Embed Size (px)
Citation preview
PACT
Risk assessment training for social workers
Chris Newman and Kate Iwi www.fsa.me.uk
Training Groundrules
• Confidentiality
• Looking after ourselves
• Respecting others and sharing the responsibility to
challenge each other respectfully
• Timekeeping
• Drink / drugs
• Distractions
2
• What problems do you face when
assessing risk in cases where domestic
violence is a child protection concern?
• And in reporting on this risk to others,
especially the courts?
• What further knowledge / skills do you
feel you need when assessing and
reporting on DV cases?
Learning outcomes
• For those who work with perpetrators and victims of DV to be able to
come to defensible conclusions about risk and vulnerability.
• To be able to make clear explanations of these conclusions in
written reports and when presenting evidence in court or other
settings.
• To be able to assess the key static (historical) risk and vulnerability
factors in a case
• To be able to assess dynamic (changeable) risk factors and explain
how these may link with treatment viability.
• To be able to assess changes in risk and to explain this assessment
to others.
4
Thinking about risk
Case 1: John
Maria, his partner of 4 years has reported a severe assault on her - (strangulation
to the point of unconsciousness) other than that no violence but some controlling
behaviour around jealousy
She has one child, had planned to live as a family with John – we are asked to
assess the risk involved in this.
John (age 45) reports happy childhood and good relationship with parents
Regular employment as carpenter
One previous marriage – still sees adult children – no reports at the start of the
case of any violence or abuse in this relationship.
No substance misuse
Some reports of fights in pubs as a younger man
But... you phone his ex-wife and she tells you that he used to grab her by
the throat to control her.
Case 2: Jamie
• Jamie (21) grew up with violence and abuse in his childhood
• Living in care home from 15 yrs – mum asked him to leave because
he was using drugs and abusive to her
• Met Anna (now 20) in hostel – she is a care leaver too
• Dozens of reports of police being called to the hostel, and
subsequently their flat – reports of shouting, smashing furniture,
Anna has been seen with black eye on two occasions
• Anna says they argue because they are both very jealous and that
she „winds him up‟
• They have one daughter and want to live together and care for her
together
• We are asked for our assessment of the risks in this.
Which case is higher risk?
Any statement about risk should be as specific
as possible and include consideration of:
Likelihood of abuse occurring
What? The nature of the potential harm – (Physical harm? Emotional harm?
Severity of abuse/violence
When? Within what time period?
Who – does the perpetrator pose a risk to – partner, child?
In what context? – when the couple are living together/apart? At time of threat to
relationship? At contact handover?
“Levels of Risk”
What does
• low
• medium
• high
• very high
mean?
These need to be defined separately for likelihood and severity
Make sure you‟re clear about this, or you get a clear definition
from the person doing the assessment
11
Example definitions of likelihood
(of harmful behaviour occurring)
very
unlikely to
occur
unlikely to
occur
may occur
(a ‘real
possibility’)
likely to
occur
very likely
to occur
Physical abuse
No Standard Moderate High
Never, or not
currently
Slapping,
pushing; no
injuries.
Slapping,
pushing;
lasting
pain or
mild, light
bruising or
shallow
cuts.
Noticeable bruising, lacerations,
pain, severe contusions, burns,
broken bones, threats and
attempts to kill partner, children,
relatives or pets. Strangulation,
holding under water or threat to
use or use of weapons; loss of
consciousness, head injury,
internal injury, permanent injury,
miscarriage.
Example definitions of severity of
abuse
For instance
In my opinion, if the couple continue to live together, Mr X poses a
moderate (may occur) risk of using physical violence severe
enough to cause injury towards Ms Y and it is very likely that he
will continue to use persistent and severe verbal aggression as
well as emotional and psychological abuse towards Ms Y.
I believe that whilst the risk of direct abuse of the child is low, the
possibility that he will be indirectly harmed during incidents cannot
be discounted. Also given his young age there is a very high
likelihood that he will be exposed to any abuse of his mother that
takes place.
Impacts of domestic violence on
children
• Trauma effects on
brain development
Perry B (1999) Incubated in terror – see course reading
Also Siegel, D (1999) The developing
mind. Toward a neurobiology of
Interpersonal experience.
New York: Guilford press.
Impacts on children
Impacts on children
• Attachment styles
(eg. Howe, D (2003) –
‘Attachment disorders’
in Attachment and
Human Development
(5:3) pp265 – 271)
Impacts on children
• Trauma bonding
(Eg. James, B 1994;
Handbook for treatment of
Attachment related
Disorder in children)
Impacts on children
• Social learning
Domestic violence and family dynamics
♀ ♂ ♂
♀
♀
♂
♀
♀
♀
♀
Children’s identities are shaped via adaptive and maladaptive coping mechanisms
• Mental Blocking, desentisation or Dissociation
• Making it Better Through Fantasy
• Physical Avoidance
• Looking for Love (and Acceptance) in all the Wrong Places
• Taking Charge Through Caretaking
• Reaching out for Help
• Crying out for Help
• Re-Directing Emotions into Positive Activities
• Trying to Explain the abuse
• Trying to Prevent or Control the abuse
Protective factors and children‟s
resilience
Rutter 1985 – protective factors against adversity
•Self esteem
•Age of child during incidents
•Frequency of incidents
•Child‟s ability to make sense of events
•Child‟s relationships with others
Rutter, M. (1985). Resilience in the face of adversity - protective factors and resistance to psychiatric-disorder. British Journal of Psychiatry, 147, 598-
611.
Children‟s Resilience
Grotberg 1997- protective factors include:
•Even and adaptable temperament
•Capacity for organised thinking and problem solving
•Physical attractiveness (influence on self esteem)
•Sense of humour
•Good social skills and supportive peer network
•Sense of autonomy and purpose
•Secure attachments to at least one parent
•Connections to wider community
•Sense of control
Grotberg, E. (1997) A guide to promoting resilience in children: strengthening the human spirit, Bernard Van Leer Foundation,
Resilience
• contact with lots of supportive adults who know about what‟s happened and will offer tonnes of love
• permission to talk to others freely
• contact with peers who‟ve gone through similar
• the best possible relationship from now on with both parents
• anything that increases their self-esteem – extra-curricular activities, running clubs, drama shows etc.
• getting therapy or specialist help to understand and express themselves better.
• Mum and dad give a clear understanding of what happened and why – no excuses
• Mum and dad get support to understand and parent them.
• helping others through voluntary work, mentoring etc.
• having a safety plan so they have some control over their safety if it happens
again. See What works in Building Resilience, Newman T 1994
Overview of risk assessment
methodology
25
Types of risk assessment (you are already doing one of these)
Clinical Assessment
Where an experienced practitioner makes the assessment, using
„practice wisdom‟ to interpret what they find out about the client.
Actuarial Assessment
• Researchers identify a range of risk factors which are
associated with harmful behaviour
• Investigate which ones show the strongest statistical
relationship with future maltreatment.
• The assessment then consists of identifying the presence or
absence of these factors and weighting them according to a set
scheme to produce a risk rating.
26
Types of risk assessment Empirically guided clinical assessment
The worker conducts a risk assessment by referring to a checklist of
factors that have a demonstrated relationship to onset and /or recidivism.
This helps the worker to focus on the relevant data to gather, so that the
final assessment, though not statistical, is well informed by the best
available research.
[Probably the best and most realistic method in our settings]
27
Exercise
What makes your alarm bells ring?
Discuss some cases you have worked with and identify as many risk factors for domestic
violence as possible.
Write down those which are to do with the victim on one colour post-its and those to do with the
perpetrator or other factors on a second colour post-its.
28
Exercise
Sort the factors into:
Static (historical) factors – the ones that can‟t be changed.
Dynamic factors (ones that are open to change, or things you might look for as evidence of reduction in risk)
Which factors or combinations of factors might be „triggers‟ i.e. signs that risk will increase suddenly or that danger is imminent?
29
Static and Dynamic factors
Risk factors can be divided into two groups:
• Static factors – those which are based in the
individual‟s past history and background
demographics, and so are not amenable to
change.
• Dynamic factors – those which can change
through treatment, interventions or the
passage of time. These include information
about the person‟s current attitudes and
beliefs gained from interview.
30
Static factors provide the backbone of any credible risk assessment.
There is a danger of overrating impressions gained from the service user in interview and underrating information about the person‟s past history and behaviour. Abusers may:
• fare well in psychological testing, often better than their victims.
• convince others that they have „learned their lesson‟ or „put their past behind them‟, overstating the deterrence value of future punishment or other consequences.
• be mild mannered and appear reasonable despite severe risk, or be noisy and intimidating with professionals despite presenting only moderate risk to their partner or child.
In contrast, victims may appear angry with services, emotionally dysregulated and difficult to work with.
(Bell, C. 2007)
31
The advice from research therefore is:
• First form a judgement of risk based on
static factors.
• Then use dynamic factors to make modest
adjustments to this estimate.
Clearly information gained from the client about
imminent, targeted risk should be acted upon.
Static + Dynamic + ‘triggers’ = Current
Risk
historical long term
characteristics
immediate
behaviours or
changes in situation
long term
risk
treatment
targets
Risk management /
supervision
A way of structuring our thinking about risk
Investigative skills
Clinical skills
monitoring
Domestic Violence Risk Factors
History of DV Other relevant
behaviour
Circumstances Mental state
•Used severe
violence, with
injuries requiring
medical
treatment
•Used
strangulation
•Used or
threatened to use
a weapon
•Threatened to kill
partner
•Used sexual
violence, such as
rape
•Assaulted other
family members,
including children
•Violence is
becoming more
frequent
•Violence is
becoming more
severe
•Current substance
misuse – notably of
alcohol or crack –
especially where it has
exacerbated the past
severity of the
violence
•Generalised
aggression, both
inside and outside the
home.
•Relevant
psychological
disorders; Anti-social
personality disorders,
eg. borderline
personality disorder, a
psychopathic
personality or beliefs
of persecution by
others
•Currently has access
to the partner
•Partner is trying to
leave or has recently
left
•Currently isolated from
support systems
•Step children in family
•Feeling of having nothing to lose
•High levels of anger and hostility
•Depression
•Recent suicide risk
•Low mental functioning
•Obsessive jealousy or control of
partner
•Obsessive thinking about partner
following separation
Life history Attitudes about the violence
•Severe abuse in the
perpetrator‟s family
of origin
Recent life stresses
•Unemployment
•Homelessness
•Bereavement
•Poverty
•Equivalent life
stresses
•Severely blaming partner
•Severely minimising or denying the
violence
•Lacking remorse
•Traditional attitudes about male
dominance
•Lacking victim empathy
•Fantasising about killing or
wreaking severe violence
•Not recognising the risk
•No motivation to change
What gets in the way of good risk
assessment?
What influences might lead practitioners to:
1.Underestimate or fail to identify risk?
2. Identify risk where it does not exist, or
overstate the level of risk?
35
• Failure to make full enough enquiries - not gathering the
evidence
• Being put off assessing a situation as risky due to „cultural
sensitivity‟,
• or being put off gathering evidence due to a scary, avoidant or
manipulative client
• Applying a rule of optimism - developing a strong optimistic
outlook in relation to a given case, which blinds you to contrary
evidence.
• Defensive practice - making overly pessimistic assessments for
fear of getting it wrong.
• Focusing on one type of abuse and not considering other types
of risk - for example putting emphasis on DV and working to
decrease that risk only to find that the parents pull together
against the child, thereby increasing the direct risk of harm.
36
• Either relying too much on a mechanistic process OR on personal judgement not supported by evidence.
• Undue weight to dynamic rather than static factors.
• Failure to identify resilience factors.
• Failing to link together information or identify clusters of
signs.
37
When thinking about risk, also be
aware that..
• Some factors are stronger indicators of risk than others. For instance, prior perpetration is the most reliable predictor of re-assault.
• Regardless of the number of ticks on a risk checklist, one or two factors alone may indicate high risk – e.g. plausible threats to kill
• The same factors may vary in significance for different people – (e.g. for some perpetrators of DV, pregnancy is a time of greatly increased risk for their partners, while for others pregnancy makes violence less likely).
• The presence of multiple factors does, however, increase risk.
• Factors do not operate in isolation, they interact. It is particularly important to consider how “volatile combinations” of particular factors may increase risk.
38
Caveats about assessment tools
• They may give an inflated impression of certainty
• When applied in practice, some assessment tools may feel restrictive and ask you to ignore risk relevant information.
• The factors they use are drawn from a particular population at a particular time – do these apply to other populations – (different areas, LGBT perpetrators, female perpetrators, ethnic minorities?)
• Also be aware of the limitations of the risk statements they make and their applicability to the field you work in.
e.g. ODARA...
cont/-
39
Ontario Domestic Assault Risk Assessment (ODARA)
• Score = 0 “7% of such wife assaulters commit another assault against an intimate partner that comes to the attention of the police within an average of about 5 years”.
• Score = 1 17%......etc
• Score = 2 22%.....
• Score = 3 34%......
• Score = 4 39%.....
• Score = 5-6 53%.....
• Score = 7-13 74%.....
(32% of whole sample reoffended)
cont/-
40
The grandmother test – “my grandma could have told you this guy was high risk”
1. Prior domestic assaults
2. Prior non-domestic assaults
3. Prior correctional sentence of 30 days or more
4. Failure on prior conditional release
5. Threat to harm or kill at the index incident
6. Confinement of victim at the index incident
7. Victim is very concerned
8. Number of children
9. Victim has children from a previous partner
10. Violence against others
11. Substance abuse
12. Assault on victim when pregnant
13. Barriers to victim support
e.g. 7 or more items, 74% of such men re-assault within 5 years.
Two ways Risk Assessment can „get it wrong‟
• the assessment may fail to identify women and children at risk (false negative),
• or it may identify risk where it does not exist, resulting in unnecessary and disruptive intervention and resulting harm to families and children (false positive).
Limitations of Risk Assessment
All assessment methods are imperfect - human behaviour, like the weather, is not perfectly predictable. A weather forecaster can get their calculations wrong and thereby make a wrong forecast, but even if they get their calculations right, the forecast may turn out wrong.
This means that even with the best methods available:
• Some high risk cases will be missed
• And some people will have social care services, or police involved in their lives unnecessarily
400 abusive
parents
600 non-abusive
parents
1000 families
Using a tool which is
75% „accurate‟*
Base rate: 40%
(400/1000)
300 (true)
positives
100
negative
results
75%
150 (false)
positives
450 (true)
negative
results
75%
Probability of a positive result being a true positive = 67%
(300+150 = 450 positive ratings, 300 of these are true positives = 67%) (*we‟ve artificially set specificity and sensitivity to the same rate)
4 abusive
parents
996 non-
abusive
parents
1000 families
Same tool
75% accurate
Base rate: 0.4%
(4/1000)
3
Positive
result
1
Negative
result
75%
253
positive
result
747
negative
result
75%
Probability of a positive result being a true positive = 1.2%
(3+253 = 256 positive ratings, 3 of these are true positives = 1.2%) (*we‟ve artificially set specificity and sensitivity to the same rate)
Why is this important?
The paradox is therefore that risk
assessment methods are least successful
where they are most needed,
when attempting to predict rare, serious
events
47
Defensible Decision Making
DEFENSIBLE DECISION MAKING IS WHERE:
• all reasonable steps are taken;
• reliable assessment methods are used;
• information is collected and thoroughly evaluated;
• decisions are recorded and carried through;
• agency processes and procedures are followed;
• managers are investigative and proactive.
Home Office, Mappa Guidance 2006.
48
Assessing the history of violence and
abuse
• Severity, frequency, persistence, escalation of violence
• Signs of imminent risk
• Severity, frequency, persistence of other forms of abuse:
– Verbal, psychological/emotional abuse
– Proprietorial behaviour
– Coercive control –perhaps key indicator
– Sexual violence and abuse
– Post separation abuse
– Exposure of children
– Direct abuse to child
– Violence in other/previous relationships
49
Weighting factors in your assessment
• Static risk factors do not all carry equal significance and cannot
be equally weighted
• It is generally the aggregated effect of multiple factors that
increases predictive power: the higher the number, the greater
the probability that harmful behaviour will occur.
• According to Gelles et al (1994), the presence of two out of ten
of the risk factors for domestic violence they identified suggests
twice the likelihood of violence than where none of the factors is
present.
• Seven or more factors, on the other hand, represents a risk
which is forty times greater.
50
Weighting factors in your assessment
• Volatile combinations -
• The combination of witnessing paternal violence to one‟s mother
and of directly suffering parental abuse is especially highly
correlated with the perpetration of domestic violence against a
partner in adulthood (Campbell 1995).
• Alcohol or drug abuse combined with major psychopathology
significantly increases the risk of violence perpetration
(Swanson et al 1990)
• Stalking, jealous rumination, suicidal „all or nothing‟ thinking
51
Weighting factors in your assessment
• Of course one or two critical factors may lead to a conclusion
that reoffending is likely
• (for instance DV perpetration in multiple relationships and a
vulnerable partner)
• And for a substantial minority of „family-only‟ DV perpetrators,
very few factors may come to light in the case history.
Working with denial
Implications for risk assessment
Long standing belief that perpetrators who deny their
violent actions are among those least likely to desist
from further violence and abuse
This applies to both domestic violence perpetrators and
child abusers.
Implications for risk assessment
Mixed evidence for a link between denial
and general criminal recidivism.
• Strongest for those men who are convinced of
their own innocence (self-deceivers)
• But for those whose denial is a conscious
attempt to avoid consequences, or to avoid
shame and embarrassment (liars), there is at
least some sense that their behaviour is
unacceptable.
„Psychological‟ vs „social‟ denial
or liars vs self deceivers
Lying is a (conscious) strategy generally employed to
avoid anticipated aversive consequences
Denial - an (unconscious) psychological defence arising
from a need to maintain a positive self-image in order
to avoid potentially overwhelming emotion, such as
shame.
In its traditional sense, being „in denial‟ is therefore a
product of self-deception: the tendency to provide
honestly held but inflated descriptions of the self.
Context is also important
Henning and Holdford (2006) found that with
domestic violence offenders at time of arrest :
„Participants who intentionally denied minor character
flaws in a possible attempt to appear socially
conforming were less likely to recidivate than
offenders who were more forthcoming‟ (pp. 123–124).
Some of those who admit at time of arrest may simply
not care about what they did – “she deserved it” or
don‟t realise that what they did was serious
So, paradoxically, „liars‟ may be less risky
than (uncaring) admitters, and self
deceivers.
Also note that some offenders cycle
between acknowledgement –>shame ->
anger->re-offending
Implications for change or treatment
• It makes it easier to continue being
abusive – if it‟s only a small problem,
why change?
• It‟s very hard to work with those in denial
• It implies that the victim needs to
change before you can stop being
abusive.
Is denial a cognitive distortion?
Maruna and Mann 2006 Yes there is a powerful correlation between
denial and offending, but which causes which?
Assumption that excuse making causes – or at
least allows for – offending.
And that treatment should therefore aim at
eliminating denial in all its forms
Is denial a cognitive distortion?
However, when challenged about having done
something wrong, all of us reasonably account for our
own actions as being influenced by multiple, external
and internal factors.
Yet, we pathologize offenders for doing the same thing.
Is denial a cognitive distortion?
In everyday use, excuses are employed
as an „aligning action indicating to the
audience that the actor is aligned with
the social order even though he or she
has violated it‟
Is denial a cognitive distortion?
Catch 22? Making excuses is „criminal thinking‟. But if I say I committed some awful offence
purely „because I wanted to‟ and because that is the „type of person‟ I am, that is just as worrying.
Is denial a cognitive distortion?
For non-depressives, failure events tend to be
external, temporary, and specific, but good
events are personal, permanent, and
pervasive.
‘If it’s bad, someone else did it to me, it’ll be
over soon, and it’s only this situation.
But if it’s good, I did it, it’s going to last
forever, and it’s going to help me in many
situations’.
For depressives the reverse is the case
Is denial a cognitive distortion?
Excuses as part of negotiated social reality
• Excuse making seems to convey a level of
respect for the victim
• Most listeners prefer accounts in which
wrongdoers excuse or justify their offending
behaviour. Accounts characterized by
preference (I did it because I wanted to) and
negligence (I did it because I didn‟t think)
made listeners angry (Weiner 1987)
Is denial a cognitive distortion?
Maruna and Mann recommend less focus on
post hoc excuses, more focus on attitudes
which are supportive of offending:
•pro-violence attitudes
•beliefs that child victims enjoy sexual contact
or are not harmed by it.
•dehumanizing victims
•seeing the world as hostile
•„labelling one‟s self as “naturally” prone to
offending
So when and why is denial
a problem?
Implications for change or treatment
• It makes it easier to continue being abusive –
if it‟s only a small problem, why change?
• It‟s very hard to work with those in denial
• minimisation of responsibility is linked to
higher treatment attrition rates
• It implies that the victim needs to change
before you can stop being abusive.
Implications for those affected by the
abuse
• It leaves the burden of dealing with the
effects of the violence with those who have been harmed.
• Others may blame or not believe the victim.
• If talking to children, it puts the blame on the non-abusive parent
• or denies their direct experience • and hinders their recovery
Implications for those affected by the
abuse
With partner abuse • it puts the blame on the non-abusive
parent • denies children‟s direct experience • and hinders their recovery
Accepting blame vs becoming
accountable, for the sake of those around you
The bedframe
Approaching the issue
• Are you worried about how you deal with anger?
• Do you ever get out of control?
• do you feel there‟s times when you‟ve gone too far? / when you haven‟t used the right methods to stand up for yourself?
• We all have disagreements at home sometimes. How do you guys resolve your conflicts? At your best? Give me an example of that. And at your worst? Give me an example of that….
• What are you like when you are angry - At you‟re best (give example)? At your worst (give example)?
• when did you first scare to her do you think? Tell me about that?
• What ways have you laid a hand on her in anger?
• How do you get physical with her when arguing?
• Have you ever hit her or pushed her around?
• What do you feel is the worst thing you‟ve done to her? What would she say to that?
71
To move away from blaming accounts
Gain their consent to focus on their part of the equation:
I am going to bring you back to talking about yourself when you seem to be focusing a lot on your partner - is that okay? I know you might go on feeling that you want to talk about her more and may feel frustrated when you‟re shut down but I‟d ask you to just stick with it for now.
72
Challenging minimisation
• Acknowledge and discuss the barriers to disclosure
• But be persistent in order to get detail and context of what happened
• use scales – on a scale of 1-5 (5 being punching her as hard as you can) how hard did you punch her?
• Ask for detail of the words that the abuser used - ask for exactly what was said and how. How loud?
• Reflecting back without the minimising e.g. he says “I just slapped her” – reflect back “you slapped her”.
• start at higher levels of violence than you believe has been used so that the client then has to work back to the actual level of violence
73
Dynamic factors
74
Expanding his/her ethical side
Listening to what he wants to tell you first
• So what brought you here? This might be a difficult process and I‟m sure you‟ve thought about that when deciding to come. What made you think it mattered enough to give it a try?
Repositioning his anger – reflecting back values
• You feel like no-one‟s heard your side - so being listened to is really important to you
• so fairness really matters to you
• so justice is something you feel really strongly about – tell me what you mean by justice….
Get his consent to ask direct questions
• I‟m going to ask some very direct questions about your family life and your relationships, is that okay?
• Would you tell me if I ask you something you don‟t want to answer - would you speak out?
Opening a window to remorse
Switching to visual memory and present–tense account
• So you are in the front room, she‟s hurt and the kids are crying.
• What are realising when you try to picture that?
Repositioning shame
• I can see that it‟s painful to look at the way you acted then
• what sort of person would you be if it wasn‟t hard/ if you didn‟t feel that?
76
Ask questions which increase his internal conflict
about his use of violence. Start with questions
about:
• the impacts of the violence on him
• his children‟s experience – you may be able to draw
parallels between his own childhood experience and his
children‟s
• You an also ask about his awareness of the effects on
his partner - but be aware that in the early stages he‟s
likely to be angry and resentful towards his partner
,
77
Contrasts with how he wanted things to be
• What did you see in your partner when you first met?
What do you think she saw in you? How do you think
you made her feel at first? And now?….
And how he wants things to be in the future
• Where do you see yourself in two/five years if you
continue to abuse? Do you think your partner will care
about you in the same way? do you think you will be in a
relationship? what do you think your relationship with
your children will be like?
• In the past have you ever said you‟ll change or won‟t do
it again? Did you keep your word? 78
Safety planning with perpetrators
• Lets assume your partner isn‟t going to change - how can you keep yourself safer over the coming weeks ?
• Do you know when it's going to happen? Is there a pattern? What are the usual trigger subjects and situations/ sore points?
• What‟s the most likely trigger situations that could come up in the next week or two?
• What could be an absolute worst case scenario in the coming weeks?
• When you were building up to being abusive before what was the first sign you were angry - what did your body feel like? What sort of faces, gestures, movements were you making? What thoughts were running through your head? In particular how did you see her? How different is this to how you see her when you‟re not angry with her?
• What do you already do to try to talk yourself down and stay safe?
• What‟s the most effective single thing you can tell yourself to wind yourself down?
• How early on in the situation do you have to start winding yourself down for it to be most effective?
• How and when would you know that this wasn‟t enough and that you were escalating towards frightening or abusive behaviour anyway?
