Frameworks for Suicide Prevention

1
Frameworks for Suicide Prevention

• Eric D. Caine, MD
• Center for the Study Prevention of Suicide
• Department of Psychiatry
• University of Rochester Medical Center
  Rochester, NY USA
• April 2008

www.rochesterpreventsuicide.org
2
Suicide Overview

• Suicide is a major public health problem.
• While psychopathology plays a central role, it is
  essential to recognize that suicide is the
  outcome of multiple risk factors and events that
  link as a chain where the end is an effort to
  kill oneself leading to death. The suicide event
  is like a punctuation mark at the end of the
  story.
• Age and gender represent two key ways of
  understanding risk factors for suicide, which
  they are and how they may work.
• Cultural forces play a critical role, with
  substantial variation one to another. It is not
  yet clear how to optimally use culture to
  modify suicide rates or patterns of risk.
• Prevention is best developed at multiple levels
  that are temporally specific and aimed at
  different links in the chain.

3
Global View of Suicide WHO
4
Preventing suicide, attempted suicide, and their
attendant mortality and morbidity will require
the development of PUBLIC HEALTH AND PREVENTIVE
PSYCHIATRY.
5
The New Public Health WHO

• Public health includes the health of the
  individual in addition to the health of
  populations.
• The health of individuals and groups depend upon
  social policies and programs (e.g., access to
  care), and national, regional, and community
  efforts that are, at once, coordinated yet
  diffuse.
• NPH promotes the building of healthy communities.
• NPH includes and far exceeds the scope of
  traditional public health (i.e., food and water
  safety communicable disease control emergency
  response).

6
(No Transcript)
7
U.S. Suicide Rates 1933-2001
Source CDC vital statistics Provided by A. Crosby
8
Suicide Rates by Age, Race, and GenderU.S.
2004
Source National Center for Health Statistics
9

• Suicide rates by ethnicity and age group United
  States, 1997-2001

Source CDC vital statistics Provided by A. Crosby
10
Suicide rates, ages 15 years by sex and
urbanization US, 1996-98
Source CDC vital statistics Provided by A. Crosby
11
Suicide by ethnicity and method United States,
2001
Source CDC vital statistics provided by A.
Crosby AI/AN Amer. Indian/Alaskan Native A/PI
Asian/Pacific Islander
12
U.S. Regional Suicide Rates
1997 Rate / 100,000 Population
8.7
11.6
17.4
8.8
12.0
12.4
12.7
11.9
13
(No Transcript)
14
(No Transcript)
15
Years of Potential Life Lost (YPLL) Following
Completed Suicide Across the Life Cycle (Males,
2000)
16
Years of Potential Life Lost (YPLL) Following
Completed Suicide Across the Life Cycle (Females,
2000)
17
Risk Factors for Suicide(Derived from US, EU,
Australia/NZ)

• Youth and young adults comorbid major
  psychopathology depression, bipolar disorder,
  or schizophrenia conduct disorder, alcohol /or
  substance abuse/dependence personal and family
  turmoil legal problems poor school or work
  performance prior attempts and family history of
  suicide.
• Adulthood comorbid depression and alcohol
  use/dependence (less in China) interpersonal
  disruptions social isolation poor work
  performance and unemployment violence and legal
  problems variable impact of marital and parental
  status, prior attempts and family history of
  suicide.

18
Risk Factors for Suicide

• Elders comorbid depression (often later onset,
  w/o a prior history of psychopathology) and
  general medical conditions, often associated with
  pain and role function decline social dependency
  or isolation widowhood personality
  inflexibility as a precursor alcohol and Rx
  substance abuse present, though not to the extent
  of younger ages. Frequent contact with primary
  care providers.
• With the exception of rural China (and perhaps
  rural India), males always predominate.

19
Frameworks for Developing Preventive
Interventions
20
Frameworks for Developing Preventive
Interventions

• Level of the intervention (universal, selective,
  indicated broad-based vs. high-risk
  population oriented prevention measures)
• Population or group of interest (e.g., youth who
  have dropped out of high school young adult
  women alcoholic men depressed elders)
• Site of implementation (site of capture)
  defining specific sample biases
• Developmental contexts for populations and
  high-risk individuals
• (Stage of disease process 1, 2, 3 prevention)

21
Terminology to Describe Populations Levels for
Prevention Programs

• Universal focused on the entire population as
  the target ? prevention through reducing risk and
  enhancing health, based on Roses Theorem
  broadly aimed, but can affect individuals as well
• Selective high-risk groups where not all of the
  members bear risks ? prevention through reducing
  specific risks among groups
• Indicated symptomatic and marked high-risk
  individuals ? clinical interventions to prevent
  full-blown disorders or adverse outcomes

22
The Language of Prevention applied to Suicide
and Attempted Suicide
23
Population vs. ClinicalApproaches

• Population-oriented prevention of disease
  expression is the ultimate desired outcome of
  population-oriented approaches (not well
  supported by insurance and systems designed for
  episode-based health care)
• Shifting the population average is central to
  prevention, and requires distinctive approaches
  in comparison to treating sick individuals
• Physicians treat an individuals disease
  (signs, symptoms, and causes), and may consider
  the larger context of illness, but not distal
  risk factors thus prevention challenges the
  cure model
• Population approaches seek out at-risk groups and
  individuals physicians typically wait for
  patients to come to the clinic door

24
Roses Theorem ...a large number of people at
small risk may give rise to more cases of disease
than a small number who are at high risk (Rose
1989)
25
The Population Distribution of Suicide Risk
Mortality threshold
Population
Low High Suicide Risk
Modified from Crosby
26
High-risk Approach
Mortality threshold
Identify and treat high-risk
Population
Low High Suicide Risk
Modified from Crosby
27
Population-oriented Approach
Mortality threshold
Move population risk
Population
Low High Suicide Risk
Modified from Crosby
28
Behavioral and Social Implications of Roses
Theorem

• Focus on shifting common (normative) behaviors
  that are present in the general population.
• Disease and illness may not be easily applied
  to targets for preventive interventions.
• Examples of common or normative conditions
  Frequent intoxication and binge drinking,
  recreational drug use, domestic distress and
  disputes, pain in the context of medical
  disorders.
• Butmost normative behaviors go unchecked
  (e.g., cell phones and driving smoking).
• For the US, this is especially complex given
  values of individual differences and restrained
  governmental intrusion into personal conduct
  moral vs. institutional authority.

29
Sites for Population-Oriented Interventions

• Broad, community-based chosen irrespective of
  risk focus upon culture and attitude change
  change in social policy and laws (population
  vaccination means control)
• Wide dispersion of information and education
  use of media to facilitate cultural
  transformation need to mitigate the risk of
  cultural transfer of suicide permissive attitudes
  
• Gatekeeper identification and education
  broadly applied training to facilitate
  identification of higher risk individuals (does
  NOT fit neatly into level terminology)
• Examples Legislation to change public practices
  media (cautiously) staff development programs in
  worksites, religious and faith-based
  organizations, community NGOs, governmental
  agencies (e.g., social services, unemployment)

30
Sites for High-Risk Reduction

• Selective and indicated interventions driven
  through points of capture for higher risk
  individuals
• Examples Courts, jails and prisons medical and
  mental health agencies chemical dependency
  programs non-medical therapeutic agencies
  (e.g., batterers programs) primary care,
  gynecology, and pediatric offices

31
Site Population Approaches
32
Site-population approaches (continued)
33
The Developmental Perspective Complementary to
Roses Theorem

• Roses Theorem fails to address defined high-risk
  groups or individuals
• Treatments of diagnoses and symptoms are
  necessary for high-risk individuals, but often
  are not sufficient. Reducing the burden of
  psychopathology does not specifically address
  turmoil-filled life situations, which in turn
  exacerbate psychiatric distress.
• Thus, treating major psychopathology
  symptomatically is a necessary step to facilitate
  further efforts to address other important risk
  factors. Too often it is not sufficient to
  prevent suicide in the longer term.

34
Suicide Prevention for Men 21-34 Years A
Developmental, Layered Model
SUICIDE
Peri-suicidal State
Despair, Hopelessness
RISK
Repeated Episodes, ? Resiliency, Social/Family
Turmoil
Therapeutic Noncompliance, Legal Difficulties,
Functional Decline
Severe Mental Illnesses, Substance/Alcohol Abuse,
Unstable Social Ecology


TIME
"Distal"
RISK FACTORS
"Proximal"
Indicated
Universal
Selective
PREVENTION STRATEGIES
Caine Conwell, 2001-04
35
Suicide Prevention for Men 25-54 Years A
Developmental, Layered Model
SUICIDE
Peri-suicidal State
Depression, Hopelessness
RISK
? Symptoms, ? Resiliency, Family Turmoil, Work
Problems
Substance/Alcohol Abuse, Role Changes, Acute
Chronic Stresses
Personality Factors, Social Ecology, Cultural
Values Perceptions

TIME
"Distal"
RISK FACTORS
"Proximal"
Indicated
Universal
Selective
PREVENTION STRATEGIES
Caine Conwell, 2001-04
36
High-risk Groups and Sites to Contact Them
37
High-risk Groups and Sites to Contact Them
(continued)
38
Indicators of Suicide Prevention Program
Effectiveness

• Process indicators most often used evidence
  of participation in programs and activities the
  type of measure least indicative of meaningful
  effect
• Impact indicators less often used evidence of
  change in program participants (their attitudes,
  knowledge, skills, and actions) or in program
  activities
• Outcome indicators least used evidence of
  change in ultimate (end point) targets ? i.e.,
  reduction of mortality and morbidity. NOTE
  Changes in suicidal ideation and depression
  frequent intermediate steps on the path to
  suicide are not specific indicators of suicide
  prevention or the prevention of serious attempts.

39
Suicide Prevention and Evaluation Universal

• Problem definition and awareness
• Surveillance (case definition) ethical
  challenges outcome indicators fundamental ??
  of attempts and deaths
• Media surveillance potential positive (Vienna)
  or negative outcomes (Hong Kong) also impact
  assessment of changes in media coverage and
  social attitudes
• Restriction of lethal means
• Coal gas paracetamol package size (UK)
  surveillance
• Education and training
• Clinicians (primary care providers, mental health
  professionals, and others)
• Public employees (police, judges, probation,
  prison personnel, social service case workers)
• Evaluation Outcome indicators of reduced
  attempts and deaths, as well as process and
  impact indicators

40
Suicide Prevention Selective

• Programs for persons with severe mental illnesses
  to assure adequate housing and consistent care
  impact indicators provision of care ultimately
  ?? ?? change in surveillance measures, though
  may NOT be detectable in terms of deaths
• Changes in program procedures regulatory
  review Require screening of patients with
  chemical dependency disorders for depression or
  mood changes, and violence (intimate
  partner/domestic) impact indicators of improved
  operations care ?? ?? change in surveillance
  measures (detectable?)

41
Potential Selective Interventions

• Integrated court-based screening and mental
  health services, covering both Family and
  Criminal Courts to overcome barriers to care
  not contingent upon adjudication, addressing the
  needs of victims, perpetrators, and children
• Workplace supported programs incentives for
  promoting mental health similar to efforts to
  promote greater physical health
• Focused efforts to support recently unemployed
  men
• Outreach programs to isolated elders
• Outreach programs to troubled street youth
• May utilize the full range of indicators
  process impact, and perhaps outcome measures if
  done at scale in large collaborative program
  networks

42
Suicide Prevention - Indicated

• Early interventions for individuals with
  recurrent psychotic episodes
• Active treatment of depressionin chemical
  dependency programs, in courts, jails, and
  prisons, in primary care settings, on campuses,
  in social care community settings for elders.
  THESE MUST BE DRIVEN BY ACTIVE OUTREACH EFFORTS,
  beyond clinic walls!
• Samples typically are too small for outcome
  evaluation assessment most often depends upon
  process and impact indicators

43
The USAF Suicide Prevention Program A
Multi-Layered Approach

• Public health-community orientation The Air
  Force Family
• Broad involvement of key leaders Medics-Mental
  Health, Public Health, Personnel, Command, Law
  Enforcement, Legal, Family Advocacy, Child
  Youth, Chaplains, CIS Walter-Reed Army Inst. Of
  Research CDC
• Consistent leadership involvement
• 11 Initiatives (reducible into key components)
• Common Risk Model

44
USAF Program Components

• Component 1 Leadership as Role Models
• and Agents of Change. Engage and educate
  leaders use existing formal (supervisor ?
  subordinate) and encourage informal (peer ? peer
  family-based) networks to convey cultural
  priorities and change values.
• Component 2 Change specific policies to support
  and assure program implementation.

45
USAF Program Components

• Component 3 Educate personnel and enhance
  community preventive programs integrate delivery
  of social and clinical services. Build skills,
  with attention to common risk issues and 11
  initiatives.
• Component 4 Assure compliance with programs and
  sustain initiative
• Newest initiative Measure implementation and
  effect (dose-response) of prevention measures!

46
ADAF Suicide Rates
AFSPP launched in late 1996
Pre-AFSPP Average 13.8/100K
10/100K Line
Post-AFSPP Average 9.6/100K
47
Knox et al., BMJ 2003
48
Key Components for Multi-Layered Prevention
General

• Sustained, focused urgent leadership at all
  levels of society
• Interactive top-down (leadership) and bottom-up
  processes (community)
• Recognition of common risk (akin to a common
  enemy) at the community level, including suicide
  and attempted suicide, alcohol and drug use,
  domestic violence, unemployment, workplace losses
  in productivity, accidental death, homicide
• Aggressive combating of stigma
• Building broadly based community coalitions

49
Key Components for Multi-Layered Prevention
Governmental

• Clarity of federal, state and local roles and
  alignment of priorities
• Collaboration in the central government,
  including research and service oriented agencies
• State-level affirmation of care for those with
  severe mental disorders, and use of regulatory
  authority
• Fundamental reorganization of funding to deal
  with similar problems in disparate settings and
  to assure universal coverage for preventative and
  therapeutic interventions

50
Framing Questions for the Designers of Suicide
Prevention Programs

• What are the broad goals and specific objectives
  of the intervention and the program?
• Where does this program fit in an overall
  framework (schema, model) of suicide
  intervention?
• Who do we expect to reach with this effort?
• Who will we miss?
• What will we be changing?

51
Framing Questions for the Designers of Suicide
Prevention Programs

• How will we measure those changes, and use that
  information to evaluate the program?
• How confident are we of our findings?
• Can the program and its results be exported
  widely?
• Can the program be sustained after its ardent
  founders have moved on?