process, this study investigated which response style clients report to be most helpful and most desired. The results suggest that while the typical response of mental health professionals to suicidal clients is most charac- teristic of the “fight” response style, clients overwhelmingly report that the contrast- ing “ideal” response style is most helpful. The findings are discussed along with im- plications for practice, research, and training.

The most recent statistics reveal that there are approximately 30,000 suicides in America every year. There are about 80 suicides and 1,500 attempted suicides in America each day: one suicide every 18 min and one attempt each minute. Suicide is the eleventh leading cause of death in the United States and the third leading

THE HUMANISTIC PSYCHOLOGIST, 33(2), 145–165 Copyright © 2005, Lawrence Erlbaum Associates, Inc.

Requests for reprints should be sent to Jill C. Thomas, Department of Psychology, Miami Univer- sity, Oxford, Ohio 45056. E-mail: [email protected].

cause of death among American youth ages 15 to 24 (American Foundation for Suicide Prevention, n.d.). For every suicide it is estimated that there are six friends, family, significant others, or loved ones the suicide has left behind. Based on this estimate and the most recent suicide statistics, there are now at least 4.5 million American survivors of suicide (Caruso, n.d.).

With these alarming statistics, it is no wonder that the response in the suicide prevention movement has historically been an active and aggressive fight to pre- vent the spread of this epidemic (U.S. Department of Health and Human Services, 2001; U.S. Public Health Service, 1999). Today’s approach to suicide prevention and intervention is rooted in the theoretical framework established in 1958 with the inception of the first suicide prevention center (the Los Angeles Suicide Preven- tion Center; Allen, 1984). This framework, known as the crisis intervention model, prescribes a standard protocol for responding to the suicidal client. After quickly establishing rapport with the client and developing an understanding of the prob- lem, the crisis intervention worker must develop options and take action, using all measures possible. These measures include: reducing immediate danger with the use of no-suicide contracts, helping to make the client’s environment safe, evaluat- ing the client’s need for medication, involving family or significant others in the in- tervention, linking the client to other community resources, helping the client to structure his or her time, and possibly voluntary or involuntary hospitalization (Fremouw, de Perczel, & Ellis, 1990).

Although the crisis intervention model explains clearly and concretely what steps are to be taken and what actions are needed, it offers little in the way of prescribing the manner in which these steps and actions are to be carried out, thus allowing for multiple possible interpretations of the model. The typical interpretation of the model is one in which the crisis interventionist is assertive, active, and aggressive. Leenaars (1994), for example, likens the activity of a crisis intervention worker to that of a cardiologist in an emergency unit. This metaphor is consistent with the be- lief that the role of the crisis interventionist is to take charge of the situation and take over for the person, who is rendered weak and/or helpless in the face of the crisis, to fix the problem or repair the person. This metaphor is consistent with the depiction of the crisis interventionist in most of the literature and results in one prototypical pro- fessional response to suicidal clients (i.e., the following “fight” response). However, as stated previously, the crisis intervention framework does offer room for other pro- cedural interpretations (i.e., the following “ideal” response).

THIS STUDY

As Maris, Berman, and Silverman (2000) have reported, “research on treatment ef- fectiveness [for suicidal people] is both difficult and rare” (p. 527). Methodological issues in the study of suicide, like confounding variables and the ethical prohibitions

146 THOMAS AND LEITNER

to using a control group in testing treatment efficacy (Berman, 2000; Frankish, 1994; Neimeyer & Pfeiffer, 1994a), make it difficult to empirically validate the crisis inter- vention approach. As such there is little empirical evidence for what works with sui- cidal clients and almost a total lack of evidence that the crisis intervention model (typically operationalized consistent with the cardiologist interpretation) is the most effective or is even effective at all (Berman, 2000; Frankish, 1994). Despite the lack of evidence, the crisis intervention model is currently believed to be the best option for the treatment of suicidal individuals, and, as such, professionals continue to teach it, and to evaluate the skills of their volunteers based on their ability to work within it (Neimeyer & Bonnelle, 1997; Neimeyer & Pfeiffer, 1994a, 1994b).

This two-part study was conducted in an effort to examine the belief that the cri- sis intervention model is the ideal treatment model for suicidal clients. Part I of the study takes a more complex look at professionals’ responses to suicidal clients both by investigating the existence of different interpretations of the crisis inter- vention model as well as the existence of ways professionals act contrary to the model. This portion of the study attempted to understand these varying interven- tion styles through interviews with community mental health professionals about the ways they have responded to suicidal clients in the past.

Part II of the study evaluated the belief that the crisis intervention model is the ideal treatment model for suicidal clients by examining which response style sui- cidal clients report through interviews to be most helpful and most desired. This portion of the study was based in the humanistic notion that clients are the experts of their own experience and therefore that their voices are a valuable part of evalu- ating the treatment process. While acknowledging the methodological difficulties inherent in suicide research, this study, by looking at the issue from the suicidal cli- ent’s perspective, offers one viable way of understanding the therapeutic and prac- tical implications of different response styles.

The data gathered in the two parts of the study were compared in an effort to an- swer the main research question: Do the responses that professionals offer suicidal clients match with what clients report to be most helpful? Given the startling sui- cide statistics and the rising rate of suicide in this country, the overall study hypoth- esis was that a mismatch exists between the response styles professionals are offer- ing and the type of response that suicidal clients desire or find most helpful.

Styles of Suicide Intervention

This examination was conducted based on a literature review that suggested the ex- istence of two problematic styles of response (“fight” and “flight”) along with an alternate more “ideal” style. Those, for example Szasz (1980, 1986), who describe the problematic response types, launch a harsh attack on many mental health pro- fessionals who intervene with suicidal clients. However, although these arguments are valuable in pointing out responses that may not be helpful to the suicidal indi-

STYLES OF SUICIDE INTERVENTION 147

vidual, they are not presented here as an attack. Most mental health professionals respond to suicidal clients in the best way they know how. The problematic “fight” and “flight” responses described in the literature are understandable responses to a very difficult situation.

“Fight.” The popular “cardiologist” interpretation of the crisis intervention model described earlier is consistent with the assertive, active, and aggressive “fight” response style, which takes power, agency, and control from the client without much forethought or discussion. (Table 1 describes the “fight” response.) It is important to make clear that the “fight” is not inherent in the actions required by the model but rather is evident in the manner in which the actions are taken. Due to the importance of the goal of suicide prevention, the professional takes the mea- sures suggested by the crisis intervention model against the client’s will. In this case, and even in other less extreme cases, although the measures themselves are not inherently “fighty,” they are reflective of the problematic “fight” response be- cause they come primarily out of the needs of the professional without first talking with, listening to, and considering the needs of the client.

For Szasz (1986), the “fight” style of suicide intervention results in a rapid abandonment of the therapeutic alliance and “coercive intervention” in which any and all means necessary to prevent the suicide are quickly taken against the client’s will. This response can even incorporate the use of fraud and force and results in a struggle for power between client and helper (Szasz, 1980). The client in this struggle wants control over his or her life. The professional, although claiming to want to help this person by the use of involuntary hospitalization, police involve-

148 THOMAS AND LEITNER

TABLE 1 Description of the “Fight” Response Type

• Incorporates the use of fraud and force (actions against the client’s will made under the false pretense that the motivation behind the action is for the client’s sake).

• May result in a struggle for power between client and helper. • May use involuntary hospitalization. • May use police involvement. • May use forced medication or shock treatments. • May advance the client’s feelings of hopelessness and despair and/or aggravate the client’s

problems. • May point out the immorality of committing suicide. • May react in a hostile or angry manner toward the suicidal individual. • May involve client’s family and/or develop another support network for the client in the

community, possibly against the client’s wishes. • May contact the client’s family physician to alert him or her to the possibility of an overdose on

prescription medications, possibly against the client’s wishes. • May aid in structuring the client’s time between sessions. • Actions come out of the needs of the professional without first talking with, listening to, and

considering the desires of the client. • Responds in an active and directive way.

ment, forced medication, or shock treatments, essentially deprives this person of basic human liberty as a means of, “gain[ing] control over the patient’s life to save himself from having to confront his doubts about the value of his own life” (Szasz, 1980, p.191). Likewise, Jobes and Maltsberger (1995) state that this “fight” re- sponse is due to the “strong countertransference wish to do something active, pow- erful, and healing so that the therapist will not have to endure the empathic pain of experiencing the patient’s despair” (p. 205). The descriptions offered by Grollman (1971) of the “helper as a moralizer” and the “angry helper” also fit the “fight” re- sponse type, and he, along with Szasz (1980), asserts that this response may actu- ally advance the client’s feelings of hopelessness and despair and aggravate the cli- ent’s problems rather than help the situation

“Flight.” Szasz (1980) elaborates the other problematic response, the “flight” response, by describing professionals who respond in this way as those who, when confronted with the issue of suicide “run for their lives” (p. 191; See Table 2). As Grollman (1971) states, suicide is

ugly for onlookers, devastating for relationships, and harrowing even for those pro- fessionally involved. So the entire subject is often studiously avoided, even when a person threatens to take his own life. Some just do not want to become entangled in the sordid predicament (p. 87).

Jobes and Maltsberger (1995) explain this type of response as being due to the thera- pist’s inability to engage with the client’s intense suffering, and say that, as a result of this inability, the therapist remains detached and “professional.” As such, they may rush through or avoid the topic of suicide altogether or even refuse to see the client.

STYLES OF SUICIDE INTERVENTION 149

TABLE 2 Description of the “Flight” Response Type

• Those who, when confronted with the issue of suicide, “run for their lives.” • May avoid the subject of suicide, even when a person threatens to take his own life. • May communicate disinterest to the suicidal client. • May feel concern about the personal ramifications of uncovering serious suicidal ideation,

including extra expenditure of time and energy. • May not inquire about suicide or may wait until the end of the interview to ask about it. • May rush through the assessment of suicidal intent, thus not creating an environment where the

client feels comfortable to share suicidal thoughts or feelings. • May ask leading questions which invite a negative response (e.g., “you’re not thinking of hurting

yourself are you?”). • May take the first “no” for an answer rather than probing further about suicidal ideation. • May take a very passive role in suicide intervention, does not respond in an active and directive

way. • May divert discussion away from powerful emotion to a more intellectualized or abstract

exchange. • May offer superficial reassurance that everything will be okay.

The “flight” response to suicide is problematic because it communicates disin- terest to the suicidal client and confirms for the client his or her belief that no one cares. Shea (1998) helps to develop a clearer picture of how this avoidance might actually look in the interaction between client and therapist. He feels that helpers are often concerned that if they uncover serious suicidal ideation, they are opening themselves up to a messy situation that will probably be very time and energy con- suming. As a result, he says that, to avoid this mess, some clinicians may not even inquire about suicide or may wait until the end of the interview to ask about it. He suggests that they may also rush through the assessment of suicidal intent, thus not creating an environment in which the client feels comfortable enough to share sui- cidal thoughts or feelings. In addition, the clinician may ask leading questions that invite a negative response (e.g., “You’re not thinking of hurting yourself are you?”) and then will take the first “no” for an answer. Shea (1998) states, “the clinician should seldom, if ever, leave the topic after a single denial” (p. 468) because many suicidal people will often deny these feelings when first asked.

TONY’S TALE

Both the “fight” and “flight” response types are illustrated by the response a cli- ent, Tony, described when interviewed by the researcher. Tony, who says the worst thing a mental health professional can do when interacting with a suicidal client is to “brush him off,” says that is exactly what happened to him. Tony re- cently called his therapist because he was feeling really down and was thinking of suicide. Tony says that when he was talking to his therapist, he felt like his therapist was not concerned about his suicidal thoughts. Tony says, “I felt like he thought I was making it up, but I wasn’t. I was thinking about suicide. All he’s concerned about is how many clients he can get and making money.” Tony says that the therapist told him that suicide was wrong and that, if he were to go through with it, his soul would be in danger of “hell and damnation.” Tony re- ports that the therapist sounded anxious at times and tried to rush him off of the phone by telling him things like, “We’ll talk about it next session.” Disgusted, Tony could not believe this response because, as he said, “you know, he didn’t know if I was going to commit suicide or not after that!” Tony also felt very let down. He now questions if he would call his therapist again if he were feeling suicidal. He says he would most likely call someone else.

Tony reported that he also tried to talk to his psychiatrist about his feelings. He was shocked and angry when the response he received was, in his words,

I’m a psychiatrist. I only dispense medicine. Your therapist is the one you should be talking to.’ She should not have said that. She’s supposed to be my doctor. I don’t un- derstand what that was all about. That was very confusing to me.

150 THOMAS AND LEITNER

When reaching out for help in a time of great need, Tony was confronted with pro- fessionals who did not seem to want to help him, or believe him, or hear him. For whatever reason, consistent with the “flight” response, these professionals seemed unavailable to be present for their client in suicidal crisis. Both professionals com- municated a disinterest to Tony and a desire to avoid the issue. Tony was also faced with a “fight-like” value judgment about the morality of considering an option like ending his life when his therapist pointed out the moral–religious implications of suicide. In addition, also consistent with the “fight” style, Tony was confronted with anger on the part of the professionals working with him, which he picked up in their tone of voice and manner of speaking. Tony did not find the “flighty” or “fighty” as- pects of these responses to be helpful. Instead, when asked what he finds to be the most helpful professional response to a suicidal client, Tony says that that he wants someone to, “Hear me out. Let me talk to them. Give me a little comfort. Not rush me off the phone … That’s what they’re supposed to be doing, helping.”

The Alternative “Ideal” Response: Neither “Fight” Nor “Flight.”

The response Tony described as most helpful more closely resembles the third re- sponse style examined in this study. This “ideal” response stands in contrast to both of the previously mentioned problematic responses (Table 3). Jobes and Maltsberger (1995) describe this type of response when they state that, “suicidal patients need the therapist’s genuine warmth, interest and respect … The therapeu- tic interest we refer to is agape: that unselfish, nonerotic unexploitative concern that loyally accepts others and seeks their well-being” (p. 209).

A description of this “ideal” response can also be found in the Personal Con- struct Psychology (PCP) literature. When talking about suicide, Kelly (1961) states, “instead of treating it as something evil, pathological or nonsensical, we can understand it far better if we look at the act itself and what it accomplishes from the point of view of the person who performs it” (p. 257). As such, rather than pathologizing suicidal behavior (i.e., asserting a Diagnostic and Statistical Manual of Mental Disorders [DSM] diagnosis as its cause), PCP searches for the meaning in it. At the same time, while taking this nonjudgmental and empathic stance, Kelly (1961) does not advocate for ultimate acceptance of the client’s desire to die. Rather, he states that the goal of therapy with a suicidal client is to “restore and ac- centuate those universal psychological processes that characteristically make life an ongoing proposition” (p. 277). Here, the PCP clinician and the “cardiologist” crisis interventionist share a common goal and, as such, some of the actions finally needed may be the same in both response styles. However, as the description of the “ideal” response style shows, the “ideal” responder aims to meet this goal in a qualitatively different manner than does the “fighty” or “flighty” responder.

This way of being with the client leaves the client with no doubt that the words, attitudes and actions of the therapist are genuine and based on caring and concern

STYLES OF SUICIDE INTERVENTION 151

for the client. Thus, if the client and the therapist should reach an impasse and they can no longer collaborate on the issue of suicide, and the therapist feels legally re- sponsible to take coercive actions to prevent the client’s suicide, it is more likely that the client will remember this genuine caring and, once the goal of the preserva- tion of life is met, will be able to reapproach the relationship with an understanding of the therapist’s actions.

ANN’S EXPERIENCE

Fortunately, not all suicidal clients have negative experiences when reaching out for help. Some clients report that they have been able to build relationships based on mutual trust and respect that are associated with a different kind of response from professionals than the “flighty” and “fighty” ones described by Tony. Ann,

152 THOMAS AND LEITNER

TABLE 3 Description of the “Ideal” Response Type

• Some of the actions taken may be the same as those seen in the “fight” response, but the actions are taken in a different manner, suicidal feelings and thoughts are explored with the client as a means to collaborate around actions to be taken, actions are not taken against the client’s will unless an impasse has been reached and the situation is such that the clinician has no other choice.

• The relationship established between the client and the therapist is described as a powerful, respectful collaboration.

• Suicide is looked at with the client in terms of what it accomplishes from the point of view of the client.

• The decision to commit suicide is understood and empathically embraced when the therapist is looking at the world through the client’s eyes.

• Takes a nonjudgmental stance towards suicide. • Emphasizes the need for an empathic understanding of the client and the client’s meaning of

suicide. • The therapist enters the relationship with the client using a credulous approach, in which the

therapist accepts everything the client says as the “truth” for the client and resists the urge to conform the client to the therapist’s world view.

• The invitational mode used is safe and respectful, inviting to move the discussion in certain directions, allowing clients to proceed at their own pace.

• Therapist acknowledges that some topics, like suicide, are painful and overwhelming to deal with. • Does not view suicide as a mental illness, does not offer a DSM diagnosis as the cause of suicidal

intent. • Affirms the client’s feelings but at the same time helps the client separate these valid feelings from

the link to self-destructive behaviors. • Therapist empathically resonates with the experiences of the client but still recognizes the

experiences as belonging to the client, not the therapist. • Client feels that the words, attitudes and actions of the therapist are genuine and based on caring

and concern for the client.

Note. DSM = Diagnostic and Statistical Manual of Mental Disorders.

for example, described her experience of the interventions she received as being very positive and, as such, is used here to illustrate the “ideal” response type. Re- cently, Ann called her case manager because she was having thoughts about want- ing to kill herself. Her case manager listened to her and told her that she had made the right choice to call. She told Ann that it was good to talk to people about how she was feeling. The case manager called Ann back later that day and told her that she would like Ann to meet with her psychiatrist the next morning. Ann agreed and was pleased that her case manager would arrange this for her. She perceived this as a very caring and responsive thing to do. Ann expressed that she has a very good re- lationship with her doctor. She says that not only does she trust him but “he trusts me too.” This seems evident in his response to her. Rather than make decisions for her, Ann states that she and her doctor had a discussion about her feelings and talked about some options. Her doctor asked for her opinion about what she thought would be most helpful. Ann said that she would have done whatever he told her she ought to do but she really just needed his attention and she felt like she got that.

The response Ann got from her treatment providers was very consistent with what Ann says is the best thing a mental health professional can do for a suicidal client, “to listen and be there for them… and to see the doctor as soon as possible.” Ann wanted to be heard and wanted attention paid to her suffering by the person she knew could help and the person she felt most connected to. Given the context of this relationship as compared to a relationship in which the client does not feel connected to the mental health professional, Ann’s statement about the worst thing a professional can do for a suicidal client is telling. Ann says,

Well, I could say, ‘send me to the hospital,’ but if that’s what it took, you know, it would have to be his way, not my way… the worst thing? … Send me home without giving me anything. And he didn’t send me home with nothing. He gave me options.

In the context of a mutually respectful and trusting relationship, Ann can re- spect and understand the decisions (e.g., hospitalization) that are ultimately made to protect her. Rather than feeling angry, let down, frustrated, discounted and con- fused, Ann feels respected and cared for, leaving the relationship with her treat- ment provider to grow deeper, more powerful, and ultimately more helpful.

SPECIFIC RESEARCH HYPOTHESES

Based on the literature describing the detrimental effects of two problematic re- sponses to suicidal clients (“fight” and “flight” responses), we predicted that the response clients describe as being the most therapeutic would be in direct contrast to these two modes. Specifically, we hypothesized that the “ideal” response would

STYLES OF SUICIDE INTERVENTION 153

most closely capture what clients describe, based on their experiences of being sui- cidal, as the most preferable mode of response. In contrast, given the predomi- nance in the literature and in the field of the “fight-like” interpretation of the crisis intervention model, we predicted that a “fight” style would be most characteristic of professionals’ responses to suicidal clients. Taken together, these predictions underlie the overall study hypothesis that a mismatch exists between the interven- tions professionals are offering and the responses that suicidal clients desire or re- port to be most helpful.

METHOD

Participants

The participants for this study were mental health professionals and clients re- cruited from community mental health agencies serving primarily metropolitan and suburban areas. All participants were paid for their participation. Initially, six agencies providing a wide range of services (e.g., case management, day treat- ment, individual therapy, residential programs, and other outreach programs) were contacted regarding participation in the study. Five of the six agreed to participate.

To qualify for Part I of the study, mental health professional participants must have had an adult client, for whom they had primary treatment responsibility, present with serious intent to commit suicide within the recent past (6 months was suggested but not required if memory for the event was good). Serious intent was defined as in- cluding situations in which danger was perceived as immediate as well as situations in which danger was not perceived as immediate but likely to occur in the near future. This definition did not include chronic suicidal ideation lacking imminent or im- pending danger. In total, 48 professionals (i.e., case managers, team leaders, super- visors, therapists, clinical nurse specialists, diagnosticians, and licensed psycholo- gists) volunteered for this study. (See Table 4 for sample demographics).

For Part II of the study, qualifying client participants were adults who had ex- pressed suicidal intent (as defined previously) while in treatment with a mental health professional in a community mental health system within the recent past (6 months was again suggested but not required). These clients, to receive treatment within the system, have been given a DSM diagnosis and have been classified by the state as “severely mentally disabled.” In all, 23 client volunteers were obtained. (See Table 5).

Procedure

After providing informed consent and relevant demographic information, the pri- mary researcher conducted semistructured interviews with professional and client

154 THOMAS AND LEITNER

155

TABLE 4 Demographics for All Professional Study Participantsa

Demographic Category n % Other Stats

Sex female 35 73 male 13 27

Race Caucasian–White 33 69 African-American–Black 12 25 other 3 6

Licensure unlicensed 32 67 licensed (LPC, LSW) 7 15 independently licensed (LISW, LPCC, CCDCIII, RN) 7 15 licensed psychologist 2 4

Education high school diploma, GED 1 2 Associate’s, 2-year degree 5 10 Bachelor’s degree 26 54 Master’s degree 15 31 postdoctoral education 1 2

Age M = 34 years

Years in field Range: 22–56 years Mdn = 5 years Range: .375–25 years

Note. CCDCIII = Certified Chemical Dependency Counselor III, GED = General Equivalency Diploma, LPC = Licensed Professional Counselor, LPCC = Licensed Prefessional Clinical Counselor, LISW = Licensed Independent Social Worker, LSW = Licensed Social Worker, RN = Registered Nurse.

TABLE 5 Demographics for All Client Study Participantsa

Demographic Category n % Other Stats

Sex female 18 78 male 5 22

Race Caucasian–White 18 78 African-American–Black 4 17 other 1 4

Age M = 42 years range: 18–59 years

aN = 23

participants, who were later debriefed and paid for participation. During inter- views with professionals, the researcher asked them to describe their experience of responding to a suicidal client, including some general information about the client and the client’s circumstances, as well as information about the clinician’s re- sponse style and actions. The researcher explored in depth with professionals what response they chose and, more important, in what manner they responded (i.e., what exactly did the professional say to the client, how did the client respond, how was the professional feeling during this interaction). In addition, the researcher in- vited professionals to discuss their past and present personal and professional views or beliefs about suicide as well as any previous personal or professional ex- periences with suicide. During client interviews, the researcher asked clients to talk about their experience during a time when they were suicidal, including the specific circumstances of the event as well as the response they received when they shared their intent with their clinician. The researcher also asked clients to describe what, in their experience, are the most and least helpful responses professionals can offer suicidal clients.

Once all data for each participant had been collected, three clinical research as- sistants, blind to the hypotheses and the other relevant data, reviewed the interview data. The research assistants used a literature-based description of the three re- sponse types (See Tables 1–3) to rate professional interviews for each of the re- sponse types professionals employed and client interviews for each of the response types clients indicated as helpful. Because it was supposed that many factors prob- ably simultaneously affect the responses of professionals in different ways, it was not expected that most professionals would exhibit one response type to the exclu- sion of the others. Rather, we thought that each professional’s overall response to a suicidal client would represent a mixture of the three response types (“fight,” “flight,” and “ideal”). As such, the research assistants were asked to record the level of each of the response types in each interview on a scale from 0 (no response of this type) to 10 (the strongest possible response of this type). An option of “Can- not Determine” was also given if rankings could not be made. The raters discussed each interview to come to a consensus on the ratings for each response type

RESULTS

Information obtained from 39 of the 48 professional participants was used in the final analysis. (See Table 6 for demographics of this subsample). The reduction in usable data was due to poor tape quality for some interviews and inability of raters to reach a consensus in some cases. Information obtained from 20 of the 23 client participants was used in the final analysis. (See Table 7 for demograph- ics of this subsample). One of the 23 clients who initially volunteered for the study chose to discontinue participation before completing the interview and one

156 THOMAS AND LEITNER

157

TABLE 6 Demographics for Subset of Professional Study Participants

Used in Data Analysisa

Demographic Category n % Other Stats

Sex female 27 69 male 12 31

Race Caucasian–White 28 72 African-American–Black 9 23 other 2 5

Licensure unlicensed 24 62 licensed (LPC/LSW) 7 18 independently licensed (LISW/LPCC/CCDCIII/RN) 6 15 licensed psychologist 2 5

Education high school diploma/GED 1 3 Associate’s, 2-year degree 4 10 Bachelor’s degree 20 51 Master’s degree 13 33 postdoctoral education 1 3

Age M = 34 yrs. range: 22–55 years

Years in field Mdn = 5 years range: .375–25 years

Note. CCDCIII = Certified Chemical Dependency Counselor III, GED = General Equivalency Diploma, LPC = Licensed Professional Counselor, LPCC = Licensed Prefessional Clinical Counselor, LISW = Licensed Independent Social Worker, LSW = Licensed Social Worker, RN = Registered Nurse.

an = 39.

TABLE 7 Demographics for Subset of Client Study Participants

Used in Data Analysisa

Demographic Category n Percentage Other Stats

Sex female 16 80 male 4 20

Race Caucasian–White 16 80 African-American–Black 3 15 other 1 5

Age M = 42 years range: 18–59 years

an = 20

other client completed the interview but did not meet the qualifications for the study. In addition, one interview was rated but could not be used, as raters were unable to come to a consensus based on the information provided in the inter- view. When comparing the data in Table 4 to Table 6 and Table 5 to Table 7, one can see that the demographics of the subsets used for data analysis do not differ substantially from those of the total sample from which they were taken. This suggests that the subsets of participants, both professional and client, are repre- sentative of the original samples.

Part I: Mental Health Professionals

In support of the hypothesis, the responses of mental health professionals are con- sistently characterized by features of the “fight” response. In the majority of inter- views, 55%, the “fight” response was rated as most characteristic of professionals’ responses (vs. “ideal,” 28%, and “flight,” 17%). On a scale from 0 (no response of this type) to 10 (most extreme possible response of this type), the average “fight” rating for professionals’ responses was 5.2, the average “flight” rating was 1.6, and the average “ideal” rating was 3.9. Overall, professionals were significantly more likely to intervene in a “fight-like” manner than they were to avoid or flee the threat of suicide (t = 6.03, p < .001). At the same time, professionals were also marginally significantly more likely to respond with “fight” than with a more “ideal,” cli- ent-friendly style (t = 1.94, p = .06). Fortunately, however, professionals overall tended to be more likely to respond to the client in an “ideal” way than with “flight” (t = 3.29, p < .01).

Part II: Clients

As with the data from the professionals, three paired sample t tests were performed on the client data obtained in the study. In support of the hypothesis, the results in- dicate that clients show a clear preference for the ideal response type. On a scale from 0 (desires no response of this type) to 10 (most extreme possible response of this type), the average “fight” rating for client preferred response was 3.05, the av- erage “flight” rating was 1.1, and the average “ideal” rating was 6.8. In fact, the ideal response was seen as most characteristic of clients’ reported intervention preferences in 80% of the interviews. In contrast, “fight-like” responses were pre- ferred by only 15% of clients, and “flight-type” responses were preferred by only 5% of clients in the study.

Results of the t tests show significant differences in all of the pairwise compari- sons. Clients prefer the “ideal” response more than twice as much as the “fight” re- sponse (t = 4.93, p < .001) and more than six times as much as the “flight” re- sponse, which they least prefer (t = 8.14, p < .001). In addition, the results show that although neither problematic response type is preferred, of the two problem-

158 THOMAS AND LEITNER

atic types, clients would rather professionals be overly protective (“fight”) than avoid or deny (“flight”) the issue of suicide (t = 4.03, p = .001). The majority (12 of 20 clients) did not include any flight-like characteristics in their descriptions of the response they feel to be most helpful but many reported a preferred response that had at least some flavor of all three response types and all were characterized at the least as a mixture of the fight and ideal responses.

DISCUSSION

Taken together, the results of the two parts of this study support our overall hypoth- esis that a mismatch exists between interventions that suicidal clients report to be most helpful and interventions that professionals tend to provide. Although clients clearly find the “ideal” response to be most helpful, professionals generally re- spond in a manner characteristic of the “fight” response. Thus, the answer to the initial research question is: No. The responses that professionals offer suicidal cli- ents generally are inconsistent with what clients report to be most helpful. Al- though ultimately the same actions may be taken, the “fight-like” manner in which professionals are responding is in many ways the opposite of the “ideal” style that clients find to be most helpful. This fundamental mismatch between the desires of suicidal clients and the actions of their treatment providers has implications for practice, research, and training. However, before turning to a discussion of the findings and their implications, limitations to the study will be discussed.

Limitations of the Study

First, it should be noted that all of the participants, clients, and professionals, came from a community mental health setting. As this type of environment is in some ways systematically different than other mental health systems (e.g., private or group practice, and inpatient treatment), the generalizability of the findings may be limited to the community mental health setting. For example, as can be seen in Table 4, most (66%) of the professionals sampled from the community mental health population had a Bachelor’s level education or less, and most (67%) were unlicensed. These proportions are consistent with the way many community men- tal health systems are structured, as most of the direct service providers are those with less training and education, whereas professionals who have less direct con- tact with clients tend to be those with more training and education. However, in other types of systems professionals with more training and education may be found in much higher proportions. This suggests that the results of this study may be specifically more reflective of the responses of community mental health pro- fessionals, who typically have less training and education, than generally represen- tative of all mental health professionals.

STYLES OF SUICIDE INTERVENTION 159

In fact, a preliminary correlational analysis conducted on the data obtained in this study indicates that one’s level of professional training may play a role in one’s response to suicidal clients. Here, higher levels of licensure tend to be as- sociated with a more “ideal” (rs = .34, p < .05, n = 41) and less “flighty” (rs = – .36, p < .05, n = 41) response but are not associated with the level of “fight” in one’s response (rs = –.13, ns, n = 41). This suggests that licensed professionals (i.e., those with more education and training), who are perhaps more readily found in other mental health settings, tend to respond in a more “ideal” manner. A similar study including mental health professionals from other settings (in- cluding other geographical locations) would be helpful in further understanding the applicability of this study’s findings.

Second, the study was methodologically limited due to the possibility for some demand characteristics with the use of the semistructured interview design. As mentioned earlier, the crisis intervention model (typically interpreted using the “fight-like” cardiologist metaphor) is currently believed to be the best option for the treatment of suicidal individuals and is the model from which professionals are taught and evaluated. Given the predominance of this model, there is a chance that many clinicians reported their actions as being congruent with the party line (i.e., the “fight” response) because they thought this is what the interviewer wanted to hear or because they felt motivated to respond in what they thought would be the most acceptable and desirable way. Although there is the possibility that the de- mand characteristics inherent in the design may have distorted the findings, the flexibility of the semistructured interview format allowed for more in depth and specific questioning. This enabled the research assistants rating the interviews to get a clearer picture of events as well as a sense of when participants may have been coloring their responses in a certain way.

Therapeutic Implications of the Mismatch

What are the potential effects of this mismatch on the client, the clinician, and the relationship? This is a valuable question to ask when considering the importance of this discrepancy in terms of suggestions for future practice. Some may wish to dismiss the finding as trivial, as many believe that it is the clinician, not the client, who knows best. In this view, it is the clinician’s view of treatment that is most ap- propriate and ultimately most effective. Further, as is implicit in this view, if the treatment is not effective, the outcome has more to do with the problematic client than with the intervention offered by the clinician.

However, professionals operating from theories of person and psychotherapy that emphasize the importance of the relationship, therefore valuing the positions of both parties in the relationship, may find that these results warrant further con- sideration. From this type of theoretical perspective, any intervention a clinician makes is not done in isolation. In other words, interventions are not things that are

160 THOMAS AND LEITNER

done by clinicians to passive, helpless clients. Rather, clients play an active role in the process by having the power to either validate or invalidate any intervention of- fered by a clinician (Leitner & Guthrie, 1993). We believe the statement made by the collective client voice brought forth in this study represents an invalidation of the style of suicide intervention that is most predominantly used. In this way, cli- ents are not only making a statement about the pain and injury they have felt as a result of such interventions but they are also communicating their sense of discon- nection in their relationships with the clinicians making these interventions.

Rather than operating from a place of optimal therapeutic distance from the cli- ent, in which the clinician is able to balance the needs of the client and the thera- pist, the disconnection in the relationship represents the stance of therapeutic strangers (Leitner, 1995). When operating from this stance, clinicians are too dis- tant from the client to experience the client’s needs and desires, perhaps as a result of being preoccupied with their own concerns or fears about the situation, resulting in interventions that are experienced as inappropriate, mistimed, and ultimately unhelpful, if not hurtful, to the client. Rather than serving to strengthen the rela- tionship between clinician and client, leading to more potentially helpful interven- tions, this stance weakens the alliance, making further effective interventions more difficult. As the relationship is weakened, clients, as they often reported in this study, are left feeling hurt and fundamentally misunderstood.

Although this therapeutic distance may ultimately be a protective move for the therapist, it has the opposite effect for the client, leaving the client feeling more vulnerable and unprotected in the relationship. As they reported, this effect is espe- cially disconcerting for suicidal clients who are already in a very difficult place. The therapeutic relationship is central in many schools of therapy no matter what the presenting issue but given the life and death nature of the particular case of the suicidal client, the relationship is perhaps even more key, as it may be one of the few things helping the client to remain connected to the world and engaged in the struggle to live. Due to this possibility, evidence of a potential disruption in the connection between the suicidal client and the clinician is concerning. As such, we believe the results of this study alert the professional community to a disconnec- tion that deserves further exploration.

Areas for Future Research

Further research in this area would be one way to continue to explore the issues raised by this study. As our previous discussion about the study limitations sug- gests, one line of further research would include replications of this study sam- pling clients and professionals from different mental health settings as well as from different parts of the country. Other future research may involve a change in study design.

STYLES OF SUICIDE INTERVENTION 161

For example, the finding that clinicians’ interventions do not match with what clients find to be most helpful was based on a comparison of results obtained from the two separate parts of this study. The comparison involved clinicians’ reports of their own interventions and clients’ reports about what interventions they find to be most helpful. However, in addition to discussing the type of intervention they de- sired from a professional when they were feeling suicidal, clients also provided in- formation about the interventions they actually received. As such, a future study could more systematically investigate the issue of disconnection from the client’s perspective by comparing what clients report about both the interventions they want and the responses they received.

In addition, as it seems to us that an important part of exploring suicide inter- vention is understanding clients’ perceptions of those interventions, it may be fruit- ful to compare professionals’ views about the responses they are giving with cli- ent’s views about those same responses. Anecdotally, the primary researcher and research assistants can agree that in general, the interventions that clients reported actually receiving matched in large part the style of interventions that mental health professional reported giving. However, this issue was not more formally in- vestigated by this study. This could be accomplished by further analysis of this data (or in study replications) by asking raters to also rate the client interviews for the levels of each of the response styles they reported actually receiving and then comparing those ratings with the ratings for professionals’ responses. Alterna- tively, it may be more useful to pair participants and compare each client’s experi- ence of the intervention he or she received with that particular clinician’s experi- ence of the intervention he or she offered. This may provide more helpful information about the connections and disconnections in the therapeutic relation- ship around the issue of suicide intervention.

Implications for Training

In addition to suggesting areas for future research, training is another area in which the findings of this study may be informative. Many professionals in this study, 45%, reported having one or fewer courses (10 or less clock hr) of training in sui- cide prevention and/or intervention. Several professionals (16%) reported having no training in this area at all. Some of the professionals interviewed had been in the field for less than a year, and as such may not have had the opportunity for in-ser- vice training to that point. However, even some of those who had been in the field for such a short time, less than 1 year, reported having already experienced a client (or more than one client) attempting suicide. The fact that clinicians enter the job immediately faced with very difficult client issues suggests that clinicians need to be prepared to deal with these issues before they begin the work. In our view, sui- cide prevention and intervention training should occur before clinicians are faced with these issues, not as a consequence of having experienced a client suicide or at-

162 THOMAS AND LEITNER

tempt. While certainly there is value in on-the-job training in many areas, this is an area where it would be more beneficial and less risky for all to have had exposure to the issues and education prior to being placed in a position of having to make quick, difficult, life and death decisions.

This, of course, does not speak to the other concern, which is the amount and content of the training. Even if the timing is more appropriate, if the amount or content are not sufficient, clinicians may still be ill prepared to handle suicidal cri- ses. The results of this study suggest that the content of training on suicide inter- vention may be particularly important. As stated earlier, the crisis intervention model most frequently advocated in the literature offers detailed prescriptions for action but little on the manner in which the actions should be taken to be most help- ful to the client. As illustrated by what clients report to be most helpful, what is more important to clients than the actions taken by their treatment providers in re- sponse their suicidality is the manner in which the interventions are given. What clients report to be most helpful is the respectful, patient, and inviting stance sug- gested in the “ideal” response type. Based on this, it seems that the professional who is able to implement the practices of the crisis intervention model within this “ideal” mode of response will be more likely to be helpful to the suicidal client than one who takes the steps outlined by the model in a more “fighty” manner. As such, if the crisis intervention model will continue to be used as it has been for more than 45 years, training in suicide intervention should emphasize not only the steps and actions involved in the model but also the manner in which the clinician implements the model.

CONCLUSION

The suicide prevention movement, now led by the Surgeon General as well as other national and local suicide prevention organizations, is focused mainly on understanding the risk factors involved for suicide and improving the profes- sional assessment of those risks as a means of suicide prevention (U.S. Depart- ment of Health and Human Services, 2001; U.S. Public Health Service, 1999). However, the results of this study serve as a reminder that suicide intervention is a very complex venture that involves much more than assessing risk. Further, from our view, the client’s voice, as expressed in this study, reminds us that sui- cide interventions are not something that professionals do to passive, helpless clients. Interventions are done with clients, which means that clients play an ac- tive role in determining the effectiveness of the intervention. Clients in this study reported that, as opposed to the response most typical of community mental health professionals, the “fight” response, they clearly preferred the “ideal” style in response to their suicidal presentation. This is fruitful information for those

STYLES OF SUICIDE INTERVENTION 163

interested in improving the efficacy of suicide intervention and reducing the alarming rates of suicide in this country.

ACKNOWLEDGMENTS

The authors would like to thank Lori Koelsch, Valerie Loeffler, and Amberly Panepinto for their work on this project. We are also grateful to Art Miller and Carl Paternite for their contributions to the thesis project on which this article is based. We would also like to express our deep appreciation to all of the study participants for sharing their time and their stories with us. Participant names and details have been falsified to respect the privacy of participants.

REFERENCES

Allen, N. (1984). The history of suicide: Suicide prevention. In C. L. Hatton & S. M. Valente (Eds.), Suicide: Assessment and intervention (2nd ed., pp. 1–15). Norwalk, CT: Appleton-Century-Crofts.

American Foundation for Suicide Prevention. (n.d.). Facts about suicide. Retrieved January 26, 2005 from http://www.afsp.org/index-1.htm

Berman, A. L. (2000, November). The assessment and treatment of persons at risk for suicide. In, M. F. Hogan (Director), Suicide: A conference on suicide prevention. Working together to save lives: A mandate for the 21st century. Symposium presented by the Ohio Department of Mental Health, Worthington, Ohio.

Caruso, K. (n.d.). Suicide statistics. Retrieved January 25, 2005 from http://www.Prevent SuicideNow.com

Frankish, C. J. (1994). Crisis centers and their role in treatment: Suicide prevention versus health pro- motion. In A. A. Leenaars, J. T. Maltsberger & R. A. Neimeyer (Eds.), Treatment of suicidal people (pp.153–165). Washington, DC: Taylor & Francis.

Fremouw, W. J., de Perczel, M., & Ellis, T. E. (1990). Suicide risk: Assessment and response guidelines (pp. 1–10, 98–107). New York: Pergamon Press.

Grollman, E. A. (1971). Suicide: Prevention, intervention, postvention (pp. 87–89). Boston: Beacon Press.

Jobes, D. A., & Maltsberger, J. T. (1995). The hazards of treating suicidal patients. In M. Sussman (Ed.), A perilous calling: The hazards of psychotherapy practice (pp. 200–214). New York: Wiley.

Kelly, G. A. (1961). Suicide: The personal construct point of view. In N. L. Farberow & E. S. Shneidman (Eds.), The cry for help (pp. 255–280). New York: McGraw-Hill.

Leenaars, A. A. (1994). Crisis intervention with highly lethal suicidal people. In A. A Leenaars, J. T. Maltsberger, & R. A Neimeyer (Eds.), Treatment of suicidal people (pp. 45–59). Washington, DC: Taylor & Francis.

Leitner, L. M. (1995). Optimal therapeutic distance: A therapist’s experience of personal construct psy- chotherapy. In R. Neimeyer & M. Mahoney (Eds.), Constructivism in psychotherapy (pp. 357–370). Washington, DC: APA Press.

Leitner, L. M., & Guthrie, A. F. (1993). Validation of therapist interventions in psychotherapy: Clarity, ambiguity, and subjectivity. International Journal of Personal Construct Psychology, 6, 281–294.

Maris, R. W., Berman, A. L., & Silverman, M. M. (2000). Comprehensive textbook of suicidology. New York: Guilford.

164 THOMAS AND LEITNER

Neimeyer, R. A., & Bonnelle, K. (1997). The Suicide Intervention Response Inventory: A revision and validation. Death Studies, 21, 59–81.

Neimeyer, R. A., & Pfeiffer, A. M. (1994a). Evaluation of suicide intervention effectiveness. Death Studies, 18, 131–166.

Neimeyer, R. A., & Pfeiffer, A. M. (1994b). The ten most common errors of suicide interventionists. In A. A Leenaars, J. T. Maltsberger & R. A Neimeyer (Eds.), Treatment of suicidal people (pp. 153– 165). Washington, DC: Taylor & Francis.

Shea, S. C. (1998). Psychiatric interviewing: The art of understanding: A practical guide for psychia- trists, psychologists, counselors, social workers, nurses and other mental health professionals (2nd ed., pp. 443–514). Philadelphia, PA: Saunders.

Szasz, T. A. (1980). The ethics of suicide. In M. P. Battin & D. J. Mayo (Eds.), Suicide: The philosophi- cal issues (pp. 185–198). New York: St. Martin Press.

Szasz, T. A. (1986). The case against suicide prevention. American Psychologist, 41(7), 806–812. U.S. Department of Health and Human Services (2001). National strategy for suicide prevention goals

and objectives for action: Summary: Inventory number SMA01-3518. Retrieved January 26, 2005 from Substance Abuse and Mental Health Services Administration. http://www.mentalhealth .samhsa.gov/suicideprevention/strategy.asp

U.S. Public Health Service (1999). The Surgeon General’s call to action to prevent suicide. Retrieved January 26, 2005 from Substance Abuse and Mental Health Services Administration. http://www.mentalhealth.samhsa.gov/suicideprevention/strategy.asp

AUTHOR NOTE

Jill C. Thomas received her MA in Clinical Psychology from Miami University in 2003. She is currently completing her PhD at Miami University. Her research in- terests included eating disorders, embodiment, action research, and other creative approaches to research and therapy.

Larry M. Leitner received his PhD in Clincal Psychology from The University of Nebraska in 1979. He is Professor of Psychology at Miami University. His re- search interests involve the application of Personal Construct Psychology to psychotherapy, psychopathology, and personal perceptions.

STYLES OF SUICIDE INTERVENTION 165