• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Training individuals in suicide prevention: Individual and organizational characteristics of effective gatekeepers

ProQuest Dissertations and Theses, 2011
Dissertation
Author: Jeffery Taylor Moore
Abstract:
Suicide remains a major public health problem in the United States. Training individuals known as gatekeepers to identify the signs and behaviors of suicide risk has been one of the most widely adopted prevention strategies. Due to financial constraints, it is not possible to train all members of a community as gatekeepers. Thus, it is more fiscally responsible to selectively train individuals within a community who possess key characteristics that will make them more effective gatekeepers. In this dissertation, the personality and organizational characteristics of effective gatekeepers were explored. Specifically, personality characteristics: Emotional Intelligence and Altruism, and organizational characteristics: social support, perceived organizational support, and gatekeeper role conflict were investigated as predictors of gatekeeper behavior. Two hundred and eighteen gatekeepers completed surveys immediately prior, immediately after, and six months following training that consisted of self-report measures of personality and organizational characteristics, and gatekeeper behavior. Poisson regression was employed to analyze the data because of the low-base rate occurrence of gatekeeper behavior. Analyses were conducted separately for the two gatekeeper training programs (ASIST & QPR). Mixed results were found in this study. Gatekeepers trained in ASIST were high in Emotional Intelligence and Altruism, resulting in range restriction. Gatekeepers trained in QPR varied more on personality characteristics, and Altruism positively predicted gatekeeper behavior six months following training. In terms of organizational predictors of gatekeeper behavior, supervisor emotional support buffered the relationship between gatekeeper role conflict and gatekeeper behavior for ASIST participants. The findings have direct implications for the training of community members as gatekeepers. Specifically, how selection principles can be applied to gatekeeper training, as well as how social support can be improved in organizations so that gatekeepers can overcome barriers (such as role conflict) in order to increase the effectiveness of trained gatekeepers. Although a number of limitations were present in the current study (i.e., low response rate and inability to generalize the findings), the potential application of these findings presents the opportunity to significantly change who is selected for training as a gatekeeper and how trainings are conducted. Future research directions include the measurement of gatekeeper behavior and effectiveness of gatekeepers.

iv     TABLE OF CONTENTS ABSTRACT ……………………………………………………………………………...ii TABLE OF CONTENTS ………………………………………………………………...iv LIST OF TABLES……………………………………………………………………......vi LIST OF FIGURES……………………………………………………………………..viii CHAPTER ONE: INTRODUCTION……………………………………………………..1 CHAPTER TWO: SUICIDE PREV ENTION GATEKEEPER TRAINING……………..5 What is a Gatekeeper Training? ………………………………………………………….5 Gatekeeper Training Programs ………………………………………………………..…5 What is a Gatekeeper’s Role in Suicide Prevention? …………………………………….6 Are Gatekeeper Training Programs Effective? …………………………………………..7 CHAPTER THREE: WHY DO PEOPLE HELP? A SOCIAL PSYCHOLOGICAL PERSPECTIVE……………………………………………………………………….....16 CHAPTER FOUR: PREDICTORS OF GATEKEEPER BEHAVIOR ………………...20 Individual Characteristics ………………………………………………………………20 Environmental Characteristics ………………………………………………………….24 Buffering Effect of Social Support ……………………………………………………....28 Research Questions and Hypotheses ……………………………………………………31 CHAPTER FIVE: METHODS ………………………………………………………….32 Participants ………………………………………………………………………….......32 Responders versus Non-Responders …………………………………………………….33 Procedure ……………………………………………………………………………......34 Measures ………………………………………………………………………………...34

v     Data Analysis …………………………………………………………………………...38 CHAPTER SIX: RESULTS ………………………………………………………….....41 Background Analyses …………………………………………………………………....41 ASIST versus QPR Participants ………………………………………………………....42 Relationship between Personality Char acteristics and Gatekeeper Behavior ……….....43 Relationship between Environmental C haracteristics and Gatekeeper Behavior ……….44 Training Effect Analysis ……………………………………………………………….....45

Follow-up Analyses: Relationship betw een Altruism and Gatekeeper Behavior ………..46 Buffering Effect of Social Support ……………………………………………………....47 CHAPTER SEVEN: DISCUSSION…………………………………………………......51 Implications ……………………………………………………………………………...53 Limitations ……………………………………………………………………………....58 Future Directions ……………………………………………………………………......63 Conclusions ……………………………………………………………………………...67 REFERENCES…………………………………………………………………………..69 APPENDIX A: Items Measuring Emotional Intelligence ……………………………....85 APPENDIX B: Items Measuring Altruism ……………………………………………...87 APPENDIX C: Items Measuring Gatekeeper Role Conflict ……………………………88 APPENDIX D: Items Meas uring Social Support ……………………………………….89 APPENDIX E: Items Measuring Pe rceived Organizational Support …………………...91 APPENDIX F: Items Measuri ng Gatekeeper Behavior ………………………………....92 TABLES………………………………………………………………………………....93 FIGURES…………………………………………………………………………….....105

vi    

LIST OF TABLES Table 1: Participant Demographics by Training

Table 2: Descriptive Statistics and Correlations for All Variables and All Participants Table 3: Poisson Regression Testing the A ssociation between Individual and Organizational Characteristics and Number of Interventions for ASIST Participants Table 4: Poisson Regression Testing the A ssociation between Individual and Organizational Characteristics and Number of Interventions for QPR Participants Table 5: Relationship between Altruism and Nu mber of Interventions for ASIST

Table 6: Relationship between Altruism and Number of Gatekeeper Behaviors for ASIST Participants Table 7: Relationship between Altruism and Nu mber of Interventions for QPR Participants Table 8: Relationship between Altruism and Numb er of Gatekeeper Behaviors for QPR Participants Table 9: Effects of Gatekeeper Role Conflic t and Supervisor Emotional Support on Gatekeeper Behavior for ASIST Participants Table 10: Effects of Gatekeeper Ro le Conflict and Supervisor Instrumental Support on Gatekeeper Behavior for ASIST Participants

vii     Table 11: Effects of Gatekeeper Role Conflic t and Supervisor Emotional Support on Gatekeeper Behavior for QPR Participants Table 12: Effects of Gatekeeper Ro le Conflict and Supervisor Instrumental Support on Gatekeeper Behavior for QPR Participants

viii    

LIST OF FIGURES

Figure 1: Proposed Interaction for Gatekeeper Role Conflict and Social Support for Gatekeeper Behavior Figure 2: Interaction of Gatekeeper Role Con flict and Supervisor Emotional Support for Gatekeeper Behavior (ASIST Participants). Figure 3: Interaction of Gatekeeper Role Con flict and Co-Worker Emotional Support for Gatekeeper Behavior (ASIST Participants).

1    

CHAPTER ONE INTRODUCTION Suicide continues to be one of the leading causes of death in the United States, with 34,598 completed suicides in 2007, the most recent year for which national data are available (Xu, Kochanek, Murphy, & Tejada-Vera, 2010). This translates into one suicide every 15.2 minutes or about 95 suicides each day (McIntosh, 2010). The mountain region (including Arizona, Colorado, Idaho, Montana, Nevada, New Mexico, Wyoming, and Utah) consistently has the highest suicide death rate (16.8 per 100,000 in 2007) compared to the other regions (McIntosh, 2010). The east south central region (including Kentucky, Tennessee, Mississippi, and Alabama) had the second highest suicide death rate (13.8 per 100,000 in 2007), while the middle Atlantic region (including New York, New Jersey, and Pennsylvania) had the lowest suicide death rate (8.5 per 100,000 in 2007) (McIntosh, 2010). In Colorado, suicide is a major public health problem. In 2007, Colorado had a suicide death rate of 16.7 per 100,000 with 811 suicide deaths. This suicide death date represented the 6 th highest suicide rate in the United States (US) with the national rate being 11.5 per 100,000 (Xu, et al., 2010). Furthermore, suicide is the second leading cause of death among Colorado youth aged 10-24 years. Between 2001 and 2005, Colorado had the 7th highest youth suicide rate in the US. A total of 296 suicide deaths occurred among Colorado youth aged 15-18 years from 1999 through 2008, a rate of 11.2

2     per 100,000, almost double the US rate of 6.76 per 100,000 (Centers for Disease Control and Prevention, 2009). On the national level, the alarming suicide statistics prompted a call to action and the development of the National Strategy for Suicide Prevention (NSSP; U.S. Department of Health and Human Services [DHHS], 2001) which established suicide prevention as a major public health priority. A number of strategies emerged from the NSSP focusing on identifying the risk and protective factors for suicide in a given population and developing action plans to decrease risk factors and increase protective factors. One common approach is training community members, known as gatekeepers, to identify signs of depression and other behaviors that increase risk for suicide (Centers for Disease Control and Prevention, 1992; Gould & Kramer, 2001; Mann et al., 2005). In theory, higher percentages of members in a given community able to recognize and refer at-risk individuals will lead to fewer suicides (Ramsay & Bagley, 1985; Tierney, Ramsay, Tanney, & Lang, 1990; Quinnett, 1995, 2005). Ideally, all members of a community would be trained in suicide prevention and the result would be a lower number of deaths. However, due to the cost of initial training, as well as retraining (due to low retention of training content), it is not practical to have all community residents trained. The average cost per trainee for a 90-minute gatekeeper training program is $25 (QPR Institute, 2010), while the estimated cost for a 16-hour, two day training can exceed $300 per trainee. Therefore, training all the residents in a community of 10,000 would cost $250,000 for a 90-minute training and more than $3,000,000 for two days of training. Hence the cost of training all members in a community is prohibitive. In fact, the total annual budget for the Colorado Office of Suicide Prevention to fund suicide

3     prevention activities state-wide is only $300,000. Thus, it is fiscally responsible to be selective about suicide preven tion training through targeting individuals in key positions to apply what they have learned. Furthermore, evaluation of gatekeeper training programs has documented that not all individuals who complete training will use what they learned (e.g., Wyman et al., 2008). According to some estimates in the training evaluation liter ature, only about ten percent of what is learned in training results in behavior change (Georgenson, 1982). In fact, in one longitudinal evaluation of gatek eeper training (Wyman, et al., 2008), less than 20% of gatekeepers had changed their behavi ors one year followi ng training, indicating that refresher training may be useful for more than 80% of those who were trained. Refresher trainings, although le ss expensive than full traini ngs, present an additional financial burden for communities hoping to train and maintain effective gatekeepers that makes it impractical to train (and retr ain) all members of the community. Furthermore, for those gatekeepers work ing within an organizational setting, the lack of behavior change af ter training programs may result from barriers that inhibit adoption of the training. For example, a lack of time and resources at work can impede a trained gatekeeper from engaging in gate keeper behaviors (M oore, Cigularov, Chen, Martinez, & Hindman, In Press) . Therefore, creating a work environment free of time or resource barriers may increase the effectiveness of gatekeepers. The Random House Dictionary online (2011) defines the word efficient as “achieving maximum productivity with minimum wasted expense” and effective as “producing the intended or expected result s.” Based on these definitions, an effective gatekeeper is an individual who recognizes the warning signs that a person may be

4     contemplating suicide and refers that pe rson for assessment and care (Quinnett, 2007). Therefore, efficiently training individuals in suicide prevention refe rs to maximizing the number of effective gatekeepers who have atte nded a gatekeeper training. In other words, wasted expenses are minimized when gatekeepers are engaging in gatekeeper behaviors regularly following training. In addition, an efficient training progr am would encourage selection of initial trainees and not require all community members to be trained. An efficient training program would also monito r the engagement and behavior of trained gatekeepers in order to determin e when retraining was needed. The identification of individual charact eristics of effective gatekeepers will inform the selection of a ppropriate individuals for ga tekeeper training, therefore increasing the overall efficiency of the pr ogram. Further, identifying organizational characteristics that hinder trained gatekeepers from helping individuals in crisis will aid in the elimination of barriers in the workplac e, allowing for more effective gatekeepers. Therefore, the aim of this di ssertation is to explore the relationship between individual and organizational characteristic s and gatekeeper behavior.  

5    

CHAPTER TWO SUICIDE PREVENTION GATEKEEPER TRAINING The Surgeon General’s Call to Action to Prevent Suicide (U.S. Public Health Service, 1999) and the National Strategy fo r Suicide Prevention ( DHHS, 2001) identified gatekeeper training as a promising suicid e prevention strategy. Suicide prevention gatekeeper is defined in this study as a comm unity member who is trained to recognize a crisis and the warning signs th at a person may be contemplati ng suicide and to refer that person for assessment and care (Quinnett, 2007). What is a Gatekeeper Training? Although gatekeeper training programs vary in length and have been developed to train a wide range of people, from clini cal professionals to the general public, the common goal in training is to teach recognition of suicide wa rning signs, ways to discuss suicidal intent, offer hope, and ab ility to refer the person in cr isis to appropriate services. The two most widely implemented gatekeeper training programs in the United Stated are reviewed below. Gatekeeper Training Programs ASIST .

Applied Suicide Intervention Skills Tr aining, (ASIST; Tierney, Ramsay, Tanney, & Lang, 1990) is a two-day intens ive, interactive ga tekeeper training program that uses a systematic model of suicide prevention in tervention. The program is based on adult

6     education principles and primarily uses discus sions and role play to help participants learn to recognize wa rning signs and how to inte rvene to prevent suicide. The model includes training to be able to accomplish six tasks. The first task is to recognize warning signs for suicide risk. The s econd task is to ask directly whether the individual is thinking about suic ide. The next step is to list en to the individual’s reasons for both living and dying (task three) before re viewing risk factors for suicide (task four). Moving to the assisting phase of the model, the fifth task is to contract a safeplan with the individual in crisis. Lastly, following up on co mmitments made in a safeplan to keep the at-risk person safe from suicid e is the sixth and final task. QPR . Question, Persuade, and Refer (QPR; Qui nnett, 2007) is a one- to three-hour training that teaches individuals how to rec ognize suicidal warning signs, ask the suicide question, persuade a suicidal person to accept he lp, and refer someone to get the help that is needed. QPR trainings are typically conduc ted in a face-to-face classroom setting and involve listening to an instru ctor, watching videos, and enga ging in discussion and role- plays with other trainees. What is a Gatekeeper’s Role in Suicide Prevention? Although gatekeeper training programs vary in length and have been developed for differing levels of involvement in suicid e prevention, from clinic al professionals to the general public, the common purpose of thes e trainings is to teach gatekeepers to recognize suicide warning signs, discuss suicidal intent, offer hope, and refer the person in crisis to appropriate services. Based on th e review of the ASIST and QPR curricula, as well as our own investigation of gatekeeper behavior (Moore, Cigularov, & Chen, In

7     Prep), six gatekeeper behaviors were identified and will be measured as outcomes in this study. These gatekeeper behaviors include: Problem Identifica tion, Interaction and Assessment of Risk and Behavior, Listen ing, Establishing a Trusting and Helping Relationship, Referral Activities, and

Follow-up Activities . Are Gatekeeper Training Programs Effective?

Early evaluations of the gatekeeper trai ning program demonstrated effectiveness using single group pretest-pos ttest designs (e.g., Grossman & Kruesi, 2000; Tierney, 1994) and trained versus untrained compar ison group designs (e.g., King & Smith, 2000). However, several recent studies have investigat ed the effectiveness of gatekeeper training programs using more vigorous evaluation strategies. In their evaluation of community QPR tr aining, Cross, Matthieu, Cerel, and Knox (2007) found that participants’ knowledge and self-efficacy increased after QPR training. Additionally, Cross and colleague s assessed participants’ gate keeper behavioral skills immediately following and six weeks after tr aining using an observational rating scale. The gatekeeper skill measure incl uded six domains: Active listening, clarifying/confirming questions, directly ask ab out suicide, persuades, and refers. More than half of the participants received sa tisfactory ratings demonstrating that they possessed adequate gatekeeper skills following training. Th e small sample size in the study and the lack of a control group, limited th e ability to generalize these findings to all community gatekeeper training programs. Howe ver, this study is important because the researchers not only measured proximate out comes (e.g., knowledge and self-efficacy) of training, but also behavioral outcomes (e.g., gatekeeper skills 6-weeks following training), acknowledging the importan ce of actual behavior change.

8     In a follow-up study, Cross, Matthieu, Lezine, and Knox (2010) again employed observational methods to assess gatekeeper skills, this time before and after training, of participants in a community QPR training. The Observational Rating Scale of Gatekeeper Skills (ORS-GS), developed previously (Cross, et al., 2007) was refined and used to measure two general and three suicide-specific skills. In addition, the researchers investigated whether pre-training participant characteristics were associated with training outcomes. Specifically, they investigated the Big Five personality factor of Openness to Experience (OTE). The researchers acknowledged that exploring the relationship between pre-training characteristics and post-training gatekeeper skills could improve the cost effectiveness of training through improving the selection of participants (Cross, et al., 2010). The researchers found that observed gatekeeper skills increased from pre- training (10% of participants met criteria for acceptable gatekeeper skills) to post-training (54% of participants met criteria). Although the increase from 10% before training to 54% following training is statistically significant, it is of concern that “46% of participants did not show the ability to ask about suicide and make an adequate referral for help” (p. 156, Cross, et al., 2010). Furthermore, sociodemographic characteristics (i.e., age, gender, education, etc.), prior experience with gatekeeper training, and Openness to Experience, either alone or together, did not significantly predict gatekeeper skills after training. One important limitation of this study was that there was no longitudinal assessment of skills to determine maintenance over time or the relationship between observed gatekeeper skills and actual application of those skills. Nevertheless, this study is important because it is the first to explore personality characteristics as a potential predictor of gatekeeper skills. Further, the researchers argued for using selection

9     as a strategy to improve the cost effectiveness of gatekeeper training. By selecting those individuals for training whom already possess characteristics that are predictive of gatekeeper skills, the likelihood that the gatekeepers will use their skills following training is greatly increased. While the previously described evaluations contribute to our knowledge of the effectiveness of gatekeeper trainings, these studies do not achieve the “gold standard” in evaluation by utilizing random assignment. In a randomized trial, Wyman and colleagues (2008) evaluated the impact of the QPR training program in one school district. Schools within the district were randomly assigned to receive training or to a wait-list control group to receive training in the future. Schools were stratified by middle or high school and high versus low rates of student crisis referrals in the previous year. Staff within the schools were stratified by job role and then randomly selected to participate in the study. Participants were followed for one year after training. The researchers found a consistent positive impact of training on participant knowledge and appraisals. However, the training impact on gatekeeper behavior depended on job class (i.e., social service staff, administrator, teacher, and support staff) and a history of communicating with distressed youth about suicide prior to training. At a one-year follow-up, the researchers found a moderate positive training impact on knowledge and large positive effects on perceived preparedness, self-evaluated knowledge, efficacy, and access to services. There was no training effect on gatekeeper behavior after one year. However, there was a significant interaction (training condition by baseline interaction) for the gatekeeper behavior “ask students about suicide”. Staff who entered the training having closer communication with students asked more students about suicide after training. However, only about 14% of

10     the staff was communicating with students about suicide before training. Thus, the training had little benefit for the remaining 86% of staff in terms of increasing gatekeeper behaviors. Also of importance was a finding of substantial school-level differences in knowledge, appraisals, and gatekeeper behaviors. The researchers attribute these differences to the “shared attitudes and commitment to suicide prevention activities because of school-level contextual influences” (p. 113, Wyman, et al., 2008). Thus, a potential obstacle for school personnel may be the culture within the school. Cross, et al. (2010) and Wyman, et al. (2008) acknowledged that the brevity of QPR training may result in a lack of behavioral change (i.e., increased communication with at risk individuals) among participants. In other words, QPR training may be too brief to teach communication skills (i.e., active listening, empathy) necessary to “engage others in emotionally charged conversations” (p. 156, Cross, et al., 2010). Thus, one weakness of QPR training, the brevity of training, may be addressed with lengthier, more in depth training programs which incorporate adult learning principles, such as role-play practice (Cross et al., 2010; Wyman et al, 2007). Based on this reasoning, the two-day ASIST training program should have a greater impact on gatekeeper behavior than QPR. Empirical evaluations of ASIST will be examined next. Initial evaluations (e.g., Tierney, 1994) of the effectiveness of a two-day gatekeeper training program were based on the Suicide Intervention Workshop (SIW; Ramsay et al., 1994), which later developed into the current gatekeeper training known as ASIST. The evidence for effectiveness in the initial investigations of ASIST was limited by the quality of the evaluations (e.g., small sample sizes, methodological weaknesses).

11     Dolev, Russell, Griesbach, & Lardner (2008), in a review of the effectiveness of ASIST, found 15 formal evaluations of the training program, a majority of them unpublished. According to quality ratings of these studies, only five were considered to be good-quality evaluations. Dolev and colleagues used Kirkpatrick’s (1959; 1994) model to summarize the evaluation findings. Kirkpatrick’s model considers the effectiveness of trainings at four levels: Reaction (i.e., how did the trainee feel about the training?), Learning (i.e., what knowledge and skills did the trainee gain?), Behavior Change (i.e., has the trainee applied the training to their work?), and Organizational Change (i.e., what have been the outcomes at the organizational/societal level?). Reactions to training were assessed in seven of the fifteen evaluations. A majority of trainees had positive reactions to training and expressed high levels of satisfaction. Thirteen of the fifteen evaluations found evidence for positive change in suicide intervention knowledge, skills and attitudes following training. Most of the evaluations (11 of 13) used self-report measures; two evaluations measured actual change in knowledge, attitudes, and skills comparing pre-training to post-training scores (Tierney, 1994). Based on the first two levels of training effectiveness, reaction and learning, it appears that ASIST is an effective training program. However, changes in knowledge, skills, and attitudes do not necessarily translate in to changed behavior (Georgenson, 1982). In fact, a meta-analysis of the literature on training outcomes found that the relationship between post-training knowledge and later performance of the trained behavior ranges from .08 to .18 (Alliger, et al., 1997), suggesting that increased knowledge is a relatively ineffective gauge of successful training programs. Keeping in mind that the ultimate goal of ASIST (or any gatekeeper training program for that matter)

12     is to create effective gatekeepers, or indi viduals who can successfully apply what they learned in training to help a person in crisis , the most important criteria for effectiveness of the ASIST training program should be act ual behavior change, the third level of Kirkpatrick’s (1959; 1994) model. Dolev and colleagues (2008) found that ten of the fifteen studies measured the extent to which trainees applied their ac quired knowledge and skills. Eight of these evaluations used a single self -report item to measure the tran sfer of knowledge and skills into practice. According to the self-report measures, nearly 50% of trainees reported using the training at least once within six months of completing training. However, the conclusions that can be drawn from these self-report measures is limited because there is likely a difference between what people say they did and what they actually did. Two evaluations used more objective measur es of knowledge and skill application. In a longitudinal evaluation of the impact of ASIST in a large community hospital in Canada, Perry and McAuliffe (2007) employe d objective measures of behavior change over a four year period: a) the number of clients that the staf f assessed for suicide risk; b) the number of people identified as at-risk for su icide; and c) the admission rate of suicidal patients presenting in the emergency depart ment. Effective knowledge and skill transfer to the workplace was evident in the results. The authors reported a 13% increase in the number of clinicians who assessed all of pa tients for suicide risk as well as an annual increase of between 14% and 21% in the num ber of patients identified as at-risk for suicide. Furthermore, there was a significant decrease in suicidal patients’ admission rate from 56% to 42%. According to the authors, the decreased admission rate was the result

13     of staff’s ability to effectively assess suicidality, which allowed some admissions to be averted. In a longitudinal evaluation of ASIST training provided to school staff, Cornell, Williams, & Hague (2006) also used objective measures to examine transfer of knowledge into skills. The researchers measured: a) the number of referrals to mental health services; b) the number of students questioned about suicide; and c) the number of contracts made with potentially suicidal individuals. The researchers found that the control group made more referrals and questioned students about suicide more often than the trainees did during the two years following training. These results suggest that the training was ineffective in changing trainees’ behavior. However, the authors did find that the trainees made more contracts with suicidal individuals than did the control group within two years of training. The findings of this study do not present a clear cut conclusion about the effectiveness of ASIST in changing participants’ behavior. Examining the impact of ASIST on the organizational and societal level, the fourth level of Kirkpatrick’s (1959; 1994) model, is a complex undertaking. Dolev and colleagues (2008) found three evaluations that attempted to measure this outcome. These evaluations only provided anecdotal evidence of the organizational and societal impact of ASIST. In two different large scale implementations of ASIST, one in a school setting and one in a community hospital, organizational change occurred (Hinbest, 2001; Perry & McAuliffe, 2007). In the school setting (Hinbest, 2001), the evaluator reported improved interactions and relationships between school and community representatives, as well as improved system-wide protocols and school policies. The community hospital gained recognition and a reputation as a leader in suicide prevention training in the

14     community, with the local community college mandating ASIST training for its nursing students (Perry & McAuliffe, 2007). Finally, at the community le vel, one study attempted to examine the impact of ASIST on the suicide rate in a school system (Cornell, et al., 2006). The schools where the staff did not receive ASIST training ha d almost three times as many students who attempted suicide compared to schools wher e staff received traini ng. However, because the schools self-selected to re ceive training, it is impossible to attribute this finding solely to the training. In summarizing their conclusions, Do lev and colleagues (2008) acknowledged that ASIST was effective in improving part icipants’ suicide inte rvention knowledge and skills, but also that there was little evidence that part icipants’ knowledge and skills resulted in changes in gatekeeper behavior s. The researchers suggested that further evaluations are needed, especially focused on transfer of learning into practice and long term impacts on communities and organizations. Isaac and colleagues’ (2009) conducted a sy stematic review of the gatekeeper literature in order to examine the state of the evidence on ga tekeeper training for suicide prevention. A search using two databases (M EDLINE and PsychINFO) and search terms suicide prevention and gatekeeper resulted in 13 studies wh ich presented evaluative evidence of suicide prevention gatekeeper tr aining programs. In general, gatekeeper training programs were found to be successf ul in positively affecting participants’ knowledge, skills, and attitudes. In additi on, Isaac and colleagues’ (2009) acknowledge that there exists a lack of evidence that gatekeeper training programs are able to positively impact suicide identification and helpi ng behaviors (i.e., gatekeeper behavior).

15     These conclusions do not mean that gatekeeper training programs should be abandoned as a means of suicide prevention. Instead, they suggest that time and money may be better spent if steps are taken to more efficiently train individuals as gatekeepers. Thus, selecting, based on pre-existing characteristics, those individuals for training who are most likely to use what they learned in training will lead to the most effective gatekeepers. Although, initial investigations of personality characteristics that predict gatekeeper behavior have been unsuccessful (Cross, et al., 2010), further exploration of these predictors will allow for training more effective gatekeepers through selection. Furthermore, evidence that the gatekeepers’ work environment impacts gatekeeper behavior (e.g., Wyman et al., 2008) suggests that further investigation of those organizational factors that facilitate gatekeeper behavior is an additional avenue for producing effective gatekeepers.

16    

CHAPTER THREE WHY DO PEOPLE HELP? A SOCIAL PSYCHOLOGICAL PERSPECTIVE The field of social psychology has been examining the question of why people help each other since the field’s inception in the early 20 th century (e.g., McDougall, 1908), with much of the work in this area occurring in the 1960’s and 1970’s. Efforts to understand what motivates people to help have focused on three explanations: (a) learning; (b) social and personal norms; and (c) arousal and emotion. Drawing from the basic principles of learning theory, the learning explanation proposes that people are motivated to help others because they have received positive reinforcement (e.g., Moss & Page, 1972) or have observed the benefits of others’ helping behaviors (e.g., Hornstein, 1970). According to the social and personal norms approach, an individual’s helping behavior is partially a result of personal values or beliefs about expected behaviors in social situations (Warburton & Terry, 2000). Finally, the arousal and emotion approach explores how different emotions can spur helping in different ways. According to the negative state relief model (Cialdini, Kenrick, & Baumann, 1982; Cialdini, Schaller, Houlihan, Arps, Fultz, & Beanman, 1987), when people experience negative emotions they are motivated to reduce them. People have learned through their socialization and experience that helping others can make them feel good (Williamson & Clark, 1989; Yinon & Landau, 1987) and good feelings derived from helping others may relieve negative emotions. In sum, much of the research on why people help has focused on

Full document contains 116 pages
Abstract: Suicide remains a major public health problem in the United States. Training individuals known as gatekeepers to identify the signs and behaviors of suicide risk has been one of the most widely adopted prevention strategies. Due to financial constraints, it is not possible to train all members of a community as gatekeepers. Thus, it is more fiscally responsible to selectively train individuals within a community who possess key characteristics that will make them more effective gatekeepers. In this dissertation, the personality and organizational characteristics of effective gatekeepers were explored. Specifically, personality characteristics: Emotional Intelligence and Altruism, and organizational characteristics: social support, perceived organizational support, and gatekeeper role conflict were investigated as predictors of gatekeeper behavior. Two hundred and eighteen gatekeepers completed surveys immediately prior, immediately after, and six months following training that consisted of self-report measures of personality and organizational characteristics, and gatekeeper behavior. Poisson regression was employed to analyze the data because of the low-base rate occurrence of gatekeeper behavior. Analyses were conducted separately for the two gatekeeper training programs (ASIST & QPR). Mixed results were found in this study. Gatekeepers trained in ASIST were high in Emotional Intelligence and Altruism, resulting in range restriction. Gatekeepers trained in QPR varied more on personality characteristics, and Altruism positively predicted gatekeeper behavior six months following training. In terms of organizational predictors of gatekeeper behavior, supervisor emotional support buffered the relationship between gatekeeper role conflict and gatekeeper behavior for ASIST participants. The findings have direct implications for the training of community members as gatekeepers. Specifically, how selection principles can be applied to gatekeeper training, as well as how social support can be improved in organizations so that gatekeepers can overcome barriers (such as role conflict) in order to increase the effectiveness of trained gatekeepers. Although a number of limitations were present in the current study (i.e., low response rate and inability to generalize the findings), the potential application of these findings presents the opportunity to significantly change who is selected for training as a gatekeeper and how trainings are conducted. Future research directions include the measurement of gatekeeper behavior and effectiveness of gatekeepers.