Risk Assesment Flashcards
what is risk assessment?
prior to 2002 - dichotomy = either dangerous or not
now - range/degree - prediction (probability and risk factors) and management (treatments etc.)
Civil settings
- Civil commitment = if person poses danger to themselves or others can be hospitalized involuntarily
- Child Protection - remove child from home to protect from abuse
- Immigration laws - prohibit admission into can if believe will engage in violence or pose risk to social, cultural, economic functioning
- school and labour - prevent any act that would endanger others
- duty to warn/ common law - patient may commit violent act
- statutory law - mandatory reporting
Criminal setting
during pretrial, sentencing and release
adolescents - only committed to secure custody if high risk
smith v jones - public safety outweighs client privilege’s
long term offenders = high risk
parole needs assessment
- NCMD only released after assessment
Smith v. Jones
when defense did not head psychiatrists warnings on future violence, filed affidavit to judge
= when clear serious and imminent danger public safety outweighs solicitor-client privilege
parole board
use many sources including risk assessment to determine if get statutory release (2/3 sentence) or early release
- statutory release can be denied with high risk
prediction outcomes
true positive - predicted to be and is violent
true negative - predicted to not be and is not violent
false positive
false negative
Base Rate problem
Base rate = percentage of ppl within population that commit criminal or violent act
- low base rate = many false positives
- easier to predict frequent events than non frequent
Baxstrom v. Herald
detained beyond sentence and ordered to be released
- released more than 300 mentally ill offenders
Steadman and Cocozza
followed 98 patients relased after Baxstrom that considered too dangerous to be released - 7%violently reoffended
- 86% false positive rate
Thornberry and jacoby
followed 400 forensic patients released after dixon v. attorney general of common… of Pennsylvania
- 15% violently reoffended
- 85% false positive rate
Ennis and Litwack
clinical expertise like “flipping coins in courtroom”
Monahan
psychiatrists active in 1/3 of predictions
Barefoot v. Estelle
US supreme court ruled mental health professionals predictions admissible and “not always wrong only most of time”
R. Moore v. the Queen
supported the role of mental health professionals in
the prediction of violent behaviour
Monahan and Steadman
3 weaknesses in violence prediciton
- limited number of risk factors studied
- how criterion variable is measured - official records underestimates violenmce - not reported/ found
- when records combined with interviews, rates increase
- base rate increases from 4.5 - 27.5% (6 times higher) - how criterion variable defined - violent or not
- change to severity, type, target, location, and motivation
heuristics
shortcuts ppl use to make decisions. lead to inaccurate decisions
illusory correlation
Belief that a correlation exists between two events that in reality are either not correlated or correlated to a much lesser degree
Desmaris, Nicholls, Read, and Brink
association btw clinicians confidence and accuracy at predicting short term in patient violence
- minimal association = overconfident bias
Methods of Risk assessment
Unstructured clinical judgement, actuarial, structured professional judgement
Unstructured clinical judgement
professional discretion and lack of guidelines
- Grove and Meehl - “ informal in the head subjective…”
Pros: flexible, widely applicable, inform treatment and prevention
Cons: highly subjective and inconsistent, questionably relevant outcome, elude scrutiny
Actuarial Approach
Decisions are based on risk factors that are selected and combined based on their empirical or statistical association with a specific outcome. No clinical judgement
pros: demonstrated link btw risk factors and outcomes, reliability btw raters, consistency across time, transparency
limits:
estimates may not generalize to new samples, predictive properties may change in different contexts, constrained use, under samples risk
ex. Violence risk appraisal guide
Tarasoff v. Regents of university of California
common law duty to warn - expressed desire to kill individual in therapy - warned campus security but not police and person killed
Dr. James Grigson
Dr. death - testified in death penalty cases
- 100% certainty would kill again etc.
estelle v. smith - smith “ severe sociopath” bc lacked remorse
- Randall Adams - declared sociopath based off 15 min interview - sentenced to death - released after another inmate confessed
- 1995 expelled from APA for ethical violations
Structured Professional Judgement
Decisions are guided by a predetermined list of risk
factors that have been selected from the research and professional literature. Judgment of risk level is based on the evaluator’s professional judgment
- final decision uses discretion
- allows idiographic risk factors - flexibility and case specific
- relevance to management and prevention
- reflects current themes in field - risk is ongoing, dynamic, and requires re-assessment
- Limitations : expertise (training), and time (comprehensive assessment not just screening)
static risk factors
Risk factors that do not fluctuate over time and
are not amenable to change (e.g., criminal history)
also historical risk factors
Dynamic risk factors
Risk factors that fluctuate over time and are amenable to change (e.g., antisocial attitude, mental state, substance use etc.)
also criminogenic needs
Dispositional risk factors
Risk factors that reflect the individual’s
traits, tendencies, or styles (e.g., negative attitudes
clinical risk factors
Types and symptoms of mental disorders (e.g., substance abuse)
contextual risk factors
Risk factors that refer to aspects of the current environment (e.g., access to victims or weapons). Sometimes called situational risk factors
Demographic risk factors
age- younger 1st offense, greater chance engage in criminal behaviour
sex - males higher risk for general offending and more serious violent acts, females less serious violence
personality risk factors
- impulsiveness
- increases likelihood of engaging in crime and violence
- lifestyle impulsivity distinguishes recidivistic rapists
- psychopathy
- moderatly related to recidivism and violence in prison
- predicts reoffending across different countries in both males and females, mental disorders, male adolescents, sexual offenders
- weakly related to violent reoffending in adolescent females
- psychopathy and deviant sexual arousal predicts sexual recidivism
deviant sexual arousal
sexual offender shows relative preference for inappropriate stimuli such as children and violent nonconsensual sex
terrorism risk factors
Monahan 2012 :
age - 20-29
gender- most males, depends on group, female able to get closer to targets
marital status - single
social class - representative of local population, suicide terr = more highly educated
prior crime - usually no record - 6% committed again (bergman)
suicidality - differ greatly from general suicidal public
mental illness - low rates
substance use - very rare
psychopathy - low rates
Malik et al. :
- social wellbeing
- economic indicators
- governance
- law enforcement
- armed conflict
past behaviour as risk factor
past violent and non violent behaviour predict future risk for violence
age of onset
earlier start = more chronic and serious offenders
farrington - 50% of boys who committed prior to 16 committed again
Elliott - 50% boys before 11, 30% btw 11-13, 10% adolescent
females not predictor
childhood history
maltreatment
- physical abuse/ neglect - increased risk for violence
- sexual abuse - no more likely
- childhood abuse predicts initiated delinquency
- prolonged abuse predicts chronic offending
- physical abuse in adolescents directly related to adolescent offending
chaiken and chaicken
severe drug users commit 15 - 20 x more robberies and burglaries
types of drugs and crimes
heroin associated most with crimes
heroin and crack greatest risk for property crimes
some cases depends on amount of drug used
drug use moderately related to recidivism
substance use
swanson survey
- men/women less than 3% with no disorder committed violence vs substance use men 22%, women 17% violence
morley et al.
- violent behaviour associated with all drugs
- high risk associated with more serious polysubstance abuse
mental disorder risk factor
- affective disorder and schiz. higher rates
- suicide attempts/ self harm history - more likely verbal/ physical aggression to others
douglas, guy, hart - psychosis 49-68% increase in violence odds
- link depends on study design, measurement, timing of symptoms/violence
social support
instrumental
emotional
appraisal
information
access to weapons/ victims
released to environment with access/ around other antisocial ppl, potential for violence increases
Static -99
10 item acturial scale to predict sexual recidivism (0-12)
- age at time of release
- lived with previous intimate partner
- prior nonsexual violence
- index nonsexual violence
- # of prior sex offenses
- # prior offenses
- male victims
- unrelated victims
- stranger victims
- noncontact sex offenses
Violence risk appraisal Guide (VRAG)
acturial predict violence from prior violent episodes - made by quincey et al. from 685 violent offenders - need to be charged to be counted ( needs record) risk factors: psychopathy checklist (PCL-R) - positive relationship, 70% of decisions elementary maladjustment DSM-III personality disorder age at index offense separated from parents before 16 failure on prior conditional release property offense marital status schizophrenia victim injury alcohol abuse female victim
Douglas, Hart, Dempster, Lyon
VRAG replication from quincey data
- results were not the same - depends on subset of individuals used for coding the VRAG
also needs to be same n to be more accurate
- estimates may not generalize to new samples,
predictive properties may change in different contexts, constrained use,
under samples risk
Paul bernardo
scarborough rapist
actuarial VRAG example - only 17-31% chance of violence
administering SPJ
- relevant information from multiple sources
- presence of risk factors
- relevance of risk factors
- formulation of violence risk / case formulation
- primary scenarios of violence
- case management plans
- final opinions - low moderate or high
HCR-20
SPJ approach to predict future violence in correctional and forensic samples
- Historical (static) documented - 10
ex violence, antisocial, relationship, employment etc.
- clinical (dynamic) present/observed - 5
ex. insight, violent intent, mental disorder, instability, treatment response
- Future (speculative) projected - 5 = RISK MANAGEMENT
ex. professional services, living situation, support, treatment response, stress
Guy (2008)
compared SPJ and actuarial
both good at predicting future violence
Recidivism process
Precipitating situation (stressor) - cognitive and emotional appraisal
- a. individual influences ( psychopathy, emotional reactivity) - criminal bevaviour
- b. available response mechanism (social support, coping) - criminal behaviour
female offenders
- less criminal behaviour
- more prostitution charges
- more likely to target family
- reoffend at lower rate
- childhood victimization is higher than men
- more serious mental illnesses
- about same risk factors + history of self-injury or attempted suicide, and self-esteem problems
- being married is a risk factor for women!
- LSI-R
- substance use more associated with recidivism than males
aboriginal offenders
culturally relevant risk factors
risk assessment tools not as accurate
Ashley Smith
suicide in prison while on suicide watch
death ruled homicide
last year transferred 17 times - no treatment plan
Ewert v. Canada
risk assessments not valid for aboriginal offenders
overturned in 2016
protective factors
and how to measure
factors that mitigate or reduce the likelihood of negative outcome
- more research in youth - prosocial involvement, social support, social orientation, attachments, intelligence
- high risk offenders - employment stability
- low risk males - family connections
- male sex offenders - professional and social support, structured group activities, goal directed living
START - assesses weaknesses and strengths
Desistance
process of ceasing to engage in criminal behaviour
Haggard et al.
why high risk offenders stop offending
- insight triggered by negative events connected to criminal lifestyle
- social avoidance
- orientation to the family
