Collaboration and a Self-Help Orientation in Therapy ... - Springer Link

P1: GDT Journal of Contemporary Psychotherapy [jcp]

ph238-jocp-476180

December 8, 2003

21:0

Style file version Nov 28th, 2002

C 2004) Journal of Contemporary Psychotherapy, Vol. 34, No. 1, Spring 2004 (°

Collaboration and a Self-Help Orientation in Therapy with Suicidal Clients Thomas E. Ellis

Working with suicidal clients historically has placed a heavy burden of responsibility on the therapist for a successful outcome. This orientation generally has left clinicians feeling stressed and in need of controlling their clients’ behavior; clients, in turn, often have developed unrealistic expectations of therapy or “resistance” of the therapist’s interventions. In this context, the chance for a potentially valuable collaboration between client and therapist is often lost. This paper builds on previous work that has introduced the notion of a collaborative working relationship with suicidal clients. It is argued that the therapeutic process ideally should not only utilize the therapist’s knowledge and skills in treating suicidal clients, but also cultivate self-efficacy by encouraging the client to adopt a selfhelp orientation to therapy. Several general principles of a collaborative/self-help orientation to working with suicidal clients are discussed and a variety of specific therapeutic strategies consistent with this orientation are described. KEY WORDS: suicide; self-help; collaboration; self-efficacy.

COLLABORATION AND A SELF-HELP ORIENTATION IN THERAPY WITH SUICIDAL CLIENTS Anyone who attempts self-therapy has a quack for a therapist and a fool for a client. Old adage Therapists and clients alike would be foolish to believe that the outcome of therapy rests solely in the hands of the therapist. Revised adage, based on current empirical evidence

True to its crisis intervention roots, psychotherapy with suicidal clients traditionally has stipulated that the therapist’s first and foremost responsibility is to Address correspondence to Thomas E. Ellis, Department of Psychology, Marshall University, Huntington, WV 25755. 41 C 2004 Human Sciences Press, Inc. 0022-0116/04/0300-0041/0 °


ph238-jocp-476180

December 8, 2003

21:0

42


Ellis

“keep the client alive.” This high therapist-responsibility orientation has translated into a significant focus in therapy sessions on ongoing risk assessment and common usage of involuntary hospitalization, as therapists feel obligated to ensure the client’s safety. A traditional “medical” model has historically cast the clinician in the role of healer and the client as the passive, hopefully compliant, beneficiary of the procedure. However, recent years have seen the beginnings of a shift in the perspectives of both therapists and clients with respect to the “ownership” of the therapeutic process. Experience teaches that clients are often not as receptive to well-intentioned interventions as therapists might wish; moreover, the importance of the client’s sense of self-efficacy is becoming increasingly apparent. With suicidal clients, the implications are especially great, in terms of therapist stress and potential control issues surrounding the client’s choices regarding self-harm or suicide. Approaches to treatment consequently are beginning to move toward assigning greater responsibility to clients and placing more emphasis upon collaboration between therapist and client (e.g., Michel, Maltsberger, Jobes, Leenaars et al., 2002; Chiles and Strosahl, 1995; Norcross, 2002). In this paper, I will begin with a brief description of the traditional therapeutic model of therapist responsibility and present an alternative, collaborative, model. I will then describe a variety of general and specific collaborative intervention strategies that encourage active participation and self-help on the part of the client. THE THERAPIST RESPONSIBILITY MODEL In the Therapist Responsibility Model (TRM), the therapist assumes the role of healer and is viewed as largely responsible for outcome. This model exists as a result of several influences. The influence of the medical model is substantial and maintains that the clinician is the knowledgeable expert who effects a cure by dispensing knowledge to the client and/or directly performing procedures upon him or her (see Wampold, 2001, for a detailed explication of the medical model). The TRM is sometimes inadvertently reinforced by trainers and supervisors, who allow themselves to be idealized by students and supervisees while failing to acknowledge fully the limitations of their procedures or the gaps in the knowledge base. Another influence is the traditional client, who, like many students, may be only too willing to view the therapist as an omnipotent healer with the power to make clients better without much effort on their part. The spectacular growth and advertising of psychopharmacological compounds in recent years has only added to the client’s sense that it is not unreasonable to expect dramatic benefits with only minimal effort. Ironically, the parallel growth of empirically supported therapies (e.g., Nathan and Gorman, 2002) has reinforced the TRM to some extent. These treatments are


ph238-jocp-476180

December 8, 2003

21:0


Collaborative Therapy with Suicidal Clients

43

highly prescriptive (tailored to specific disorders), structured, directive, manualized, time-limited, and supported by significant empirical evidence. The implicit (albeit unintended) message, especially to novice therapists and clients, is that these are “surgical” procedures that are performed upon the client and whose effectiveness hinges largely upon the talent and technology of the therapist. Further contributing to the sense that the therapist is responsible for the client’s outcome are external influences from business and legal arenas. Managed care, introduced in an attempt at health care cost control, places responsibility squarely on the therapist to obtain results in a brief period of time and with limited resources. Little if any allowance is made for complex clients who do not follow the course of improvement seen in controlled laboratory studies of single-diagnosis clients and therapists with specialty expertise. At stake are managed care decisions that affect both the therapist’s income and the client’s well-being. Of course, few, if any, factors have exerted more influence than liability actions, which transpire in a system explicitly designed to assign blame. Malpractice lawsuits seek to establish that the provider directly caused harm to the plaintiff due to some dereliction of duty. The therapist is viewed as responsible for adverse client outcomes. Therapists are acutely aware of their potential culpability with high-risk clients; unfortunately, horror stories told in the malpractice avoidance literature and in continuing education workshops often serve only to intensify therapists’ fears. Consequences of the TRM for the treatment of suicidal clients are substantial. It is well established that therapists consider suicidal clients to constitute the most stressful aspect of their practices (e.g., Deutsch, 1984). It has also been shown that the loss of a client to suicide has an impact on the therapist on a par with the death of a relative (Chemtob, Bauer, and Hamada, 1988). When the therapist feels responsible for a client’s life, at least two other undesirable side-effects occur. First, the often considerable strengths and assets that the client can potentially apply to the situation are overlooked. Second, the more responsible the therapist feels, the greater is his or her perceived need to control the client. The client’s understandable response to such efforts often is noncompliance or “resistance” to therapeutic interventions. In the resulting scenario, the client fights to maintain autonomy and keep options open while the equally well-meaning therapist struggles ever harder to save the client from himself or herself. The therapeutic alliance often is the first casualty. This situation reflects the difference between TRM-oriented “management” of the suicidal client and collaborative therapy. Hendin (1995) is particularly critical of the former: [Suicide “management” techniques]. . . seem reasonable, but in practice they reflect a state of mind and a way of relating to suicidal clients that often make treatment unsuccessful. . . All the precautions and all the management may result in encouraging one of the most lethal aspects of the suicidal individual, that is, his tendency to make someone else responsible for his staying alive (p. 185, emphasis added).


ph238-jocp-476180

December 8, 2003

21:0


44

Ellis

Jobes (2000) also speaks of what is lost when the clinician, feeling responsible for the client’s life and fearing possible liability should the client inflict self-harm, assumes a controlling posture: It is widely understood that clinicians must protect clients from themselves, even if it requires breaking the sacred bond of confidentiality and of withholding basic civil liberties through involuntary hospitalization. Nothing about this imminent-danger standoff seems therapeutic. In truth, the natal or even well-established clinical alliance can be suddenly transformed from a shared vehicle of healing into a battle ground where the professional’s “clinical wisdom” is authoritatively pitted against the patient’s personal liberty (p. 9).

The Aeschi Group, a working group of concerned clinicians, has studied and commented on this and related issues in detail, with particular emphasis on the experience of the client (Michel, Maltsberger, Jobes, Leenaars, Orbach, Stadler, Dey, Young, and Vlach, 2002). The interested reader is referred to the group’s web site: http://www.aeschiconference.unibe.ch/.

THE COLLABORATIVE MODEL Is there a viable alternative to the Therapist Responsibility Model? It is sometimes believed that the only alternative is a Client Responsibility Model (CRM). Otherwise known as the “blame the victim” approach, it is less explicit than the TRM and apparent mainly in the private comments of frustrated therapists, who maintain that treatment failures are usually caused by clients who are resistant, uncooperative, unmotivated, or invested in remaining sick. At best, treatment failure is attributed to the client’s “noncompliance;” at worst, the client is accused of not being interested in getting better or having a “need to suffer.” In any case, many clients are only too willing to accept such explanations for their lack of improvement. The quality of service to suicidal clients is severely compromised when delivered by clinicians with a CRM mindset. It is not uncommon to hear novice (and sometimes experienced) therapists comment, “If a client is intent on committing suicide, there’s nothing you can do to stop them.” The shortage of studies showing significant benefits from suicide treatment and prevention programs are sometimes cited as justification for this view. Even more dangerously, when therapists view clients as solely responsible for outcomes and themselves as potentially vulnerable in those outcomes, strong negative feelings toward the client can result, described by Maltsberger and Buie (1973) as “countertransference hate.” However, viewing the CRM as the only alternative to the TRM is a clear example of a false dichotomy (the belief that there are only two extreme options when other options do, in fact, exist). As is often the case with dichotomies, kernels of truth are contained in both models. It is certainly appropriate that clinicians


ph238-jocp-476180

December 8, 2003

21:0



45

be viewed as responsible to some extent for client outcomes. Ethical codes of conduct dictate this view, as well as numerous therapy outcome studies showing that what the clinician does during sessions does, in fact, make a difference in outcomes (Nathan and Gorman, 2002). However, there is plentiful evidence that client behaviors also make a difference. In the health care arena, for example, comparison of the leading causes of death at the start of the last two centuries shows a dramatic shift from infectious diseases to illnesses that are related to behavioral choices such as diet, exercise, and tobacco and alcohol usage (Straub, 2001). Similarly, in the psychotherapy arena, it is clear that active participation by the client in therapy is associated with superior outcomes. For example, Burns and Spangler (2000) showed that therapeutic homework completion by depressed clients was significantly associated with improvement, independent of severity of depression or other mediating variables. An alternative to the TRM-CRM dichotomy, which we shall call the Collaborative Model, has emerged in recent years, influenced significantly by the introduction of collaborative approaches such as Beck’s cognitive therapy (e.g., Beck, Rush, Shaw, and Emery, 1979). Here, responsibility is shared by clinician and client. For example, Judith Beck (1993) provides a list of questions to assess collaboration in therapy, such as, “Have the client and I been truly collaborating? Are we functioning as a team? Are we both working hard? Do we both feel responsible for progress (p. 304)?” Another example of collaboration in therapy is Linehan’s treatment program for individuals with borderline personality disorder (Linehan, 1993). Treatment planning in this program includes a very specific contract outlining respective roles and responsibilities of therapist and client. For example, the therapist agrees to be available to the client and to “make every reasonable effort to be effective,” while the client agrees to attend therapy regularly, work on therapy-interfering behaviors, and so on. In perhaps the most pointed departure to-date from the TRM (or CRM), Chiles and Strosahl (1995) place great emphasis on the limitations of therapeutic interventions and the importance of collaboration with the client. In a chapter aptly subtitled, “The Medium is the Message,” they state: Your most important general attribute is your desire to build a collaborative set with your client early in the process of addressing suicidal behavior. Often, both you and your client have exaggerated ideas about your capacity to alleviate your client’s sense of suffering. You need to undermine idealized expectations in a way that has a rapport-building feature to it. Building a collaborative set involves recognizing that, by virtue of your training, you possess special sets of information about how people change in difficult circumstances. However, the actual process of behavior change will have to be a team effort, a “we” effort. . . . Your role [is] to act as a resource when your client is ready to use your knowledge in some constructive way (pp. 117–118).

Collaboration, which might be conceptualized as a key component of the therapeutic alliance, enjoys considerable empirical support in studies of therapeutic


ph238-jocp-476180

December 8, 2003

21:0


46

Ellis

process and outcome. In their review of the literature on goal consensus and collaboration in psychotherapy, Tryon and Winograd (2002) describe a variety of studies showing these variables to be significantly associated with initial engagement in therapy (including whether the client returned for second and subsequent sessions) and therapeutic outcome as measured by client, therapist, and observer ratings. They concluded that, “Results of these studies provide considerable support for the positive impact of collaborative involvement on treatment outcome (p. 119).” In his comprehensive critical review of the psychotherapy outcome literature, Wampold (2001) makes the case for “general” factors, such as the therapeutic alliance, as the key ingredients in therapeutic change, as opposed to the specific, theoretically-based interventions to which outcomes are generally ascribed. After a broad review of the research literature on the therapeutic alliance, of which collaboration is a key component, he concludes that, “The alliance appears to be a necessary aspect of therapy, regardless of the nature of the therapy. . . [I]t appears that the relationship accounts for dramatically more of the variability in outcomes than does the totality of specific ingredients (p. 158).” Regardless of whether one concurs with Wampold’s conclusions, it seems clear that a collaborative alliance deserves a prominent place in the therapeutic endeavor, perhaps especially so with clients who are suicidal. In the next section, I will describe some general principles of collaborative therapy with suicidal clients. The sections that follow will present various examples of general and specific interventions consistent with these principles. GENERAL PRINCIPLES OF COLLABORATIVE PROCESS Self-help Orientation from Day One Therapists often miss a prime opportunity to cultivate a collaborative mindset in the client at the time of first contact. The “intake” session is often framed as a strictly evaluative meeting, concluding with a statement such as, “Come back next week and we’ll get to work.” By then, the client may have already adopted a passive treatment orientation. Rather than having to work to “undo” this orientation, it can be helpful to invite the client to collaborate during and immediately after the initial session. Asking the client for his or her own assessment of the problem and possible remedies or recommending reading and monitoring activities as “homework” can communicate from the start the expectation that the client will be actively involved in his or her therapy. In fact, I recommend a telephone contact in advance of the initial session, in which the clinician can obtain preliminary information, describe his or her approach, and even suggest reading, symptom monitoring, or filling out assessment forms ahead of time.


ph238-jocp-476180

December 8, 2003


21:0


47

Fading Directiveness Over Time Of course, some clients, especially those who are severely depressed, may not be able to participate actively in the beginning. Asking a depressed client, “What would you like to work on today?” is comparable to a golf pro asking the same question of a beginning golfer: The response may be little more than a perplexed look. For these individuals, it is advisable to determine (collaboratively, of course) what might be realistic for the individual to undertake. For example, rather than asking the client to read a self-help book, or even a chapter, an agreement might be reached to read just a few pages or to ask a friend or family member to obtain a copy of the book. In any event, it is generally agreed—by cognitive and psychodynamic therapists alike (see Ellis, 2001)—that suicidal clients require a high degree of activity and directiveness by the therapist early in the course of therapy. However (and this can be discussed very explicitly with the client), it is important to planfully decrease directiveness over the course of therapy, while encouraging the client reciprocally to take increasing responsibility and become more active. This includes such matters as determining the agenda for the day’s session, finding solutions for current problems, and planning self-help homework activities. For many, if not most, clients, the goal of “becoming your own therapist” can be more than just a clich´e. Shifting Responsibility for Crisis Management Early in the course of therapy with a suicidal individual, it is often necessary for the therapist to take a highly active role in finding solutions to problems and resolving crises. However, consistent with the principle of fading directiveness, it is also useful for the therapist to proceed with an assumption that, over time, the responsibility for crisis management will gradually shift to the client. Psychodynamic writers speak of providing sustaining resources while helping the client to develop good internal objects (Maltsberger and Goldblatt, 1996); cognitivebehavioral therapists strive to shift their role from solving problems for clients to teaching problem-solving skills (Chiles and Strosahl, 1995; Clum and Lerner, 1990). In any event, it should be emphasized that any such shift must be made in active collaboration with the client, lest this be perceived by the client as withholding or abandonment. Transparent Agenda It is surprisingly common in conventional therapy for the therapist to fail to share his or her formulation and treatment plan with the client. This, ironically, makes it difficult or impossible for even minimally resistant clients to cooperate


ph238-jocp-476180

December 8, 2003

48

21:0


Ellis

fully with the therapist’s agenda. The collaborative model allows for few if any secrets between therapist and client, in much the same manner that a personal trainer would discuss strengths, weaknesses, and improvement strategies with a fitness client. The client who is aware of the whys and wherefores of therapy are in a much better position to collaborate than one who feels “in the dark” about what is going on during sessions. Mutual Goal-Setting and Agenda-Setting Consistent with a transparent agenda is the practice of collaboratively setting goals with the client (Tryon and Winograd, 2002). Goals may be conceptualized in three categories: Therapy goals, session goals (agenda), and mini-contracts. Setting therapy goals is much easier said than done. Certainly, it is obvious that the overarching goal is to “help the client get better,” but what this means to therapist and client may diverge if it is not discussed openly and very specifically. This is an extremely common issue with suicidal clients; for clinicians often assume that the goal of therapy, first and foremost, is to prevent suicidal behavior. Yet, if it is discussed with the client, it is often found that this is not a primary goal; in fact, the client may very specifically wish to maintain self-harm and suicide as options, “in case of an emergency.” Proceeding with therapy without resolving this conflict is in my view one of the leading sources of resistance in the treatment of suicidal clients. Finding common ground and agreeing on mutual goals (such as reducing suffering) can increase collaboration immensely. Similarly, the practice in cognitive therapy of collaboratively setting a clear agenda at the beginning of each session (Beck, 1995) increases the client’s level of involvement and reduces chances of working at cross-purposes during the session. It is important to note here that simply asking the client for his or her agenda and letting this alone dictate the session is not consistent with a collaborative orientation. Instead, the therapist can say, “Here are some things I thought we might want to focus on today. How does that sound to you? What would you like to add to the agenda?” A negotiation might then follow as to whether all of the items can realistically be addressed within the session and which items might be able to wait until a future session. An important therapist role in this process is ensuring that session goals fit reasonably well within the framework of the overall therapy goals (above). Even collaboratively planned sessions can be expected sometimes to go offtrack. This is especially true when the therapist neglects to get “permission” from the client for specific activities within the session. This might be as simple as assuming (incorrectly) that the client is ready to move from one agenda item to another. Affectively laden interventions are another common sticking point. Simply directing the client into the activity often leads to passive participation or even outright resistance. A “mini-contracting” intervention, such as, “I’m thinking


ph238-jocp-476180

December 8, 2003


21:0


49

that a chair exercise might be more helpful than just talking in the abstract about this loss. How does that sound to you?” is more likely to lead to a collaborative outcome (including the possibility of skipping the exercise). Coach/Trainer Model It is safe to say that most clients enter therapy viewing the therapist through the lens of the Therapist Responsibility Model, based on a lifetime of experiences with treatment by family doctors, surgeons, dentists, and the like. One device for reorienting the psychotherapy client is through the use of analogies and metaphors, in particular regarding the therapist and therapy. Depending on the background of the client, it can be helpful to compare oneself to a piano teacher, ski instructor, or foreign language tutor. Clients typically are quick to recognize that, while the instructor plays an important role, skill acquisition and improvement comes only with active participation (and hard work!) on the part of the learner. The importance of this skill acquisition/coaching model (as opposed to a disease/healer model) in cultivating a collaborative/self-help orientation can hardly be overemphasized. GENERAL INTERVENTION STRATEGIES Socratic Dialogue Perhaps the quickest way to put a client into a passive, or even resistant, mode is to give advice, correct “cognitive errors,” supply answers to problems, or otherwise fail to involve the client in discovering solutions. A Socratic approach is utilized in some therapies, and cognitive therapies in particular, as a means of “guided discovery” in order to keep the client energized and engaged in the therapeutic process (DeReubeis, Tang, and Beck, 2001). In the process, an important message is communicated: “I (the therapist) can give you some guidance, but you have considerable strengths beyond mine to help you reach your goals.” It is not uncommon, when the therapist “gets out of the way,” that clients come up with solutions that had not occurred to the therapist. The client’s self-efficacy is thus reinforced rather than undermined. Homework Perhaps the most powerful way to underscore the client’s role in the healing process (as well as generalizing treatment gains from the therapy session to the client’s outside life) is through the planful use of self-help homework activities (see Kazantzis and Lampropoulos, 2002, for a recent review). As stated above, this can and should begin as early in the course of therapy as possible. The array of


ph238-jocp-476180

December 8, 2003

21:0

50


Ellis

possible activities is broad and limited only by the resourcefulness and creativity of therapist and client. A common place to start is suggested readings that are consistent with the therapist’s treatment approach (e.g., Greenberger and Padesky, 1995). As therapy progresses, homework should be designed to be consistent with work done during the session. A routine (Socratic) question near the end of each session might be, “What do you think you might do between now and next session to continue our work today?” A variety of resources exist to provide ideas and structure for homework activities (e.g., Burns, 1999; Ellis and Newman, 1996). Specific examples of homework activities are discussed below.

Teaching Problem-Solving Skills Deficient problem-solving skills repeatedly have been shown to characterize suicidal individuals relative to other populations (e.g., Schotte and Clum, 1987; Clum and Lerner, 1990). Consistent with the use of Socratic questioning, teaching problem-solving skills (as opposed to supplying suggested solutions to problems) serves to activate the client’s own resources in dealing with adverse situations. A number of structured programs for teaching problem-solving skills exist, notably, that of D’Zurilla and Nezu (2001). In addition, problem-solving training plays a prominent role in treatment programs developed specifically for suicidal individuals (Clum and Lerner, 1990; Linehan, 1993; Rudd, Joiner, and Rajab, 2001).

Suicide Acknowledged as an Available Option It is not possible to adopt a truly collaborative relationship with a suicidal individual without a willingness to explore self-harm and suicide as real, available behavioral options. It is not uncommon for therapists, novice and veterans alike, to assume an “antisuicide” position whether or not the client is there yet. This often has the effect of the client’s feeling like the therapist is trying to “take away” this option and thus clinging to it all the more. On the other hand, if the therapist is willing to discuss sincerely the possible benefits of suicidal behavior, this can effectively open the door to cooperatively exploring the negative consequences as well. In other words, acknowledging suicide as an available option does not preclude the therapist’s having a distinct “bias” toward nonsuicidal options. It is more a matter of gaining credibility with the client by “facing facts” that both individuals know: that suicide is there as an available option, which, like all other options, has a set of potential benefits and consequences. Perhaps the frankest exploration of this controversial issue to-date is that of Chiles and Strosahl (1995). They counsel therapists to avoid value judgments regarding suicidal behavior and “beware of power struggles,” in which the therapist takes a prohibitive stance toward self-harm behavior, often with the implicit


ph238-jocp-476180

December 8, 2003


21:0


51

threat of involuntary hospitalization. A matter-of-fact problem-solving framework is recommended, in which therapist and client collaboratively explore legitimate benefits and costs of self-harm behavior vis-`a-vis other coping options. This framework is applied by Ellis and Newman (1996) in a section entitled, “Advantages of Suicide: Are you Kidding?” Here, an exercise invites the reader to list possible benefits of suicidal behavior, but then to consider “the catch” associated with each benefit. It is pointed out, for example, while one may, in fact, end one’s pain by committing suicide, this choice also eliminates all future opportunities to reduce pain through other, nonlethal, means.

SPECIFIC INTERVENTION STRATEGIES The following specific strategies are consistent with my training and experience as a cognitive-behavioral therapist. However, this is not to suggest that therapists necessarily must use cognitive-behavioral interventions if they wish to work collaboratively with their clients and cultivate a self-help orientation. Most of these strategies are fully compatible with a variety of therapeutic orientations. Cultivating Reasons for Living Marsha Linehan has aptly redefined work with suicidal individuals from “preventing suicide” to “creating a life worth living” (1993). This approach is based in part on her studies showing that suicidal individuals are able to identify significantly fewer “reasons for living” than nonsuicidal individuals (Linehan, Goodstein, Nielsen, and Chiles 1983). Using this framework with suicidal clients carries with it both promise and pitfalls. Indeed, suicidal individuals often are so focused on what is wrong with their lives that they overlook many aspects of their lives that do, in fact, bring them gratification and meaning. In this situation, the therapist must take care lest he or she is placed in the position of finding reasons for living for the client. It is important here to maintain a collaborative stance in providing guidance but also eliciting the client’s participation in identifying reasons for living that he or she might be overlooking. A self-help exercise to this end appears in Ellis and Newman (1996; p. 36). On the other hand, it must be acknowledged that some individuals experience so much tragedy and loss in their lives that one wonders whether their reasons for dying do, in fact, outweigh their reasons for living. Most experienced therapists have worked with at least one individual whose combination of catastrophic health problems, losses (occupational, interpersonal, etc.), financial and legal problems, and absence of social support have contributed to such a low quality of life that efforts to identify countervailing reasons for living prove fruitless. Interestingly,


ph238-jocp-476180

December 8, 2003

52

21:0


Ellis

Linehan (personal communication, 2001) recommends acknowledging this openly by agreeing, “You’re right, your life, as it is, is not worth living,” but then continuing, “Now, are you willing to work with me to make the changes that would make your life worth living?” Of course, in extreme cases (such as painful, terminal medical conditions), the therapist ultimately must struggle with the very personal issue of whether “rational suicide” is a valid construct and whether he or she is willing to “go there” with a client (Werth, 1996).

The “Victory List” Because of the cognitive error of “selective abstraction (Beck et al., 1979),” suicidal clients commonly lose awareness of their positive accomplishments or things that have gone right for them. Even recent accomplishments in therapy (such as getting through a crisis without a suicide attempt or hospital admission) can be forgotten, as the client is heard to say, “This isn’t helping—I’m not getting anywhere.” It can be helpful to encourage such clients to begin a running tabulation of successes, in particular, those that reinforce a sense of self-efficacy. That is, rather than focusing simply on good things that “happen” (e.g., the weather cooperating for a picnic), emphasis is placed on what the client has brought about (e.g., pursuing pleasurable activities or resolving a dispute with a family member). The Victory List serves not only to reinforce self-help activities when they occur, but also can become a vital part of an “emotional first-aid kit,” to remind the client of his or her successes after disappointments and when feeling discouraged.

Written Cognitive Restructuring Cognitive therapists routinely train clients in a set of skills that allow them to articulate, examine, and modify automatic thoughts, erroneous beliefs, and maladaptive interpretations of events that lead to distress and maladaptive behaviors. These skills can be taught using a variety of devices, the most rudimentary of which is the two-column technique. Here, the client learns to write down thoughts associated with a specific episode of upset and then consider whether there might be other, less upsetting ways of viewing the situation. The technique is first demonstrated during therapy sessions, as the therapist shows the client how to work with distressing emotions triggered during the session by articulating associated thoughts, writing them down, and then considering whether various alternate thoughts might be preferable. It can be helpful to provide names for the two ways of thinking. For example, for the individual upset by self-critical thoughts, two “modes” might be referred to as the Critic and the Ally. The section below illustrates how such work might be done through a written dialog exercise:


ph238-jocp-476180

December 8, 2003

21:0

Collaborative Therapy with Suicidal Clients Critic: Ally: Critic: Ally: Critic: Ally: Critic: Ally:


53

This report really stinks. You’re such an idiot! It’s not my best work, but my supervisor said it was fine. It should have been better. It was full of mistakes. Actually, it had three typos. Besides, who says everything I do has to be perfect? You’ll never get promoted at this rate. I have a chance if I keep working at it. But if I don’t, so what? If you don’t get promoted, it will prove what a loser you are. At worst, it would prove that I don’t excel at my job. But my job is not me. No matter what happens, I can refuse to condemn myself for it.

Another format is the Dysfunctional Thought Record (Beck et al., 1979), a multi-column form that allows more advanced clients to identify trigger events and cognitive distortions as well as quantify the degree of distress before and after cognitive restructuring. In any case, training in cognitive restructuring skills can be an effective, collaborative strategy for cultivating self-help skills and combating helplessness and hopelessness.

Chair Dialogue The Gestalt therapy chair dialog method is an invaluable means of bringing affect into what otherwise can be an over-intellectualized process of examining and changing cognitions. The client is invited to allow chairs to represent various parts of the self (say, depressed child and a supportive parent) and to move between (or among) chairs to allow these sides of the self to process issues. This is essentially the same work as the cognitive restructuring process described above, although the physical movement and verbalization generally brings more energy and affect into the session. With experience, clients can be guided toward considerable autonomy in this technique, to the point that many are able to use it on their own in-between sessions and after therapy has ended.

Homework Activities Once considered only a helpful adjunct, homework is now considered by some to be an indispensable component of therapy (Kazantzis and Lampropoulos, 2002). Homework is viewed as a means of both generalizing learning from the therapy environment to “real life” and cultivating the client’s sense of responsibility, selfefficacy, and control. The nature of homework activities depends upon what will be most helpful for a given client in a specific therapeutic context and is limited only by the creativity and resourcefulness of the therapist and client. A few specific examples follow. Reading. Self-help reading serves a variety of purposes in therapy. Clients who are well educated about their condition and the form of therapy being used


54

ph238-jocp-476180

December 8, 2003

21:0


Ellis

are in a better position to work collaboratively with the therapist toward established therapy goals. Moreover, therapy session time can be used more efficiently when focused on therapeutic change rather than education. The recent proliferation of empirically supported therapies has been accompanied by the appearance of numerous self-help manuals (e.g., Burns, 1999; Greenberg and Padesky, 1995; Ellis and Newman, 1996). Clinicians concerned about the quality of self-help books might wish to consult the Authoritative Guide to Self-Help Resources in Mental Health (Norcross, Santrock, Campbell, Smith, Sommer, and Zuckerman, 2000). Monitoring. Because retrospective self-report is known to have serious limitations, it is advisable to encourage clients to keep written records of various phenomena of interest. The Victory List, discussed above, is one example of such monitoring. Other examples include regular monitoring of severity of depression, using a standardized scale such as the Beck Depression Inventory (which appears in modified form in Burns, 1999); suicidal thoughts and behaviors, pleasurable activities, headaches, sleep habits, etc. In addition to reinforcing a sense of responsibility and control in the client, monitoring has the important added benefit of establishing correspondence between specific events (environmental, cognitive, physiological, etc.) and target problems such as crying episodes or suicidal impulses. The result typically is a decreased sense of mystery about episodes of distress and an increased sense of control. Listening to Session Tapes. Once considered an unorthodox practice in psychotherapy, it has become an increasingly common practice to invite clients to tape record sessions to listen to at home (Burns, 1989). While often hesitant at first, clients commonly report that listening to session tapes at home (and preferably taking notes) reveals significant information that was missed during the session. The therapist can explain that memory studies have shown that large amounts of spoken information are lost within the first few hours, in part because we are distracted by reflections on what was said previously or thinking about what to say next. The idea of listening to session tapes can be presented as a way of getting “two sessions for the price of one.” It is not uncommon for clients to save their “greatest hits” tapes for use during later crises, including after therapy has ended. Therapy Journal. Much as the therapist keeps progress notes, the client can be encouraged to function as his or her own therapist by keeping a therapy journal. This can be a spiral notebook or pocket folder for keeping handouts and worksheets from therapy sessions. Clients can be encouraged to write down therapy goals, notes from session tapes, self-progress notes, questions for future sessions, etc. Many clients use this notebook as their first self-help resource prior to calling the therapist whenever problems arise. As with the session tapes, clients sometimes use their therapy journals for months or years after therapy is concluded.


ph238-jocp-476180

December 8, 2003


21:0


55

CONCLUSION In their important paper, the Aeschi Group (Michel et al., 2001) maintain that the traditional medical model, used in the initial assessment of suicidal clients in crisis, “often leaves the needs of suicidal clients unmet (p. 424).” They recommend a more collaborative approach to client interviewing that can not only meet the clinician’s need for obtaining information and assessing risk, but meet the client’s needs for connection and acceptance as well. In this paper, I have sought to extend the collaborative framework beyond risk assessment to therapeutic work with suicidal individuals. While justified on humanitarian grounds, it is also argued that a collaborative, rather than “therapist-responsibility,” model is likely to result in a) increased activation of the client’s own resources (and therefore improved outcomes) and b) decreased stress for (and perhaps higher quality service from) the therapist. Perhaps most importantly, it is a plausible remedy for the dangerous situation in which the client makes others responsible for his or her staying alive. While appealing on the surface, the implications of this model are potentially controversial. Is it, in fact, realistic (or safe, for that matter) to expect a person with substantial emotional distress and instability to be “responsible” for his or her therapy (and life)? Is there a danger of the therapist’s taking inadequate precautions or not aggressively intervening in life-threatening situations? Do suicidal individuals, in fact, possess the inner resources to apply within a self-help framework, or is this expecting too much and setting the client up for failure? Many questions such as these can only be answered through further clinical experience and systematic research. However, it should be noted that the dangers may lie more in the two models at the extremes rather than in the intermediate collaborative model. In other words, the therapist operating from the Therapist Responsibility Model is more apt to do too much, and perhaps over-utilize involuntary commitment, while the sins of omission (e.g., failure to intervene when needed) are more likely to be committed from within the model at the other extreme (the Client Responsibility Model). The collaborative model would appear to represent a more balanced approach, in which the relative responsibility of each party (therapist and client) can be viewed as flexible, changing with the circumstances. Viewed from another angle, a collaborative, self-help model encourages utilization of numerous procedures that have been shown empirically to be beneficial in a variety of contexts. These empirically-supported procedures make prominent use of the client’s own energies as well as the therapist’s. Their success with other clinical populations makes it reasonable to predict their effectiveness with the suicidal client as well. Preliminary results with suicidal populations (e.g., Linehan, 1993; Rudd, Joiner, and Rajab, 2001) suggest that further exploration is warranted.


ph238-jocp-476180

December 8, 2003

56

21:0


Ellis

REFERENCES Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression. New York: Guilford. Beck, J. S. (1995). Cognitive therapy: Basics and beyond. New York: Guilford. Burns, D. D. (1999). The feeling good handbook (2nd ed.). New York: Viking Penguin. Chambless, D. L., Sanderson, W. C., Shoham, V., Johnson, S. B., Pope, K. S., Crits-Christoph, P., Baker, M., Johnson, B., Woody, S. R., Sue, S., Beutler, L., Williams, D. A., & McCurry, S. (1996). An update on empirically validated therapies. The clinical psychologist, 49, 5–18. Chemtob, C., Hamada, R., Bauer, G., Kinney, B., & Torigoe, R. (1988). Patients’ suicides: Frequency and impact on psychiatrists. American Journal of Psychiatry, 145, 224–228. Chiles, J. A & Strosahl, K. D. (1995). The suicidal patient: Principles of assessment, treatment, and case management. Washington, DC: American Psychiatric Press. Clum, G. A., & Lerner, M. (1990). A problem solving approach to treating individuals at risk for suicide. In Lester, D. (Ed.). Current concepts of suicide. Philadelphia, PA: The Charles Press. Deutsch, D. J. (1984). Self-reported sources of stress among psychotherapists. Professional Psychology: Research and Practice, 15, 833–845. D’Zurilla, T. J., & Nezu, A. M. (2001). Problem-solving therapies. In Dobson, K. S. (Ed.). Handbook of cognitive-behavioral therapies (2nd ed.) (pp. 211–245). New York: Guilford. DeReubeis, R. J., Tang, T. Z., and Beck, A. T. (2001). Cognitive therapy. In Dobson, K. S. (Ed.). Handbook of cognitive-behavioral therapies (2nd ed.) (pp. 349–392). New York: Guilford. Ellis, T. E. (2001). Psychotherapy with suicidal patients. In D. Lester (Ed.). Suicide prevention: Resources for the millennium (pp. 129–151). Philadelphia: Brunner-Routledge. Ellis, T. E., & Newman, C. F. (1996). Choosing to live: How to defeat suicide through cognitive therapy. Oakland, CA: New Harbinger. Greenberger, D., & Padesky, C. A. (1995). Mind over mood. New York: Guilford. Hendin, H. (1995). Suicide in America: New and expanded edition. New York: Norton. Jobes, D. A. (2000). Collaborating to prevent suicide: A clinical-research perspective. Suicide and Life-Threatening Behavior, 30, 8–17. Kazantzis, N., & Lampropoulos, G. K. (2002). Reflecting on homework in psychotherapy: What can we conclude from research and experience? Journal of Clinical Psychology, 58, 577–585. Linehan, M. M. (1993). Cognitive-behavioral treatment of borderline personality disorder. New York: Guilford. Linehan, M. M., Goodstein, J. L., Nielsen, S. L., & Chiles, J. A. (1983). Reasons for staying alive when you’re thinking of killing yourself: The Reasons for Living Inventory. Journal of Consulting and Clinical Psychology, 51, 276–286. Maltsberger, J. T., & Buie, D. H. (1973). Countertransference hate in the treatment of suicidal patients. Archives of General Psychiatry, 30, 625–633. Maltsberger, J. T., & Goldblatt, M. J. (1996). Essential papers on suicide. New York: New York University Press. Michel, K., Maltsberger, J. T., Jobes, D. A., Leenars, A. A., Orbach, I., Stadler, K., Dey, P., Young, R. A., & Vlach, L. (2002). Discovering the truth in attempted suicide. American Journal of Psychotherapy, 56, 424–437. Nathan, P. E., & Gorman, J. M. (2002). A guide to treatments that work. New York: Oxford Universities Press. Norcross, J. C. (Ed.) (2002). Psychotherapy relationships that work: Therapist contributions and responsiveness to patients. New York: Oxford. Norcross, J. C., Santrock, J. W., Campbell, L. F., Smith, T. P., Sommer, R., & Zuckerman, E. L. (2000). Authoritative guide to self-help resources in mental health. New York: Guilford. Rudd, M. D., Joiner, T. E., & Rajab, M. H. (2001). Treating suicidal behavior: An effective, time-limited approach. New York: Guilford. Schotte, D., & Clum, G. (1987). Problem-solving skills in suicidal psychiatric patients. Journal of Consulting and Clinical Psychology, 55, 49–54. Straub, R. O. (2001). Health psychology. New York: Worth.


ph238-jocp-476180


December 8, 2003

21:0


57

Tryon, G. S., & Winograd, G. (2002). Goal consensus and collaboration. In J.C. Norcross (Ed.). Psychotherapy relationships that work: Therapist contributions and responsiveness to patients. New York: Oxford. Wampold, B. (2001). The great psychotherapy debate: Models, methods, and findings. Mahwah, NJ: Lawrence Erlbaum Associates. Werth, J. L. (1996). Rational suicide? Implications for mental health professionals. Washington, DC: Taylor and Francis.