THE EFFICACY OF NO-SUICIDE CONTRACTS WITH CLIENTS IN COUNSELING ON AN OUTPATIENT BASIS Except where reference is made to the work of others, the work described in this dissertation is my own or was done in collaboration with my advisory committee. This dissertation does not include property or classified information. _______________________ Mary L. Bartlett Certificate of Approval: John Dagley Jamie Carney, Chair Assistant Professor Professor Counselor Education, Counseling Counselor Education, Counseling Psychology, and School Psychology Psychology, and School Psychology Jill Salisbury-Glennon Stephen McFarland Associate Professor Dean Educational FLT Graduate School THE EFFICACY OF NO-SUICIDE CONTRACTS WITH CLIENTS IN COUNSELING ON AN OUTPATIENT BASIS Mary L. Bartlett A Dissertation Submitted to the Graduate Faculty of Auburn University in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy Auburn, Alabama August 7, 2006 iii THE EFFICACY OF NO-SUICIDE CONTRACTS WITH CLIENTS IN COUNSELING ON AN OUTPATIENT BASIS Mary L. Bartlett Permission is granted to Auburn University to make copies of this dissertation at its discretion, upon requests of individuals or institutions and at their expense. The author reserves all publication rights. Signature of Author Date of Graduation iv VITA Mary L. Bartlett, daughter of Dennis Ratajczyk and Sharon (Inkmann) Ratajczyk, was born November 20, 1967 in Milwaukee, Wisconsin. She graduated from Pulaski High School in Milwaukee, Wisconsin in 1985. She received a Bachelor of Arts degree in Communication and Mass Communication from the University of Wisconsin-Milwaukee in December of 1990, and a Master of Arts degree in Counseling and Personnel Services from the University of Maryland?College Park in August of 1997. She began Auburn University?s doctoral program in Counseling Education, Counseling Psychology, and School Psychology in August of 2003. She is a Licensed Professional Counselor in Alabama, Maryland, and Virginia; a Certified Alabama Counseling Supervisor; a Nationally Certified Counselor; a Certified Family Life Educator; and a Therapist at GrandView Behavioral Health Centers in Montgomery, Alabama. v DISSERTATION ABSTRACT THE EFFICACY OF NO-SUICIDE CONTRACTS WITH CLIENTS IN COUNSELING ON AN OUTPATIENT BASIS Mary L. Bartlett Ph.D., Auburn University, August 7, 2006 (M.A., University of Maryland?College Park, 1997) (B.A., University of Wisconsin-Milwaukee, 1990) 139 Typed Pages Directed by Jamie Carney Clients at two outpatient facilities who self-identified as having expressed suicidal thoughts or engaged in suicidal behaviors during some point in their counseling treatment were invited to participate in a survey to assess their perception of the relative effectiveness of 22 suicidal behavior treatment methods in order to determine the perceived efficacy of no-suicide contracting as compared with other treatment methods. Sixty-six participants anonymously completed surveys to provide demographic data and to indicate their perception of the relative effectiveness of the treatment methods. Results of multiple regression analysis and criterion-coding techniques indicated that the use of the no-suicide contract was perceived by the participants of this study to be the least vi effective treatment method of those assessed. Seven of the treatment methods assessed, including the use of medication, discussion of contributing stress factors, improvement in lifestyle health, increase in the number of appointments, open discussion of suicidal thoughts, improvement in problem-solving skills, and increase in social activities formed a statistically significant cluster of perceived highly effective treatment methods. No positive correlation between the demographic factors of age, gender, time in treatment, or number of suicide attempts and the perceived relative effectiveness of the treatment methods assessed was found. Limitations of the study, comparisons with similar studies, implications of the results, and future research directions are noted. vii ACKNOWLEDGEMENTS First and foremost, I would like to thank my husband Wayne H. Bartlett for his selfless and unwavering support throughout all my educational pursuits. His enthusiasm for life and commitment to education have inspired and carried me through to this highest achievement. Many thanks are extended to Dr. Jamie Carney for serving as my advisor and chair. Her experience, can-do attitude, humor, and hospitality made my doctoral experience meaningful. My thanks are also extended to my committee members Dr. John Dagley and Dr. Jill Salisbury-Glennon who guided and supported me through the process, as well as to Dr. Carole Zugazaga for serving as my outside reader. Thanks to the many faculty, administrative staff, and library specialists who all had a part in contributing to my success and completion of the Counselor Education program; especially Altamese Stroud-Hill who formatted this enormous project. I am grateful to Dr. Jeane B. Lee who not only encouraged me to pursue this degree, but was patient and kind as I did. Thank you for believing in my potential to fulfill this endeavor. Additionally, support from various family members, colleagues, and friends were essential to this achievement. In particular I would like to thank Dr. Chieko Koyama who spent hours proofing my work, Dr. Donna Starkey who answered my countless questions about the process, as well as the dedicated staff and brave clients of viii GrandView Behavioral Health Centers--without whom this research would not have been possible. Finally, this dissertation is dedicated to the memory of D. B. whose suicide intensified my interest in the research and education of counselors in the field of suicidology. He always reminded me that a little compassion goes a long way and that suicide is an unfortunate reality of the counseling profession. My experience with him will forever impact and guide my work. ix Style manual used: Publication Manual of the American Psychological Association, Fifth edition (2001) Computer software used: Microsoft Word (2001), Statistical Package for the Social Sciences (SPSS) 11.0 (2001) x TABLE OF CONTENT LIST OF TABLES..................................................................................................... xii LIST OF FIGURES ................................................................................................... xiv I. PURPOSE AND ORGANIZATION............................................................. 1 Introduction.................................................................................................... 1 Rationale ........................................................................................................ 1 Significance of the Study............................................................................... 5 Purpose of the Study ...................................................................................... 6 Research Questions........................................................................................ 7 Operational Definitions.................................................................................. 7 II. REVIEW OF LITERATURE ........................................................................ 9 Introduction.................................................................................................... 9 Historical Development of No-Suicide Contracts ......................................... 9 Movement Away from No-Suicide Contracts ............................................... 28 Alternative Approaches ................................................................................. 39 Current Status of No-Suicide Contracts......................................................... 51 Summary........................................................................................................ 57 III. METHODS AND PROCEDURES ............................................................... 59 Introduction.................................................................................................... 59 Research Questions........................................................................................ 59 Participants..................................................................................................... 60 Procedures...................................................................................................... 60 Measures ........................................................................................................ 62 Data Analysis................................................................................................. 63 IV. RESULTS ...................................................................................................... 64 Introduction.................................................................................................... 64 Validity and Reliability.................................................................................. 65 Demographics ................................................................................................ 66 xi Research Question One.................................................................................. 71 Research Question Two ................................................................................. 72 Research Question Three ............................................................................... 74 Highest Rated Effective Treatment Method: Medication.................. 75 Second Highest Effective Rated Treatment Method: Discuss Contributing Stress Factors ............................................ 76 Third Highest Rated Effective Treatment Method: Improvement in Lifestyle Health (Diet, Sleep, Etc.) ................... 77 Fourth Highest Rated Effective Treatment Method: Increase Number of Appointments .............................................. 78 Fifth Highest Rated Effective Treatment Method: Openly Discuss Suicidal Thoughts .............................................. 79 Summary........................................................................................................ 81 V. DISCUSSION, LIMITATIONS, RECOMMENDATIONS, AND IMPLICATIONS.................................................................................. 82 Introduction.................................................................................................... 82 Discussion...................................................................................................... 82 Limitations of This Study .............................................................................. 89 Recommendations for Future Research......................................................... 92 Implications for the Field of Counseling and Training.................................. 94 Summary........................................................................................................ 97 REFERENCES .......................................................................................................... 99 APPENDICES ........................................................................................................... 113 Appendix A?IRB Approval Letter............................................................... 114 Appendix B?Letter of Agreement ............................................................... 116 Appendix C?Script for Receptionist............................................................ 118 Appendix D?Informed Consent Letter ........................................................ 120 Appendix E?Demographics and Treatment Intervention Effectiveness Scale ................................................................ 122 Appendix F?No-Suicide Contract Survey ................................................... 126 xii LIST OF TABLES Table 1: Demographic Information of all Participants ......................................... 67 Table 2: Treatment Information of all Participants................................................ 68 Table 3: Responses to Demographics and Treatment Intervention Effectiveness Scale Question 11.............................................................. 73 Table 4: Summary of Preliminary Multiple Regression Correlation Assessment on Top Five Treatment Methods.......................................... 74 Table 5: Summary of Preliminary Criterion-Coding Procedure on Demographic Variable Gender ................................................................ 75 Table 6: Age, Time in Treatment, and Number of Suicide Attempts on Treatment Method Medication............................................ 76 Table 7: t-Test for Gender Effect on Treatment Method Medication.................... 76 Table 8: Age, Time in Treatment, and Number of Suicide Attempts on Treatment Method Discuss Contributing Stress Factors ......................... 77 Table 9: t-Test for Gender Effect on Treatment Method Medication.................... 77 Table 10: Age, Time in Treatment, and Number of Suicide Attempts on Treatment Method Improvement in Lifestyle Health (Diet, Sleep, etc.) ..................................................................................... 78 Table 11: t-Test for Gender Effect on Treatment Method Improvement in Lifestyle Health (Diet, Sleep, etc.)................................ 78 Table 12: Age, Time in Treatment, and Number of Suicide Attempts on the Treatment Method of Increase Number of Appointments............ 79 Table 13: t-Test for Gender Effect on the Treatment Method Increase Number of Appointments ........................................................................ 79 xiii Table 14: Age, Time in Treatment, and Number of Suicide Attempts on Treatment Method Openly Discuss Suicidal Thoughts ...................... 80 Table 15: t-Test for Gender Effect on Treatment Method Openly Discuss Suicidal Thoughts....................................................................... 80 xiv LIST OF FIGURES Figure 1 Mean Responses of Demographics and Treatment Intervention Effectiveness Scale Question 11 ......................................... 72 1 I. PURPOSE AND ORGANIZATION Introduction This study explored the practice of using a no-suicide contract within the mental health profession as a technique to prevent suicide by clients. An analysis was conducted to determine the rank order of treatment option effectiveness assigned by the participants and to determine whether a relationship existed between selected demographic factors and the rank ordering. The purpose was to investigate whether evidence existed to validate the use of no-suicide contracts in providing adequate protection to suicidal clients and success in avoiding suicidal actions. More specifically, the purpose of the present study was to examine information obtained from clients in counseling and psychiatric care settings on an outpatient basis in an effort to determine if their experience with no-suicide contracts was helpful in reducing or preventing suicidal thoughts and behaviors, or whether alternate approaches were more helpful. Rationale Suicide is at an all-time epidemic rate in the United States. In 2003, there were 31,484 suicides in the United States, or one suicide every 17 minutes each day. More people die by suicide each year than by homicide, and suicide is ranked as the eleventh cause of death among Americans; homicide is ranked fourteenth (American Association 2 of Suicidology, 2006; Suicide Prevention Action Network USA, 2006). Suicide significantly impacts youth in this country and is the third leading cause of death for young adults between the ages of 15 and 24 (American Association of Suicidology, 2006; Centers for Disease Control, 2006). Youth suicide has risen more than 300 percent over the last forty years (Capuzzi, 2002; The Jason Foundation, 2006). Suicide rates increase with age and are the highest among Caucasian elderly men (American Association of Suicidology, 2006). These national averages apply to the state of Alabama as well. Suicide rates exceed homicide rates annually; it is the eleventh leading cause of death overall and third among youth ages 15 and 24. In a survey of Alabama youth, approximately one in twelve high school students reported having attempted suicide (Alabama Department of Public Health, 2006). As these numbers continue to increase annually, efforts to reduce suicide rates in the United States continue. Part of the effort involves identifying efficacious and ethical treatment techniques. The no-suicide contract is a technique that is commonly used by counselors treating clients for suicidal behaviors (Bongar, 1991; Weiss, 2001). The technique is referred to by many names, including anti-suicide promise, the suicide prevention contract, the binding promise, the no-suicide decision, a commitment for safety, an alliance for safety, and the no-harm agreement (Farrow, 2002; Kelly & Knudson, 2000; Maher & Bongar, 1993; Miller, Jacobs, & Gutheil, 1998; Range et al., 2002). It is a therapeutic agreement negotiated between the client and counselor for the client to agree not to act on any thoughts or impulses which might harm or cause the death of the client, usually identifying a specified time and contingencies in case the client becomes unable 3 to keep the agreement (Assey, 1985; Buelow & Range, 2001; Hipple & Cimbolic, 1979; Miller, 1999; Range et al., 2002; Twimane, 1981). Despite the popularity of this technique, there appears to be little empirical evidence to support the effectiveness of these contracts (Davidson, Wagner, & Range, 1995; Kelly & Knudson, 2000; Maltsberger, 1991; Stanford, Goetz, & Bloom, 1994; Weiss, 2001). This applies not only to practices within the United States, but also to those in other countries (Pfaff, Acres, & Wilson, 1999; Takahashi, 1993). Miller et al. (1998) found both in literature and by surveying clinicians that the use of no-suicide contracts seemed to be based on beliefs regarding their effectiveness rather than on objective data or formal training. Additionally, the no-suicide contract is not a legal document and will not exculpate a practitioner from malpractice liability if the client does commit suicide, yet it continues to be used (Clark & Kerkhof, 1993; Range et al., 2002; Simon, 1999; Weiss, 2001). The use of treatments including no-suicide contracts are supposed to be absent of a negative impact. Counselors tend to use no-suicide contracts because of the generally negative perception of suicide in our culture and the shared cultural meaning of suicide in Western society that suicide is shameful, and a sign of mental illness and weakness. Since the no-suicide contract is not seen as directly harmful, its use may be a step taken to reduce the potential for the client to commit suicide, even if the connection between the corrective act and the result is not well understood (Counts, 1991; Lee & Bartlett, 2005; Mothersole, 1996). Suicide of a client is the most common and most challenging of clinical emergencies for mental health professionals (Ewalt, 1967; Knapp & VandeCreek, 1983; 4 Shein, 1976), is consistently rated by counselors as a highly stressful experience (Deutsch, 1984; Farber, 1983), has a significant emotional impact on the treating clinician (Chemtob, Hamada, Bauer, Torigoe & Kinney, 1988; Knapp & VandeCreek, 1983), and has become a frequent basis for malpractice suits against counselors over the past fifteen years (Berman & Cohen-Sandler, 1983; Bongar, 1989; Jobes & Berman, 1993; Knapp & VandeCreek, 1983; Reid, 2004). Despite these concerns, most professionals working with suicidal clients are under-trained in the management of suicidal clients (Farrow, 2002; Jobes & Berman, 1993; Maltsberger, 1991; Neimeyer, 2000; Neimeyer & Pfeiffer, 1994; Range et al., 2002; Shein, 1976). In the absence of effective training, counselors may rely on this technique which ?has acquired a status akin to folklore knowledge; widely known and accepted, yet rarely questioned or tested? (Mahrer, 1993, p. 12). Likewise, Farrow (2002) indicated that knowledge and a proper rationale for how and why to use the no-suicide contract as an assessment tool is often passed on in an ad hoc manner rather than for reasons with sound underpinnings. In addition to the fact that there is little evidence to support the effectiveness of no-suicide contracts, it is believed that their use may actually interfere with an effective assessment of the client?s likelihood and ability to commit suicide. If no-suicide contracts are used in lieu of a more effective treatment that has empirical support, the client may not receive adequate care and the desired outcome of protection may be limited (Farrow, 2002; Farrow, Simpson, & Warren, 2002; Miller et al., 1998; Mothersole, 1996; Stanford et al., 1994). Techniques that have not been empirically demonstrated, such as the use of the no-suicide contract, must be used cautiously, if at all (Farrow et al., 2002; Weiss, 2001). Therefore, no-suicide contracting without a thorough assessment of risk is 5 inadvisable (Farrow et al., 2002; Goldblatt, 1994; Weiss, 2001). In our increasingly litigious society, counselors are better served by demonstrating that a comprehensive plan, including an aggressive, empirically based assessment and treatment, is used rather than relying on a no-suicide contract (Clark & Kerkoff, 1993; Goldblatt, 1994; Jobes & Berman, 1993). Research focusing on the efficacy of the no-suicide contract at preventing suicidal behaviors has been conducted in various populations, including high school adolescents (Hennig, Crabtree, & Baum, 1998), psychiatrically hospitalized adolescents (Jones, O?Brien, & McMahon, 1993), college students (Buelow & Range, 2001), and adults in psychiatric inpatient units (Busch, Clark, Fawcett & Kravitz, 1993; Drew, 2001; Hawton, 2000). However, inconsistent results were found in these studies due to various methodological limitations, indicating that further research is necessary to demonstrate the efficacy of no-suicide contracts (Farrow, 2002). There remains a paucity of research into the efficacy of no-suicide contracts with clients in an outpatient setting (Farrow et al., 2002; Hawton & van Heeringen, 2000; Rudd, Joiner, Jobes & King, 1999). This lack of research and the varying clinical opinions makes the continued use of no-suicide contracts questionable (Davis, Williams & Hayes, 2002; Bongar, 1991; Hillard, 1990; Mahrer & Bongar, 1993; Miller et al., 1998). Significance of the Study This study obtained empirical evidence in an effort to determine the effectiveness of using no-suicide contracts with clients in counseling settings on an outpatient basis. This study helped to establish whether no-suicide contracts should continue to be used as 6 a mechanism to assist clients who are suicidal within the outpatient setting. Obtaining data to determine whether this historically and widely used technique has efficacy in this setting is the first step in determining its value in the helping professions. If data suggests that no-suicide contracts are not demonstrated to be effective, or as effective as currently believed, perhaps counseling professionals should stop using no-suicide contracts and begin identifying and using a more comprehensive approach that may ensure higher levels of client safety and counselor protection. In working with this high-risk client group, the most effective treatment should be provided; therefore the perceptions of clients are important and the outcome of this study adds empirical evidence to the validity of the use and the effectiveness that the no-suicide contract has in relation to other methods. Purpose of the Study The purpose of this study was to examine how clients rated the effectiveness of no-suicide contracts; how clients rated the effectiveness of other suicide prevention interventions used; how no-suicide contracts compared to other interventions; and whether there was a relationship between the ratings of effectiveness of suicidal interventions and specific treatment variables such as gender, age, time in treatment, and number of attempts. 7 Research Questions It was expected that clients would perceive the use of a no-suicide contract as less effective when compared to other methods that were used to prevent or intervene in their suicidal experiences. Three questions were explored in this study: 1. How did clients rate the effectiveness of no-suicide contracts? 2. How did clients rate the effectiveness of other interventions? 3. Was there a relationship between ratings of effectiveness of suicidal intervention methods and specific treatment variables (such as gender, age, time in treatment, and number of attempts)? Operational Definitions The term no-suicide contract is operationally defined throughout literature as an explicit verbal or written statement of intent in which the client agrees to refrain from causing self-harm for a designated period of time; the no-suicide contract usually includes contingencies in case the client experiences self-harm urges so that the suicidal behavior may be avoided. The no-suicide contract constitutes the client?s promise of safety in the context of suicidal thoughts or behaviors (Assey, 1985; Hipple & Cimbolic, 1979; Maher & Bongar, 1993; Range et al., 2002; Twimane, 1981). Additional terms on this topic are used throughout literature, but are not specifically defined. For the purposes of this study, the following terms were defined as follows: Suicidal thoughts was defined as developed intentions or plans to commit harm to one?s self or to commit suicide. Suicidal behaviors was defined as instances when a person takes action to harm or kill him or herself. Outpatient clients was defined as individuals who are currently seeing a 8 counselor or psychiatrist at a counseling agency on a routine basis and are not hospitalized. Other interventions was defined as counseling techniques and approaches designed and used to assist an outpatient client to manage suicidal thoughts or prevent an outpatient client from following through on suicidal behaviors or both. While there are many mental health professional titles, including clinician, counselor, therapist, psychotherapist, and psychologist, for the purpose of this research, counselor was selected as the term used to identify mental health professionals providing counseling services to clients on an outpatient basis. 9 II. REVIEW OF LITERATURE Introduction This chapter describes the historical development of the no-suicide counseling technique, and various studies that have been conducted in an attempt to determine the efficacy of no-suicide contracts. A review of the literature describes the recent movement away from their use, and alternative approaches are examined. Finally, the current status of no-suicide contracts and the concerns surrounding their continued use are presented. Historical Development of No-Suicide Contracts The proper assessment and treatment of suicidal clients has long been researched and documented (Ewalt, 1967; Schwartz & Errera, 1963; Ungerleider, 1960; Whitley & Denison, 1963). While referrals to inpatient units were appropriate and necessary in order to protect clients who were a clear and immediate threat to themselves (Drew, 1999; Kreitman, 1986; Robins, Murphy, Wilkinson, Gassner & Kayes, 1996), it was also recognized early on, that mental health providers need to be trained to understand how to effectively treat clients whose conditions do not warrant such a referral (Greenblatt, Moore, Albert & Solomon, 1963). It has been suggested that hospitals could actually provide more of an opportunity for clients to commit suicide and therefore, should not be relied on too heavily (Ewalt, 1967). This observation may have been what led to the 10 search for practical techniques in assessing and making referral decisions. The first identified reference regarding the concept of contracting with suicidal clients appeared in psychiatric literature in 1967 when Ewalt stated ?In no instance should the person be cut off, nor should he be asked to promise not to commit suicide? (p. 1183). He went on to offer a list of suggestions a practitioner could use both in keeping the client out of the hospital and ensuring client safety. These suggestions included making special arrangements with the suicidal client so that the counselor could be reached by telephone, at home if necessary, and clarifying when the counselor could not be reached this way. Another suggestion was to tell the client when the counselor would be out of town and how he or she could be reached; in doing so the client would feel that the counselor was genuinely concerned about the client?s problems. It was suggested that every suicide threat should be taken seriously, even if the counselor saw the threat as manipulative, until a strong enough relationship was fostered so that the counselor could make an adequate determination for care, and the client believed the counselor had an adequate appreciation for what he or she was going through. These sorts of arrangements were used regularly throughout the mental health profession as the appropriate method of working with suicidal clients until 1973, when Drye, Goulding & Goulding presented data which supported and encouraged the use of no-suicide contracts. The no-suicide contract was proposed as part of an assessment process whereby suicidal clients were asked to make the following statement, ?No matter what happens, I will not kill myself, accidentally or on purpose, at any time? (Drye et al., 1973, p. 172). Clients were then asked to describe their internal responses to the statement. If the clients 11 reported feeling confidence in the statement they made, without qualifying it, it was suggested that the evaluator could dismiss suicide as a management problem. If clients objected to or qualified the statement about suicide, then it was believed they were at risk. Drye et al., (1973) claimed that over the course of five years, 600 patients made no-suicide decisions and were successful in keeping the agreement, although two clients refused to participate, changed providers, and did commit suicide. Questionnaires about use of the practice were sent to 31 therapist trainees who reported on 609 clients; of those clients, 266 were considered seriously suicidal. During the course of the study, two clients died: one died in an automobile accident and one overdosed on sleeping pills. Of the same 31 therapist trainees who did not use no-suicide contracts with clients, 20 suicides or serious attempts at suicide were reported. At first glance the results of this study suggest that the use of no-suicide contracts was successful; however, the study was informal, no information on the reliability or validity of the questionnaire used was reported, superficial information about how participants were chosen was presented, and the time frame of the study was never reported, all of which indicates serious flaws in the study (Davidson et al., 1995; Drye et al., 1973; Stanford et al., 1994). It appeared that the authors intended for the no-suicide contract to be used for assessment purposes only, and that it was not intended to be used in its present form today. Despite the fact that inadequate information made statistical analysis of the data impossible and that the efficacy of the technique was still in question, this report was pivotal in the acceptance of the no-suicide contract, and the contracting process began to find its way into regular 12 practice among several helping professions (Clark & Kerkoff, 1993; Davidson et al., 1995; Lee & Bartlett, 2005; Standford et al., 1994). Contemporaneously with Drye et al., patient contracting appeared in nursing literature beginning in the early 1970s. Contract negotiation with patients and clients was developed on the principles of environmental contingencies for the purpose of involving clients in their treatment. Contingency contracts were used as a means of facilitating learning and to change behavior, which evolved from early behavior modification treatment programs. Also widely reported were treatment contracts. These contracts differed from contingency contracts in that they included treatment goals, a time limit, treatment methods, personnel to be involved, and the parameters of client involvement, concluding with the contract being signed by the client and counseling personnel (Boehm, 1989; Rosen, 1978). Preliminary work by Loomis (1982;1985), distinguished the treatment contract as being derived from a transactional analysis framework, whereas behavioral and reinforcement theory were the conceptual frameworks of contingency contracting. As the nursing profession began to integrate these contracts into practice, differences in treatment contracts were seen throughout nursing literature at this time. A health-care contract was described for use in a primary care setting (Hayes & Davis, 1980), behavioral contracts were being used with borderline personality disordered clients (McEnany & Tescher, 1985; Selzer, Koenigsberg & Kernberg, 1987), and for use with patients suffering with specific medical conditions (Thompson & Willis, 1982; White, 1986). In examining the various studies regarding theoretical frameworks guiding the use of these two kinds of contracts, Boehm (1989) pointed out that confusion and a 13 lack of organized theoretical framework on which to base the use of a contract in working with clients, required future study. Nonetheless, these contracts continued to be widely used in the nursing profession despite not being tested with any scientific rigor (Assey, 1985; Drye et al., 1973; Jones et al., 1993; Loomis, 1985; Silverman, Berman, Bongar, Litman & Maris, 1994; Smith & Bope, 1986). Given that various contracting methods were a practice already in use by nurses, it is not surprising that no-suicide contracts were introduced into the nursing literature as an option to use in working with suicidal clients. Twimane (1981) discussed how nurses learn about lethality assessment during psychiatric nursing courses, and questioned whether a standard lethality assessment was enough in helping a nurse to decide if a person was at risk for suicide. She referenced Drye, et al. (1973) and indicated that their tool, the no-suicide contract, could help nurses further assess suicide potential by providing data about the degree to which a person is at risk and how long a client could be trusted not to take his or her life. The article explained what no-suicide contracts were and how to use them with clients. While Twimane made it clear that the no-suicide contract was not a replacement for a standard lethality assessment, she did support the use of the no-suicide contract as a tool to help gather more information. No consideration or discussion was included regarding demonstrated efficacy of the tool other than the author?s personal testimony about its effectiveness, and only two references were cited to support her contention. Maluccio and Marlow (1974) indicated that the client-worker contract was a dynamic tool that could contribute substantially to effective outcomes in social work 14 intervention. These authors suggested that contracts could be used to clarify objectives and encourage clients to participate in their intervention process more fully. They reasoned that the contract assists the client and social worker to explore and reach agreement on treatment goals, gives both the practitioner and client a sense of immediate involvement, signifies mutual commitment to assume responsibility, and provides a baseline to a periodic review of accomplishments while assessing progress and re- examining conditions of the contract. Despite these identified benefits, the authors point out that contracts in social work are inadequately formulated and incompletely incorporated into practice, as well as having minimal clarification of theoretical foundation, delineation of use, and tests of validity. Maluccio and Marlow indicate that the contract?s restricted application to social work can be a factor that contributes to a clash in perspective between a worker and client, client discontinuance, and frustration between them, yet go on to indicate that the purpose of the article is to ?stimulate interest in examining, conceptualizing, and using the contract? (p.28). While no specific mention is made regarding a no-suicide contract, the promotion of contracting between client and social worker for treatment purposes was significant. In 1979, Hipple and Cimbolic wrote a book focusing on helping counselors deal with suicide. In it, they devoted a chapter to the concept of no-suicide contracts. After describing several critical characteristics of contracts for use in the therapy process, the authors described the process of using no-suicide contracts and identified perceived benefits, including getting a client to slow down thoughts about the suicide in order to think more clearly, getting the client to postpone the act of suicide for a specified time, 15 and reinforcing that the client has ultimate responsibility for his or her own life. The chapter quotes work from Drye et al. (1973), but does not cite or list them as a reference despite multiple references to their technique. The authors state, ?Contracts are very effective during all phases of treatment? (p. 72), referring to no-suicide contracts, but this assertion is not substantiated with any empirical data to support their claim. By the 1980s, the no-suicide contract ?seemed to have become firmly established in the literature with minimal empirical base? (Stanford et al., 1994, p. 545). Assey (1985) supported and encouraged the use of the no-suicide contract, identifying three major components to the contract, including balance, expression of caring in an open manner, and reinforcement of the contract. She described at great length these three components and even provided a short vignette in which the counselor works with a client who is suicidal. The counselor asks the client to make a promise not to kill herself. Assey writes in the first person describing how effective this promise- making concept is based on her own use of the no-suicide contract, but nowhere in the article is any reference made to data obtained which empirically validates the procedure as a viable technique. In 1988, no-suicide contracting was introduced into family practice literature despite the continued lack of scientific evidence to support is usefulness (Kelly & Knudson, 2000; Pary, Lippmann, & Tobias, 1988; Smith & Bope, 1986; Standford et al., 1994; U. S. Department of Health and Human Services, 1990). Soon thereafter, no- suicide contracts received widespread acceptance in mental health literature and practice, and reliance on them continued to increase. However, due to the absence of empirical 16 support and efficacy data as well as legal and ethical factors worthy of consideration, disagreement about their value began to surface (Weiss, 2001) and literary discussion about the advantages and disadvantages of their use continued to be published. Mothersole (1996) explored a core facet of transactional analysis-based psychotherapy, which is the use of tapping into the adult ego state of a client in early stages of treatment to emphasize to the client that he or she can only change if they stay alive, rather than choosing an escape method such as suicide. By appealing to the internal adult of a client, the approach communicates that the client can take control of his or her life and whether to exist or not exist. The author explained that inviting the client to commit to staying alive is a profound intervention ?highlighting one?s individual existential choice over one?s life? (p.151), but that the intervention must be predicated on a strong therapeutic bond. If the intervention is predicated on the strong counselor and client bond, it will most likely be perceived by the client as one based on empathic understanding on the part of the provider. However, if the suggestion for a no-suicide contract is introduced by the counselor in a routine way, it may hinder the exploration of further self-destructive thoughts and feelings, and the client may agree to the contract only to satisfy the counselor and not as a result of true intent. If the no-suicide contract is predicated on the counselor and client bond it may help the client move toward a full reconsideration of redecision about his or her life. Mothersole explained how transference, over-adaptation, and distortions impact the way a client receives the request from a counselor for the client to stay alive, emphasizing the importance of the therapeutic alliance. However, Mothersole explained that his concern regarding the no- 17 suicide contract is that it may downplay the importance of having the client examine his or her existential realities related to life and death boundaries. In addition, the introduction of a no-suicide contract may stem from the cultural blind spot a counselor may have regarding his or her own views about suicide. This blind spot may be influenced by the counselor?s experience of increased anxiety, which is commonly provoked when a client introduces the concept of suicide into treatment. Nonetheless, Mothersole indicated that he supports the use of no-suicide contracts to help in the process of assessment to determine the degree to which a client is suicidal, whether the client can continue counseling on an outpatient basis, and to hold the client accountable while the client continues work on his or her self-destructive inclinations. Mothersole concluded by suggesting that when a no-suicide contract is introduced, it is helpful for the counselor to explain that the contract is not intended to stop the client from feeling suicidal, but rather to strengthen the alliance and give the client cognitive control. While this article offers a thoughtful consideration of how a no-suicide contract fits into transactional-analysis of psychotherapy, Mothersole?s conclusions are premised on observations of the technique rather than empirical data to support his positions. Given the high concentration of actively suicidal clients sent to psychiatric units at general hospitals, the likelihood of suicide is higher for psychiatric inpatients than for the general population (Drew, 1999; Freidman, 1989). Drew (1999) conducted a study to describe the use of no-suicide contracts in all state-licensed psychiatric inpatient settings in Ohio. The study identified contact people at 102 hospitals and resulted in a final response rate of 82% from 84 hospitals. No-suicide contracts were used by 79% of the 18 responding hospitals, primarily by nurses, with patients who were admitted for making a suicidal gesture or attempt, expressed suicidal ideation, or after an admitted patient attempted to commit suicide. Fifty-one percent of the respondents reported an incidence of 10 or more suicidal behaviors each year, despite the use of no-suicide contracts. Drew explained that this finding reinforced the ?need to more thoroughly evaluate assessment strategies and interventions used to maintain patient safety? (p. 27). The study was limited by a small respondent base, estimates of suicide behavior were relied on to minimize the burden of records review for the responders, and therapeutic interaction was not identified as a separate suicide prevention measure, so many respondents classified that intervention in the ?other? category, which impacted results. While this study confirmed that the use of no-suicide contracts are common among nurses in psychiatric inpatient settings in Ohio, Drew pointed out that a lack of guidance in an ideal contract form, best conditions for contracting, relevant characteristics of client and counselor, and the nature of the therapeutic relationship to provide a context for the contract require future research. Another study conducted to evaluate the perceptions of written no-suicide contracts with psychiatric inpatients was conducted by Davis et al. (2002). In this study, 135 psychiatric inpatients, each of whom were admitted due to suicidal danger, completed a survey instrument developed for the study. Patients responded to 23 items derived from a literature review on no-suicide contracts, using a Likert-type scale; higher scores indicated more positive endorsements of each item. The survey was given to patients near discharge from the hospital to allow retrospective consideration of the 19 advantages and disadvantages of the no-suicide contracts each of them signed during their hospitalization period. Results indicated that these patients reported positive attitudes toward written no-suicide contracts regardless of level of suicidal danger, age, gender, and presence or absence of Axis II disorders. The patients? responses suggested that they considered the contracts useful because they helped them to cognitively evaluate aspects of their suicidal crises, accept responsibility for living, and to keep their word, which was considered honorable to them. However, patients who had attempted suicide multiple times rated the therapeutic factors of no-suicide contracts less favorably. Internal limitations to this study included low inter-rater reliability checks on DSM-IV diagnoses, no structured interviews with the participants, and the fact that suicide crises and suicide history were obtained through self-reports only. An external validity concern was the fact that there was a greater distribution of younger women participants. Given that older males pose a greater risk of suicide, the results may not generalize to a larger population. Nonetheless, the authors indicate that ?The strength of this study was the collection of perceptions from actual consumers of written no-suicide agreements? (p. 63), and concluded that abandoning the use of no-suicide contracts because of overvaluing of the technique by counselors and the lack of research on them may be premature. Questions regarding the use of no-suicide contracts with clients of various ages led to a study by Davidson et al. (1995), which asked 46 licensed psychologists of a state psychological association to respond to questions regarding their beliefs and attitudes about no-suicide contracts and the effectiveness of contracting with clients of varying ages. Results indicated that clinicians believed no-suicide contracts to be appropriate with 20 adults and adolescents, and clinicians rated the use of no-suicide contracts with children ages 6-11 and 9-12 as slightly pessimistic or neutral. The respondents also rated no- suicide contracts as helpful with moderately suicidal clients, but only slightly helpful with mildly or severely suicidal clients. Furthermore, results indicated that the clinicians believed that no-suicide contracts could help a client postpone a decision to commit suicide until after a crisis has passed, help a therapist feel more in control and less anxious, but that clients are still at risk for suicide, and the use of a no-suicide contract did not reduce legal liability in the event that a client did sign a contract and then committed suicide. It was pointed out that while the respondents gave opinions about the use of no-suicide contracts with children, only one-third of the respondents had occasional experience working directly with them. This study conducted by Davidson et al, indicated that clinicians believed in using no-suicide contracts with adolescents and adults and that an advantage of the use of no-suicide contracts is that they were not seen to weaken the therapeutic relationship, and so were potentially helpful. In light of these results, it was suggested that surveying presently or previously suicidal adults and children to assess if they agree on the value of such an agreement would be an appropriate avenue for future research. Jones and O?Brien (1990) conducted a study of 39 psychiatrically hospitalized children who had attempted, talked about, or engaged in self-mutilation behaviors. Contingency contracts were developed with the children in which they would receive privileges based on meeting the terms of their agreements. A variety of contracts were used including no-suicide contracts. After the exercise, the children each completed a 21 questionnaire that assessed the efficacy of the various treatments they received, which included contracts. The children rated their experience of contracting very high in helping them to change their behaviors. However, there were numerous flaws with the study, including no random assignments, which raised the question of whether only the healthiest children agreed to participate; the hospital staff changed over time; and confidentiality of the children?s responses was not addressed, so social desirability may have impacted ratings. This study was significant in that it examined the attitudes of suicidal and potentially suicidal individuals rather than the professionals who deal with them. Another study that attempted to gather empirical data about no-suicide contracts was conducted by Beulow and Range (2001). These authors evaluated the use of three no-suicide contracts that differed in length and specificity with 112 mixed-ethnicity college students at a mid-sized university. Forty percent of the students admitted to having previous suicidal ideation, and 54% reported having some form of previous counseling. The students read each of the contracts and rated them on seven factors that might decrease suicidal desire, including use of the no-suicide contract, medication, improved job satisfaction, client and counselor relationship, fear of death, improved living situation, and stronger coping skills. The students were asked to rank order the three contracts from best to worst. The respondents rated no-suicide contracts last as a factor that would contribute to preventing suicide, and ranked the other factors more highly, with fear of death, an improved living situation, and stronger coping skills among the highest. Results also indicated that when given a choice, college students prefer a 22 specific and detailed no-suicide contract over a simple one and believed that no-suicide contracts may be helpful to some people. Despite these results, Beulow and Range commented that no-suicide contracts are only one aspect of treatment for suicidal clients, cautioned against using no-suicide contracts in lieu of an appropriately developed therapeutic relationship, and indicated that no-suicide contracts can interfere with an effective therapy process. The significance of this study is that it assessed the perceptions of the no-suicide contracts by clients who experienced suicidal ideation and gathered empirical data about the treatment method, which until then continued to be mostly conjecture. Kroll (2000) conducted a study in which 514 psychiatrists in Minnesota were sent postcard questionnaires inquiring about their practices and experiences with no-suicide contracts. Of the 267 returned responses (52% response rate) results indicated that 57% of the respondents used no-suicide contracts, yet 41% of the psychiatrists who used them had patients who committed suicide or made serious attempts after entering into a contract. These results suggested that no-suicide contracts were not universally accepted as a standard of practice among the psychiatrists, underscored the tenuousness of relying on the contract as an effective suicide prevention tool, and suggested both that no-suicide contracts had limited efficacy and that more data was needed to determine their effectiveness in preventing suicide. The number of articles that were published in which advantages and disadvantages of no-suicide contracts were examined continued to grow. One advantage consistently discussed within the literature was that the no-suicide contract served to 23 develop the therapeutic relationship (Assey, 1985; Davidson et al., 1995; Drew, 1999; Fine & Sansome, 1990; Hipple & Cimbolic, 1979; Miller, 1999; Mothersole, 1996; Range et al, 2002; Selzer et al., 1987; Simon, 1991; Stanford et al., 1994; Weiss, 2001). Stanford et al. (1994) explained that a mental health professional often has no prior experience or relationship with a suicidal client, and the no-suicide contract can be used to initiate a therapeutic alliance. In these cases, these researchers urged ?if the no- suicide contract is presented as an open expression of caring, this beginning alliance could be strengthened? (p. 346). Mothersole (1996) claimed that the therapeutic power of a no-suicide contract must be predicated on a strong therapeutic bond between the counselor and client. When a strong bond exists, the invitation to contract to stay alive ?is likely to be experienced by clients as arising from their needs and from a position of empathic understanding on the part of the psychotherapist? (p. 151). In this way, the no- suicide contract could be a powerful aid to the counseling process and to the therapeutic relationship, serving as the anchor that allows for a safe exploration of destructive psyche impulses. On an interpersonal level, the no-suicide contract could help to initiate and establish a therapeutic alliance. It could serve to reinforce a suicidal client?s need for active collaboration in treatment. Clients who believe their counselor had asked them to sign a no-suicide contract out of genuine concern for their safety may perceive the counselor to be empathic. Using the no-suicide contract may strengthen their relationship and lead to positive treatment outcomes (Drew, 1999, Miller, 1999; Sills, 1997; Stanford et al., 1994). 24 Another reported advantage of using the no-suicide contract was to gather diagnostic information regarding severity (Drye et al., 1973; Stanford et al., 1994; Range et al, 2002; Twimane, 1981; Weiss, 2001). Drye et al. (1973) originally conceived the use of a no-suicide contract as a mechanism to assess a person?s likelihood to commit suicide. If the client could not make a firm statement that he or she would not commit suicide, the contention was that the person was more likely at risk to follow through on the suicidal fantasies. The degree to which a person qualified his or her response would give the counseling professional a method to classify the severity of risk. Stanford et al. (1994) also believed that the no-suicide contract could be useful for diagnostic purposes. By asking a potentially suicidal client in a crisis situation to make a statement to not commit suicide, in both the verbal content and in the affect and body language in which the reply was given, the professional could make a more concrete clinical decision regarding the votility of the client. The no-suicide contract could further assist in uncovering specific issues that precipitated the suicidal thoughts which these authors stated ?provides insight to the clinician into the severity or reversibility of the precipitating stressors? (p. 346). As a diagnostic tool, the no-suicide contract could be successful in helping to explore various aspects of suicidality (Stanford et al., 1994). Twimane (1981) suggested that diagnostically, the no-suicide contract was not a replacement for a lethality assessment, but that it could be used as a tool to provide more information and a larger margin for safety. A third proposed advantage of using the no-suicide contract was that it provided time for the suicidal person to more thoroughly discuss and evaluate a choice of suicide 25 (Davidson et al., 1995; Getz et al., 1983; Hipple & Cimbolic, 1979; Range et al., 2002; Weiss, 2001). According to Hipple and Cimbolic, helping a client to slow down his or her thought processes could lead that person to clearer thinking, and help to change the emotional state of the client. The confused thoughts that a suicidal client experiences can be difficult to assimilate; through the process of developing a clear contract, in addition to helping slow down the person?s thought processes, the counselor helps the client to articulate his or her feelings, which requires the client to organize thoughts and feelings (1979). The behavioral manipulation function of the no-suicide contract gives clients time to intellectually and emotionally understand the true essence, significance, and permanence of his or her choice regarding suicide (Getz et al., 1983; Stanford et al., 1994). Selzer et al. (1987) believed that when a client was resistant to participate in a no- suicide contract, the counselor could use confrontation to explore and to clarify the resistance. It was further concluded that the no-suicide contract can have an important role in helping clients to hold themselves together and to remain in counseling while working through self-destructive periods (Davidson et al., 1995; Mothersole, 1996; Range et al., 2002). The time, effort, and commitment that go into a no-suicide contract can give a client the space necessary to develop positive coping and action plans. It can serve as a passage through a challenging period while clients work out negative emotions, difficult circumstances, and various therapeutic issues. As a result, some clients are more willing and able to discuss and work on deeper issues than if the topic of suicide had not been addressed directly (Lee & Bartlett, 2005). The process of having a client sign a no-suicide 26 contract seems to naturally lend itself to the process of discussion of the issue of suicide at greater length, which can lead to a decrease in the desire to follow through with the act (Jacobs, 1992; Mothersole, 1996). The introduction and use of a no-suicide contract is reported to reduce both client and counselor anxiety about the choice of suicide (Drye et al., 1973; Farrow, 2002; Hipple & Cimbolic, 1979; Mahrer & Bongar, 1993; Simon, 1991; Stanford et al., 1994). Reaffirming that the counselor and client are working on a common goal can have a calming effect for both parties. Both the counselor and client may experience a sense of relief once a no-suicide contract is completed because the immediate pressure of the threat of death has been removed (Miller et al., 1998). Drye et al. indicated that by using a no-suicide contract, the anxiety of the evaluator is decreased. The fact that the evaluator collaborates with and shares the evaluation task with the client enables the client to share in the burden of assessment and reinforces that the client must assume responsibility for his or her choices (1973). It can be difficult for mental health providers to recognize that they are only part of the equation in helping the client to choose living over dying; in fact, it is the client who has the ultimate control. When a client agrees to a no-suicide contract, the message of shared responsibility is reaffirmed for both parties, and the exercise helps to reduce the counselor?s feelings of frustration, anger, and resentment in dealing with the suicidal client. When those feelings exist and are not addressed or recognized, they can impair the clinical decisions of the evaluator?the no-suicide contract can help to keep these emotions in check (Mothersole, 1996). 27 Using a no-suicide contract may help clients to feel more in control because they will have participated in setting their own terms in the contract. This feeling of control can contribute to continued effective counseling because the client may take more responsibility for his or her own therapy (Maluccio & Marlow, 1974; Miller, 1999; Range et al., 2002). Stanford et al. (1994) suggested that having a concrete instrument such as a no-suicide contract helps evaluators to more objectively assist the suicidal client, which helps to minimize the stress on the evaluator. While these advantages were proposed, in the absence of adequate, scientifically rigorous testing, the value of no-suicide contracts in the process of working with suicidal clients continued to remain mostly conjecture, and disadvantages discussed in the literature appeared to parallel the advantages. For example, Egan (1997) pointed out that in emergency settings there is often little opportunity to establish a therapeutic relationship, and so reliance on no-suicide contracts could be dangerous in making decisions about hospital admission. While it is believed that the therapeutic alliance forms the basis for reliance on a no-suicide contract, Simon (1999) suggested that the therapeutic alliance is a dynamic interaction and is in constant flux, therefore making reliance on it tenuable at best. The client may perceive the invitation to sign a no-suicide contract as protection for the counselor rather than out of care for the client which could negatively affect the therapeutic alliance the counselor had hoped to preserve in using it (Farrow, 2000; Miller et al., 1998; Range et al., 2002; Reid, 2004; Weiss, 2001). While the use of a no-suicide contract can decrease the anxiety level of the counselor, this altered state of anxiety may falsely reassure the counselor that the client 28 will not commit suicide, which can lead the counselor to lower his or her vigilance regarding the risk involved and inadvertently failing to complete a comprehensive assessment (Stanford et al., 1994; Miller, 1999; Reid, 2005; Weiss, 2001). No-suicide contracts may be used to alleviate counselor discomfort regarding a clinical area they are under-trained in or in which they experience counter-transference (Bongar & Harmatz, 1989; Standford et al., 1994; Miller et al., 1998; Mothersole, 1996). This can create a barrier for clients to further discuss their suicidal thoughts and feelings which is an important outlet for them (Mahrer, 1993; Miller, 1999; Miller et al., 1998). It may be these reasons and many others, that contributed to alerting professionals that use of no- suicide contracts warranted reconsideration, and a shift in literature occurred. Movement Away from No-Suicide Contracts Reid (2003) expressed his position against the use of no-suicide contracts describing his amazement at the number of staff and counselors whose decisions about client safety rely on the client?s statements, whether verbal or written, that they are not suicidal. He pointed out that suicidal patients do not always tell nurses or counselors the truth about their plans, so to rely on the testimony of a client to that fact is unsafe both for the client and the counselor. Since there are many other methods available to counselors to guide them in making clinical decisions about the care of suicidal clients they are strongly urged to consider alternate techniques rather than a no-suicide contract. Another point that Reid (2005) made was his observation that an urban myth was created that dictated counselors or hospitals must have a no-suicide contract in the chart of a client in 29 order to diminish liability in the unfortunate event that a suicidal tragedy occurs with a client. Reid (1998) identified several court cases in which various clients had been asked to sign no-suicide contracts and despite the agreements committed suicide. He claimed to have no quarrel with the utility of no-suicide contracts from a therapeutic context as a mechanism to convey concern, foster client participation in his or her treatment, and encouraging positive behavior, but questioned the reliability of no-suicide contracts and cautioned against using them to assuage concerns about real danger. The author discussed the cross examination of a doctor whose client had signed a contract but then committed suicide. Reid wrote that the plaintiff?s lawyer stated, ?She [the client] wasn?t thinking about the so-called ?contract? when she felt like her whole life was over ?. and she wasn?t thinking straight, was she, doctor?? (p. 317). Reid made the point that counselors consider the unreliability of clients when making other promises, such as to follow medication regimes, and indicated that it makes sense to view no-suicide contracts the same way, concluding, ?If a promise or a contract were sufficient, we?d [counselors] be unnecessary? (p. 318). The no-suicide contract provides little protection for counselors in suicide malpractice suits, and in fact can result in a malpractice suit if a no-suicide contract is located in the medical record of a person who committed suicide (Egan, 1997; Simon, 1999). Bongar et al., (1989) explained that compared to fifteen years ago, mental health professionals have acquired new legal duties, and through the use of the court system, 30 society as a whole began scrutinizing ?the manner in which psychotherapists exercise appropriate duty of care owed to a suicidal patient? (p. 54). In 2001, Drew conducted a study which reviewed the hospital medical records of 650 discharged inpatients to examine how the use of no-suicide contracts affected the likelihood of self-harm behavior in this setting. The charts of discharged patients who had a diagnosis of a major mood disorder, schizoaffective disorder, or schizophrenia were used, since individuals with these diagnoses have a higher risk of suicide. A five-day length of hospital stay was another factor among the charts because this period offered more of an opportunity for the development of therapeutic relationships. Levels of restriction that the patients experienced were also factored in: restrictions included the use of physical restraints; room, seclusion, and unit location restriction; and accompanied and unaccompanied privileges. Stricter restrictions were given to patients who exhibited a greater likelihood to commit suicide. Results indicated that contracting was a common practice with 33% of the patient charts reviewed, and that patients with no-suicide contracts and higher level of restrictions had a significantly higher likelihood of self-harm behaviors. In this study, the prevention of self-harm behaviors through the use of contracting was not shown. The findings in the study by Drew suggested that if no- suicide contracts are used because they are considered to have therapeutic value, then ?modifications in suicide prevention measures should be based on other methods for assessment of risk of self-harm? (p. 105). Another factor contributing to the movement away from using no-suicide contracts was a lack of consistent training in their use. Miller et al. (1998) surveyed 112 31 mental health professionals who attended a course on suicide for the Harvard Medical School faculty about the use of no-suicide contracts. It was discovered that the majority of participants, most of which were psychiatrists and doctoral-level psychologists, had never received any formal training on the topic. Formal training was defined as attending one or more lectures that described the history, use of, indications, contraindications, risks, and benefits of the contracts. Sixty-one percent of the psychiatrists and 71% of the psychologists indicated that during their own residency or internship training, they had not received any formal training on the use of no-suicide contracts, yet 61% of the psychiatrists and 83% of the psychologists indicated that they used the contracts with their suicidal patients at least half the time. Miller et al wrote, ?The results of this survey show a disjunction between the widespread use of suicide-prevention contracts and the relative absence of formal training? (p. 80). The authors went on to point out that while the use of the contracts is widespread, their use is based on a subjective belief rather than on objective data or formal training. As managed-care systems grew to reduce the cost of medical care, the number of seriously ill and seriously suicidal clients and their care on an outpatient basis increased simultaneously (Bongar, 2002; Maltsberger, 1994; Rudd & Joiner, 1998). The shift from fee-for-service to managed care resulted in an erosion of pre-existing standards of care for suicidal clients. As the number of suicidal clients being treated in outpatients settings increased, the number and availability of resources to monitor their care decreased (Bongar, 2002), and the number of malpractice claims against mental health professionals treating outpatients who committed suicide increased exponentially (Berman & Cohen- 32 Sandler, 1983; Jobes & Berman, 1993; Reid, 2004). Reid (2004) pointed out that while psychiatrists are less likely than other physicians to be sued, filed malpractice suits against psychiatrists and other mental health providers was significantly growing, and in fact was, ?the most common cause of action against mental health care professionals? (p. 1). At a time when more outpatient counselors needed to be trained to deal with suicidal clients and appropriate intervention measures, psychology training programs remained remiss in educating trainees regarding correct management of these mental health emergencies (Bernstein, Feldberg & Brown, 1991; Bongar & Harmatz, 1989; Ellis & Dickey, 1998; Rudd & Joiner, 1998). Additionally, while traditional inpatient settings had historically defined protocols for handling suicidal clients, outpatient settings did not; as a result, outpatient counselors became increasingly subject to litigation, making the establishment of such protocols even more necessary (Berman, 1990; Jobes & Berman, 1993; Rudd & Joiner, 1998). Regardless of the new constraints counselors were experiencing as a result of managed-care growth, Bongar (2002) pointed out that ?It is imperative that clinicians realize that they are legally, ethically, and professionally responsible for determining appropriate patient care? (p. 141). Farrow (2002) conducted a qualitative study that explored why nurses use no- suicide contracts when it runs against their better clinical judgment rather than relying on their own expertise. Nine registered nurses in current practice in crisis teams participated in in-depth interviews. Results consistently revealed that nurses used no-suicide contracts to protect themselves, because the use of no-suicide contracts had become widespread 33 enough to generate an expectation for their use, despite a lack of empirical evidence to justify this practice, and because an absence of more appropriate resources existed. None of the nurses could recall any formal training in the use of no-suicide contracts, and many participants mentioned that they relied on them because they had difficulty allocating time to spend with patients. This resourcing deficit compelled them to use no-suicide contracts rather than implementing what they considered to be interventions that met more appropriate standards of care. Essentially, they perceived no-suicide contracts as a stop-gap method of managing suicidal crises. Farrow concluded that in an effort to protect themselves, nurses often subjected patients to no-suicide contracts for reasons other than good practice standards. Among the many challenging aspects of working with a suicidal client on an outpatient basis is determining whether the client requires hospitalization or if the symptoms can be managed outside this more restrictive action. It has been pointed out that if a counselor makes the decision not to seek commitment for a client and the client commits suicide, the counselor may be held liable (Simon, 1988). In the case of outpatient suicide, courts typically struggle with two central issues, forseeability and causation; forseeability is defined as the degree to which the counselor could have predicted the suicide, and causation is defined as the degree to which sufficient evidence existed to permit an identification of risk of harm and whether the counselor (or the institution) exhibited appropriate care to protect the client (Bongar, 2002; Knapp & VandeCreek, 1983; Simon, 1991; VandeCreek, Knapp, & Herzog, 1987). 34 Jobes and Berman (1993) submitted that the fundamental issues of good practice rested with forseeability and reasonable care, similarly defining forseeabilty as the reasonable and comprehensive assessment of risk, and reasonable care as involving the reliable and appropriate implementation of interventions or precautions based on foreseeability. These researchers further went on to explain that the standard of care is ?idiosyncratically defined by competing experts on a case-by-case basis?, and that standards of care are differentially evaluated in relation to opinions that are developed throughout the context of any given case (p. 3). This suggests that no two cases are alike and therefore, while some practices in the assessment and treatment of suicidal clients are advisable and remain consistent, each case must be developed and the plan executed in a way that meets the specific needs of the client at risk and which evidences that foreseeability and reasonable care were considered (Jobes & Berman, 1993). Black (1979) explained that in clinical practice the term standard of care is interpreted broadly, whereas courts specify it as, ?that degree of care which a reasonably prudent person or professional should exercise in same or similar circumstances? (p. 1260). Deviations from the standard of care are referred to as negligence (Bongar et al., 1998); and negligence is ?described as doing something which he or she should not have done or omitting to do something which he or she should have done? (Simon, 1988, p. 3). Standards of care are constantly changing and there is no one agreed upon way to act in every situation; therefore, potential risks and benefits exist for every decision made. Since these circumstances are ever changing, periodic re-evaluation of assessment and treatment methods are essential given the increasingly litigious nature of our society 35 (Bongar, 2002; Jobes & Berman, 1993; Simon, 1991). The best defense against liability should a client commit suicide is for the counselor to have provided good clinical care that followed acceptable standards of practice (Bongar et al., 1998; Gutheil, 1992). In light of these circumstances and the growing complexity in working with suicidal clients, an emphasis on appropriate risk management is central to a preventative approach and to the issues of liability after the suicide of a client (Bongar et al., 1998). The debate about whether a no-suicide contract factored into what would be considered an appropriate risk management plan continued. Questions that counselors who use no-suicide contracts began asking was whether the contracts have any legal authority and whether such an agreement with a client immunized the counselor from a liability if the client subsequently attempted or committed suicide (Bongar, 1991; Simon, 1999; Fine & Sansome, 1990). There is a consensus that a no-suicide contract is not a legal document and that it provides no legal protection against possible litigation (Miller, 1999; Miller et al., 1998; Range et al., 2002; Reid, 1998; Simon, 1991; Weiss, 2001). A contract is only legally binding if the parties are viewed as competent to participate, valuable consideration is included, mutual obligation exists, and the contract is consistent with public policy. A no-suicide contract does not meet the criteria of a legally binding contract, and using it could make the counselor legally liable should a counselor fail to demonstrate that a comprehensive assessment was completed (R. E. Poundstone, personal communication, September 28, 2005). 36 A case that provided some clarification about the use of no-suicide contracts and how courts viewed them was Stepakoff v Kantar (1985). In this case, a psychiatrist had his manic-depressive client sign a pact to contact him if the client felt suicidal. On several occasions the client did contact the psychiatrist, and during one such call the psychiatrist assessed that the client was unlikely to commit suicide; however, the client did. The Massachusetts Supreme Court ruled that the psychiatrist met the required standard of care. Having a pact with the client to not commit suicide didn?t seem to affect the standard by which the decision was made; however, it was significant to note that the Court did not express an opinion either way about no-suicide contracts. What the Court seemed to place a higher consideration on was the other elements of care that the psychiatrist performed to meet the standard of care, including frequent contact and periodic assessment of the client, documented consideration of involuntary hospitalization, and the use of a substitute therapist to assist the client when the primary psychiatrist was on vacation. This decision reinforced that had the psychiatrist relied solely on the no-suicide contract as a mechanism to predict and prevent suicide rather than utilizing adequate clinical and risk benefit assessment protocols, the psychiatrist may have been at more risk of liability in this case. Simon (2004) used strong language in opposing the use of no-suicide contracts stating, ?There is little or no basis for relying on a suicide prevention contract obtained from a severely mentally ill patient. For these reasons, suicide prevention contracts are of little or no utility in emergency settings? (p. 68). The author explained that the use of a no-suicide contract in clinical treatment and planning tends to be an event, whereas 37 suicide risk assessment is an ongoing process, and that obtaining a no-suicide contract establishes that a client is at risk for suicide, not that risk has been assessed. Simon further pointed out that no standard exists that requires a written contract. He cited two cases, Olson v Molzea (1997), and Porubiansky v Emory University (1981) to explain that a no-suicide contract is an example of an exculpatory clause, which means that a person cannot enforce a contract that would relieve him or her of legal liability for harm caused by negligence. In other words, negligence cannot be contracted away. Simon concluded by explaining that a safety management plan is more appropriate than a no- suicide contract. Similarly, Gutheil (1992) explained that appropriate risk management is the core of a preventive approach to the reality of liability for suicidal clients, and reinforced that the best defense of liability is for the counselor to have provided sufficiently good and well documented clinical care that followed acceptable standards of practice. He went on to explain that the possibility of suicidal clients is a reality of the mental health profession; suicidal patients are difficult to work with and can discourage treatment providers. However, Gutheil stressed that counselors must not avoid situations such as working with suicidal clients for fear of malpractice suits, else client bases would be reduced. Instead, counselors must become educated in working with this challenging client base and use sound risk management planning, which provides safety for the client and peace of mind for the counselor. In the late 1970s a shift in case law began allowing for the filing of malpractice suits against clinicians when a client died by self-injury while under the care the treating 38 professional in both inpatient and outpatient settings. As a result of this landmark decision, standards of care for assessment, treatment, and management of suicidal clients evolved. However, standards of care remained ambiguous, and currently there is no uniform set of standards for the care of suicidal clients (Berman & Cohen-Sandler, 1982; Bongar, Maris, Berman & Litman, 1998; Bongar, Peterson, Harris, & Aissis, 1989; Wettstein, 1989). The standard of care is defined by the opinions of experts used as consultants in tort actions, and is based on mythical average practitioner practices. In reality, counselors identify and practice from their own set of standards. Case law and out-of-court settlements are most responsible for dictating standards of care for counselors, and they are usually based on claims of omission (Bongar et al., 1998). Given this fact, it is easy to understand why counselors remain eager to continue using no- suicide contracts although they are not proven to be effective. Counselors may believe that their use demonstrates action to include measures in assessing the level of client suicidality. Part of the challenge in developing a uniform standard of care, aside from the fact that suicide is recognized to be unpredictable and that each client?s situation is unique, is that counselors practice from different theoretical orientations, which dictate varying interventions (Bongar et al., 1998; Knapp & VandeCreek, 1983). Therefore, standards of care vary based on the degree to which a counselor understands, practices, and documents risk. Decisions are determined by the information gathered by the counselor. Failure to obtain as much data as possible through various means falls below the standard of care (Bongar et al., 1998). In malpractice cases involving suicide, the courts recognized that mental health professionals cannot predict the act of suicide in 39 clients; however, risk detection must be demonstrated to establish a reasonable standard of care (Chiles & Strosahl, 1995; Gutheil, 1992; Jobes & Berman, 1993; Kleepspies, Deleppo, Gallagher & Niles, 1999; Mahrer & Bongar, 1993; Maltsberger, 1986). While various aspects of the information-gathering process in risk detection are subjective, and courts recognize this (Knapp & VandeCreek, 1983), risk management protocols are offered, and sufficient empirically based assessment tools exist, making it unnecessary for counselors to rely on tools not yet empirically confirmed (Range et al., 2002). Alternative Approaches Understanding treatment planning with suicidal clients is complicated by limitations in the ability to predict client behavior and by the difficulty and complication of counseling suicidal clients. It was the desire among mental health practitioners to understand and assist this perplexing client base that most likely contributed to the tendency to use devices such as no-suicide contracts when they were first suggested (Maltsberger, 1986; Miller, 1999). While some benefits of the no-suicide contract have been discussed throughout the literature, there is minimal empirical evidence to validate those claims; in fact, concerns exist that reliance on no-suicide contracts can cause a counselor to conduct a less-than-thorough assessment than the situation warrants. This can lead to poorly conceived and potentially dangerous decisions that impact the standard of care, which is the legal yardstick by which a counselor?s actions are measured (Jobes & Berman, 1993; Stanford et al., 1994). These concerns seemed to facilitate a more critical examination of the use of no-suicide contracts, which resulted in alternative risk 40 management approaches being proposed that does not include the use of a no-suicide contract. Farrow (2000) stated, ?I strongly suggest that rather than use unproven and flawed concepts such as no-suicide contracts, we emphasize the skills and knowledge we already possess? (p. 4). Miller (1999) described an alternative approach to using no-suicide contracts with suicidal clients, which among many other things, emphasized the importance of documentation and assessment of a client?s capacity to participate in informed consent. He suggested that a key element in the management of suicidal risk is sharing the clinical burden with the client. The following list outlines the partnership process: avoiding pro forma interventions, relying on informed consent to create a framework for appraisal of treatment options, educating the client about medical treatment uncertainty, discussing and sharing the burden of managing waxing and waning suicidal thoughts, educating the client about treatment options and voluntary participation, and reviewing the risks and benefits of those treatment options. Miller concluded by saying, ?Although some clinicians may still choose to use the suicide-prevention contract, many may find that the exercise of avoiding the contract leads to more robust, direct, and grounded management of suicidal risk? (p. 479). Mahrer and Bongar (1993) wrote, ?A number of authorities believe that there may be more effective and more appropriate ways to respond to the crisis moment of extreme suicidality than the standard no-suicide contract? (p. 287). In examining several works, these authors compiled a list in which they suggest the following: rather than asking the client to sign a no-suicide contract, ask the client what they are likely to do to reduce 41 suicidal feelings, ask the client to commit to contacting the counselor if he or she experiences a loss of control over suicidal impulses, create a list of names and numbers of people to be called if needed for support, increase activity of the counselor during sessions, increase frequency and length of sessions, review alternate responses to the overwhelming suicidal impulses, and include family and friends into ongoing assessment and treatment. Kleespies et al., (1999) listed recommendations for containing the emotional turmoil experienced by clients and indicated that having such a procedure for dealing with suicidal clients is imperative to good practice. The suggestions of these researchers included having an interviewing strategy to evaluate imminent suicide risk, developing a working alliance in which to communicate empathy and foster trust while remaining aware of and using internal stress and frustration with the situation to promote interpersonal engagement, estimating risk level while recognizing that accurate prediction is impossible, using diagnosis-specific profiling associated with high risk for suicide as a guide to the estimation, completing a lethality assessment, which can include a gradation of risk, determining whether inpatient or outpatient management is best suited for the situation, and completing thorough documentation and consultation to ensure that one is proceeding with reasonable and prudent practices. In an outpatient setting, Jobes and Berman (1993) offered a comprehensive clinical risk management plan that consisted of the following: making sure that all practicing counselors know and understand legal statutes relevant to suicide, confidentiality and informed consent, protocols for involuntary confinement, and ethical 42 guidelines for working with suicidal clients; having a detailed written policy specifying risk assessment, treatment, and referral guidelines; assuring the clinical competency of staff members, including training for novice counselors and continuing education opportunities for experienced counselors; maintaining thorough and detailed written documentation of assessments and treatments by using on-going progress notes, or by using forms that are specific to suicidal clients, including assessments used in the existing record; implementing a formal tracking system throughout the course of a client?s treatment, which, while cumbersome, forces counselors to remain clinically responsible for ongoing assessment and care; and establishing and maintaining relevant resources for staff, including external clinical consultation relationships with other counseling professionals and legal contacts, understanding malpractice coverage terms, developing a resource library, and maintaining current lists of outpatient, inpatient, and emergency resources. Joiner, Walker, Rudd & Jobes (1999) explained that assessing suicide risk in such a way that emphasizes history of attempts and current symptoms produces a better framework for risk assessment, which leads to sound clinical decisions and effective activity in working with suicidal clients on an outpatient basis. These authors also suggested using seven domains relevant to risk to make these decisions, including previous suicidal behavior, which can be broken into three groups: suicide ideators, single attemptors, and multiple attemptors (since differences exist in baseline risk for all three categories) the nature of current suicidal symptoms in order to determine which are particularly worrisome; precipitant stressors that may be facilitating or exacerbating the 43 symptoms; general symptomatic presentation in order to ascertain the presence of Axis I and II symptomatology and diagnostic comorbidity; impulsivity and self-control to determine which group the client is best placed in; and protective factors, such as ability to write about the suicidal thoughts and the ability to participate in social support networks, self-control planning and problem-solving exercises. Once these factors are assessed, the counselor can rate the client risk severity on a continuum ranging from nonexistent to extreme in order to decide what treatment methods, including hospitalization, might be most effective for that client. Joiner, et al., contented that by assessing client history of past attempts and the nature of current suicidal symptoms in combination with evaluating these other risk factors, a more objective categorization schema is established so that clinical decision making and appropriate activities can be clarified. Joiner et al., (1999) indicated that activities for symptom remediation can include making regular statements that remind the client to put previously identified self-control strategies into action; discussion of suicidal thoughts or impulses with the counselor if available or a colleague if the counselor is not available or the network of social supports; and going to the emergency room if unable to control the impulses for those clients who may be in a mild-risk category. For those in the moderate-risk category, Joiner, et al., suggested similar protocols to those previously outlined by Mahrer and Bongar (1993). In addition Joiner, et al., (1999) suggested the following specific actions be taken to manage moderate-risk clients: provide 24-hour availability of the counselor, frequently reevaluate suicide risk, note specific changes that are contributing to the elevation of risk, refer for 44 medication if not already being used, and periodic telephone check-in with the counselor; these recommendations are consistent throughout the literature (Bongar, 1991; Bongar et al., 1989; Lee & Bartlett, 2005; Mahrer, 1993). Another suggestion provided by Joiner et al (1999) for clients in the moderate-risk category is to create a detailed emergency plan, which can be written on a small card to be kept in a wallet or purse for easy access if the patient needs directions to follow. For clients in severe and extreme-risk categories, around-the-clock monitoring, which may include active involvement of family members, and consideration of hospitalization are in order. Regardless of risk category, regular and detailed documentation and consultation with peers is strongly emphasized, and the use of a no- suicide contract is not included (Joiner et al., 1999). Chiles and Strosahl (1995) explained that no-suicide contracts have been used as a requirement for clients to be transferred from one clinical setting to another. These authors suggested that asking a client who is in outpatient counseling to sign a no-suicide contract is like requiring a depressed person not to be depressed in order to go home from session. Clearly, if the client were able to not be depressed as a result of such pressure the client would have done it already and would not be seeking the help of a counselor. Rather than the emphasis being on what the client should not do, Chiles and Strosahl suggest a shift in emphasis to helping the client understand what they should do, while always keeping in mind that no strategy guarantees removal of suicidal potential. These researchers suggest using a positive action plan that asks the client to engage in constructive behaviors during the crisis period, such as increasing pleasant reinforcing 45 events like attending a movie, re-engaging in activity that will likely result in success, initiating social contacts despite its emotional challenge, and increasing physical exercise; then praise the client for engaging in positive problem solving. By reframing the suicidal thoughts and behaviors into a problem-solving context, the client?s focus becomes solving real-life problems, which gets the person to consider symptoms from a different perspective and reduces anxiety about signs of abnormality related to behavior. Chiles and Strosahl stated that, ?this tactic alone will often defuse a suicidal crisis? (p. 129). Berman and Cohen-Sandler (1983) made several suggestions for establishing standards of care and maximizing quality of care given to suicidal clients, which simultaneously reduces malpractice actions against mental health providers. Berman and Cohen-Sandler?s recommendations included using risk assessment scales with demonstrated predictive value and documenting clinical judgments in client records whenever observation is necessary, but particularly during times of transitional stress, with an emphasis on termination or discharge periods. It is pointed out that documentation helps to ensure that staff members conform to orders regarding particular client care as well as to safeguard against successful litigation. Documentation should include that an evaluation for hospitalization and medication was made, and if less restrictive measures were decided on, the reasons for those measures should be documented as well. Additional measures to be taken are to establish support system involvement, including informing close relatives and friends that suicide is presently a risk for that client, and to put into place out-of-office availability measures. Awareness of 46 personal emotional responses to suicidal clients should be monitored, considered, and discussed with colleagues to help ensure that counselor bias does not interfere with best practices. Finally, Berman and Cohen-Sandler indicated that offering follow-up treatment for bereaved survivors is encouraged because provider empathy for survivors can facilitate healthy resolution of grief work and reduce emotionally driven malpractice lawsuits. Similarly, Motto (1979) indicated that the period surrounding the loss or change of a therapist is an especially vulnerable time for suicidal clients, as well as during periods of counselor absence, and suggested that increased support be instituted during those times. Throughout the suicidal intervention period, ensuring that the client is getting adequate sleep is recommended, giving frequent and smaller prescriptions is preferable, and asking family members to dispense the medication is useful. Obtaining advance permission to respond to inquiries made by friends and relatives and communicating the client?s increased thoughts of suicide to them as needed is advised to help avoid an awkward situation regarding confidentiality. Enrolling the client into local day treatment or residential care programs are also options to consider, as well as providing the client with contact information for suicide prevention lines and crisis center telephone networks. Motto concluded by emphasizing that detailed recording of rationale regardless of approach taken is essential. Rudd et al., (1999) identified twenty-two recommendations for the outpatient psychotherapeutic treatment of suicidality in adults and adolescents. Rudd et al., contended that recommendations need to have an empirical base rather than the existing 47 litigation-determined standard of care that has emerged over the last few decades, and emphasized that empirically driven practice guidelines are the most valuable resource a counselor can have when treating suicidal clients because these guidelines provide the adequate structure and concrete support necessary to advocate for long-term treatment of chronically suicidal clients. Among the recommendations were many that have already been discussed within this section and which agree with the authors previously cited. Additionally, Rudd et al. (1999) suggested that cognitive-behavioral therapy strategies be used, which integrate problem solving as the core short-term intervention method to reduce ideation, depression, and hopelessness. In order to reduce suicide attempts, longer-term treatment that targets specific deficits such as emotion regulation, impulsivity, anger management, interpersonal assertiveness within relationships, and self- image disturbance should be used. Regardless of therapeutic orientation, an educational component is necessary to help the client understand both direct and indirect treatment goals. Assessing treatment outcomes at regular intervals using psychometrically sound instruments should be used to compliment and balance patient self reports during treatments, and sending letters and making telephone calls to follow up on clients who terminate prematurely may help to reduce future attempts. The authors indicated that acknowledging the limits of psychotherapeutic treatment practice recommendations is important because it assists researchers in pursuing a scientific investigation to advance the field, and assists in professional debate about how to best care for people in greatest need of treatment for suicidality and the continued evolution of standards of care for them. 48 It is well understood that the client-counselor relationship is an important factor in the ongoing assessment and therapy of a suicidal client (Dulit & Michels, 1992; Kleespies et al., 1999; Maltsberger, 1986; Mothersole, 1996; Rudd et al., 1999; Simon, 1988). The clinical relationship can help to prevent suicide based on the degree to which the counselor is able to express genuine concern for the client in distress (Assey, 1985). A well established relationship increases the counselor?s ability to obtain information from the client to increase treatment options and to assist the client through a painful period when no other sources of emotional support are evident (Motto, 1979). The importance of developing a strong therapeutic alliance with the suicidal client is that it is central to the treatment plan of a client in outpatient services. This rapport is accomplished by using the relationship as a source of safety and support during the crisis, by attending to the profound loneliness the client may be experiencing, and by working toward a reduction in symptoms collaboratively (Rudd et al., 1999). Perhaps one of the best ways to determine what to do is to examine what plaintiff attorneys point out that counselors do not do which puts counselors and their clients at risk. Bongar et al. (1998) identified a dozen common failure situations, including failure to properly evaluate the need for or failure to deliver suitable psychopharmacological treatment, failure to specify for and follow through with hospitalization, failure to maintain appropriate client-counselor relationships, failures in supervision and consultation, failure to evaluate for suicide risk during intake, failure to evaluate for suicide risk at management transitions, failure to secure records of previous treatment and poor history taking, failure to conduct a mental status exam, failure to diagnose, failure to 49 create a treatment plan, failure to ensure that the outpatient environment was safe, and failure to adequately document clinical judgments, rationales, and observations. By keeping this list and referring to it when working with a suicidal client, counselors may be better reminded of what to do. In reviewing the list by Bongar et al., (1998) and all the suggestions and protocols previously mentioned, it appears to be agreed on and abundantly clear that documentation is essential to the process of working with suicidal clients (Berman, 1990; Berman & Cohen-Sandler; 1983; Bongar, 2002; Bongar et al., 1998; Chiles & Strosahl, 1995; Gutheil, 1980; Hillard, 1990; Jacobs, 1992; Jobes & Berman, 1993; Joiner et al., 1999; Kleespies et al., 1999; Motto, 1979; Reid, 2004; Simon, 1992). While this issue has been discussed throughout, it is so critical that it warrants additional elaboration. Maintaining strong documentation on a client during a suicidal crisis period is critical because it provides the counselor with an opportunity to refer back to and consider what has already been done and what may be left to do. This serves as a roadmap to provide the reasonable care that peers, professional ethics boards, clients, and their family members expect, as well as to serve as a protective measure should a client commit suicide and a malpractice suit be filed (Reid, 2004). Berman (1990) wrote, Given the war zone of the courtroom, it is nothing short of playing Russian roulette with your professional life should you not document your decisions, your rationale for your judgements [sic], and for the procedures you choose to employ. This is not defensive practice, it is reasonable, prudent and competent practice. (p. 39) 50 Berman (1990) went on to explain that while a counselor does not have to conform to any one strategy of assessment and treatment because there is no universally agreed upon and accepted mode, the best and only defense for any particular practice is documentation. The only way for a counselor to establish that he or she acted competently is by keeping adequate session notes and client records, which unfortunately many caregivers do not do. Reid (2004) pointed out that asking questions such as ?Are you suicidal?? and ?Do you have the means?? is inadequate. Documentation that indicates these questions constituted an entire assessment is an admission that poor care was provided. While a counselor or psychiatrist may have provided reasonable care to a client who committed suicide, a jury may award the family compensation because it was unclear in the documentation that the care was actually given. However, if all suicide risk assessments and risk benefit notes are well documented, this is less likely to happen (Bongar et al., 1998; Simon, 1992; VandeCreek et al., 1987). VandeCreek et al., (1987) suggested that a model risk-benefit note includes the following: an assessment of risk, information alerting the counselor to the risk, which high-risk factors were present during the situation and in the past for that client, which low-risk factors were present, what questions were asked and what responses were given, and how all the compiled information directed the action of the counselor as well as why other actions were rejected. Jacobs (1992) explained that good documentation does not mean write more, which is often both impractical and ineffective. He indicated that effective documentation 51 focuses on three main areas: the risk-benefit analysis of the intervention or approach chosen by the provider, the exercise of a counselor?s judgment at critical decision points and how well they are explained or rationalized, and the client?s capacity or competence to participate in the decision making process and whether the client was able to weigh the risks and benefits of the information he or she provided to the counselor. Jacobs discussed this final area of focus at it related to the use of a no-suicide contract, indicating that the concept of a no-suicide contract is particularly concerning when used by inexperienced counselors who may view the use of a no-suicide contract as a guarantee or warranty that the client will not attempt or commit suicide if signed. Jacobs stated that an unfortunate result is that the competence or capacity of a client to participate is rarely assessed and recorded as must be for all decisions made. As stated by Gutheil (1980), perhaps the best guide to follow as it relates to the treatment management of a suicidal client is that if it is not written down, it will be assumed that it did not happen (Gutheil, 1980). Current Status of No-Suicide Contracts Current research indicates that no-suicide contracts continue to be widely used by mental health professionals despite little empirical evidence of their effectiveness (Bongar, 1991; Buelow & Range, 2001; Egan, 1997; Kelly & Knudson, 2000; Maltsberger, 1991; Miller et al., 1998; Stanford et al., 1994), despite a lack of training with the specific technique (Farrow, 2002; Miller et al., 1998; Neimeyer, 2000; Range et al., 2002; Weiss, 2001), and despite a lack of training in suicidology as a whole 52 (Bernstein, Fedberg & Brown, 1991; Bongar & Harmatz, 1989; Bongar et al., 1989; Ellis & Dickey, 1998; Farrow, 2002; Jobes & Berman, 1993; Kleespies et al., 1999; Maltsberger, 1991; Range et al., 2002; Shein, 1976). Community mental health textbooks continue to introduce the concept of no- suicide contracts as a viable option when dealing with suicidal clients. Seligman (2004) suggested that developing a verbal or written contract is a useful action because it reassures clients that the counselor is concerned about him or her and is trying to help. Howatt (2000) suggested a five-step suicide intervention model which includes asking a client to promise not to kill him or herself on purpose or accidentally, and that encouraging a client to make a written contract can lead the suicidal person to a no- suicide decision. If the client agrees to such a contract, the contention is that the danger for suicide is decreased. Gladding and Newsome (2004) also suggested that in an effort to do whatever is necessary to keep a suicidal client safe, asking the client to sign a no- suicide contract is part of the necessary protocol. Conversely, in reviewing effective crisis intervention strategies, Echterling, Presbury, and McKee (2005) indicate that there is no evidence to support the effectiveness of a no-suicide contract, and while the technique may help the counselor sleep better at night, the use of a no-suicide contract is an ?inadequate substitute for comprehensive treatment for a person who poses a serious risk for suicide? (p. 157). They suggest that rather than using a legalistic contract that might seem coercive, helping a client to develop a plan for choosing to live is more appropriate. A growing concern regarding the continued use of no-suicide contracts in working with suicidal clients is the liability of using a technique that is not empirically 53 demonstrated to be effective (Bongar, 1991; Clark & Kerkoff, 1993; Farrow, 2002; Goldblatt, 1994; Jobes & Berman, 1993; Reid, 2004). A further concern is whether a client who is suicidal can give informed consent to enter a no-suicide contract (Ayd & Palma, 1999; Clark & Kerkhof, 1993; Farrow & O?Brien, 2003; Gutheil, 1992; Jacobs, 1992; Hillard, 1990; Miller et al., 1998; Reid, 1998, 2005; Simon, 2004; Stanford et al., 1994), particularly if the client is inebriated or psychotic is also an important concern (Drye et al., 1973; Egan, 1997; Mahrer, 1993; Weiss, 2001). Hillard (1990) explained that informed consent exists when a client is competent to give the consent, the client has been given adequate information to make the consent possible, and when the consent is given voluntarily. Each state has a statute that defines competence, but in general, in order to be competent to make a specific clinical decision, which includes informed consent, a client should be aware of the clinical situation, have some understanding of the issues involved in the decision, and be able to interpret information rationally. In providing adequate information so a client can give informed consent, the client should be told the risks and benefits of the treatment recommendation, what alternative treatment options exist, and the positive and negative consequences of the alternative options. Hillard pointed out that voluntary consent occurs only when the client feels free to accept or decline the treatment option. Exceptions to the necessity of obtaining informed consent occurs when immediate treatment is required, as in an emergency, when a client knowingly and voluntarily waives his or her right to be informed, when it is determined that a complete disclosure might negatively effect the client?s well being, and when it is assessed that the client is 54 incompetent to give consent, which is often the case during suicidal episodes (Bongar, 1991; Hillard, 1990). In cases where it is determined that the client is psychotic, mentally retarded, inebriated, or suffering from severe depression, it is more clearly evident that he or she would not meet the criteria of competence (Drye et al., 1973; Egan, 1997; Egan et al., (1997); Goulding & Goulding, 1979; Mahrer, 1993; Weiss, 2001). Miller et al. (1998) explained that clinical phenomena such as intoxication and psychotic symptoms vary across time. These researchers stated that ?These variables often change abruptly and alter the level of suicide risk; they are impossible to assess by strict objective measures and in many cases are unknowable? (p. 79). However, in cases where the physical and mental conditions are less definitive, clinical judgments becomes more circumstantial. It has been suggested that for clients with Borderline Personality Disorder, who are often prone to attempting suicide, determining competence is critical; often, given the irrationality of their thinking, the use of contracts of any sort with these clients is tenuous at best (Jacobs, 1992; Shea, 1999; Weiss, 2001). Gutheil (1992) indicated that no matter how informal no-suicide contracts are, the client?s capacity to participate in them is a natural prerequisite for their use. He indicated that a client?s capacity or competence to participate in treatment decisions, to appropriately weigh risks and benefits of this treatment option, and to provide reasonable information about his or her own suicidal condition is rarely specific, is rarely adequately assessed and is often unreliable. If the personal evaluation and self-testimony of a client is relied on by a counselor, this information should be explicitly recorded. 55 Stanford et al. (1994) addressed the issue of informed consent by explaining that treatment in an outpatient mental health setting may include the use of a no-suicide contract, whereas in dealing with the same client, a court might find the person to be incompetent to provide informed consent regarding his or her treatment options in an inpatient setting. This discrepancy forces mental health providers to critically evaluate whether a client who is irrational enough to consider suicide as an option is competent enough to sufficiently weigh the risks and benefits of alternative treatments, including the question of hospitalization versus outpatient care. Therefore, it is advised that competence be clearly addressed in the medical record. Stanford et al. claimed that ?Assessing competency, evaluating risk factors for suicide, and assigning responsibility for potential suicidal behavior must be part of the comprehensive clinical evaluation of the suicidal patient? (p. 347). Assessment of competency should occur independent of the fact that the client may have already given informed consent to participate in a counseling relationship. Farrow and O?Brien (2003) discussed informed consent from the perspective of a client entering mental health treatment initially versus a time when the same person is impaired by major mental illness. These authors pointed out that during times of crisis, if a decision to enforce a civil commitment is made, it has usually been determined that the person is incompetent to make informed decisions about his or her safety and the safety of others, and therefore, the formulation of a no-suicide contract would be inappropriate. If a counselor determines that the client is at low risk of suicide in the short term, and the 56 client is assessed to be competent, the process of informed consent should be undertaken if a no-suicide contract is formulated as part of the treatment protocol. Farrow and O?Brien (2003) reported results from a larger study that involved interviewing crisis nurses and clients on the effects of the use of the no-suicide contract in community crisis situations. In the article, information from eight clients was presented. Given that informed consent requires competence, being fully informed, and voluntary consent, these elements were examined specifically. In terms of competence, some interviewees did not comprehend what was being asked of them. One interviewee stated, ?My thinking was so confused [because of life stressors]?I didn?t understand what they were suggesting? (p. 203). As for being fully informed, several clients indicated feeling compelled to accept entering a no-suicide contract because they thought it was their only choice. This suggested that the no-suicide contract reduced their ability to receive full information regarding treatment options and their right to refuse or accept these other treatments. All interviewees believed that refusal to enter the no-suicide contract would have resulted in unwanted interventions, including involuntary commitment to an inpatient unit. One interviewee commented, ?It?s like they [the crisis team] pretend to give you a choice [when one doesn?t exist]? (p. 204). While it is not clear whether the crisis team members intended for the introduction and formulation of the no-suicide contract to be coercive, the perception of the clients? was that it was not a voluntary decision for them. Farrow and O?Brien concluded, ?Doubts must be cast upon the ability of patients to give true informed consent to enter into a no-suicide contract in community crisis situations? (p. 206). 57 While some research has been conducted to explore the use of and efficacy of no- suicide contracts in inpatient settings, more research in this area is warranted (Davis et al., 2002; Drew, 2001). Interestingly, even less research has been conducted to explore their efficacy and usefulness to clients in outpatient settings (Bongar, 1991; Farrow et al., 2002; Jobes & Berman, 1993; Rudd et al., 1999), or with potentially suicidal individuals who have used them regardless of setting (Range et al., 2002). Considering that lives are at stake, it is surprising that such a technique without strong support of its effectiveness continues to be used. Davis, et al., (2002) suggested that while counselors may view no- suicide contracts as having value, the experience of a client may be different. Shea (1999) suggested that people are complex, and responses to this kind of contracting vary; in fact, while contracting may be useful with some people, it could be useless or even counterproductive with others. It has been suggested that assessing whether clients who have been suicidal and have used a no-suicide contract found them to be helpful is warranted (Beulow & Range, 2001; Jones & O?Brien, 1990; Lee & Bartlett, 2005; Range et al., 2002). The apparent paucity of research in this area as well as the aforementioned issues surrounding the no-suicide contract make their use controversial and worthy of further exploration (Bongar, 1991; Egan, 1997; Farrow, 2002; Goldblatt, 1994; Mahrer & Bongar, 1993; Miller et al., 1998; Mothersole, 1997; Rudd, 1996, Rudd & Joiner, 1998). Summary Despite the lack of empirical evidence, legal and ethical considerations, controversies of their use, and the growing number of alternate more effective 58 approaches, the no-suicide contract continues to be in wide use in the treatment of suicidal clients. The research in this study is motivated by the numerous recommendations that were located throughout the literature suggesting that pertinent information can be gained in asking clients who have used no-suicide contracts whether they believe them to be an effective intervention. This research can provide some of the most valuable data in determining if these contracts actually work and whether they should be included in risk management planning (Beulow & Range, 2001; Jones & O?Brien, 1990; Range et al., 2002). Therefore, this study attempts to identify whether adults who have had suicidal ideation or have engaged in suicidal behaviors, who were asked to sign a no-suicide contract during that time and who are currently participating in outpatient counseling agree on the value of no-suicide contracts by gathering and evaluating empirical data to support or refute this contention, and to discern whether these contracts are an effective treatment device that warrant their continued use in this setting. 59 III. METHODS AND PROCEDURES Introduction This study was designed to assess how clients rate the effectiveness of no-suicide contracts, and how clients rate the effectiveness of other interventions used in suicidal situations. Because clients reported on the no-suicide contract and alternate interventions, the perceived effectiveness of no-suicide contracts in comparison to alternate interventions was evaluated. This study also sought to determine whether there was a relationship between the perceived effectiveness of suicidal interventions and specific demographics and treatment variables, such as gender, age, time in treatment, and number of attempts. Research Questions It was expected that clients would perceive the use of a no-suicide contract as less effective when compared to other methods that were used to prevent or intervene in their suicidal experiences. Research questions to be examined were: 1. How did clients rate the effectiveness of no-suicide contracts? 2. How did clients rate the effectiveness of other interventions? 60 3. Was there a relationship between ratings of effectiveness of suicidal intervention methods and specific treatment variables (such as gender, age, time in treatment, and number of attempts)? Participants The sample for this study were comprised of adult clients (19 years or older) of two community mental health outpatient treatment facilities; one located in Montgomery, Alabama and the other in Opelika, Alabama. The sample consisted of sixty-six clients who were currently receiving counseling or medication management services, or both for the stabilization of their mental health-related symptoms. Participation in this study was restricted to people who self-identified as having expressed suicidal thoughts or engaged in suicidal behaviors during some point in their counseling treatment history. Procedures After receipt of approval of the Auburn University Institutional Review Board (Appendix A), packets were distributed to the identified treatment facilities that are part of the GrandView Behavioral Health Centers. Consent to include these centers had been obtained (Appendix B). As clients checked in at the front desk for their scheduled appointment, they were handed a packet of information in an unsealed envelope, and were asked, according to a script (Appendix C), to review the contents to determine if they were eligible and if they wanted to participate in a research project. These materials included the consent information and a description of the study (Appendix D). After review of the package (which included the informed consent and survey materials), the 61 clients determined whether they were eligible to participate and if they wanted to participate. If the client chose not to participate, he or she returned the uncompleted package to the receptionist. If the client agreed to participate, then he or she kept the letter of informed consent and completed the Demographics and Treatment Intervention Effectiveness Scale (Appendix E). Participants who indicated having past experience with a no-suicide contract as part of their treatment responded to items on a second survey, the No-Suicide Contract Survey (Appendix F). This survey asked them to rate their last experience with the no-suicide contract. The time required for the participant to complete the survey was not measured, but the survey-completion process did not interfere with any scheduled appointments and was completed during the period between client arrival and client appointment, typically about fifteen minutes. Once completed, they put the completed contents into the same envelope, sealed it, and returned the package to the receptionist before meeting with their designated mental health professional. To protect participants, the responses to the information form and surveys were anonymous and no identifiable information was collected from the participants. All data collected was kept in a locked file cabinet on the site of the treatment facility. Only the front receptionist (at each location) who distributed and collected the research packets had access to the cabinet. All completed packets were returned to the receptionist sealed. Survey packets were picked up weekly by the researcher. The researcher maintained the data in a locked filing cabinet at her residence; however, it should be reiterated that all data was anonymous and participants could not be linked to specific data. 62 Measures The Demographics and Treatment Intervention Effectiveness Scale was developed by the researcher. Participants were asked to respond to a series of questions to gather initial demographic information such as gender, age, years in treatment, and number of suicide attempts. Participants were also asked to identify intervention treatment methods that were used during their last suicidal episode, to rate those interventions, and to provide brief explanations of the methods they perceived as most and least effective. The Demographic and Treatment Intervention Effectiveness Scale was developed for the purposes of this study and was based on research of Ewalt (1967), Mahrer and Bongar (1993), Kleespies et al., (1999), Joiner et al., (1999), and Chiles and Strosahl (1995), which helped identify alternative methods used in intervening with suicidal clients (Appendix E). The Commitment to Safety Survey (Davis et al., 2002) was developed for use with psychiatric inpatients admitted for suicidal danger. It gathers perceptions of the benefits and limitations of written no-suicide agreements. For this research project, the Commitment to Safety Survey was modified to assess the views of clients who had participated in the use of no-suicide contracts in past treatment. Modifications to the Commitment to Safety Survey, which included changing the title from Commitment to Safety Survey to No-Suicide Contract Survey (Appendix F), were made with author?s approval. Both instruments used were self-report measures. The Demographics and Treatment Intervention Effectiveness Scale included multiple-choice responses, yes-and- no responses, a series of statements to which the client responded in a Likert-scale choice 63 ranging from very useful to not used, and open-ended questions. The No-Suicide Contract Survey included multiple choice items using a Likert scale, with responses ranging from strongly agree to strongly disagree. The responses to the original Commitment to Safety Survey were evaluated for validity and reliability through principal component factor analysis with orthogonal rotation and coefficient alpha. Results indicated a three-factor structure with all factors demonstrating moderate to high reliability (ranging from 0.67 to 0.95). The factors and their respective reliability estimate were as follows: Factor I: Therapeutic Features with coefficient alpha of 0.95, Factor II: Coercive Features with coefficient alpha of 0.67, and Factor III: Detached Features with coefficient alpha of 0.68. Data Analysis Data was analyzed using the Statistical Package for Social Services (SPSS) computer software. Descriptive analysis among samples were used to determine how frequently methods were used and the overall perceived effectiveness of methods. Descriptive analysis was used to determine which of the five most frequently used methods clients indicated were most helpful. A multiple regression was used to examine whether there was a relationship between ratings of the therapeutic variables (e.g., gender, age, time in treatment, and number of attempts) and methods used and overall rankings. 64 IV. RESULTS Introduction The purpose of this study was to explore the practice of using a no-suicide contract within the mental health profession as a technique to prevent suicide by clients. More specifically, the purpose of the present study was to examine information obtained from clients in counseling and psychiatric care settings on an outpatient basis in an effort to determine if their experience with no-suicide contracts was helpful in reducing or preventing suicidal thoughts and behaviors, or whether alternate approaches were more helpful. It was expected that clients would perceive the use of a no-suicide contract as less effective when compared to other methods that were used to prevent or intervene in their suicidal experiences. Research questions examined were: 1. How do clients rate the effectiveness of no-suicide contracts? 2. How do clients rate the effectiveness of other interventions? 3. Is there a relationship between ratings of effectiveness of suicidal intervention methods and specific treatment variables (such as gender, age, time in treatment, and number of attempts)? The participants in this study were adult clients (19 years or older) of two community mental health outpatient treatment facilities; one located in Montgomery, 65 Alabama, and the other in Opelika, Alabama. The sample consisted of 66 clients who were currently receiving counseling or medication management services, or both, for the stabilization of their mental health-related symptoms. Participation in this study was restricted to people who self-identified as having expressed suicidal thoughts or engaged in suicidal behaviors during some point in their counseling treatment history. Seventeen of the 66 participants indicated that they had signed a no-suicide contract as part of their treatment during their most recent suicidal experience. The remainder of this chapter includes a presentation of the reliability and the information revealed based on the research questions. Validity and Reliability The Demographics and Treatment Intervention Effectiveness Scale was developed by the researcher. Participants were asked to respond to a series of questions to gather initial demographic information such as gender, age, years in treatment, and number of suicide attempts. Participants were also asked to identify intervention treatment methods that were used during their last suicidal episode, to rate those interventions, and to provide brief explanations of the methods they perceived as most and least effective. The Demographic and Treatment Intervention Effectiveness Scale was based on research of Ewalt (1967), Mahrer and Bongar (1993), Kleespies et al., (1999), Joiner et al., (1999), and Chiles and Strosahl (1995), which that identified alternative methods used in intervening with suicidal clients (Appendix E). The No-Suicide Contract Survey was modified from a study conducted by Davis et al. (2002). The original Commitment to Safety Survey was developed for use with 66 psychiatric inpatients admitted for suicidal danger. It measured perceptions of the benefits and limitations of written no-suicide agreements. For this research project, the Commitment to Safety Survey was modified to assess the views of clients who had participated in the use of no-suicide contracts in past treatment. Modifications to the Commitment to Safety Survey were made, which included changing the title from Commitment to Safety Survey to No-Suicide Contract Survey (Appendix F). Demographics There were 66 participants from GrandView Behavioral Health Centers who participated in this study. Of these 66 participants, 17 completed the No-Suicide Contract Survey. Descriptive statistics were utilized to examine the demographic information of all the 66 participants. These data are reported in Table 1. 67 Table 1 Demographic Information of all Participants Demographic Category N % Gender Male 18 27.3 Female 46 69.7 Not Indicated 2 3.0 Age 19-25 10 15.2 26-35 11 16.7 36-45 14 21.2 46-55 17 25.8 56-65 12 18.2 66 and older 2 3.0 The population assessed during this research was predominately female, with ages spanning the range of adulthood; no children were assessed as a part of this research. Of the 66 participants, 18 (27.3%) were male, 46 (69.7%) were female; two participants (3.0%) did not indicate their gender. The age of the 66 participants spanned from 19 years old to more than 65; 10 (15.2%) were between the range of 19 and 25; 11 (16.7%) were between 26 and 35; 14 (21.2%) were between 36 and 45; 17 (25.8%) were between 46 and 55; 12 (18.2%) were between 56 and 65; and 2 (3.0%) were 65 and older. Descriptive statistics were utilized to present the treatment information of the 66 participants in Table 2. 68 Table 2 Treatment Information of all Participants Factor N % Time in Counseling Treatment (Years) 1-5 34 51.5 6-10 14 21.2 11-15 9 13.6 16-20 2 3.0 20 or more 7 10.6 Thoughts of Committing Suicide Yes 60 90.9 No 6 9.1 Gesture of Suicide (Without Intent) Yes 31 47.0 No 35 53.0 Suicide Attempts Yes 27 40.9 No 39 59.1 (table continues) 69 Table 2 (continued) Factor N % Number of Attempts 0 32 48.5 1 14 21.2 2 14 21.2 3 4 6.1 4 1 1.5 5 or more 1 1.5 Agreed to a No-Suicide Contract Yes 17 25.7 No 51 74.3 Number of Contract Agreements 0 51 77.3 1 10 15.2 2 1 1.5 3 2 3.0 4 0 0.0 5 or more 2 3.0 (table continues) 70 Table 2 (continued) Factor N % Agreement Made while Inpatient or Outpatient Inpatient 5 29.4 Outpatient 12 70.6 Satisfaction of Care Yes 42 63.6 No 8 12.1 No Response 16 24.2 The population assessed during this research had significant personal experience with suicide, with counseling services in general, and with the use or offer of use of no- suicide contracts. Of the 66 participants, 34 (51.5%) had been in counseling services between 1 and 5 years; 14 (21.2%) between 6 and10 years; 9 (13.6%) between 11 and15 years; 2 (3.0%) between 16 and 20 years; and 7 (10.6%) had been in counseling services between 20 and more years. Sixty of the participants (90.9%) had previous thoughts of committing suicide, and 6 (9.1%) indicated having no prior thoughts of committing suicide. Thirty-one (47.0%) indicated having made a gesture without intent to complete, and 35 (53.0%) reported no previous gesture. Twenty-seven participants (40.9%) had attempted suicide with intent to complete, and 39 (59.1%) had no prior suicide attempts. Thirty-two (48.5%) indicated no attempt to commit suicide; 14 (21.2%) reported 1 attempt; 14 (21.2%) reported 2 attempts; 4 (6.1%) reported 3 attempts; 1 (1.5%) reported 71 4 attempts; 1 (1.5%) reported 5 or more attempts. Seventeen participants (25.7%) reported agreeing to a no-suicide contract, and 51 (74.3%) reported having never agreed to a no-suicide contract. Fifty-one participants (77.3%) had never signed a no-suicide contracts; 10 (15.2%) had signed 1; 1 (1.5%) had signed 2; 2 (3.0%) had signed 3; 0 (0.0%) had signed 4; and 2 (3.0%) had signed 5 or more contracts. Of the seventeen participants who indicated signing a no-suicide contract at some point in their treatment history, five participants (29.4%) were in an inpatient psychiatric unit at the time they signed a no-suicide contract, and 12 (70.6%) were being treated on an outpatient basis for mental health reasons the last time they signed a no-suicide contract. Forty-two participants (63.6%) indicated that they were satisfied with the way their counseling professional dealt with their most recent suicidal episode, 8 (12.1%) indicated they were not satisfied, and 16 (24.2%) did not respond. Research Question One Research question one focused on how clients rate the effectiveness of no-suicide contracts. Descriptive analysis was conducted on the responses to Question 11 of the Demographics and Treatment Intervention Effectiveness Scale to answer this question. As indicated in Graph 1, this treatment method?s mean (M = 1.58, SD = 1.21) was the lowest of the range of 22 treatment methods assessed in this study. Response Index Legend 1. No-Suicide Contract 2. Medication 3. Hospitalization 4. Increase Number of Appointments 5. Establish a Check-in System 6. Increase Counseling Session Length 7. Collaboratively Revise Treatment Plan 8. Use of Formal Survey(s) 9. Contact Family Members 10. Openly Discuss Suicidal Thoughts 11. Discuss Contributing Stress Factors 12. Improvement in Problem-Solving Skills 13. Identification of Alternate Responses to Suicide 14. Improvement in Self-Control Planning Skills 15. Participation in Group Counseling 16. Create a List of Emergency Contacts 17. Frequent Evaluations of Suicide Risk 18. Improvement in Lifestyle Health (Diet, Sleep, etc.) 19. Increase in Social Activities 20. Assessment of Living Environment Safety 21. Establishing a 24-hour Availability of Counselor 22. Participation in Faith-Based Activities 1.5 2 2.5 3 3.5 4 4.5 1 2 3 4 5 6 7 8 9 10111213141516171819202122 Response Index Number Graph 1: Mean Responses of Demographics and Treatment Intervention Effectiveness Scale Question 11 Research Question Two Research question two considered how clients rate the effectiveness of other treatment interventions. Descriptive analysis was conducted on the responses to Question 11 of the Demographics and Treatment Intervention Effectiveness Scale to answer this question. The results are presented in Table 3. The five most effective treatment methods indicated by the participants were: Medication (M = 4.08, SD = 1.05), Discuss Contributing Stress Factors (M = 3.94, SD = 1.01), Improvement in Lifestyle Health (Diet, Sleep, etc.) (M = 3.72, SD = 1.23), Increase Number of Appointments (M = 3.52, SD = 1.42), and Openly Discuss Suicidal Thoughts (M = 3.45, SD = 1.26). This data 72 73 indicates that of the treatment methods assessed in this study, medication was considered the most effective, and a no-suicide contract was considered the least effective. Table 3 Responses to Demographics and Treatment Intervention Effectiveness Scale Question 11 Treatment Methods M SD Medication 4.08 1.05 Discuss Contributing Stress Factors 3.94 1.01 Improvement in Lifestyle Health 3.72 1.23 Increase Number of Appointments 3.52 1.42 Openly Discuss Suicidal Thoughts 3.45 1.26 Improvement in Problem-Solving Skills 3.39 1.41 Increase in Social Activities 3.36 1.26 Improvement in Self-Control Planning Skills 3.00 1.50 Identification of Alternate Responses to Suicide 2.98 1.50 Participation in Faith-Based Activities 2.95 1.59 Hospitalization 2.89 1.59 Establishing 24-hour Availability of Counselor 2.76 1.67 Assessment of Living Environment Safety 2.69 1.51 Collaboratively Revise Treatment Plan 2.57 1.51 Increase Counseling Session Length 2.56 1.51 Frequent Evaluation of Suicide Risk 2.55 1.43 Contact Family Members 2.44 1.48 Create a List of Emergency Contacts 2.41 1.50 Participation in Group Counseling 2.25 1.34 Establish a Check-in System 2.23 1.48 Use of Formal Survey(s) 1.80 1.13 No-Suicide Contract 1.58 1.21 Research Question Three Research question three examined whether there is a relationship between ratings of effectiveness of suicidal intervention methods assessed in this study and the specific demographic variables of gender, age, time in treatment, and number of attempts. A preliminary multiple regression correlation analysis was conducted to explore this research question; the top five treatment methods identified by participants were examined at length. The results of this examination are presented in Table 4. Given that multiple regression analysis deals only with continuous variables, a criterion-coding procedure was employed to include the categorical variable gender in the multiple regression analysis, and the results of this procedure are presented in Table 5. Table 4 Summary of Preliminary Multiple Regression Correlation Assessment on Top Five Treatment Methods Medication Discuss Contributing Stress Improvement in Lifestyle Health Increase Number of Openly Discuss Suicidal Time in Treatment Age Number of Suicide Attempts r = -0.012 p = 0.926 r = 0.111 p = 0.382 r = -0.007 p = 0.955 r = -0.077 p = 0.542 r = 0.290 p = 0.019 r = 0.068 p = 0.597 r = -0.141 p = 0.268 r = -0.051 p = 0.692 r = -0.046 p = 0.717 r = 0.111 p = 0.378 r = -0.011 p = 0.932 r = -0.042 p = 0.740 r = -0.051 p = 0.692 r = 0.195 p = 0.120 r = -0.010 p = 0.940 74 Table 5 Summary of Preliminary Criterion-Coding Procedure on Demographic Variable Gender Gender t = -0.058 df = 59 p = 0.797 Improvement in Lifestyle Health Medication Openly Discuss Suicidal Increase Number of Appointments Discuss Contributing Stress Factors t = -1.466 df = 60 p = 0.148 t = -0.140 df = 60 p = 0.890 t = 0.215 df = 61 p = 0.831 t = 1.968 df = 61 p = 0.054 A primary correlation assessment analysis was performed on each of the demographic variables of interest (gender, age, time in treatment, and number of suicide attempts) to assess the relationship between the intended dependent variable and each individual independent variable; the results of these assessments are presented in Tables 6 through 15 below. Only statistically significant correlations with this intervention were included in the multiple regression models. Highest Rated Effective Treatment Method: Medication Results from preliminary analysis were age (r = -0.012, p = 0.926), time in treatment (r = 0.068, p = 0.597), and number of suicide attempts (r = -0.011, p = 0.932). Among these three variables, none of them had a significant correlation with the intended variable. Gender effect was assessed through an independent sample t-test. The result of this procedure indicated that there was no difference in responses based on gender (t = -0.258, df = 59, p = 0.797). Results of correlation analysis and independent sample t-test are presented in Table 6 and Table 7. 75 76 Table 6 Age, Time in Treatment, and Number of Suicide Attempts on Treatment Method Medication Demographic Category r p Age -0.012 0.926 Time in Treatment 0.068 0.597 Number of Suicide Attempts -0.011 0.932 Table 7 t-test for Gender Effect on Treatment Method Medication Demographic Category t df p Gender -0.058 59 0.797 Second Highest Effective Treatment Method: Discuss Contributing Stress Factors Results from preliminary analysis were age (r = 0.111, p = 0.382), time in treatment (r = -0.141, p = 0.268), and number of suicide attempts (r = -0.042, p = 0.740). None of these three variables had a significant correlation with the intended variable. Gender effect was assessed through an independent sample t-test. The result of this procedure indicated there was no difference in responses based on gender (t = -1.466, df = 60, p = 0.148). Results of correlation analysis and independent sample t-test are presented in Table 8 and Table 9. 77 Table 8 Age, Time in Treatment, and Number of Suicide Attempts on Treatment Method Discuss Contributing Stress Factors Demographic Category r p Age 0.111 0.382 Time in Treatment -0.141 0.268 Number of Suicide Attempts -0.042 0.740 Table 9 t-test for Gender Effect on Treatment Method Medication Demographic Category t df p Gender -1.466 60 0.148 Third Highest Rated Effective Treatment Method: Improvement in Lifestyle Health (Diet, Sleep, etc.) Results from preliminary analysis were age (r = -0.007, p = 0.955), time in treatment (r = 0.083, p = 0.514), number of suicide attempts (r = -0.051, p = 0.692). None of these three variables had a significant correlation with the intended variables. Gender effect was assessed through an independent sample t-test. Results of this procedure indicated there was no difference between responses based on gender (t = -0.140, df = 60, 78 p = 0.890). Results of the correlation analysis and independent sample t-test are presented in Table 10 and Table 11. Table 10 Age, Time in Treatment, and Number of Suicide Attempts on Treatment Method Improvement in Lifestyle Health (Diet, Sleep, etc.) Demographic Category r p Age -0.007 0.955 Time in Treatment -0.051 0.692 Number of Suicide Attempts -0.051 0.692 Table 11 t-Test for Gender Effect on Treatment Method Improvement in Lifestyle Health (Diet, Sleep, etc.) t df p Gender -0.140 60 0.890 Fourth Highest Rated Effective Treatment Method: Increase Number of Appointments Results from preliminary analysis were age (r = -0.077, p = 0.542), time in treatment (r = -0.046, p = 0.717), and number of suicide attempts (r = 0.195, p = 0.120). None of these three variables had a significant correlation with the intended variables. Gender effect was assessed through an independent sample t-test. The result of this 79 procedure indicated there was no difference in responses based on gender (t = 0.215, df = 61, p = 0.831). Results of the correlation analysis and independent sample t-test are presented in Table 12 and Table 13. Table 12 Age, Time in Treatment, and Number of Suicide Attempts on the Treatment Method of Increase Number of Appointments Demographic Category r p Age -0.077 0.542 Time in Treatment -0.046 0.717 Number of Suicide Attempts 0.195 0.120 Table 13 t-Test for Gender Effect on the Treatment Method Increase Number of Appointments t df p Gender 0.215 61 0.831 Fifth Highest Rated Effective Treatment Method: Openly Discuss Suicidal Thoughts Results from preliminary analysis were age (r = 0.290, p = 0.019), time in treatment (r = 0.111, p = 0.378), and number of suicide attempts (r = 0.010, p = 0.940). None of these three variables had a significant correlation with the intended variable. 80 Gender effect was assessed through an independent sample t-test. The result of this procedure indicated there was no difference in responses based on gender (t = 1.968, df = 61, p = 0.054). Results of the correlation analysis and independent sample t-test are presented in Table 14 and 15. Table 14 Age, Time in Treatment, and Number of Suicide Attempts on Treatment Method Openly Discuss Suicidal Thoughts Demographic Category r p Age 0.290 0.019 Time in Treatment 0.111 0.378 Number of Suicide Attempts 0.010 0.940 Table 15 t-Test for Gender Effect on Treatment Method Openly Discuss Suicidal Thoughts t df p Gender 1.968 61 0.054 81 Summary This research focused on the views of clients regarding the effectiveness of treatment methods for preventing suicide, with special attention given to the relative perceived effectiveness of the no-suicide contract. The participants in this study viewed the use of no-suicide contracts as least effective of the treatment methods assessed. It was also determined that this assessment was constant across demographic variables of gender, age, time in treatment, and number of suicide attempts. This analysis indicated that of the treatment methods assessed, the use of medication was viewed by the participants as being the most effective option, followed closely by discussing contributing stress factors, improvement in lifestyle health, increase in number of appointments, and an open discussion of suicidal thoughts. It was clear that there is a wide degree of perception of effectiveness or lack thereof among the treatment methods assessed by this research, which presents significant implications for the manner in which counselors treat suicidal clients. 82 V. DISCUSSION Introduction The purpose of this study was to explore client responses to the practice of using a no-suicide contract within the mental health profession as a technique to prevent suicide by clients. More specifically, the purpose of the present study was to examine information obtained from clients in counseling and psychiatric care settings on an outpatient basis in an effort to determine if their experience with no-suicide contracts was helpful in reducing or preventing suicidal thoughts and behaviors, or whether alternate approaches were more helpful. Data was collected by means of participant survey completion from two outpatient mental health counseling centers of the same organization. A total of 66 clients participated, 17 of whom indicated that a no-suicide contract was used at some point during a suicidal experience. In this final chapter, the findings will be explored, the limitations of this study will be examined, recommendations for future research will be presented, and implications for counseling practice and training will be examined. Discussion The responses of participants to the research questions indicated that the no- suicide contract was perceived as the least effective of the 22 treatment methods assessed, 83 that the 21 remaining treatment methods were seen as variously effective, with a general correlation between the degree to which the treatment method addressed life skills issues and that method?s perceived effectiveness, and that no relationship existed between the ratings of the effectiveness of suicidal intervention methods and demographic variables of age, gender, time in treatment, or number of suicide attempts; the relative rankings of the effectiveness of treatment methods remained constant across demographic variables. Davis et al. (2002) evaluated the perceptions of 135 psychiatric inpatients admitted for suicidal danger with regard to the advantages and disadvantages of the no- suicide contract, but Davis? experimental design did not compare the use of the no- suicide contract to other therapeutic techniques. While results of this study indicated that the patients reported positive attitudes towards no-suicide contracts, Davis et al. suggested that while counselors may view no-suicide contracts as having value, the experience of clients may be different when the use of no-suicide contracts is compared to other therapies. The present study confirmed this suggestion by finding that clients perceived the use of the no-suicide contract as of little effectiveness when compared to the other treatment methods assessed. Beulow and Range (2001) evaluated the perceptions of 112 college students with regard to three no-suicide contracts that varied in length and specificity and included a choice of treatment options, including no-suicide contracts, medication, improved job satisfaction, fear of death, and other factors. Results of this study indicated that the use of the no-suicide contract as a technique of therapy was rated last among seven factors that might decrease suicidal desire. The present study strongly confirmed Beulow and 84 Range?s results by finding that clients perceived the use of the no-suicide contract as of little effectiveness as compared to the other treatment methods assessed. Jones and O?Brien (1990) conducted a study of 39 psychiatrically hospitalized children who had attempted, talked about, or engaged in self-mutilation behaviors. Contracts were developed in which the children would receive privileges based on meeting the terms of their agreements. No-suicide contracts were included in the types of contracts used. Results of this study indicated that the children rated their experience of contracting as very high in helping them to change their behaviors. However, there were many flaws in the Jones and O?Brien study, including the absence of random assignments, rotation of hospital personnel, and an absence of confidentiality in the responses. While the Jones and O?Brien study was limited to children, and therefore its results may not necessarily be transferable to adults, there is no apparent reason to believe that the experience of children on the issue of efficacy of suicide treatment methods must be different than that of adults. In the absence of evidence to the contrary, the present study contradicted Jones and O?Brien?s findings by assessing client perceptions of the use of the no-suicide contract as of little effectiveness as compared to the other treatment methods assessed. Maluccio and Marlow (1974) suggested that contracts of various kinds, including no-suicide contracts, might be used to contribute to effective outcomes in social work intervention, and reasoned that the contracting process could be used to assist the client and social worker to explore and reach agreement on treatment goals, producing a sense of immediate involvement, mutual commitment, and a basis for future review of accomplishments. However, they did not study the perceptions of the no-suicide contract 85 by suicidal clients. The present study, finding that the use of no-suicide contracts is perceived as of little effectiveness compared to the other treatment methods assessed, contradicts Maluccio and Marlow?s unsupported suggestion. Hipple and Cimbolic (1979) speculated that the use of no-suicide contracts were ?very effective during all phases of treatment? (p. 72), but the assertion is not substantiated with data. The present study contradicted Hipple and Cimbolic?s assertion, finding that the use of the no-suicide contract is perceived as of little effectiveness compared to the other treatment methods assessed. When other methods of intervention were considered, the present study found that the clients assessed had specific preferences for the type and nature of intervention. The use of medication as a treatment method was strongly identified as the most effective by the participants. Interestingly, participants indicated that medication was the most helpful treatment method, which was consistent with recommendations throughout the literature (Bongar, 1991; Bongar et al., 1998; Joiner et al., 1999; Mahrer, 1993). When assessing all of the interventions, it was found that interventions were clustered together based on perception of effectiveness by the clients. This resulted in five clearly defined clusters of treatment methods. The first cluster consisted of the following seven treatment methods: medication, discussion of contributing stress factors, improvement in lifestyle health, increased number of appointments, open discussion of suicidal thoughts, improvement in problem-solving skills, and increase in social activities. These treatment options were rated statistically significantly higher than the next group of highest rated treatment options, and represented the clients? perception of the most effective treatment options of those assessed in the present study. 86 The factors identified in the first cluster represent not only common techniques in use for outpatient treatment, but also some of the basic coping skills necessary for all persons to maintain mental health (ability to articulate problems, ability to manage stress, ability to relate to others, ability to solve problems, etc.), and so it seems logical that those sorts of skills would be found by participants to be effective in addressing their ability to manage their tendency toward suicide as a mechanism for dealing with problems. The most highly rated factor of this cluster (and of the entire set of 22 treatment factors), the use of medication, may suggest an acknowledgment of the improving nature of medication specificity and effectiveness across the range of the mental health professions. Bongar (1991), Bongar et al. (1998) and Rudd et al. (1999) indicated that psychotropic medication may often be necessary and is advised as a treatment option because it ensures the stability necessary for continuing outpatient care. Mothersole (1996) indicated that a no-suicide contract could be helpful in strengthening the counselor-client alliance and giving the client cognitive control. The participants in this research indicated that the use of the no-suicide contract is not needed to accomplish this goal; they rated direct discussion of contributing stress factors and open discussion of suicidal thoughts as among the most effective treatment methods of those available to them, contradicting Mothersole?s suggestion. Beulow and Range (2001) assessed the opinions of college students regarding the use of no-suicide contracts; these students rated development of stronger coping skills as among the highest rated factors. The participants in the present study included improvement in problem-solving skills among this first cluster of highly effective interventions, confirming Beulow and Range?s work. 87 Participants in the present research indicated that increasing the number of appointments was among the most helpful treatment techniques, confirming a previous recommendation of Ewalt (1967) and Mahrer and Bongar (1993). Miller et al. (1998) indicated that discussing and sharing the burden of managing waxing and waning suicidal thoughts is a more effective treatment strategy than the use of a no-suicide contract; this finding was confirmed by the participants in the present research, who included both open discussion of suicidal thoughts and discussion of stress factors as among the most helpful interventions. The second clearly delineated cluster of effective treatment methods identified by the research participants consisted of six treatment methods which all were rated within a statistically significant grouping: improvement in self-control planning skills, identification of alternate responses to suicide, participation in faith-based activities, hospitalization, establishing 24-hour counselor availability, and an assessment of living environment safety. The factors identified in this cluster also addressed fundamental life skills (improvement in self-control planning skills), and begin to focus on actions taken to address not the internal factors contributing to the tendency toward suicide-based thoughts and activities but instead as those related to the suicide-based activity itself (identification of alternate responses to suicide, hospitalization, establishing 24-hour availability of counselor). The third cluster of treatment methods identified by this research consisted of collaboratively revising the treatment plan, increasing the session length, and frequent evaluation of suicide risk. Research participants rated this cluster within a statistically significant grouping. The focus of all the treatment methods in this third, narrowly rated 88 cluster reflect approaches made toward making the counseling experience more effective instead of making the client?s ability to manage stress and problems in the world more effective; this is potentially the impetus behind the relatively low effectiveness rankings for these treatment methods. The next cluster of treatment methods consisted of contacting family members, creating a list of emergency contacts, participation in group counseling, and establishing a check-in system. Research participants rated this cluster within a statistically significant grouping. These factors relate to structure, management, and degree of participation of family members in the treatment, and were viewed as of poor and declining effectiveness by the research participants. The final cluster produced by the participants? ranking of 22 treatment methods was well delineated within a statistically significant grouping as the least effective of the treatment options assessed in the study. These two ranked below the others in a statistically significant way, indicating that there were viewed as quite effectively by the research participants. As observed with regard to the poorly rated treatment methods in the previous clusters, the treatment methods in this last cluster were ones which focus on the mechanics of the treatment provided and choice of treatment options. It appears that research participants considered choices related to the mechanics of the treatment to be of little effectiveness as compared to treatment methods that addressed their life skills concerns. The results of this study did not support a relationship between demographic factors such as age, gender, time of treatment, and number of attempts and the relative rankings of the treatment methods. Like the study by Davis et al. (2002), the present 89 research found that age and gender did not impact the responses of participants regarding their rating of no-suicide contracts. However, Davis et al. also found that the number of suicide attempts was related to the evaluation of the no-suicide contract as an intervention technique; that there was an inverse relationship between the number of suicide attempts and the perceived effectiveness of the no-suicide contract. The present research failed to confirm this finding. Limitations of This Study The primary limitation of this study was the potentially reduced reliability of the results based on the relatively small number of participants. While the reliability of the present results meets requirements of statistical significance, an increased number of participants would have helped to confirm the results. Potential participants were sought from the flow of clients through two satellite locations of a mental health agency from roughly 1,100 active clients during the ten week collection period; it was thought that a large number of participants would be available, and of those, a significant number of participants would volunteer to participate. However, this was not seen. In addition, the study was limited to those participants who self-identified for inclusion in the study?the consequent results necessarily neglect the opinions of those under treatment for a tendency toward suicide who did not self-identify for participation. Another limitation of the study involved the choice of treatment methods presented for the participants to evaluate; the data collected pertaining to research question one strongly indicated that clients considered the effectiveness of no-suicide contracts to be lowest of the treatment options available for clients to evaluate. Increased 90 reliability of the research may have been obtained by offering a treatment method which would have been rated as lower than the use of no-suicide contracts, and therefore avoiding the situation of having the treatment method of interest develop as an outlier of the participants? evaluation process. A limitation may have occurred as a result of the time lapse between the application of the instrument to measure the participants? view of effective treatment methods and the actual implementation of the treatment method. It is recognized that some change in judgment may be produced as a result of reflection on a past event and the ebbing of emotional involvement that occurs with the passage of time?the participants? view of what was effective at the time of treatment may be influenced by subsequent events. Reducing the time lapse between application of treatment and assessment of efficacy may have produced more valid results. The population who volunteered for participation in the study was 70 percent female; since older males statistically pose a greater risk of suicide (American Association of Suicidology, 2006), the results of this study may not generalize to the larger population, and this represents an additional limitation of the study. The pool of potential participants from which the population was obtained consists of chronically mentally ill individuals who may have tended towards more serious psychological diagnoses. As indicated in Chapter four, 25% percent of the population had identified themselves as having been in counseling for more than a decade. Therefore, it is not necessarily evident that the results of the present study can be transferred to the population at large. In addition, there were potentially effective treatment methods that were not included on the list of treatment methods assessed in the study, such as 91 establishing a meaningful counselor-client relationship; including this option as an assessable treatment option would have expanded the range of treatment options available for participant assessment. The no-suicide contract was operationally defined as a verbal or written agreement, but that distinction was not made on the instrument used to collect the participants? perspectives on the effectiveness of treatment methods. It is possible, therefore, that some participants may have assumed that the question concerning no- suicide contracts referred only to written contracts; one participant indicated that she had made a verbal commitment and therefore did not complete the second instrument, that gathered information about the participant?s experience with no-suicide contracts. If the distinction had been made clear, this participant?s perspective of her experience with the verbal no-suicide contract could have been included in the research results. In addition, the instrument did not contain a definition of a no-suicide contract, but it did contain and refer to some homonyms for no-suicide contract such as no-harm contract, commitment to live, and others. This researcher intended to treat verbal and written no-suicide contracts similarly, but that intention was not communicated to the participants on the instrument, and it may be ineffective to treat these different types of arrangements in a similar way. There was an ambiguity of terms on the instrument that may have influenced the participants? understanding of what was being asked. For example, on the Demographics and treatment Intervention Effectiveness Scale, one of the treatment options was Establishing 24-hour Availability of Counselor. It is possible that this treatment method might have been better understood by the participant if some other verb was used to 92 communicate the essence of the treatment method from the participant?s point of view. Another example of ambiguity in the instrument was Use of Formal Surveys. This term, while understood in the profession, may not have been immediately apparent to participants. Other limitations inherent in the use of a written instrument include the necessity that the participant be able to read the English language, be able to assess their own experiences, be able to maintain the necessary attention span to complete the instrument, and be able to write sufficiently well to indicate their responses. These factors may have contributed to the low number of participants, since those who could not complete the instrument or who suspected that they might not have sufficient reading, self-assessment, or writing skills to complete the instrument would be influenced to decline the invitation to participate. The No-Suicide Contract Survey, which was the instrument used to collect information from the participants who indicated that they had completed a no-suicide contract at some point in their treatment, was adapted for use from an existing survey. This existing instrument had only been used once previously?while reliability and validity had been established, an increased degree of reliability and validity may have been available with a more recognized or better established tool had been used instead. Recommendations for Future Research Future research that uses clients undergoing outpatient treatment who have struggled with suicide ideation to evaluate the effectiveness of treatment methods should include a wider range of treatment methods, some of which the researcher believes will 93 be rated very low, in the choices offered for evaluation so as to prevent or reduce the tendency towards the creation of statistical outliers with regard to the treatment methods of interest. In addition, repeating the study with an increased number of participants would contribute to the reliability and validity of the results. This study examined client perceptions of the use of no-suicide contracts; an avenue for future research may be to conduct the same kind of study regarding counselors? perceptions of the effectiveness of various treatment options. Given that the debate of no-suicide contracts continues, performing an evaluation of counselors? perceptions of the effectiveness of the no-suicide contract may produce data demonstrating that counselors agree with the findings of this study, that the use of no- suicide contracts is seen by clients as minimally effective, compared to other treatment options assessed. Future research might include an evaluation of the clients? perception of effectiveness of treatment method not only after their treatment, as was done in the present research, but before their treatment. An analysis of the perception of the clients as to what might be an effective treatment method before the treatment, as compared to their perceptions of what was effective for them after the treatment, may help future practitioners manage the treatment requests of clients for whom they are planning treatment as well as illuminate the degree to which the expectations of clients entering treatment matches the experience of similar clients who have completed treatment. This study limited its attention to adults; Davis et al. (2002) suggested that surveying previously suicidal adults and children to assess their agreement on the value or lack thereof of no-suicide contracts may be an appropriate avenue for future research. 94 Duplicating this study with children may be useful in determining the degree to which children?s perception of the effectiveness of treatment methods agrees with that of adults. The results of the present study imply that clients potentially perceive the use of no-suicide contracts as an ineffective part of a program of treatment of suicidal thoughts or behavior. The clients who participated in this study rated the effectiveness of treatment methods in a way that suggests that the perception of effectiveness may be related to the degree to which the treatment method provides, develops, or expands necessary and basic life skills and abilities used to address and resolve problems. Research designed to distinguish between treatment methods that improve life skills and ones that impact the mechanics of the treatment session may provide data to confirm or contradict this result. Implications for the Field of Counseling and Training It is well documented that the attempted or completed suicide is the most common and most challenging of clinical emergencies for mental health professionals (Ewalt, 1967; Knapp & VandeCreek, 1983; Shein, 1976). This study helped to gather information on treatment methods directly from the clients that will be served by mental health counseling professionals, therefore, this data may be appropriate to integrate into training materials. This research demonstrated that clients perceive the no-suicide contract technique to be of poor effectiveness, and yet, community mental health textbooks continue to suggest their use for treating suicidal clients (Gladding & Newsome, 2004; Howatt, 2000; Seligman, 2004). In the absence of additional training, counselors may rely on the technique of the no-suicide contract, which was demonstrated by this study to be perceived as least helpful 95 in relation to other treatment options by suicidal clients. The implication is that the mental health professions may be well served in reconsidering whether the no-suicide contract should be presented as a potential treatment option for this population. Few counselor education, psychology, or psychiatry programs train students to deal with suicidal clients (Foster & McAdams, 2000; Miller et al., 1998), yet we know that most mental health counselors will experience a client who attempts suicide (Schwartz & Rogers, 2004). This disjoint between training and actual practice represents an avenue for change in the field of counseling and training. It may also be useful to seek a way to improve the comfort level that counseling students develop toward suicidal clients; while it may not be possible as a result of time restraints to develop and execute courses that focus on suicide and suicide treatment methods alone, these issues can be incorporated into existing curriculum throughout the course of a counseling program. The results of this research included an identification of a cluster of seven treatment options that were perceived as most effective by the clients that participated in the study; the no-suicide contract technique was not among them, but the identification of these multiple and roughly equally highly effective treatment options suggests that a multiple-option approach might be considered by counselors working with suicidal clients. Ewalt suggested as early as 1967 that asking clients to promise to not commit suicide was not a viable treatment option, and that instead, treatment options similar to those identified in the present study as of high effectiveness be considered. Similarly, Mahrer and Bongar (1993), Kleespies et al. (1999), Joiner et al. (1999) presented lists of potential treatment options that were considered effective for the treatment of suicidal clients that did not include the use of the no-suicide contract. 96 In view of the continued practice of no-suicide contracting in the treatment of suicidal clients and the results of the present study that suggests that no-suicide contracting is seen as being of little effectiveness as compared to the other treatment options assessed, it would seem appropriate to consider if the initial training of counselors includes recommendations to use the no-suicide contracting technique as a treatment method. If so, perhaps this should change. Legally, the no-suicide contract cannot be made binding, since those clients who enter into it are by definition not fully mentally capable of participating as is required by the law (R.E. Poundstone, personal communication, September 28, 2005). Ethical mental health practice requires that counselors re-evaluate treatment practices periodically to incorporate advances in research; the no-suicide contract has been demonstrated to be of little effectiveness. To continue to use the technique in view of these results may be a questionable ethical practice. Instead, students in counseling should be trained on how to appropriately assess suicidality and how to use treatment options that are of demonstrated effectiveness in dealing with these clients and their needs. Effective treatment options might include those identified by the present study in the highly effective cluster of seven options described above. The use of treatment options should be supplemented via the use of a formal survey mechanism to assess the degree of suicidality; in doing so, a counselor can choose treatment options tailored to the client?s particular situation. Incorporating assessment surveys also has the effect of producing best practices data which can be utilized by other counselors. While the participants in the present study rated the use of formal surveys as of low effectiveness, the survey would be used not to provide treatment for the suicidal 97 condition but instead as a way to monitor the effectiveness of treatment methods that are used. Additionally, training of counselors must focus on how to engage in a meaningful conversation to lead the client to a consideration on his or her suicidal thoughts and ways in which the client can manage the impulses and establish an ability to self-resolve the suicidal desires in a non-lethal manner. The present study found that a discussion of the suicidal thoughts and of contributing stress factors was perceived as among the highest of the treatment options assessed. Given the increasing number of suicides in this nation, the ambiguity that currently exists among the judicial system regarding a unified definition of standard of care, and the emotional devastation that family members experience when a loved one commits suicide, it seems fitting that action at the national level be directed toward establishing a national standard of care to protect the psychologically wounded who consider suicide an option, and to those who provide services to them. Whether the use of a no-suicide contract is included as part of the assessment process and that national standard remains to be seen. The continued research and collection of empirical data regarding no-suicide contracts may impact that decision process, and therefore, needs to continue. Summary The current study revealed that among the many techniques available for use with suicidal clients, the no-suicide contract may be among the least useful. This study did not provide or generate evidence that the use of a no-suicide contract is harmful; that issue was not in question nor was it being researched. If the use of the no-suicide contract can 98 be identified as of poor effectiveness, as this research seemed to indicate, then counselors can avoid wasting limited time and energy with the no-suicide contract and instead focus on the use of treatment methods that were seen to be of improved effectiveness. As this research was limited to clients in counseling on an outpatient basis, it was distinguished from previous research involving inpatient clients. However, this research supplemented the existing work and by seeking and analyzing the viewpoints of current clients, added their perspective on the issue of the suitability of the no-suicide contracting technique. The relative rankings of treatment methods were seen by this study to be fixed across age, gender, time in treatment, and number of suicide attempts graduations, and so it is thought that the use of no-suicide contracts will be of little effectiveness in the treatment of suicidal adults. 99 REFERENCES Alabama Department of Public Health. Retrieved June 30, 2006, from http://adph.org/suicideprevention American Association of Suicidology. Retrieved May 8, 2006, from http://www.suicidology.org Assey, J. L. (1985). The suicide prevention contract. Perspectives in psychiatric care, 23(3), 99-103. Ayd, F. J., & Palma, J. M. (1999). Suicide: Risk recognition and prevention. Psychiatric Times, May, 38-41. Berman, A. L, (1990). Standard of care in assessment of suicidal potential. Psychotherapy in Private Practice, 8(2), 35-41. Berman, A. L., & Cohen-Sandler, R. (1983). Suicide and malpractice: Expert testimony and the standard of care. Professional Psychology: Research and Practice, 14(1), 6-19. Bernstein, R. M., Feldberg, C., & Brown, R. (1991). After-hours coverage in psychology training clinics. Professional Psychology: Research and Practice, 22, 204-208. Black, H. C. (1979). Black?s law dictionary. St. Paul, MN: West. Boehm, S. (1989). Patient contracting. Advances in Nursing Science, 7, 143-153. 100 Bongar, B. (2002). The suicidal patient: Clinical and legal standards of care (2nd ed.). Washington, DC: American Psychological Association. Bongar, B. (1991). The suicidal client: Clinical and legal standards of care. Washington, DC: American Psychological Association. Bongar, B., & Harmatz, M. (1989). Graduate training in clinical psychology and the study of suicide. Professional Psychology: Research and Practice, 20, 209-213. Bongar, B., Maris, R. W., Berman, A. L., & Litman, R. E. (1998). Outpatient standards of care and the suicidal patient. In B. Bongar et al. (Eds.), Risk management with suicidal patients (pp. 4-33). New York: Guilford Press. Bongar, B., Peterson, L. G., Harris, E. A., & Aissis, J. (1989). Clinical and legal considerations in the management of suicidal patients: An integrative overview. Journal of Integrative and Eclectic Psychotherapy, 8(1), 53-67. Buelow, G., & Range, L. M. (2001). No-suicide contracts among college students. Death Studies, 25, 583-592. Busch, K. A., Clark, D. C., Fawcett, J., & Kravitz, H. M. (1993). Clinical features of inpatient suicides. Psychiatric Annals, 23, 256-262. Capuzzi, D. (2002). Legal and ethical challenges in counseling suicidal students. Professional School Counseling, 6, 36-45. Centers for Disease Control. Retrieved June 30, 2006, from http://www.cdc.gov/ncipc/factsheets/suifacts.htm Chemtob, C. M., Hamada, R. S., Bauer, G. B., Torigoe, R. Y., & Kinney, B. (1988). Patient suicide: Frequency and impact on psychologists. Professional Psychology: Research and Practice, 19, 421-425. 101 Chiles, J. A., & Strosahl, K. D. (1995). The suicidal patient: Principles of assessment, treatment, and case management. Washington, DC: American Psychiatric Press. Clark, D. C., & Kerkhof, J. F. M. (1993). No-suicide decisions and suicide contracts in therapy. Crisis, 14(3), 98-99. Counts, D. (1991). Suicide in different ages from a cross-cultural perspective. In A. A. Leenaars (Ed.), Life span perspectives of suicide: Time-lines in the suicide process (pp. 215-228). New York: Plenum Press. Davis, S., Williams, I. S., & Hays, L. W. (2002). Psychiatric inpatients? perceptions of written no-suicide agreements: An exploratory study. Suicide and Life- Threatening Behavior, 32(1), 51-66. Davidson, M. W., Wagner, W. G., & Range, L. M. (1995). Clinician?s attitudes toward no-suicide agreements. The American Association of Suicidology, 25(3), 410-414. Deutsch, C. J. (1984). Self-report sources of stress among psychotherapists. Professional Psychology: Research and Practice, 15, 833-845. Drew, B. L. (2001). Self-harm behavior and no-suicide contracting in psychiatric inpatient settings. Archives of Psychiatric Nursing, 15(3), 99-106. Drew, B. L. (1999). No-suicide contracts to prevent suicidal behavior in inpatient psychiatric settings. Journal of the American Psychiatric Nurses Association, 5(1), 23-28. Drye, R. C., Goulding, R. L., & Goulding, M. E. (1973). No-suicide decisions: Client monitoring of suicidal risk. American Journal of Psychiatry, 130(2), 171-174. 102 Dulit, R. A., & Michels, R. (1992). Psychodynamics and suicide. In D. Jacobs (Ed.). Suicide and clinical practice (pp. 147-167). Washington, DC: American Psychiatric Press. Echterling, L. G., Presbury, J. H., & McKee, J. E. (2005). Crisis intervention: Promoting resilience and resolution in troubled times. Upper Saddle River, NJ: Pearson Education. Egan, M. P. (1997). Contracting for safety: A concept analysis. Crisis, 18(1), 17-23. Egan, M. P., Rivera, S. G. Robillard, R. R., & Hanson, A. (1997). The no-suicide contract: Helpful or harmful? Journal of Psychosocial Nursing Mental Health Services, 3, 31-32. Ellis, T. E., & Dickey, T. O. (1998). Procedures surrounding the suicide of a trainee?s patient: A national survey of psychology internships and psychiatry residency programs. Professional Psychology: Research and Practice, 29(5), 492-497. Ewalt, J. R. (1967). Other psychiatric emergencies. In A. M. Freedman, H. I. Kaplan, & H. S. Kaplan (Eds.), Comprehensive textbook of psychiatry (pp. 1179-1187). Baltimore: Williams & Wilkins. Farber, B. A. (1983). Psychotherapist?s perceptions of stressful patient behavior. Professional Psychology: Research and Practice, 14, 697-705. Farrow, T. L. (2002). Owning their expertise: Why nurses use no suicide contracts rather than their own assessments. International Journal of Mental Health Nursing, 11, 214-219. Farrow. T. L. (2000). No suicide contracts are potentially dangerous. Nursing New Zealand, 6(9), 3-4. 103 Farrow, T. L., & O?Brien, A. J. (2003). No-suicide contracts and informed consent: An analysis of ethical issues. Nursing Ethics, (10)2, 199-207. Farrow, T. L., Simpson, A. I. F., & Warren, H. B. (2002). The effects of the use of no- suicide contracts in community crisis situations: The experience of clinicians and consumers. Brief Treatment and Crisis Intervention, 2(3), 241-246. Fine, M. A., & Sansome, R. A. (1990). Dilemmas in the management of suicidal behavior in individuals with borderline personality. American Journal of Psychotherapy, 44, 160-171. Foster, V., & McAdams, C. (1999). The impact of client suicide in counselor training: Implications for counselor education and supervision. Counselor Education and Supervision, (39)1, 22-33. Freidman, R. S. (1989). Hospital treatment of the suicidal patient. In D. G. Jacobs & H. N. Brown (Eds.), Suicide: Understanding and responding: Harvard Medical School perspectives on suicide (pp. 379-402). Madison, CT: International Universities Press. Getz, W. L., Allen, D. B., Myers, R. K., & Lindner, K. C. (1983). Brief counseling with suicidal persons. Lexington, KY: Lexington Books. Gladding, S. T., & Newsome, D. W. (2004). Community and agency counseling (2nd ed.). Upper Saddle River, NJ: Pearson Education. Goldblatt, M. J. (1994). Hospitalization of the suicidal patient. In A. A. Leenaars, J. T. Maltsberger, & R. A. Neimeyer (Eds.), Treatment of suicidal people (pp. 153- 165). Washington, DC: Taylor & Francis. 104 Goulding, M., & Goulding, R. L. (1979). Changing lives through redecision therapy. New York: Brunner/Mazel. Greenblatt, M., Moore, R. F., Albert, R. S., & Solomon, M. H. (1963). The prevention of hospitalization: Treatment without admission for psychiatric patients. New York: Grune & Stratton. Gutheil, T. G. (1992). Suicide and suit: Liability after self-destruction. In D. Jacobs (Ed.). Suicide and clinical practice (pp. 147-167). Washington, DC: American Psychiatric Press. Gutheil, T. G. (1980). Paranoia and progress notes: A guide to forensically informed psychiatric recordkeeping. Hospital and Community Psychiatry, 31, 479-482. Hawton, K. (2000). General hospital management of suicide attempters. In K. Hawton & K. van Heeringen (Eds.), The international handbook of suicide and attempted suicide (pp. 519-537). New York: Wiley. Hawton, K., & van Heeringen, K. (2000). The international handbook of suicide and attempted suicide. New York: Wiley. Hayes, W. S., & Davis, L. L. (1980). What is a health care contract? Health Values, 4(2), 82-86, 89. Hennig, C. W., Crabtree, C. R., & Baum, D. (1998). Mental health CPR: Peer contracting as a response to potential suicide in adolescents. Archives of Suicide Research, 4, 169-187. Hillard, , J. R. (1990). Suicide. In J. R. Hillard (Ed.). Manual of clinical emergency psychiatry. Washington, DC: American Psychiatric Press. 105 Hipple, J., & Cimbolic, P. (1979). The counselor and suicidal crisis. Springfield, IL: Charles C. Thomas. Howatt, W. A. (2000). The human services counseling toolbox: Theory, development, technique, and resources. Belmont, CA: Brooks/Cole. Jacobs, D. G. (1992). Evaluating and treating suicidal behavior in the borderline patient. In D. G. Jacobs (Ed.), Suicide and clinical practice (pp. 115-130). Washington, DC: American Psychiatric Press. Joiner, T. E., Walker, R. L., Rudd, D. M., & Jobes, D. A. (1999). Scientizing and routinizing the assessment of suicidality in outpatient practice. Professional Psychology: Research and Practice, 30(5), 447-453. Jobes, D. A., & Berman, A. L. (1993). Suicide and malpractice liability assessing and revising policies, procedures, and practice in outpatient settings. Professional Psychology: Research and Practice, 24(1), 91-99. Jones, R. N., & O?Brien, P. (1990). Unique interventions for child inpatient psychiatry. Journal of Psychosocial Nursing, 28, 29-31. Jones, R. N., O?Brien, P., & McMahon, W. M. (1993). Contracting to lower precaution status for child psychiatric patients. Journal of Psychosocial Nursing, 31, 6-10. Kelly, K. T., & Knudson, M. P. (2000). Are no suicide contracts effective in preventing suicide in suicidal clients seen by primary care physicians? Archives of Family Medicine, 9(10), 1119-1123. Kleespies, P. M., Deleppo, J. D., Gallagher, P. L., & Niles, B. L. (1999). Managing suicidal emergencies: Recommendations for the practitioner, 30(5), 454-463. 106 Knapp, S., & VandeCreek, L. (1983). Malpractice risks with suicidal patients. Psychotherapy: Theory, research and practice, 20(3), 274-280. Kreitman, N. (1986). The clinical assessment and management of the suicidal patient. In A. Roy (Ed.), Suicide (pp. 181-195). Baltimore: Williams & Wilkins. Kroll, J. (2000). Use of no-suicide contracts by psychiatrists in minnesota. American Journal of Psychiatry, 157, 1684-1686. Lee, J. B., & Bartlett, M. L. (2005). Suicide prevention: Critical elements for managing suicidal clients and counselor liability without the use of a no-suicide contract. Death Studies, 29, 1-19. Loomis, M. (1985). Levels of contracting. Journal of Psychosocial Nursing, 23(3), 9-14. Loomis, M. (1982). Contracting for change. Transactional analysis Journal, 12(1), 51- 55. Mahrer, J. (1993). The use of no-suicide contracts and agreements with suicidal patients. Doctoral dissertation, Pacific Graduate School of Psychology. Dissertation Abstracts International, 54 (6-B), 3345. Mahrer, J., & Bongar, B. (1993). Assessment and management of suicide risk and the use of the no-suicide contract. Innovations in Clinical Practice: A Source Book, 12, 277-293. Maltsberger, J. T. (1994). Calculated risk-taking in the treatment of suicidal patients: Ethical and legal problems. In A. A. Leenaars, J. T. Maltsberger, & R. A. Neimeyer (Eds.), Treatment of suicidal people (pp. 195- 205). Washington, DC: Taylor & Francis. 107 Maltsberger, J. T. (1991). The prevention of suicide in adults. In A. A. Leenaars (Ed.), Lifespan perspectives of suicide: Time-lines in the suicide process (pp. 295-307). New York: Plenum Press. Maltsberger, J. T. (1986). Suicide risk: The formulation of clinical judgment. New York: New York University Press. Maluccio, A. N., & Marlow, W. D. (1974). The case for the contract. Social Work, 19(1), 28-36. McEnany, G. W., & Tescher, B. E. (1985). Contracting for care: One nursing approach to the hospitalized borderline patient. Journal of Psychosocial Nursing, 23(4), 11- 18. Miller, M. C. (1999). Suicide-prevention contracts: advantages, disadvantages, and an alternative approach. In D.G. Jacobs (Ed.), The Harvard Medical School guide to suicide assessment and intervention (pp. 463-481). San Francisco: Jossey-Bass. Miller, M. C., Jacobs, D. G., & Gutheil, T. G. (1998). Talisman or taboo: The controversy of the suicide-prevention contract. Harvard Review of Psychiatry, 6(2), 78-87. Mothersole, G. (1997). Contracts and harmful behavior. In C. Sills (Ed.), Contracts in Counselling (pp.113-124). London: Sage Publications. Mothersole, G. (1996). Existential realities and no-suicide contracts. Transactional Analysis Journal, 26(2), 151-159. Motto, J. A. (1979). Guidelines for the management of the suicidal patient. Weekly Psychiatric Update Series, 20(3), 2-8. 108 Neimeyer, R. A. (2000). Suicide and hastened death: Toward a training agenda for counseling psychology. The Counseling Psychologist, 28, 551-560. Neimeyer, R. A., & Pfeiffer, A. M. (1994). The ten most common errors of suicide interventionists. In A. A. Leenaars, J. T. Maltsberger, & R. A. Neimeyer (Eds.), Treatment of suicidal people (pp. 207-224). Washington, DC: Taylor & Francis. Olson v Molzea, 558 S.W.3d 429 (Tenn.1997) Pary, R., Lippmann, S., & Tobias, C. R. (1988). A preventive approach to the suicidal patient. Journal of Family Practice, 26, 185-189. Pfaff, J. J., Acres, J., & Wilson, M. (1999). The role of general practitioner?s in parasuicide: A Western Australia perspective. Archives of Suicide Research, 5, 207-214. Porubiansky v Emory Univ., 156 Ga. App. 602, 275 S.E.2d 163 (1981) Range, L. M., Campbell, C., Kovac, S. H., Marion-Jones, M., Aldridge, H., Kogas, S. et al. (2002). No suicide contracts: An overview and recommendations. Death Studies, 26, 51-74. Reid, W. H. (2005). Contracting for safety redux, 11(1), 54-57. Reid, W. H. (2004). Avoiding the malpractice snare: Documenting suicide risk management. Journal of Psychiatric Practice, 10(3), 1-5. Reid, W. H. (2003). Risk assessment, prediction, and forseeability. Journal of Psychiatric Practice, 9(1), 82-86. Reid, W. H. (1998). Promises, promises: Don?t rely on patients? no-suicide/no-violence contracts. Journal of Practical Psychiatry and Behavioral Health, 4, 316-318. 109 Robins, E., Murphy, G. E., Wilkinson, R. H., Gassner, S., & Kayes, J. (1996). Some clinical considerations in the prevention of suicide based on a study of 134 successful suicides. In J. T. Maltsberger & M. J. Goldblatt (Eds.), Essential papers on suicide (pp. 142-159). New York: New York University Press. Rosen, B. (1978). Written treatment contracts: Their use in planning treatment programs for inpatients. British Journal of Psychiatry, 133, 410-415. Rudd, M. D. (1996). The role of no-suicide agreements in the treatment of suicidal ideation and behavior: A comment. Suicide and Life-Threatening Behavior, 26, 215. Rudd, M. D., & Joiner, T. (1998). The assessment, management, and treatment of suicidality: Toward clinically informed and balanced standards of care. American Psychological Association, 5(2), 135-150. Rudd, M. D., Joiner, T. E., Jobes, D. A., & King, C. A. (1999). The outpatient treatment of suicidality: An integration of science and recognition of limitations. Professional Psychology: Research and Practice, 30(5), 437-446. Schwartz, M. D., & Errera, P. (1963). Psychiatric care in a general hospital emergency room: II Diagnostic features. Archives of General Psychiatry, 9, 113. Schwartz, R., & Rogers, J. (2004). Suicide assessment and evaluation strategies: A primer for counseling psychologists. Counseling Psychology Quarterly, 17, 89- 97. Seligman, L. (2004). Technical and conceptual skills for mental health professionals. Upper Saddle River, NJ: Pearson Education. 110 Selzer, M. A., Koenigsberg, H. W., & Kernberg, O. F. (1987). The initial contract in the treatment of borderline patients. American Journal of Psychiatry, 144, 927-930. Shea, S. C. (1999). The practical art of suicide assessment: A guide for mental health professionals and substance abuse counselors. New York: Wiley. Shein, H. M. (1976). Obstacles in the education of psychiatric residents. Omega, 7, 75- 82. Sills, C. (1997). Contracts and contract making. In C. Sills (Ed.), Contracts in counseling (pp. 11-35). Thousand Oaks, CA: Sage. Silverman, M. M., Berman, A. L., Bongar, B., Litman, R. E., & Maris, R. W. (1994). Inpatient standards of care and the suicidal patient Part II: An integration with clinical risk management. Suicide and Life-Threatening Behavior, 24, 152-169. Simon, R. I. (1988). Concise guide to clinical psychiatry and the law. Washington, DC: American Psychiatric Press. Simon, R. I. (2004). Suicide risk: Guidelines for clinically based risk management. Washington, DC: American Psychiatric Publishing. Simon, R. I. (1999). The suicide prevention contract: Clinical, legal, and risk management issues. Journal of the American Academy of Psychiatry Law, 27(3), 445-450. Simon, R. I. (Ed.) (1992). Review of clinical psychiatry and the law, vol. 3. Washington, DC: American Psychiatric Press. Simon, R. I. (Ed.) (1991). Review of clinical psychiatry and the law, vol. 2. Washington, DC: American Psychiatric Press. 111 Smith, C. W., & Bope, E. T. (1986). The suicidal patient: The primary care physician?s role in evaluation and treatment. Postgraduate Medicine, 79, 195-202. Stanford, E. J., Goetz, R. R., & Bloom, J. D. (1994). The no harm contract in the emergency assessment of suicidal risk. Journal of Clinical Psychiatry, 55(8), 410- 414. Stepakoff v. Kantar, 473 N.E.2d. 1131 (Mass.1985) Suicide Prevention Action Network USA. Retrieved June 30, 2006, from http://spanusa.org Takahashi, Y. (1993). Suicide prevention in Japan. In A. A. Leenaars (Ed.), Suicidology (pp. 324-334). Northvale, NJ: Jason Aronson. Thompson, P. S., & Willis, J. C. (1982). Compliance challenges in a black lung clinic. Nursing Clinics of North America, 17, 513-521. Twimane, G. B. (1981). No-suicide contract for nurses. Journal of Psychiatric Nursing and Mental Health Services, 19(7), 11-12. Ungerleider, J. T. (1960). The psychiatric emergency: Analysis of six months? experience of a university hospitals? consultation service. Archives of General Psychiatry, 3, 593. U. S. Department of Health and Human Services (1990). Healthy people 2000: National health promotion and disease prevention objectives. Washington, DC: Author. VandeCreek, L., Knapp, S., & Herzog, C. (1987). Malpractice risks in the treatment of dangerous patients. Psychotherapy: Theory, Research, and Practice, 24, 145-153. Weiss, A. (2001). The no suicide contract: Possibilities and pitfalls. American Journal of Psychotherapy, 55(3), 414-419. 112 Wettstein, R. M. (1989). Psychiatric malpractice. In A. Tasman, R. E. Hales, & A. J. Frances (Eds.), Annual review of psychiatry (Vol. 8, pp. 392-408). Washington, DC: American Psychiatric Press. White, J. H. (1986). Behavioral intervention for the obese client. Nurse Practitioner, 11(1), 27-28, 30, 32. Whitley, J. S., & Denison, D. M. (1963). The psychiatric casualty. British Journal of Psychiatry, 109, 488. 113 APPENDICES 114 APPENDIX A IRB APPROVAL LETTER 115 116 APPENDIX B LETTER OF AGREEMENT 117 118 APPENDIX C SCRIPT 119 Script for Receptionist Handing out Research Packets Hello. GrandView Behavioral Health Centers are participating in a research project. All of our clients are being given this packet of information. While you are waiting for your appointment, please review and complete the contents of the package if you are eligible and would like to participate in this research project. This research is not part of the treatment or services you are receiving. If you have already participated, please do not complete another survey. Thank you. 120 APPENDIX D INFORMED CONSENT LETTER 121 122 APPENDIX E DEMOGRAPHICS AND TREATMENT INTERVENTION EFFECTIVENESS SCALE 123 Demographics and Treatment Intervention Effectiveness Scale Please indicate your response to the following questions by circling the appropriate answer. 1. Gender: Male Female 2. Age: 19-25 26-35 36-45 46-55 56-65 Older than 65 3. How long have you been receiving mental health counseling? 1-5 years 6-10 years 11-15 years 16-20 years More than 20 years 4. Have you ever had thoughts about committing suicide? Yes No 5. Have you ever made a gesture of suicide without actually planning to kill yourself? Yes No 6. Have you ever attempted suicide with the intent to kill yourself? Yes No 7. If you have ever attempted suicide, how many times have you attempted? 0 times 1 time 2 times 3 times 4 times 5 or more times 8. Have you ever agreed to a No-Suicide Contract? Yes If your answer is Yes, please proceed to Question #9 No If your answer is No, please proceed to Question #11 Note: A No-Suicide Contract can also be referred to as a No-Harm Agreement, Commitment to Live, Suicide Prevention Contract, No-Suicide Decision, or a Commitment for Safety, among many things. 9. How many times have you agreed to a No-Suicide Contract? 1 time 2 times 3 times 4 times 5 or more times 10. The last time you agreed to a No-Suicide Contract, were you: In an Inpatient Facility An Outpatient (being treated for mental health reasons) 11. Listed below are treatment methods your counseling professional may have used during your most recent suicidal episode. Please circle the response that best indicates how effective each method was in helping you to avoid harming yourself. If the identified method was not used, please circle Not Used. Very Not Not Useful Useful Neutral Useful Used No-Suicide Contract ------------------------------------- 5 4 3 2 1 Medication ------------------------------------------------ 5 4 3 2 1 Hospitalization -------------------------------------------- 5 4 3 2 1 Increase Number of Appointments -------------------- 5 4 3 2 1 Establish a Check-in System --------------------------- 5 4 3 2 1 Increase Counseling Session Length------------------- 5 4 3 2 1 Collaboratively Revise Treatment Plan---------------- 5 4 3 2 1 Use of Formal Survey(s) -------------------------------- 5 4 3 2 1 Contact Family Members-------------------------------- 5 4 3 2 1 Openly Discuss Suicidal Thoughts--------------------- 5 4 3 2 1 Discuss Contributing Stress Factors-------------------- 5 4 3 2 1 Improvement in Problem-Solving Skills-------------- 5 4 3 2 1 Identification of Alternate Responses to Suicide ---- 5 4 3 2 1 Improvement in Self-Control Planning Skills--------- 5 4 3 2 1 Participation in Group Counseling--------------------- 5 4 3 2 1 Create a List of Emergency Contacts------------------ 5 4 3 2 1 Frequent Evaluation of Suicide Risk------------------- 5 4 3 2 1 Improvement in Lifestyle Health (Diet, Sleep, etc.)- 5 4 3 2 1 Increase in Social Activities----------------------------- 5 4 3 2 1 Assessment of Living Environment Safety------------ 5 4 3 2 1 Establishing 24-hour Availability of Counselor------ 5 4 3 2 1 Participation in Faith-Based Activities---------------- 5 4 3 2 1 Other ____________________________________ 5 4 3 2 1 Other ____________________________________ 5 4 3 2 1 Other ____________________________________ 5 4 3 2 1 124 125 Please answer the following questions in the space provided. 12. What was the most effective treatment method used in helping you avoid harm to yourself, and why? ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ 13. What was the least effective treatment method used in helping you avoid harm to yourself, and why? ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ 14. Were you satisfied with the way your counseling professional dealt with your most recent suicidal episode? Yes No 15. If not, what would you have preferred your counseling professional had done? ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ If you have used a No-Suicide Contract, please complete the No-Suicide Contract Survey next. Upon completing the survey, seal the sheets in the envelope you received them in, and return the packet to the receptionist. If you have never been asked to participate in a No-Suicide Contract, then your participation is complete. Please place all forms back into the envelope, seal it, and return it to the receptionist. Thank you for participating in this study. 126 APPENDIX F NO-SUICIDE CONTRACT SURVEY 127 No-Suicide Contract Survey This survey is about your thoughts and feelings concerning the most recent No-Suicide Contract you agreed to. Please indicate how helpful or unhelpful you perceived that No-Suicide Contract to be. Indicate your response by placing an X in the column that reflects your response to each statement. Statement Strongly Disagree Disagree Neutral Agree Strongly Agree 1. I believed that writing out and signing my No- Suicide Contract was helpful to me. 2. I had confidence in my ability to keep the No- Suicide Contract that I participated in. 3. I made the No-Suicide Contract for myself. 4. I made the No-Suicide Contract for other people. 5. Writing out and signing my No-Suicide Contract gave me a greater sense of control. 6. I believed my No-Suicide Contract would stop my thoughts of suicide. 7. Writing out my No-Suicide Contract helped me better understand my thoughts and feelings about suicide. 8. I believed that making my No-Suicide Contract would help me feel less upset. 9. I believed my No-Suicide Contract helped by giving me the chance to think about my suicidal thoughts and feelings 10. Signing my No-Suicide Contract helped me to slow down and think about the things which had been upsetting me. 11. My participation in my No-Suicide Contract helped me realize I was responsible for living. 12. I believed I was asked to sign my No-Suicide Contract because I was cared about by the person(s) treating me. 13. I believed I had a good relationship with my provider(s) at the time I entered into my No- Suicide Contract. 14. I felt pressured into making my No-Suicide Contract. 15. I believed I was asked to sign my No-Suicide Contract as legal protection for those treating me. 16. My No-Suicide Contract probably helped the person(s) treating me feel less upset about my suicidal thoughts and feelings. 128 Statement Strongly Disagree Disagree Neutral Agree Strongly Agree 17. I believed participation in a No-Suicide Contract was requested as part of a routine. 18. I believed that participation in a No-Suicide Contract was given to me because of my unique problems. 19. I believed a written No-Suicide Contract was more helpful to me than a verbal No-Suicide Contract would have been. 20. My No-Suicide Contract helped me because I saw that others were involved in my treatment. 21. My No-Suicide Contract helped me to stay alive because I was committed to keeping my word. Steven E. Davis ? 2002, Modified and used by permission.