Risk Assessment

Question 1

key definitions

Answer 1

• In general terms the criminal justice system has defined risk as:
• the risk of future re-offending & reconviction - the probability that an offender / prisoner will offend, be arrested, and be reconvicted within two years; and
• the risk of serious harm - if reconvicted, the probability that the offence will be one of “serious harm”
  Section 224 of the Criminal Justice Act 2003 defines serious harm (in the sentencing context, when determining whether an offender presents a significant risk to the public of serious harm by the commission of further offences) as “death or serious personal injury, whether physical or psychological”.

Question 2

what is a risk assessment

Answer 2

• Systematic effort to estimate and evaluate adverse outcome
  
  • Try to make predictions about future behaviours- comes with its own challenges
  • Forensic risk assessments draw on information about the offender and their circumstances in order to reach judgements about their future behaviour
    Concept of risk encompasses “the nature of what adverse outcomes will be, severity, imminence, and frequency or duration of harm – as well as its likelihood” (Litwack et al., 2006, pp.493)

Question 3

what is seriousness

Answer 3

Somethings are extremely serious e.g. homicide, serial killer. On the lower end of the scale are offences such as shoplifting

Question 4

what is likelihood

Answer 4

• They can be used to make a prediction about the future
  The likelihood someone will reoffend

Question 5

adverse outcomes

Answer 5

who is it going to happen to and how undesirable the outcomes are
- How undesirable these would be

Question 6

what is dangerousness

Answer 6

Propensity to cause significant harm and lasting damage
Someone who is dangerous is deemed to be high risk

Question 7

offender may be labelled as dangerous:

Answer 7

• on the basis of a high probability of committing many offences OR
  
  • because of the potential to perpetrate a small number of very serious offences OR
    more commonly an unknown mix of the two

Question 8

the risk of harm posed by offenders can be seen as having 2 key dimensions

Answer 8

• the relative likelihood that an offence will occur; and
  
  • the relative impact or harm of the offence - what exactly might happen, to what or whom, under what circumstances, and why.
  • It is important to understand these two dimensions of risk. Some crimes (eg shoplifting) have relatively little impact or harm but, statistically, are the most common.
    Others (eg homicide) are rare but cause maximum damage.

Question 9

risk is categorised as risk to:

Answer 9

• It is important to identify the person or groups of people who are specifically at risk; this allows resources and protective measures to be applied effectively
• the public: either generally or a specific group such as the elderly, vulnerable adults (for example, those with a learning disability), women or an ethnic minority group;
• a known adult: such as a previous victim or partner;
• prisoners: within a custodial setting;
• children: either specific children or children in general
• staff: anyone working with the individual whether from Probation, the Prison Service, police or other agency (eg. health);
  self: the possibility that the individual will commit suicide or self-harm.

Question 10

level of risk

Answer 10

• LOW: current evidence does not indicate a likelihood of causing serious harm;
• MEDIUM: there are identifiable indicators of serious harm. The offender has the potential to cause such harm, but is unlikely to do so unless there is a change in circumstances - for example, failure to take medication, loss of accommodation, relationship breakdown, drug or alcohol misuse;
• HIGH: there are identifiable indicators of serious harm. The potential event could happen at any time and the impact would be serious;
  VERY HIGH: there is an imminent risk of serious harm. The potential event is more likely than not to happen as soon as the opportunity arises and the impact would be serious.

Question 11

a good risk assessment will:

Answer 11

• Make a prediction of risk along both dimensions- likelihood and impact of harm
• Identify the likelihood or reoffending
• Identify the risk of harm (what harm and to whom)
• Identify the key risk factors that led to the offence under consideration and that should be addressed
  To reduce the likelihood of further reoffending

Question 12

what can risk assessments be used for?

Answer 12

• Assess likelihood that someone will:
• Reoffend and for what type of offence
• Cause harm to themselves
• Cause harm to others
• Treatment needs of that person
• Individuals treatability and treatment readiness
• Identify the particular conditions in which a specific individual will behave violently or criminally.
• Zahid Murabek- young offenders institute, due to be released in 6 weeks due to petty theft, placed in a cell with Robert Stewart (known violent racist) and over a 6 week period Stewart abused and bullied Zahid- the prison service did not do anything about it. Hours before he was due to be released, Stewart murdered Zahid unprovoked.
  Now, the prison service undertake a cell sharing risk assessment so they are not placed with someone that could pose risk to them.

Question 13

risk assessments are used by:

Answer 13

• Police
• Courts
• Prisons
  Treatment providers

Question 14

risk assessments are carried out at various stages of an offender pathway

Answer 14

• During investigation (RF lecture Prof Alison)
• Pre-sentence
• At admissions
• Pre treatment
• Post treatment
• Pre-release
  Post-release

Question 15

what are the consequences of a false negative (wrongly said that someone is not risky)

Answer 15

Andrew Dawson- known local thief who turned into a murderer when he was very young, after Dawson was released (deemed to be safe), five days later killed two of his neighbours and then ran. When he was arrested he was armed with 7 knives and a letter that confessed his actions, said in courts he had an ‘urge to kill’, probation officers had deemed him low risk on the basis that there was no signs he would commit another offence. All the way through prison he was planning to do this when he was released. This is where an inappropriate level of risk has been assigned
Jordan McSweeney- sexually assaulted and then kicked to death Zara Aleena, all of his actions were viewed in isolation so deemed him low risk.

Question 16

what are the consequences of a false positive

Answer 16

• Harder to display- someone who has deemed to be of risk but do not actually reoffend e.g. parole is rejected and are kept in prison for longer than necessary
• Damage to a particular person and their family, a waste of public resources (costs a lot to keep someone in a prison)
  In America, if you do 3 federal sentences it is an automatic life sentences- a person had a small amount of drugs three times so got a life sentence. Obviously posed low risk and was silly to keep her in prison for life

Question 17

overarching aim of a risk assessment

Answer 17

• is to maximise the proportion of correct predictions
  whilst minimising the false positives and negatives

Question 18

risk factors

Answer 18

• Are the conditions that increase the likelihood, frequency and length of offending behaviour and form the basis of all risk assessments
• Static factors
  
  • Cannot be changed or influenced and are based on historical events or characteristics e.g. brain injury, ethnicity, age, offending history
• Dynamic risk factors
  
  • Dynamic risk factors relate to the attitudes, circumstances and behaviours that underpin or support offending.
    Can change through interventions (theoretically!)

Question 19

clinical risk assessment

Answer 19

• Clinicians use their own experience and intuition
• Collect background information
• Observations of the offender by professionals
• Interviews with the offender
• Comes from medical field and seen as diagnostic method of assessment
• Strengths
  
  • the debate centres on the accuracy of the clinician’s ability to predict violence.
  • the clinician can assess emotional state throughout the interview- they can pick up on nuances e.g. body language, no empathy or remorse.
  • For example, emotional traits such as lack of empathy and anger (Menzies, Webster, & Sepejak, 1985) and physiological and behavioural traits such as chanting, flared nostrils, flushed face and clenching of the jaw/hands (Berg, Bell, & Tupin, 2000
• Limitations
  
  • Thornberry and Jacoby (1979) 65% who were subsequently released into the community, only 11% were re-arrested for violent offences.
  • Monahan’s (1984) claim that two thirds of all clinical predictions of violence are incorrect
  • Susceptible to decision-making heuristics and biases: very difficult to change your view and opinion about someone
  • High percentage of false positive errors
    Risk assessment is completely dependent on one clinician and the decision they make

Question 20

actuarial risk assessment

Answer 20

• Uniform rules and strict decision making criteria are applied
• Based on longitudinal studies which assess associations between risk factors at the outcome being measured
• These require analysis of substantial samples from which generalisations are drawn
• Use algorithms or statistical equations often of static risk factors
• Give risk score or risk level
• Strengths and weaknesses
  
  • Eliminate the subjective errors present in clinical judgement
  • Douglas, Ogloff, and Hart (2003) have criticised actuarial tools for being rigid, lacking sensitivity to change and failing to aid risk management and the prevention of violence.
  • Difficulty in generalising from a group to an individual (Scott & Resnick, 2006),
  • Oversimplify the complexity of factors involved because cases are diverse in nature
    rely solely on an actuarial tool could cause potentially important indicators of violent behaviour to be missed.

Question 21

Webster and Hucker, 2007

Answer 21

• “to yield statistically significant effects summarized across large numbers of people, particular factors seen in isolation or even limited combinations, have to be very powerful to ‘show through’.
  Researchers have sometimes failed to realize that their studies rest on what they are presently able to measure, that a good deal of hard-to-index information necessarily remains crucial in the making of individual release decisions”.

Question 22

structured clinical judgement

Answer 22

• Structured clinical format
• A focus on dynamic factors (e.g. attitudes, level of future planning, stress etc.)
• Dispositional factors (e.g. personality)
• Trait factors (e.g. attitudes, deviant sexual interests, impulsivity)
• Contextual factors (e.g. potential destabilisers)
• Enables systematic judgements
• Improve accuracy of predictions
• Assisting in planning and delivery of treatment and management
• Used regularly in practice and is often seen as a compromise using clinical and actuarial measures to help predict future crimes
• Hanson and Thornton (1999), the predictive accuracy of professional risk assessments (both actuarial and clinical) is only slightly better than chance.
• Combining empirically established risk factors with clinical judgement
• The use of both actuarial tools and of structured risk assessment instruments in conjunction with clinical interviews has therefore been judged to be most effective (Scott & Resnick, 2006).
• This recommendation harvests the benefits of both actuarial tools and clinical judgement (structured or unaided) while minimizing their limitations in practice.
• clinician is given some level of flexibility and discretion for when cases contain idiosyncrasies but yet they can still provide an empirically based individual risk assessment
  Completing risk assessments with a person results in a far more meaningful and better adhered to plan of action

Question 23

popular measures

Answer 23

• Offender Assessment System (OASys)-
  
  • Main tool used by Prison and Probation
• Level of Service Inventory - Revised (LSI-R)
  
  • Primary actuarial measure for non-mentally ill offenders (usually used for sentencing and parole decisions)
• Psychopathy Checklist - Revised (PCL-R)
  
  • Primary measure for psychopathy, which is highly correlated with recidivism and risk of violence
• Structured Assessment of Violence Risk in Youth (SAVRY)
  
  • For adolescents 12-18 years old
• HCR-20 (Historical, Clinical, and Risk Management)
  Primary risk assessment measure for mentally ill offenders

Question 24

structured professional judgement strengths

Answer 24

• Grounded in actuarial variables
• Good validity
  
  • Combination of static and dynamic factors increases power
• More useful than purely actuarial as can help define strategies to manage risk
• Useful in identifying specific factors that increase or decrease risk
• Does have a ‘clinical override’ component
• Risk assessment informs risk management
  How to keep this person and other people safe?

Question 25

structured clinical judgement weaknesses

Answer 25

• Need more research
  
  • outcome studies
• Problems trying to combine actuarial & clinical measures, particularly if they contradict each other (Dawes et al, 1989)
  HCR-20 good at predicting risk in ‘high scorers’ but less good in the ‘middle or low’ range scorers (Strand et al, 1999)

Question 26

HCR-20

Answer 26

• Historical, Clinical, Risk management and the number 20 refers to the number of items
• The HCR-20 takes account of the individual/client’s current mental, emotional and behavioural functioning
• Using information from a number of sources such as face-to-face interviews, observation, clinical notes, ward notes, and psychological and neurological testing.
• Historical (H)- history of problems with:
  
  • Violence
  • Other antisocial behaviour
  • Relationship instability
  • Employment problems
  • Substance use
  • Major mental disorder
  • Personality disorder
  • Traumatic experiences
  • Violent attitudes
  • Treatment or supervision response
• Clinical- recent problems with:
  
  • Insight
  • Violent ideation or intent
  • Symptoms of major mental disorder
  • Instability in their life
  • Treatment of supervision response
• Risk- future problems with
  
  • Professional services and plans
  • Living situation
  • Personal support
  • Treatment or supervision response
    Stress or coping

Question 27

Hares psychopathy checklist

Answer 27

• The Psychopathy Checklist – Revised (PCL-R) was published by Robert Hare in 1991.
• This is a 20-item checklist that purports to measure Psychopathy
• It requires the administration of a semi-structured interview that can last between 90 and 120 minutes.
• Assesses a range of demographic, criminological, social and psychological information in a systematic manner.
• The PCL-R has been shown to be a strong predictor of recidivism and violence in offenders and psychiatric individuals, even though it is not a risk assessment device.
  However, it is heavily oriented towards the forensic context and tells us nothing about risk to self, mental instability or vulnerability

Question 28

Systematic review and meta analysis of structured clinical tools (Oganah, Seyedsalehi & Fazel, 2023)

Answer 28

• Systematic review of 50 articles, 36 tools, 10,460 participants, 12 countries
• 94% of studies used post release convictions or recidivism
• Overall predictive power was mixed
• 98% of studies had high risk of bias mainly due to poor analytical approaches
• Meta-analysis on violence recidivism from 19 studies with at least 100 participants
• Outcomes ranged from 0.72 for H10, 0.69 for HCR-20 and VRAG to 0.64 for Static 99
  Mixed evidence for predictive power, so important practitioners select the correct tool

Question 29

summary

Answer 29

• Risk assessment occur for a number of reasons, throughout the criminal process and are conducted by a number of different agencies
• There are different types of risk assessments which utilise static and dynamic risk factors
• The main aim of a risk assessment it to maximising the number of correct predictions whilst minimising the number of false negatives and false positives
  There are strengths and weaknesses to all risk assessments

Question 30

the internet and ILOC

Answer 30

• What are the key features of internet offending and IIOC?
  
  • There is no regulatory body to ‘police’ the internet (Bainbridge & Berry, 2011).
  • Control of the enormous amount of available content is limited (Beech et al., 2008).
  • Global usage of the internet recently topped 4 billion users in 2017, which is over 50% of the worlds population (Internet World Stats, 2017).
• The Internet…
  
  • Provides Accessibility, Affordability & Anonymity – ‘Triple A Engine’ (Cooper 2002).
  • Allows individuals to engage with others who share the same pro-offending attitudes and enables easy access to IIOC (Quayle & Taylor, 2002).
• Google processes over 40,000 search queries every second on average (Internet Live Stats, 2017).
• This translates to over 3.5 billion searches per day and 1.2 trillion searches per year worldwide.
• In just 60 seconds, 28,258 internet users are viewing pornography (Internet Filter Review, 2006).
• Approximately 20% of all internet pornography is child sexual abuse (National Center for Missing and Exploited Children, 2013).
• One ISP reported blocking more than 20,000 attempts to access IIOC on the internet in one day (Quayle, 2010).
  The trade in IIOC is argued to be between a £2 (Ropelato, 2006) and £13 billion (Bourke & Hernandez, 2009) industry annually.

Question 31

offenders and ILOC use

Answer 31

• There is a considerable body of academic research that has looked at offenders convicted of IIOC offences.
• In trying to understand offending behaviour, a number of key questions …
• How do offenders use IIOC within their offending?
• How prevalent are contact sexual abusers within IIOC offending population?
  What are the key features of IIOC & dual offending?

Question 32

the use of ILOc in offending

Answer 32

• Diverts from contact abuse: ‘compensatory’ model
  
  • Use of IIOC acts as a diversion to contact offending (Riegel, 2004).
  • This model suggests that viewing and reaching arousal to IIOC enables offenders to use this as a means to inhibit the urge to act upon fantasies (Babchishin et al., 2011).
  • However, this model has little empirical support.
• Encourages contact abuse: ‘facilitation’ model
  
  • IIOC facilitates the ‘spiral of sexual abuse’.
  • Offenders begin with lower level images and progress through the levels from grooming/inciting/producing to the contact sexual abuse of a child (Sullivan, 2002).
• In occurrence alongside contact abuse: ‘concurrent’ model
  
  • IIOC is used as part of an already established paraphilic lifestyle (Bourke & Hernandez, 2009).
    Individuals use IIOC to justify their paedophilic interests (i.e. if it exists it must be ok to view) rather than them becoming a paedophile because they have viewed IIOC (Sheenan & Sullivan, 2010).

Question 33

prevalence of dual offenders within ILOC samples

Answer 33

• Key question: What proportion of IIOC offenders are also committing contact sexual offences (or have previously)?
• A meta-analysis that considers a wide range of research suggests that contact sexual offending against children among IIOC offenders ranges from:
• As low as 5% (Seto & Eke, 2008);
• 55% when using self-report data;
• To as high as 84.5% - but this is considered a statistical outlier (Bourke & Hernandez, 2009);
• Research conducted as part of the FIIP & FIIP 2 projects suggests a best guess of 1 in 6.
  Source: (Seto, Hanson & Babchishin, 2011).

Question 34

levels of images

Answer 34

• L1: Images of erotic posing, with no sexual activity;
• L2: Non-penetrative sexual activities between children, or solo masturbation by a child;
• L3: Non-penetrative sexual activity between adults and children;
• L4: Penetrative sexual activity involving a child or children, or both children and adults;
• L5: Sadism or involving the penetration of, or by, an animal.

Question 35

key features of ILOC and dual offending

Answer 35

• There are number of factors found to differentiate between offenders according to risk (IIOC vs. dual):
• Previous behaviour
  
  • is an indicator of future behaviour
  • anti-social tendencies
  • previous criminal history (not just sexual offences).
• Access to children
  
  • greater opportunity for contact offending.
• Behavioural facilitators
  engaging in risky behaviours.

Question 36

ILOC risk prioritisation tools

Answer 36

• Operation Ore
  
  • Commenced in 1999 and was the UK’s biggest ever computer crime investigation of its time.
  • Referrals sent to the UK from the US containing details of individuals who had paid for access to IIOC online using credit cards.
  • Leading to 7,250 suspects identified, 4,283 homes searched, 3,744 arrests, 1,848 charged, 1,451 convictions, 493 cautioned and 140 children safeguarded.
• Following Operation Ore
  
  • Continually growing number of IIOC investigations and access to the internet.
  • Inconsistencies in workload prioritisation and risk assessment.
  • Absence of an academically validated way to prioritise the most dangerous offenders (i.e. those most likely to also commit contact sexual abuse against a child).
  • Lead to the idea to develop a risk prioritisation tool.
    The Kent Internet Risk Assessment Tool (KIRAT) was developed.