Crisis management Suicide loss in schools

Transcription

1 Crisis management Suicide loss in schools Madelyn S. Gould, Ph.D., M.P.H. Columbia University/ 3 rd Australian Postvention Conference Sydney, Australia June 28, 2012

2 Crisis management Suicide loss in schools By promoting the mental health of survivors, postvention thus becomes prevention. (Webb, 1986) Moreover, Postvention needs to be guided by our understanding of prevention. (Gould, today)

3 HOW DOES A SUICIDE OCCUR? UNDERLYING VULNERABILTY e.g.: Mood Disorder / Substance Abuse / Aggression / Anxiety / Impulsivity/ Sexual Orientation/ Abnormal Serotonin Metabolism/ Family Characteristics, including history of suicidality/ Sexual Abuse/Physical Abuse/Social adversity STRESS EVENT (often caused by underlying condition) e.g.: In Trouble With Law or School / Loss/ Bullied INHIBITION e.g.: Family cohesion/ Available Support Religiosity ACUTE MOOD CHANGE e.g.: Anxiety Dread / Hopelessness / Anger FACILITATION e.g.: Method/Weapon available / Recent example / Media displays SURVIVAL SUICIDE

4 Casefinding Find affected individual Provide effective treatment Screening, Gatekeeper training PREVENTION STRATEGIES UNDERLYING VULNERABILTY Casefinding/ Crisis intevention Crisis hotline STRESS EVENT Casefinding/ Crisis intevention Crisis hotline Risk Factor Reduction Media education Restriction of firearms/lethal means Resilience development/ skills training Postvention/ Crisis intervention INHIBITION SURVIVAL ACUTE MOOD CHANGE FACILITATION SUICIDE

5 Scope of the Problem

6 Scope of the Problem: Youth Suicide Deaths, Australia In 2010, suicide ranked as the leading cause of death for young people (ages and 20-24). 296 people between the ages of 15 and 24 died by suicide. In 2010, suicides accounted for only 1.6% of all deaths in Australia, however they made up 21.4% of deaths among young people ages and 24.9% of deaths among people ages Australian Bureau of Statistics (ABS cat. no , 2010)

7 Scope of the Problem: Suicide Deaths, U.S. In 2009, suicide ranked as the third leading cause of death for young people (ages and 20-24); only accidents and homicides occurred more frequently. 4,371 people between the ages of 15 and 24 died by suicide. Suicide rates among American Indian/Alaska Native adolescents and young adults are 1.8 times higher than the national average for that age group.

8 Leading Causes of Death, Australia Ages: 15-24, Both Sexes, 2010 Intentional self-harm 9.4 Car occupant injured in transport accident 6.5 Event of undetermined intent 2.5 Accidental poisoning by/exposure to noxious substances Motorcycle rider injured in transport accident Pedestrian injured in transport accident Other forms of heart disease Assault Malignant neoplasms of lymphoid/haematopoietic/related tissue Episodic and paroxysmal disorders Rates per 100,000

9 Suicide Deaths Among Youth, Australia and U.S. Ages ( ) Columbia University Rate per 100,000 USA Total 10 Australia Year (ABS cat. no , 2010) (WISQARS Fatal Injury Reports, National and Regional, )

10 Suicide Deaths Among Youth, Australia and U.S. Ages 15-19, by Sex ( ) USA Boys USA Girls Australia Boys Australia Girls Rate per 100, Year (ABS cat. no , 2010)

11 Suicide Deaths Among Youth, Australia and U.S. Ages ( ) Columbia University 20 Rate per 100, USA Australia Year

12 Suicide Deaths Among Youth, Australia and U.S. Ages 20-24, by Sex ( ) Rate per 100, USA Boys USA Girls Australia Boys Australia Girls Year

13 Youth Suicide Risk Factors

14 Fleischmann et al. Completed Suicide and Psychiatric Diagnoses in Young People: A critical Examination of the Evidence. American Journal of Orthopsychiatry. Vol 75 (4):

15 Fleischmann et al. Completed Suicide and Psychiatric Diagnoses in Young People: A critical Examination of the Evidence. American Journal of Orthopsychiatry. Vol 75 (4):

16 PRIOR SUICIDAL BEHAVIOR Between 1/4 to 1/3 of suicides have made a prior attempt completed 30 fold increased risk for boys 3 fold increased risk for girls

17 INTERRELATIONSHIP BETWEEN SUICIDE AND OTHER RISK BEHAVIORS Flisher et al (2000) MECA (N = 1,285) King et al* (2001) MECA (N = 1,285) Garrison et al (1993) SC YRBS (N = 3,764) Physical fights Alcohol Cigarettes Marijuana Intercourse *OR s adjusted for demographic and psychiatric Dx.

18 PERSONALITY/COGNITIVE FACTORS Hopelessness Poor interpersonal problemsolving ability Aggression/Impulsivity

19 SUICIDE ATTEMPTS AND SAME-SEX SEXUAL ORIENTATION AUTHOR N CONTROLS AGE OR ATTEMPT Garafalo et al ,365 MA 95 YRBS HS 2.28 *** Faulkner & Cranston 3,054 MA 93 YRBS HS 2.5*** 1998 Fergusson et al ,265 Christ Church *** Cohort Remafedi et al MN Ad Hlth MS & HS 7.1*** male van Heeringen & 396 Belgian School *** females Vincke 2000 (m=20) Russell & Joyner, ,940 Add Health grades * (adjusted for other suicide risk factors (Compiled by Greenberg and Gould, 2002)

20 SEROTONIN DYSREGULATION 1. Many studies showing decreased serotinergic activity in suicides 2. Low serotonin associated with excitable, impulsive, and violent behavior 3. Serotonin lower in males, after alcohol intake in the elderly

21 GENETIC FACTORS Focus has been on genes that play important roles in the regulation of serotonin.

22 MOST COMMON FAMILY RISKS Cases Controls Family history of suicidal behavior 17% 5% 2 Parental psychopathology 68% 32% 1 Parent-child relationships conflict 39% 20% 1 poor communication 30% 12% 2 Parental divorce 48% 33% 2 ( 1 Brent et al., 1994; 2 Gould et al., 1996)

23 MOST COMMON STRESSFUL LIFE EVENTS Cases Controls Interpersonal losses 46% 29% Legal/disciplinary crises 42% 13% (averaged Brent et al., 1999 and Gould et al., 1996)

24 SUICIDE AND PHYSICAL ABUSE Association between physical abuse and suicide completion (Brent et al., 1994) Childhood physical abuse associated with increased risk of suicide attempts in late adolescence or early adulthood, even after adjusting for demographic, child psychiatric and parental characteristics. (Johnson et al., 2002) C.RF1

25 SUICIDE AND SEXUAL ABUSE No evidence of association between sexual abuse and suicide completion (Brent et al., 1994) 33% of sexually abused children are suicidal at ages 16 to 18 (Fergusson et al. 1996) Individuals with history of sexual assault are 6 times more likely to report a suicide attempt, controlling for demographic characteristics, PTSD and depression (Davidson et al., 1996) (Compiled by Greenberg and Gould, 2002) C.RF1

26 Bullying and its impact on suicide

27

28 Are teens bullied to death? What is the relationship between bullying and suicidal ideation/behavior? What are the implications of media messages? Can bullying behaviors be prevented?

29 Definition Although definitions vary, most agree that bullying involves: Attack or intimidation with the intention to cause fear, distress, or harm that is either physical, verbal, or psychological/relational A real or perceived imbalance of power between the bully and the victim Violence Repeated attacks or intimidation Bullying between the same children over time (Centers for Disease Control and Prevention, 2011; Olweus, 1991) 29

30 Types of Bullying Physical e.g., hitting, kicking Verbal e.g., name calling Social Exclusion Spreading Rumors Cyberbullying e.g., via s, texts, web sites

31 Characteristics of Cyberbullying Difficulty of escaping from it Large potential audience Anonymity of the cyberbully Cyberbully may be less aware of consequences of his/her actions Fewer opportunities for empathy (Hinduja & Patchin., 2009; Kowalski & Limber., 2007; Smith et al., 2008; Sourander et al., 2010) 31

32 Youth involved in bullying Bullies Victims Bully-Victims Bystanders behavior 32

33 100% 80% 60% 40% 20% 0% Percentage of Students Being Involved in Bullying as Victims only, Bullies only or Bully-Victims Columbia University 41 Nation Study of 11, 13 & 15-Year-Old School Children (Craig et al., 2009) Czech Republic Lithuania Latvia Estonia Greece Greenland Romania Turkey Ukraine Bulgaria Austria Russia Israel Belgium-French Germany Switzerland Luxembourg Portugal Poland France Canada US Macedonia Italy Netherlands Slovenia Denmark Belgium-Flemish England Croatia Norway Malta Scotland Ireland Wales Finland Hungary Ice;amd Spain Sweden Boys Girls From Olweus, 2011

34 Prevalence of Cyberbullying Columbia University 15.8% of high school students reported cyberbullying in past 12 months. A majority (59.7%) of cyberbullying victims were also school bullying victims. Victimization was higher among nonheterosexual identified youths. Schneider et al.

35 Associated Risks The past decade has witnessed a surge in research on association between bullying behaviors and depression, suicidal ideation and behavior. Cross-sectional Studies: Kaltiala-Heino et al., 1999 Finland; Kim et al., 2005 Korea; Rigby & Slee, 1999 Australia; Cleary, 2000 USA; Van der Wal et al., 2003 Netherlands; Eisenberg et al., 2003 USA; Roland, 2002 Norway; Brunstein Klomek et al., 2007 USA; Brunstein Klomek, 2008 USA; Kaminski & Fang, Longitudinal Studies: Kim et al., 2009 Korea; Brunstein Klomek et al., 2008 Finland; Brunstein Klomek et al., 2009 Finland; Brunstein Klomek et al., 2011 USA.

36 U.S. Cross-sectional study Funded by NIMH Prevalence of bullying behavior in and out of school Association of bullying behavior with depression, suicidal ideation and suicide attempts by gender Impact of the co-occurrence of bully-victims students years old 9 th -12 th grade in 6 high schools in NY State (Gould et al., 2005; Klomek et al., 2007) 36

37 Question samples: bullying behavior GENERAL: Bullied at school? Bullied away from school? Bullied others in school? Bullied others away from school? not at all 1-2 times SPECIFIC VICTIMIZATION: How often has someone bullied you in these ways: Made fun of you because of religion or race Made fun of you because of your looks or way you talk Hit, slapped or punched you Spread rumors or mean lies about you. Made sexual jokes, comments or gestures to you Used or internet to be mean to you (Gould et al., 2005; Nansel et al., 2001) times a few times a week most days

38 Prevalence of Frequent Bullying Behavior In and Out of School (Klomek et al., 2007) % Male Female Victim Bully Victim Bully In School Out Of School

39 Prevalence of Infrequent Bullying Columbia University Behavior In and Out of School (Klomek et al., 2007) % Male Female Victim Bully Victim Bully In School Out Of School

40 Frequent Victimization by Gender (Klomek et al., 2007) Columbia University % Male Female Religion/Race Looks/Speech Physical Rumors/Lies sexual comments or Internet

41 Gender differences There appears to be gender differences associated with frequency of involvement in bullying either as a bully or a victim and adverse psychological outcomes. Females: any involvement in bullying is associated with adverse outcomes. Males: frequent involvement in bullying is primarily associated with adverse outcomes. Gender Paradox : females are less likely to be bullies but when they are, they have a more severe impairment than their male counterparts (Kim, et al., 2006; Tiet et al., 2001; Wasserman et al., 2005) 41

42 Important points Bullying others, and not only being victimized, is associated with depression, suicidal ideation and attempts. (Forero et al., 1999; Kaltiala-Heino et al., 1999, 2000; Roland., 2002) The strongest association between involvement in bullying and depression/suicidal ideation/attempts is found among those who are both bullies and victims (bully-victims). (Kim et al., 2005; Kim & Leventhal., 2008; Klomek et al., 2007) 42

43 Limitations of cross-sectional studies Bullying experience and suicidal ideation/behavior (SI/SB) are measured at the same time and often through self-report. Even if the reporting of bullying experience and SI/SB are statistically correlated, this does not infer causality. There might be mediating variables. For example: Bullying Depression Suicidal ideation/behavior Another option may be that mental health problems and/or other community/social risk factors could increase the risk of bullying experience and suicidal ideation/behavior (Arseneault et al., 2010; Kumpulainen et al., 2000; Wang et al., 2011) 43

44 Longitudinal studies U.S. (Klomek et al., 2011) Funded by AFSP and CDC Three groups of high schools students at outset: a. Frequent bullying experience, no depression or SI/SB b. Frequent bullying experience and depression, SI/SB c. Experiencing depression, SI/SB but no bullying experience Students who only reported bullying experience (as bullies, victims, or both) did not develop later depression or SI/SB and had fewer psychiatric problems than students identified as at-risk for suicide. Students who reported bullying experience and depression, SI/SB were more impaired 4 years later than those who only reported depression or SI/SB. 44

45 Longitudinal Studies - Finland (Sournader et al., 2005; Klomek et al., 2008, 2009) Columbia University Bullying behavior at age 8 Suicidal ideation at age 18 among males N=2348 Suicide attempts and suicide at age 25 among both genders N=5302 Suicidal Ideation Males- Bullying behavior is not associated with suicidal ideation when controlling for baseline depression 45

46 Boys suicide attempts and suicide Bullying at age 8 and suicide attempts and suicide until 25 Frequent bullying and victimization are associated with later suicide attempts and completed suicides, but not after controlling for conduct and depression symptoms Once psychopathology was controlled bullying no longer significantly predicted suicide attempts and completed suicides 46

47 Girls suicide attempts and suicide Bullying age 8 and suicide attempts/suicide until age 25 Frequent victimization is associated with later suicide attempts and completed suicides, even after controlling for conduct and depression symptoms Frequent childhood victimization puts girls at risk for later suicidal behavior, regardless of childhood psychopathology 47

48 Discrepancies between studies Columbia University Correlational vs. Longitudinal studies Differences in the participants age Different definitions of bullying Experience can vary by type, frequency, intensity, duration Different assessment of bullying. (e.g., self-report surveys, asking peers to identify those who are bullies or bullied) Different outcomes: suicidal ideation- any/severe suicidal behavior- any attempt/severe attempt/suicide Controlling for baseline psychopathology (Kim & Leventhal, 2008) 48

49 Conclusions (I) Complex relationship exists between bullying and risk of suicidal ideation/behavior. Bullying and peer victimization puts adolescents at increased risk of suicidal ideation and behavior, especially when other psychopathology is present. It is not necessarily the bullying per se. There are important mediating variables. Suicidal ideation and behavior is usually not attributed to just one event or factor. 49

50 Conclusions (II) Bully, victims and bully-victims are different groups. Bully-victims are at high risk. It is the frequent involvement that is most concerning. Each gender has a markedly different risk profile. Girls may be more at risk of adverse psychological outcomes than boys. Students involved in frequent bullying behavior should be screened for suicidal ideation/behavior. More longitudinal studies are necessary to establish causality. 50

51 Bullying Media Messages (I) Typical media message: bullying causes suicide. This does not tell the full story. Suicide risk may be substantially mediated by other factors. (Gould et al. 2003, Shaffer et al., 1996; Brent et al., 1993). 51

52 Media Messages: Unintended Consequences Columbia University

53

54 OLWEUS BULLYING PREVENTION PROGRAM (OBPP) School level Class level Individual level

55 OLWEUS BULLYING PREVENTION PROGRAM (OBPP) The first evaluation of the effects of the Olweus BP Program was based on data from approximately 2500 students in 42 primary and junior high schools (grades 5-8) in Bergen, Norway, followed over a period of 2 years ( ). Marked reductions by approx. 50% - in bully/victim problems for the period studied, measured after 8 and 20 months of intervention Clear reductions in antisocial behavior such as vandlism, theft, drunkenness and truancy Marked improvement in various aspects of the social climate of classrooms From Olweus, 2011

56 Resilience development/skills training Columbia University Good Behavior Game Not designed as suicide prevention program; rather behavior management technique for reducing disruptive behavior Universal implementation in elementary schools (for students aged six through 12 years) Competitive teams losing points or privileges for rule-breaking or disruptive behavior

57 Resilience development/skills training Good Behavior Game Short-term efficacy trials have shown reduced aggressive and disruptive behavior, a potential risk factor for suicide (Ialongo et al., 1999; Kellam et al., 1994) Epidemiologically-based randomized trial involving 41 classrooms in 19 schools in a U.S. inner city showed long-term effect (15 years) on young adult suicide ideation. (Wilcox et al., 2008)

58 Resources on Bullying Prevention StopBullying.gov Tip sheets and other resources for multiple audiences FindYouthInfo.gov Interagency resources on range of youthrelated topics Blueprints for Violence Prevention es/pdf/bullyingsuicideprevention.pdf

59 More Youth Suicide Risk Factors

60 Access to lethal means Agricultural chemicals Sri Lanka, rural India, Samoa, China China Self-poisoning accounts for 80% of completed suicides Sri Lanka and Samoa, banning of toxic pesticides coincided with a decline in suicide rates (Gunnell et al 2007, Bowles et al 1995)

61 Firearm Availability as a Suicide Promoter in U.S. Parallel in firearm ownership and suicide rates (ecological) Case control studies show greater prevalence of guns and less securely stored guns in suicide than control homes

62 Household gun ownership levels and rates of firearm and non-firearm suicide mortality: United States, % yearly decline in firearm ownership rates over 22 year study period For each 10% decline in percentage of households containing both children and firearms, firearm suicide among 0-19 year olds dropped 8.3%, Miller, M et al. Inj Prev 2006;12: Copyright 2006 BMJ Publishing Group Ltd.

63 SUICIDE CONTAGION Process by which knowledge of one suicide facilitates the occurrence of a subsequent suicide. Viewed within the larger context of behavioral contagion or social learning theory

64 SUICIDE CONTAGION/ MODELING - Sources of evidence - Impact of media Suicide clusters Impact of exposure to suicidal peer

65 Media Magnitude of the Effect of Suicide Contagion/Imitation 12% increase in suicides following Marilyn Monroe s death ~75% decrease in subway-suicides following implementation of media guidelines in Vienna Exposure to Suicidal Peer OR s range from (attempted suicide) Suicide Clusters 1-5% of teen suicides occur in clusters (Cerel et al., 2005; Gould et al., 1990; Grossman et al., 1991; Harkavy Friedman et al., 1987; Phillips, 1974; Shafii et al., 1985; Sonneck et al., 1994)

66 PROTECTIVE FACTORS Family cohesion Religiosity

67 Family Cohesion as a protective factor Longitudinal study of middle school students (McKeown et al., 1998) Cross-sectional community studies of high school students (Rubenstein et al., 1989; 1998); and college students (Zhang and Jin, 1996). Students who described family life in terms of a high degree of mutual involvement, shared interests, and emotional support were 3.5 to 5.5 times less likely to be suicidal than adolescents from less cohesive families who had the same levels of depression or life stress.

68 Religiosity as a protective factor (I) Suicide as sin: Biblical teaching that life is not man s to take. Template for life: Religion provides basic moral rules to follow; Durkheim s social integration model. Social support: Religion provides connectedness to others; skeptics suggest social support fully mediates relationship between religiosity & psychological functioning, while others disagree (McClain et al., 2003) Meaning: Religion provides sense that life has meaning no matter how great the suffering.

69 Religiosity as a protective factor (III) Suicidal Behavior (Hilton et al., 2002; Siegrist, 1996; Walker & Bishop, 2005; Zhang and Jin, 1996) Depression (Miller et al., 1997) However, these studies have not controlled for potential confounders, such as substance abuse, which may be less prevalent among religious youths.

70 HOW DOES A SUICIDE OCCUR? UNDERLYING VULNERABILTY e.g.: Mood Disorder / Substance Abuse / Aggression / Anxiety / Impulsivity/ Sexual Orientation/ Abnormal Serotonin Metabolism/ Family Characteristics, including history of suicidality/ Sexual Abuse/Physical Abuse/Social adversity STRESS EVENT (often caused by underlying condition) e.g.: In Trouble With Law or School / Loss/ Bullied INHIBITION e.g.: Family cohesion/ Available Support Religiosity ACUTE MOOD CHANGE e.g.: Anxiety Dread / Hopelessness / Anger FACILITATION e.g.: Method/Weapon available / Recent example / Media displays SURVIVAL SUICIDE

71 Prevention strategies that can be adapted for postvention circumstances

72 POSTVENTION/CRISIS INTERVENTION -Efficacy- Family survivors: No controlled studies School-based programs: Few evaluation studies; methodologic difficulties Community-based programs: No evaluation studies From Gould, 2001

73 Casefinding Find affected individual Provide effective treatment Screening, Gatekeeper training PREVENTION STRATEGIES UNDERLYING VULNERABILTY Casefinding/ Crisis intevention Crisis hotline STRESS EVENT Casefinding/ Crisis intevention Crisis hotline Risk Factor Reduction Media education Restriction of firearms/lethal means Resilience development/ skills training Postvention/ Crisis intervention INHIBITION SURVIVAL ACUTE MOOD CHANGE FACILITATION SUICIDE

74 Screening

75 SCREENING - Background - Strategies to identify and refer suicidal youth are based on the valid premise that suicidal adolescents are underidentified (Kashani et al., 1989; Shaffer et al., 1990; Shaffer & Craft, 1999; Velez et al., 1988). Youth suicide occurs in the context of an active, often treatable, mental illness (Brent et al., 1999; Groholt et al., 1998; Shaffer et al., 1996). Potent risk factors have been established that can identify high risk youth (Gould et al., 2003).

76 SCREENING : DIRECT CASE-FINDING BY SELF- ADMINISTERED QUESTIONNAIRE M O D E L SCREEN FOR Mood disorder Suicidal ideation Suicide attempts Substance and alcohol abuse CASE-MANAGE (Shaffer & Craft, 1999) TREAT

77 SCREENING - Examples of Programs - Signs of Suicide (SOS)- Hybrid of student educational component and screen (Aseltine, 2003; Asentline & DeMartino, 2004;Aseltine et al., 2007) Teen Screen (Shaffer & Craft, 1999; Shaffer et al., 2004) U.S. College Screening Project Web-based service (Haas et al., 2003)

78 Screening - Beneficial Effects - Clinical validity and reliability findings of school-based screening procedures are encouraging (Aseltine, 2003; Aseltine & DeMartino, 2004; Thompson & Eggert, 1999; Reynolds, 1991; Shaffer & Craft, 999; Shaffer, 2004). Shown to identify high risk students - very good to excellent sensitivity 75% - 100% few false negatives Many high risk teens were not otherwise known (Scott et al., 20 SOS found short-term decrease in attempts (Aseltine et al., 200 Facility-level risk of serious suicide attempts reduced by screening in juvenile justice facilities (Scherff et al., 2005)

79 Screening - Limitations - Poor specificity - many false positives second-stage evaluations necessary Suicide risk waxes and wanes multiple screenings may be necessary Implementation meets resistance by HS principals and superintendents Treatment resources need to be available

80 SCREENING - A Major Concern - Asking about suicide will put ideas into kids heads

81 EVALUATING IATROGENIC RISK OF SUICIDE SCREENING PROGRAMS Madelyn S. Gould, Ph.D., M.P.H. Frank Marrocco, Ph.D. Marjorie Kleinman, M.S. Graham Thomas. B.S. Jean Cote, M.S.W. Kate Moskoff, M.S.W. Mark Davies, M.P.H. -Project funded by the National Institute of Mental Health- (R01 MH64632) (Gould et al., JAMA, 2005)

82 Specific Aims To answer Does asking about suicidal ideation or behavior during the course of a screening program create distress or increase suicidal ideation among high school students? Does asking about suicidal ideation or behavior create distress or increase suicidal ideation in some high-risk students, such as depressed or substance-abusing youngsters or those with a past history of suicide attempts?

83 Evaluating Iatrogenic Risk of Suicide Screening Programs Columbia University Overall Research Design Timeline Experimental Group Control Group Day 1 (survey 1) Baseline POMS-A1 Beck Depression Inventory With suicide question Without suicide question Drug Use Screening Inventory Suicidal Ideation Questionnaire Suicide attempt history Immediate POMS-A2 Day 3 (survey 2) Persistent POMS-A3 Interim depression question Suicide questions Interim suicide item Suicidal Ideation Questionnaire Suicide attempt history Beck Depression Inventory suicide item

84 Flow of Participants Through Study

85 Evaluating Iatrogenic Risk of Suicide Screening Programs (Gould et al., JAMA, 2005) Summary of Our Project No differential dropout after Day 1 Asking about suicidal ideation/behavior does NOT create distress Asking about suicidal ideation/behavior does NOT increase suicidal ideation High-risk students are NOT more distressed or suicidal after being asked about suicidal ideation or behavior; on the contrary, depressed students and previous suicide attempters appeared LESS distressed and suicidal after being asked about suicidal ideation or behavior SCREENING IS SAFE

86 SCREENING Another concern A suicide screening procedure is only as effective as its ability to get at-risk students the care they need. There has been little systematic assessment of whether at risk youth have accessed services after their identification by the screen and whether their health status has improved.

87 Service Use by At-Risk Youth After School-Based Suicide Screenings Madelyn S. Gould, Ph.D., M.P.H. Kimberly Hoagwood, Ph.D., Frank Marrocco, Ph.D. Lia Amakawa, B.A. Marjorie Kleinman, M.S. Elizabeth Altschuler, M.S. Ed. Project funded by Centers for Disease Control R49 CE (Gould et al., JAACAP, 2009)

88 Service Use by At-Risk Youth After Suicide Screenings (Gould et al., JAACAP 2009) DESIGN A 2-year follow-up study of 317 youth identified as at-risk for suicidal behavior was conducted. At-risk cohort included those youth who reported recent or past suicidal behavior, prominent current suicidal ideation, moderate to severe depression and/or substance use impairment. A face-to-face clinical evaluation was conducted with all suicidal youth to validate their responses on the screening surveys. The parents of each at-risk student were contacted by a project clinical social worker to initiate case-management and referral procedures.

89 Service Use by At-Risk Youth After Suicide Screenings (Gould et al., JAACAP 2009) Columbia University CONCLUSIONS (I) Vast majority of at-risk youth were not in treatment at time screen. Approximately two thirds of those referred to treatment had used a new service by the follow-up 2 years later. Services accessed were mainly outpatient services, incorporating some form of psychotherapy rather than pharmacotherapy. Access to services by youths depends on the recognition and action of key adults, as well on their own perception of a problem.

90 Service Use by At-Risk Youth After Suicide Screenings (Gould et al., JAACAP 2009) Columbia University CONCLUSIONS (II) Screening seems to be effective in enhancing the likelihood that students at risk for suicidal behavior will get into treatment. However, nearly one third of at-risk students still do not get into treatment. Screening and referral services need to be coordinated to facilitate timely access into treatment. Need to enhance the engagement of parents and youths and address their perceptions about mental health problems.

91 Gatekeeper Training

92 Gatekeeper Training Underlying rationale Suicidal adolescents are under-identified Even professionals are reluctant to ask about suicide Community helpers can be among the first to detect signs of suicidality Major Aims Develop knowledge, attitudes and skills to inquire directly about distress, persuade suicidal youth to accept help, and provide referrals

93 Gatekeeper Training Example of programs

94 Gatekeeper Training Group-based randomized trial with 32 schools examined impact of Question, Persuade, Refer (QPR) training on a stratified random sample of 249 staff with 1-year average follow-up. (Wyman et al., 2008) Increase in knowledge, SELF-RATINGS of preparation to intervene and refer to treatment services; however the increases in asking about suicide occurred primarily among school staff who reported they inquired about suicide prior to gatekeeper training (14%). Behavioral rehearsal with role play practice results in higher gatekeeper skills in RCT of QPR. (Cross et al., 2011)

95 Crisis Hotlines

96 CRISIS CENTERS AND HOTLINES Theoretical rationale: Suicidal behavior is often associated with a crisis Suicide is usually contemplated with psychological ambivalence Cry for help dealt with by those with special training Practical advantages: Ubiquitous source of help worldwide Convenient, accessible and available outside usual office hours Anonymous

97 CRISIS CENTERS AND HOTLINES Columbia University Goals: Evaluate imminent suicidal risk Reduce crisis and suicidal state Identify alternative coping strategies and develop action plan /safety plan including formal and/or informal community resources

98 CRISIS CENTERS AND HOTLINES Increasing role in the armamentarium of suicide prevention efforts A notable advent in the past decade in the U.S. has been the implementation of National Suicide Prevention Lifeline (NSPL) - a national network of suicide Prevention crisis lines: TALK and SUICIDE. The national hotline numbers are currently providing back-up resources for a myriad of suicide prevention programs, including: public awareness messaging campaigns, school-based suicide prevention programs, and federal-, community- and advocacyinformation/referral documents and internet sites.

99 EFFECTIVENESS Evaluations of telephone crisis services have included: caller feedback/satisfaction, rates of follow up with referrals assessments of helping processes changes in community suicide rates (distal) changes in caller crisis or suicidal status (proximal)

100 EVALUATION OF PROXIMAL OUTCOMES (I) Gould, Kalafat et al.(2007) studied adult suicidal (n=1085) and nonsuicidal crisis (n=1617) callers from eight crisis hotlines across the U.S. We employed callers' own ratings of their mental state and suicidality, in response to a standardized set of inquiries by the crisis counselors at the beginning and end of the call, to assess the immediate proximal effect of the crisis intervention. A follow-up assessment, two to four weeks later, was also conducted to assess the duration of an effect and the telephone intervention's impact on future suicidal risk and behavior.

101 Evaluation of Proximal Outcomes (II) Key findings of our study: Seriously suicidal individuals were calling telephone crisis services - 8% in midst of attempt, 58% had made prior attempt. Significant reductions in callers' self reported crisis and suicide states from the beginning to the end of the calls. 11.6% of suicidal callers reported at follow up that the call prevented them from harming or killing themselves.

102 BUT. Of callers who were rated as non-suicidal crisis callers by crisis staff, 12% reported at follow up that they were either feeling suicidal during or since their calls to the center. Half of these callers had been suicidal at time of initial call, but this was not known or recognized by counselor.

103 Led to NSPL Practices 100% of crisis centers adopted NSPL risk assessment standards (Suicidal Desire, Capability, Intent; Buffers/Connectedness) as a result of missed suicidal callers 100% of crisis centers were offered ASIST training as a result of missed suicidal callers. Conducted pilot T4T March, 2007 for center trainers, NSPL and evaluation staff

104 Pilot Study of ASIST training Key Findings Marked variability among centers in addressing suicide Variability is related to amount of suicide training ASIST pilot showed some positive impact on counselors addressing suicide (but, analysis does not examine thoroughness of assessment) Demonstrated need for more rigorous trial of ASIST s effectiveness

105 Subsequent ASIST Intervention RCT January 2008 June 2008 July 2008 January 2009 July 2009 February 2010 (N=4) (N=6) (N=7) Fidelity Development Phase (N=2) Assessment Start Date X refers to ASIST trainings at centers Assessment End Date

106 Core elements of ASIST Intervention Trial ASIST Training for Trainers (T4T) for selected center staff ASIST Training for all center telephone counselors Application of ASIST by all center telephone counselors Response of caller to intervention Impact on short-term caller outcome Transfer of ASIST Transfer/Fidelilty of ASIST (Videotaping of Trainings) Silent Monitoring of calls Silent Monitoring of calls Research Follow Up Assessment with callers Work in progress

107

108 Follow-Up Procedures Quantitative Caller Feedback (I) Were there things about the follow-up call(s) from the crisis center that were helpful to you? N* % Yes % No % Do not remember % * N=600

109 Follow-Up Procedures Quantitative Caller Feedback (II) To what extent did the counselor s calling you stop you from killing yourself? N* % A lot % A little % Not at all % It made things worse 1 0.2% * N=527

110 Follow-Up Procedures Quantitative Caller Feedback (III) To what extent did the counselor s calling you keep you safe? N* % A lot % A little % Not at all % It made things worse 1 0.2% * N=527

111 Qualitative Data Coding: Preliminary Results Types of responses from callers who answered Yes to the question Were there things about the follow-up call(s) from the crisis center that were helpful to you? N %* Just getting the call(s) / knowing someone cares % Building rapport / qualities of the counselor(s) % Content of the call / therapeutic intervention % Information and referrals / promotion % of treatment engagement Other % * Percentages add up to >100%, since individual answers could receive more than one code

112 Examples of Caller Feedback (I) What was it about the follow-up calls that stopped you / that kept you safe? What stopped me was that someone who doesn't know me had interest in me, cared about me. I've lost so many people in my life, in such a hard way, and I stopped caring about my life. I haven't had anyone support me that way, and them calling me gave me a boost. The follow-up calls really gave me the message that they really did care, and that it wasn't just a one-time resource if I needed to turn to them again. That was really what kept me from continuing with my [suicidal] thoughts. Without those calls, I would have gone the other way. She gave me something to work on, something to look forward to.

113 Examples of Caller Feedback (II) What was it about the follow-up calls that stopped you / that kept you safe? [T]hey gave me a better grounding in reality Talking to them made me more conscious of what I really wanted which was relief from pain rather than to kill myself which is a rather permanent solution. One counselor came up with an action plan, as far as what I could do to keep the suicidal thoughts at bay. She didn t tell me what to do, but she helped me come up with things I could do on my own to take care of myself. The concern and the willingness to try to get me in somewhere different after the first [treatment] option that I was offered didn't work out.

114 FOLLOW-UP PRELIMINARY CONCLUSIONS The majority of callers consider follow-up to have saved their lives and kept them safe a lot. Callers experience connection with crisis counselors as a major source of support and stability. Centers consider follow-up valuable enough to mandate change in crisis intervention model

115 CRISIS CENTERS AND HOTLINES Challenges for Intervening with Youth Getting teens to call Linking youth to services

116 THE EVALUATION OF INTERVENTIONS TO PREVENT SUICIDE: AN EVALUATION OF TELEPHONE CRISIS SERVICES FOR ADOLESCENTS Madelyn S. Gould, Ph.D., M.P.H. Jimmie Lou Munfakh Ted Greenberg, M.P.H. Marjorie Kleinman, M.S. Keri Lubel, Ph.D. -Project funded by the Centers for Disease Control and Prevention- (U81/CCU216089) Gould, et al., SLTB 2006

117 To answer: Specific Aims (1) Are adolescents using hotlines significantly less than other formal sources of help? (2) What are the reasons adolescents state for not using hotlines? (3) Are the reasons for nonuse of hotlines different from the reasons for nonuse of other formal sources?

118 Are adolescents using hotlines less than other sources of help? (N = 519) + EVER USED LAST YEAR USED Source n % n % MHP %*** %*** School Counselors %*** %*** Internet %*** %*** Clergy %** %* OHP % % Substance Abuse Program % % Hotline % 9 1.8% + Due to missing responses, the denominator of "ever used" ranges from 508 to 515 *p<.05, ** p<.01, *** p<.001 (Gould et al., JAACAP 2002)

119 Summary of Study Findings When in crisis: Few adolescents use telephone hotlines Internet use (chat rooms) is prevalent

120 Contributed to New SAMHSA Initiative Columbia University Aim: To enhance the outreach and effectiveness of crisis center interventions by the use of emerging technologies such as chat and texting. New evaluation: The aim is to develop and pilot assessment tools and procedures for use with these emerging technologies.

121 Resilience development/ skills training

122 Resilience development/skills training Major Aims Prevent suicide through the enhancement of problem-solving, coping, cognitive skills, and help-seeking behaviors. Enhance protective factors to immunize students against suicidal feelings. These skills may prevent suicide risk factors such as depression, hopelessness and drug abuse.

123 Resilience development/skills training Example Designed to enhance socialecological protective factors Aims to increase connectedness between trusted adults and students, transform school climate using peer influence and positive messaging. Peer leadership training

124 Resilience development/skills training Example Randomized controlled trial in 18 high schools in three U.S. states. (Wyman et al., 2010) Improved peer leaders adaptive norms re suicide, their connectedness to adults, and their school engagement, with the largest gains for those entering with the least adaptive norms. Larger randomized controlled trial focusing on suicide outcomes is in progress (funded by U.S. NIMH)

125 School-based Suicide Prevention Strategies Effectiveness data mounting particularly from randomized trials and longitudinal studies But, need to overcome challenges of : Winning support from school personnel Competing for time with academic programing Ensuring no iatrogenic effects Maintaining skills and effects over time

126 Postvention Resource

127 POSTVENTION/CRISIS INTERVENTION Underlying rationale: Broad range of psychological sequelae identified Assume timely response will reduce morbidity/mortality Major Aims: Assist survivors in grief process Identify and refer at-risk survivors

128 Free download from and

129 After a Suicide: A Toolkit for Schools The American Foundation for Suicide Prevention (AFSP) and the Suicide Prevention Resource Center (SPRC), two of the U.S. s leading suicide prevention organizations, have collaborated to produce this toolkit to assist schools in the aftermath of a suicide (or other death) in the school community. It is a highly practical resource for schools facing real-time crises. While designed specifically to address the aftermath of suicide, schools will find it useful following other deaths as well. The toolkit includes an overview of key considerations, general guidelines for action, do s and don ts, templates, and sample materials applicable to diverse populations and communities. It also provides links to additional information.

130 After a Suicide: A Toolkit for Schools Guiding Principles Schools should strive to treat all student deaths in the same way Be aware of the risk of suicide contagion: do not oversimplify, glamorize or romanticize suicide Emphasize connection between suicide and underlying mental health issues Help is available

131 After a Suicide: A Toolkit for Schools Specific Areas Addressed Crisis Response Helping Students Cope Working with the Community Memorialization Social Media Suicide Contagion Bringing in Outside Help Going Forward

132 After a Suicide: A Toolkit for Schools Table of Contents Introduction and Executive Summary 6 Get the Facts First 9 Crisis Response 10 Tools for Crisis Response 13 Helping Students Cope 29 Working with the Community 32 Memorialization 35 Social Media 40 Suicide Contagion 43 Bringing in Outside Help 47 Going Forward 48

133 Crisis Response Form/activate an Emergency Response Team Arrange for sensitive & consistent notification of students, parents, staff Convene staff and parent meetings Disseminate accurate information about suicide Prepare a media statement & designate a media spokesperson Designate a community liaison Monitor activities throughout school, paying attention to students who may be having particular difficulty

134 Helping Students Cope Help students identify and express their emotions Encourage students to think about specific coping strategies they can use when intense emotions such as worry or sadness begin to well up Reach out to parents; offer guidance about how to talk with youth about suicide

135 Working with the Community Because schools exist within the context of a larger community, it s very important that in the aftermath of a suicide or other death they establish and maintain open lines of communication with community partners such as the coroner/medical examiner, police department, mayor s office, funeral director, clergy, and mental health professionals. A coordinated approach can be especially critical when the suicide death receives a great deal of media coverage and the entire community becomes involved. In particular, it is advisable that the school establish an ongoing relationship with a community mental health center that can see students in the event of a psychiatric emergency. In the aftermath of a suicide death, schools will want to notify the center to ensure seamless referrals if students show signs of distress.

136 Memorialization (I) Because adolescents are especially vulnerable to the risk of suicide contagion, it is important to memorialize the student in a way that doesn t inadvertently glamorize or romanticize either the student or the death. Wherever possible, schools should both meet with the student s friends and coordinate with the family, in the interest of identifying a meaningful, safe approach to acknowledging the loss. Schools can play an important role in channeling the energy and passion of the students (and greater community) in a positive direction. Suggestions for safe memorialization are provided on the following slide.

137 Memorialization (II) Suggestions for safely memorializing students who have died by suicide: Students who wish to memorialize the deceased can be encouraged to: hold a day of community service or create a school-based community service program in honor of the deceased sponsor a mental health awareness day volunteer at a community crisis hotline Schools may wish to make posterboard and markers available so that students can gather and write messages. After a few days, the posters can be removed and offered to the family. Schools are advised to set time limits for spontaneous memorials (up to 5 days, or until after the funeral) and to monitor them for for messages that may be inappropriate (hostile or inflammatory) or that indicate students who may themselves be at risk.

138 Social Media In the aftermath of a suicide death, social media can be used to: disseminate important and accurate information to the school community, identify students who may be in need of additional support or further intervention, share resources for grief support and mental health care, and promote safe messages that emphasize suicide prevention and minimize the risk of suicide contagion. Working in partnership with student leaders will enhance the credibility and effectiveness of social media efforts.

139 Suicide Contagion (I) Key considerations for school communities facing possible contagion: Identifying other students at possible risk for suicide Connecting with local mental health resources Managing heightened emotional reactions at school Monitoring media coverage Building a community coalition

140 Suicide Contagion (II) Identifying other students at possible risk for suicide: School-based mental health screening Remind all staff of the important role they may play in identifying changes in behavior among the students they know and see every day, and discuss plan for handling students who are having difficulty. In addition, schools can encourage the local medical community, including primary care doctors and pediatricians, to screen for depression, substance abuse, and other relevant disorders in the youth they see.

141 Bringing in Outside Help Particularly when dealing with possible suicide contagion, school crisis team members should remain mindful of their own limitations, and consider bringing in trained trauma responders from other school districts or local mental health centers to help them as needed.

142 Going Forward In the months following a suicide, schools should consider implementing programs in: Suicide awareness for teachers and staff Mental health awareness for students Gatekeeper training for teachers and staff Additional recommendations (not specifically mentioned in this section of the toolkit): Skills training for students Ongoing mental health screening and referral Maintaining ongoing relationships with community mental health centers/providers Maintaining the emergency response team

143 After a Suicide: A Toolkit for Schools Crisis Response Tools Sample Agenda for Initial All-Staff Meeting (14) Sample Death Notification Statement for Students (15-16) Sample Death Notification Statement for Parents (17-18) Sample Media Statement (19-20) Key Messages for Media Spokesperson (21) Sample Agenda for Parent Meeting (22-23) Talking About Suicide (24-25) Facts about Suicide and Mental Disorders in Adolescents (26-28)