View metadata, citation and similar papers at core.ac.uk
brought to you by
CORE
provided by London Met Repository
October 2015
Violence, abuse
and mental health
in England
Population patterns
Responding effectively to violence and abuse
(REVA project) Briefing 1
Summary
• New analysis of national survey data shows that a quarter
of the population have had experience of violence and abuse.
• This research highlights strong links between different types
of violence and abuse and a range of mental disorders.
• These experiences are major and under-acknowledged factors
shaping people’s service needs.
• Services must address both mental health and violence and abuse
if they are to respond effectively to service users’ needs.
Introduction and key findings
In 2006 the Department of Health
introduced routine enquiry about
peoples’ experience of violence
and abuse as part of adult mental
health assessments. This was
in response to evidence that such
experience was a significant
aspect of the histories and
difficulties of many service users.
This research extends the evidence base by showing
how mental illness is linked with experience of abuse
and violence. It is based on a representative
community-based sample rather than a patient
population and examines experiences of both
women and men across the life course.
Key findings
• Experiences of violence and abuse are strongly
related to subsequent mental health and the
services people need and use.
• Six distinct groups in the population were identified
through analysis of data from the Adult Psychiatric
Morbidity Survey (APMS) on reported experiences
of violence and abuse in childhood and adulthood.
• Three quarters of the population have little
experience of violence or abuse, but the remaining
quarter consisted of people with five distinct
profiles of violence and abuse. Each group differed
in terms of their socioeconomic circumstances,
health, mental health and use of treatment and
services. Poverty, disability, poor health and health
risk behaviours were much
more common in those groups characterised
by extensive violence and abuse.
• One of the groups, representing 1 in 25 of the
population (around 1.5 million adults) had
experienced extensive forms of both physical and
sexual violence, with an abuse history extending
back to childhood. Nearly everyone in this group
had, at some point in their life, been pinned
down, kicked or hit by a partner. Half had been
threatened with death. Most had been sexually
abused as a child and some severely beaten by
a parent or carer. Many had also been raped
as an adult. Over half the members of this group
had a common mental disorder such as clinical
depression or anxiety. However, only 10% were in
receipt of counselling or a talking therapy when
they were interviewed.
• A further group – representing 1 in 50 of the
population - were characterised by their
experience of extensive physical violence and
coercive control by a partner (but not by other
kinds of abuse). They also had very high levels of
common mental disorder. Our analysis shows an
extremely strong relationship between partner
violence and mental health which has previously
received little attention.
01
Relevance for service provision
Mental health professionals
Public health
People who have experienced extensive physical
and sexual violence are far more likely than those
with little experience of violence or abuse to have
a common mental disorder, psychosis, PTSD or an
eating disorder. Effectively supporting survivors of
violence and abuse should be ‘core business’ for
mental health services.
Sexual and domestic violence are major public
health issues. Violence and abuse are experienced
in all socioeconomic groups but those with the most
extensive experience are more likely to also have to
cope with disability, low-income and the challenges
of poorer health, housing and neighbourhoods. The
evidence clearly links the experience of extensive
physical and sexual abuse with alcohol dependency,
smoking, and obesity.
Health care commissioning
Despite being 15 times more likely to have multiple
mental disorders, people with extensive experience
of physical and sexual violence were just four times
more likely to discuss their mental health with a GP
and only three times more likely to access community
mental health services. Only 10% were receiving any
kind of talking therapy. However, they were 12 times
more likely to have spent time as an in-patient on
a mental health unit.
Crisis and emergency services
Suicide attempts are 15 times more likely among
people who have experienced extensive physical and
sexual abuse. They are also 5 times more likely ever
to have self-harmed than those with little experience
of abuse. A&E staff, paramedics, police and fire
officers are likely to encounter survivors of violence
and abuse at their most distressed.
3
02
The research
The findings presented here
were part of wider research into
effective mental health service
responses to the long-term
consequences of violence and
abuse, conducted by researchers
from NatCen Social Research,
DMSS Research, and the Child
and Women Abuse Studies Unit at
London Metropolitan University.
The analysis presented here uses data from the
Department of Health funded Adult Psychiatric
Morbidity Survey (APMS). This is the only
representative, national, community-based data
source to use mental assessments to generate rates
of treated and untreated mental illness and risk
factors in England. APMS 2007 included long
interviews with 7,400 adult men and women of
all ages from across England. They were asked
about their:
• Experience of a range of types of violence
and abuse
• Current mental health
• Current social and economic circumstances
• Use of health and community mental health
services.
We used latent class analysis, a statistical approach
which divides the population into non-overlapping
groups of people with similar profiles, in order to
increase our understanding of how people with
different histories of violence and abuse go on to
have different mental health outcomes and service
needs.
4
03
Lifetime experience of abuse
and violence: the six groups
The population can be divided into six different profiles based
on lifetime experience of a wide range of different types of abuse
and violence.
5
03 Life time experience of abuse and violence: the six groups
Group 1
Group 5
Relatively little experience of violence or abuse
Three quarters (76%) of the population belonged to
this group and had relatively little experience of any
of the types of abusive experience asked about.
However, even in this group, 3% had been prevented
by a partner from seeing friends and family, and 2%
had household finances withheld. 14% reported some
experience of having been bullied.
Sexual violence as an adult (and sometimes also
in childhood)
Group 2
Physical violence from a partner
One in ten (10%) belonged to this group. 81% of
whom had been ‘pinned down, slapped, pushed’ by a
partner, and 61% had been ‘kicked, bit, hit’. There was
some evidence of coercive control. For example, a
third (33%) had been prevented from seeing friends/
family and 39% were threatened with hurt. However,
sexual abuse and the more ‘severe’ types of physical
abuse (choking, use of weapon, threatening to kill)
were rarely reported by members of this group.
Group 3:
Extensive physical violence from a partner
3% of the population was in this group. The abusive
experiences that defined this group were almost
entirely sexual. Virtually all (99%) had experienced
non-consensual sexual touching as an adult, and
almost a third had non-consensual intercourse in
adulthood (30%). Childhood sexual abuse was also
evident in this group: half had been ‘talked to in a
sexual way’ and 40% had been touched sexually
as a child.
Group 6
Extensive physical and sexual violence as adult
and child
One in twenty-five (4%) of the population was in this
group. Almost all had been ‘pinned down, slapped,
pushed’, as well as ‘kicked, bit, hit’ by a partner, and
had experienced high levels of coercive control. In
addition, they had suffered very high levels of severe
sexual violence: 38% had non-consensual sex in
adulthood and 23% had experienced this as a child.
One in fifty (2%) were assigned to this group. 81% of
people in this group had been threatened with death
and 53% had weapons used against them. Almost
all had been ‘pinned down, slapped, pushed’ and
‘kicked, bit, hit’ by a partner, and two-thirds (65%)
had also been choked. They had experienced high
levels of coercive control. However, members of this
group had experienced very little sexual violence.
Group 4
Sexual violence as a child (not in adulthood)
One in twenty (5%) belonged to this group. 85% had
been ‘talked to in a sexual way’ and 66% had been
touched sexually as a child. 13% had experienced
non-consensual sex in childhood. 13% of people in this
group had been beaten by a parent. Although almost
half had experienced non-consensual sexual talk in
adulthood, few had experienced abusive sexual
contact as an adult.
6
04
Profiles of the six groups
Proportion of each violence and abuse group who are female
84
80
73
67
60
46
%
Little
violence/abuse
Physical from
partner
Extensive
Sexual: only as
physical/coercion,
child
partner
Sexual: adult,
sometimes child
Extensive
physical, sexual
as child/adult
Sex, age and ethnicity
Women were more likely than men to be in every
abuse group, especially those groups where such
experiences were most extensive. 84% of those in the
‘extensive physical and sexual violence’ group were
women. However, men were present in every group.
The people most likely to be in a group characterised
by violence and abuse were divorced women aged
35 to 54.
group. The low prevalence of older people in the
groups characterised by violence and abuse may
be partly due to under reporting of violence by this
age group.
In terms of ethnicity, white people were
overrepresented in one of the two physical violence
only groups, and black people were overrepresented
in the ‘extensive physical and sexual violence’ group.
Divorced and separated people were overrepresented
in groups characterised by extensive violence and
abuse, and older people (and the widowed) were
overrepresented in the ‘little violence or abuse’
7
04 Profiles of the six groups
Socioeconomic profile
Extensive experience of violence and abuse was
more common among; people with a lower
household income, a low level of educational
qualification (GCSE or equivalent) and living in rented
households, in the most deprived neighbourhoods.
However, this pattern did not hold for all groups
characterized by violence and abuse. Having a
degree, for example, was associated with being
in one of the two sexual violence only groups.
While there were associations between socioeconomic
factors and abuse and violence, what was even more
pronounced was the fact that violence and abuse are
experienced in all socioeconomic groups. 29% of
people with extensive experience of physical and
sexual abuse live in households with an income in
the highest tertile.
Proportion of each violence and abuse group living in a
household in the lowest income tertile
48
42
31
32
27
23
%
Little
violence/abuse
Physical from
partner
Extensive
physical/coercion,
partner
Sexual: only as
child
Sexual: adult,
sometimes child
Extensive
physical, sexual
as child/adult
Health, wellbeing and disability
Violence and abuse were strongly and consistently
associated with poor health and disability. People in
the two groups characterized by extensive violence
and abuse were most likely to describe their health
as ‘fair’ or ‘poor’. People in all the violence and abuse
groups were less likely to regard themselves as
‘happy’ compared with people in the ‘little violence
or abuse’ group. Most violence and abuse groups
were associated with people having difficulty with
one or more ‘activities of daily living’. (ADLs provide
an indication of disability). People in the five groups
characterized by experience of abuse and violence
were also more likely to be providing care for
someone else due to the other person’s ill-health
or disability.
8
04 Profiles of the six groups
Proportion of each violence and abuse group that needs help
with multiple activities of daily living (2+ ADLs)
52
%
17
Little
violence/abuse
20
Physical from
partner
23
22
16
Extensive
Sexual: only as
physical/coercion,
child
partner
Sexual: adult,
sometimes child
Extensive
physical, sexual
as child/adult
Common mental disorders
Violence and abuse are strongly associated with
common mental disorders (CMD) such as anxiety
or depression. People in the ‘extensive physical and
sexual’ violence group were five times more likely
than those with little experience of violence to have
a CMD. More than half of this group met the
threshold for a CMD.
This same pattern held true for individual disorders
(such as phobias, depressive disorder and
generalized anxiety disorder). Phobias were present
in 10% of people in the extensive physical violence
group and 14% of people in the extensive physical
and sexual group. This compared with a prevalence
of 1% among people with few such experiences.
Other mental disorders
Violence and abuse are not just predictive of common
mental disorders. A wide range of different mental
disorders, including screening positive for psychosis,
post-traumatic stress disorder (PTSD) and eating
disorders, showed strong and consistent associations
with such experiences.
There was also a strong link with having more than
one disorder. People in the ‘extensive physical and
sexual’ group were about 15 times more likely than
those with little experience of violence and abuse to
have three or more disorders present at the time of
interview.
9
04 Profiles of the six groups
Proportion of each violence and abuse group with 3 or more
mental disorders
17
14
10
%
5
5
1
Little
violence/abuse
Physical from
partner
Extensive
Sexual: only as
physical/coercion,
child
partner
Sexual: adult,
sometimes child
Extensive
physical, sexual
as child/adult
Suicide and self-harm
There is a strong association between experience of
violence and abuse and suicidal behavior. People in
all the groups characterised by violence and abuse
were at least five times more likely than those with
little experience to have attempted to take their own
life. People in the ‘extensive physical and sexual
group’ were fifteen times more likely to have done so,
and 4% had made an attempt in the last year. 56% of
people in this group had self-harmed at some time
– compared to 10% of those with little experience of
violence and abuse.
Proportion of each violence and abuse group who have
attempted suicide
29
%
10
12
10
10
2
Little
violence/abuse
Physical from
partner
Extensive
Sexual: only as
physical/coercion,
child
partner
Sexual: adult,
Extensive
sometimes child physical, sexual
as child/adult
Mental health treatment and service use
Use of community mental health services was not
as common as might be expected given the rates of
poor mental health in the violence and abuse groups.
Only 10% of people in the ‘extensive physical and
sexual’ group were currently in receipt of a talking
therapy, despite more than half this group having a
CMD and 16% screening positive for posttraumatic
stress disorder. However, 12% of people in the
extensive physical and sexual group had been
admitted to a unit specialising in mental health. Such
high rates of use of secondary health care services
may indicate a failure of primary and community
care providers to meet the needs of people with
experience of violence and abuse.
10
04 Profiles of the six groups
Proportion of each violence and abuse group admitted to a
ward specialising in mental health
12
%
4
4
3
3
1
Little
violence/abuse
Physical from
partner
Extensive
Sexual: only as
physical/coercion,
child
partner
Sexual: adult,
Extensive
sometimes child physical, sexual
as child/adult
Health risk behaviours
Violence and abuse are strongly associated with
a range of health risk behaviours. People in the
two groups characterized by extensive violence
and abuse were more than twice as likely to be
dependent on illegal drugs and to be smokers as
those with little such experience.
Alcohol dependence was also more common among
people in the two ‘extensive violence’ groups. 38% of
people in the ‘extensive physical and sexual’ group
had a problematic pattern of alcohol consumption.
Proportion of each violence and abuse group who are regular
smokers
48
45
29
%
24
20
Little
violence/abuse
19
Physical from
partner
Extensive
Sexual: only as
physical/coercion,
child
partner
Sexual: adult,
Extensive
sometimes child physical, sexual
as child/adult
Experience in the early years and parenting
People in the two groups characterised by extensive
violence and abuse were less likely to have lived with
both natural parents to the age of 16. They were
more likely to have spent time in an institution or
local authority care. We do not know whether
the childhood experiences of violence and abuse
preceded going into care or happened while they
were in care.
People in the groups characterised by physical
violence were more likely to have children of their
own (compared with those with little experience or
in a group characterised by sexual violence only).
For full tables of results and details of the methodology
used, see the full report of the statistical analysis:
www.natcen.ac.uk/REVAstrand1report
11
This is the first of five briefings based
on the REVA study:
• Violence, abuse and mental health in England
(REVA Briefing 1)
www.natcen.ac.uk/revabriefing1
• Guidance for Trust managers: Implementing
and sustaining routine enquiryabout violence
and abuse in mental health services
(REVA Briefing 2)
www.natcen.ac.uk/revabriefing2
• A briefing for mental health professionals: Why
asking about abuse matters to service users
(REVA Briefing 3)
www.natcen.ac.uk/revabriefing3
• A briefing for commissioners: What survivors
of violence and abuse say about mental health
services (REVA Briefing 4).
www.natcen.ac.uk/revabriefing4
• A briefing for service providers and
commissioners: Measuring outcomes for
survivors of violence and abuse
(REVA Briefing 5).
www.natcen.ac.uk/revabriefing5
This briefing reports independent research
commissioned and funded by the Department
of Health Policy Research Programme (Effective
Responses to Long-Term Consequences of
Violence, Trauma and Abuse, 115/0005). The
views expressed in this briefing are those of
the author(s) and not necessarily those of the
Department of Health.
12
The REVA research was conducted
by the following team:
Dr Sara Scott and Dr Jennie Williams
DMSS Research
Dr Carol McNaughton Nicholls
Truth Consulting
Sally McManus, Ashley Brown
and Shannon Harvey
NatCen
Prof Liz Kelly and Joanne Lovett
CWASU, London Metropolitan University