Media Analysis

BackgroundBackground Since de-institutionalis-Since de-institutionalis-

ation, muchhasbeenwritten abouttheation, muchhas beenwritten aboutthe

riskposedtothe communityby thosewithriskposedtothe communityby thosewith

severemental illness.However, violentseveremental illness.However, violent

victimisation of peoplewithmentalvictimisation of peoplewithmental

illnesses has received little attention.illnesses has received little attention.

AimsAims To establishthe1-year prevalenceTo establishthe1-year prevalence

of violent victimisation in community-of violent victimisation in community-

dwellingpatientswith psychosis and todwellingpatientswith psychosis and to

identify the socio-demographic andidentify the socio-demographic and

clinical correlates of violent victimisation.clinical correlates of violent victimisation.

MethodMethod Atotal of 691subjectswithAtotal of 691subjectswith

establishedpsychotic disorderswereestablishedpsychotic disorderswere

interviewed.The past-year prevalence ofinterviewed.The past-year prevalence of

violent victimisationwas estimated andviolent victimisationwas estimated and

comparedwithgeneralpopulationfigures.comparedwithgeneralpopulationfigures.

Thosewho reported being violentlyThosewho reported being violently

victimisedwere comparedwiththosewhovictimisedwerecomparedwiththosewho

didnoton a range of social and clinicaldidnoton a range of social and clinical

characteristics.characteristics.

ResultsResults Sixteenper centof patientsSixteenpercentof patients

reported being violently victimised.reportedbeing violently victimised.

Victimsof violencewere significantlymoreVictimsof violencewere significantlymore

likely to report severe psychopathologicallikely to report severe psychopathological

symptoms, homelessness, substancesymptoms, homelessness, substance

misuse andprevious violentbehaviour andmisuse andpreviousviolentbehaviour and

weremore likely to have a comorbidweremore likely to have a comorbid

personalitydisorder.personalitydisorder.

ConclusionsConclusions Thosewith psychosis areThosewith psychosis are

atconsiderable riskof violent victimisationatconsiderable riskof violent victimisation

inthe community. Victimisationinthe community. Victimisation

experience should be recorded intheexperience should be recorded inthe

standardpsychiatric interview.standardpsychiatric interview.

Declaration of interestDeclaration of interest None.None.

Since de-institutionalisation, much has beenSince de-institutionalisation, much has been

written about the risk posed to members ofwritten about the risk posed to members of

the public by those with severe mentalthe public by those with severe mental

illness (Mullenillness (Mullen et alet al, 2000). The overall evi-, 2000). The overall evi-

dence, however, is that the contributiondence, however, is that the contribution

made by those with psychosis to violentmade by those with psychosis to violent

crime in society is small and is accountedcrime in society is small and is accounted

for by a small minority of patients (Walshfor by a small minority of patients (Walsh

et alet al, 2002). Conversely, little attention, 2002). Conversely, little attention

generally is paid to the risk posed to thisgenerally is paid to the risk posed to this

vulnerable group of people. Cross-sectionalvulnerable group of people. Cross-sectional

surveys have reported the prevalence ofsurveys have reported the prevalence of

criminal victimisation to be high amongcriminal victimisation to be high among

mentally ill people (Hidaymentally ill people (Hiday et alet al, 1999;, 1999;

BrekkeBrekke et alet al, 2001; Hiday, 2001; Hiday et alet al, 2002)., 2002).

The only case–control study to date hasThe only case–control study to date has

found those with severe mental illness atfound those with severe mental illness at

significantly increased risk of violent victi-significantly increased risk of violent victi-

misation compared with neighbourhoodmisation compared with neighbourhood

controls, after controlling for socio-controls, after controlling for socio-

economic disadvantage and the individual’seconomic disadvantage and the individual’s

own violent behaviour (Silver, 2002).own violent behaviour (Silver, 2002).

BackgroundBackground

Criminal victimisation of those with severeCriminal victimisation of those with severe

mental illness has been associated withmental illness has been associated with

more severe clinical symptoms (Brekkemore severe clinical symptoms (Brekke etet

alal, 2001; Hiday, 2001; Hiday et alet al, 2002), substance mis-, 2002), substance mis-

use (Hidayuse (Hiday et alet al, 1999; Brekke, 1999; Brekke et alet al, 2001),, 2001),

transient living conditions (including home-transient living conditions (including home-

lessness) (Hidaylessness) (Hiday et alet al, 1999), lower func-, 1999), lower func-

tioning, lack of social support and ationing, lack of social support and a

history of previous victimisation (Hidayhistory of previous victimisation (Hiday etet

alal, 2002). However, most studies examin-, 2002). However, most studies examin-

ing associated factors have failed to distin-ing associated factors have failed to distin-

guish between being the victim of aguish between being the victim of a

violent or a non-violent crime. Only oneviolent or a non-violent crime. Only one

study to date has examined socio-study to date has examined socio-

demographic and clinical correlates ofdemographic and clinical correlates of

violent victimisation separately; this studyviolent victimisation separately; this study

found that one-third of patients dischargedfound that one-third of patients discharged

from psychiatric hospitals and living infrom psychiatric hospitals and living in

hostels had been the victims of crime inhostels had been the victims of crime in

the preceding year. Victims of violencethe preceding year. Victims of violence

were younger, more socially active,were younger, more socially active,

reported more psychopathology and lessreported more psychopathology and less

satisfaction with their lives and engaged insatisfaction with their lives and engaged in

more criminal behaviour than both non-more criminal behaviour than both non-

victims and the victims of non-violent crimevictims and the victims of non-violent crime

(Lehman & Linn, 1984).(Lehman & Linn, 1984).

The aims of the present study are two-The aims of the present study are two-

fold: to establish the 1-year prevalence offold: to establish the 1-year prevalence of

violent victimisation in community-violent victimisation in community-

dwelling patients with psychosis and todwelling patients with psychosis and to

compare this with the official statisticscompare this with the official statistics

concerning prevalence in the generalconcerning prevalence in the general

population; and to examine the socio-population; and to examine the socio-

demographic and clinical correlates ofdemographic and clinical correlates of

violent victimisation in the largest sampleviolent victimisation in the largest sample

of patients with psychosis to date.of patients with psychosis to date.

METHODMETHOD

SubjectsSubjects

A total of 708 subjects were recruited fromA total of 708 subjects were recruited from

four inner-city areas in England as part offour inner-city areas in England as part of

the UK700 case management trial (UK700the UK700 case management trial (UK700

Group, 1999). Subjects were identified byGroup, 1999). Subjects were identified by

systematic review of in-patient and out-systematic review of in-patient and out-

patient registers and fulfilled the followingpatient registers and fulfilled the following

inclusion/exclusion criteria:inclusion/exclusion criteria:

(a)(a) aged 18–65 years;aged 18–65 years;

(b)(b) a diagnosis of psychosis: defined as thea diagnosis of psychosis: defined as the

presence, according to Research Diag-presence, according to Research Diag-

nostic Criteria (RDC; Spitzernostic Criteria (RDC; Spitzer et alet al,,

1978), of delusions, hallucinations or1978), of delusions, hallucinations or

thought disorder;thought disorder;

(c)(c) hospitalised for psychotic symptomshospitalised for psychotic symptoms

at least twice, with the most recentat least twice, with the most recent

admission within the past 2 years;admission within the past 2 years;

(d)(d) absence of a primary diagnosis ofabsence of a primary diagnosis of

substance misuse;substance misuse;

(e)(e) absence of organic brain damage.absence of organic brain damage.

In this way, we set out to collect a sample ofIn this way, we set out to collect a sample of

patients with established illness typical ofpatients with established illness typical of

those receiving multi-disciplinary psychiatricthose receiving multi-disciplinary psychiatric

care in the community.care in the community.

Data collectionData collection

All subjects were interviewed between 1994All subjects were interviewed between 1994

and 1996 using a battery of instruments;and 1996 using a battery of instruments;

these baseline assessments provided thethese baseline assessments provided the

data that were analysed for the purposesdata that were analysed for the purposes

of this study. The interviewers were eitherof this study. The interviewers were either

senior trainee psychiatrists or psychologysenior trainee psychiatrists or psychology

graduates, all of whom participated in angraduates, all of whom participated in an

initial 2-day training course andinitial 2-day training course and

completed five pilot interview schedules.completed five pilot interview schedules.

Training materials included lectures, jointTraining materials included lectures, joint

patient interviews, case vignettes andpatient interviews, case vignettes and

video interviews. Completed interviewvideo interviews. Completed interview

schedules were inspected regularly on siteschedules were inspected regularly on site

for errors and inconsistencies (UK700for errors and inconsistencies (UK700

Group, 1999).Group, 1999).

2 3 32 3 3

BR I T I SH JOURNAL OF P SYCHIATRYBR IT I SH JOURNAL OF P SYCHIATRY ( 2 0 0 3 ) , 1 8 3 , 2 3 3 ^ 2 3 8( 2 0 0 3 ) , 1 8 3 , 2 3 3 ^ 2 3 8

Prevalence of violent victimisation in severePrevalence of violent victimisation in severe

mental illnessmental illness

ELIZABETH WALSH, PAUL MORAN, CHARLES SCOTT, KWAME McKENZIE,ELIZABETH WALSH, PAUL MORAN, CHARLES SCOTT, KWAME McKENZIE, TOM BURNS, FRANCIS CREED, PETER TYRER,* ROBIN. M. MURRAY andTOM BURNS, FRANCIS CREED, PETER TYRER,* ROBIN. M. MURRAY and TOM FAHYon behalf of the UK700 GROUPTOM FAHYon behalf of the UK700 GROUP

*This paper was accepted before the appointmentof P.T.*This paper was accepted before the appointmentof P.T. as Editor of theas Editor of the JournalJournal..

WALSH ET ALWALSH ET AL

Outcome variablesOutcome variables

The primary outcome of interest wasThe primary outcome of interest was

violent victimisation in the year prior toviolent victimisation in the year prior to

interview. As part of the Lancashireinterview. As part of the Lancashire

Quality of Life Profile (OliverQuality of Life Profile (Oliver, 1991), sub-, 1991), sub-

jects were asked the following question:jects were asked the following question:

‘In the last year have you been assaulted,‘In the last year have you been assaulted,

beaten, molested or otherwise the victimbeaten, molested or otherwise the victim

of violence?’ Respondents answered ‘yes’of violence?’ Respondents answered ‘yes’

or ‘no’ to this question. Perceived vulner-or ‘no’ to this question. Perceived vulner-

ability to victimisation was measured byability to victimisation was measured by

asking how satisfied subjects were withasking how satisfied subjects were with

their own personal safety and the safety oftheir own personal safety and the safety of

their neighbourhood. Responses, scoredtheir neighbourhood. Responses, scored

on a seven-point Likert scale, were cate-on a seven-point Likert scale, were cate-

gorised into a binary outcome of satisfied/gorised into a binary outcome of satisfied/

dissatisfied.dissatisfied.

Explanatory variablesExplanatory variables

Possible correlates of violent victimisation,Possible correlates of violent victimisation,

chosenchosen a prioria priori on the basis of previouson the basis of previous

research, were measured using the follow-research, were measured using the follow-

ing instruments.ing instruments.

(a)(a) UK700 Socio-demographic Schedule:UK700 Socio-demographic Schedule:

age, gender, ethnicity (interviewer-age, gender, ethnicity (interviewer-

assigned as White, African–Caribbeanassigned as White, African–Caribbean

or Other), marital status, social classor Other), marital status, social class

(by occupation of father at birth),(by occupation of father at birth),

educational achievement, age at onseteducational achievement, age at onset

and at first admission for psychosisand at first admission for psychosis

and length of illness.and length of illness.

(b)(b) Comprehensive PsychopathologicalComprehensive Psychopathological

Rating Scale (CPRS; AsbergRating Scale (CPRS; Åsberg et alet al,,

1978): measures the reported and1978): measures the reported and

observed psychopathology in theobserved psychopathology in the

previous week.previous week.

(c)(c) Scale for the Assessment of NegativeScale for the Assessment of Negative

Symptoms (SANS; Andreasen, 1984):Symptoms (SANS; Andreasen, 1984):

an observer-rated measure of negativean observer-rated measure of negative

symptoms.symptoms.

(d)(d) Disability Assessment Scale (DAS;Disability Assessment Scale (DAS;

JablenskyJablensky et alet al, 1980): measures the, 1980): measures the

level of disability.level of disability.

(e)(e) Camberwell Assessment of Need –Camberwell Assessment of Need –

Research version (CAN–R; PhelanResearch version (CAN–R; Phelan etet

alal, 1995): a measure of the total, 1995): a measure of the total

number of ‘unmet’ needs for care.number of ‘unmet’ needs for care.

(f)(f) WHO Life Chart (World HealthWHO Life Chart (World Health

Organization, 1992Organization, 1992aa): information): information

collected relates to the previous 2collected relates to the previous 2

years and includes homelessness,years and includes homelessness,

number of psychiatric admissions,number of psychiatric admissions,

months living independently andmonths living independently and

physical assault (self-report).physical assault (self-report).

(g)(g) Substance Misuse Questionnaire: sub-Substance Misuse Questionnaire: sub-

jects were asked about their use ofjects were asked about their use of

alcohol and illegal drugs in the previousalcohol and illegal drugs in the previous

year; variables collected includedyear; variables collected included

alcohol misuse (greater than 2 units ofalcohol misuse (greater than 2 units of

alcohol per day for women and 3alcohol per day for women and 3

units per day for men) and illegal drugunits per day for men) and illegal drug

use (coded as none, at least one anduse (coded as none, at least one and

more than one).more than one).

(h)(h) Operational Criteria Checklist for Psy-Operational Criteria Checklist for Psy-

chotic Illness (OCCPI; McGuffinchotic Illness (OCCPI; McGuffin et alet al,,

1991): used to generate RDC diagnoses1991): used to generate RDC diagnoses

from case notes.from case notes.

(i)(i) Personality Assessment Schedule, RapidPersonality Assessment Schedule, Rapid

version (PAS–R; van Hornversion (PAS–R; van Horn et alet al, 2000):, 2000):

derived from the original PAS schedulederived from the original PAS schedule

(Tyrer(Tyrer et alet al, 1979), the PAS–R is a, 1979), the PAS–R is a

direct development of the PAS anddirect development of the PAS and

allows for a rapid screen for the presenceallows for a rapid screen for the presence

of ICD–10 (World Health Organiza-of ICD–10 (World Health Organiza-

tion, 1992tion, 1992bb) personality disorder.) personality disorder.

Scoring for each category of personalityScoring for each category of personality

disorder on the PAS–R is on a three-disorder on the PAS–R is on a three-

point scale from 0 to 2 (0, absence ofpoint scale from 0 to 2 (0, absence of

any dysfunction associated with theany dysfunction associated with the

personality trait; 1, personality diffi-personality trait; 1, personality diffi-

culty; 2, personality disorder). For theculty; 2, personality disorder). For the

purposes of this study, the PAS–R datapurposes of this study, the PAS–R data

were regrouped into a dichotomouswere regrouped into a dichotomous

variable with two categories: person-variable with two categories: person-

ality disorder and no personalityality disorder and no personality

disorder. (Personality disorder wasdisorder. (Personality disorder was

defined as a PAS–R score of 2 on anydefined as a PAS–R score of 2 on any

personality disorder category.)personality disorder category.)

(j)(j) Mental Illness Needs Index (MINI;Mental Illness Needs Index (MINI;

GloverGlover et alet al, 1998): this is based on, 1998): this is based on

postcode, designed to predict thepostcode, designed to predict the

number of people likely to receive in-number of people likely to receive in-

patient care in a defined area, derivedpatient care in a defined area, derived

from socio-demographic variablesfrom socio-demographic variables

(social isolation, poverty, unemploy-(social isolation, poverty, unemploy-

ment, permanent sickness andment, permanent sickness and

temporary and insecure housing); it istemporary and insecure housing); it is

used to adjust for deprivation of areaused to adjust for deprivation of area

of residence.of residence.

(k)(k) Lancashire Quality of Life ProfileLancashire Quality of Life Profile

(LQoLP; Oliver, 1991): apart from the(LQoLP; Oliver, 1991): apart from the

primary outcome, this questionnaireprimary outcome, this questionnaire

also measured intensity of familyalso measured intensity of family

contact.contact.

(l)(l) Offenders Index: a computerised data-Offenders Index: a computerised data-

base that holds criminal history database that holds criminal history data

for more than six million offendersfor more than six million offenders

since 1963 in England and Wales. Forsince 1963 in England and Wales. For

the purposes of this study, criminalthe purposes of this study, criminal

convictions were divided into violentconvictions were divided into violent

and non-violent. The offence categoriesand non-violent. The offence categories

considered to constitute violence andconsidered to constitute violence and

listed under ‘violence against thelisted under ‘violence against the

person’ were murder, attemptedperson’ were murder, attempted

murder, threat or conspiracy tomurder, threat or conspiracy to

murder, wounding or other act endan-murder, wounding or other act endan-

gering life, assault, common assault,gering life, assault, common assault,

intimidation and molestation, andintimidation and molestation, and

violent disorder. Non-violent offencesviolent disorder. Non-violent offences

included all other recorded offences.included all other recorded offences.

Official criminal records were soughtOfficial criminal records were sought

for all subjects, who were subsequentlyfor all subjects, who were subsequently

coded as being either positive or nega-coded as being either positive or nega-

tive for violent and non-violenttive for violent and non-violent

offences.offences.

Statistical analysisStatistical analysis

The proportion of subjects reportingThe proportion of subjects reporting

violent victimisation and perceived threatviolent victimisation and perceived threat

of victimisation was estimated. Possibleof victimisation was estimated. Possible

socio-demographic and clinical correlatessocio-demographic and clinical correlates

of violent victimisation were examinedof violent victimisation were examined

using logistic regression. Initially, theusing logistic regression. Initially, the

association between violent victimisationassociation between violent victimisation

and each explanatory variable was examin-and each explanatory variable was examin-

ed unadjusted for other variables. All vari-ed unadjusted for other variables. All vari-

ables in the univariate analysis significantables in the univariate analysis significant

atat PP¼0.05 were then entered into a multi-0.05 were then entered into a multi-

variate model and stepwise methods werevariate model and stepwise methods were

used to identify the final model bestused to identify the final model best

associated with violent victimisation. Allassociated with violent victimisation. All

other variables then were added to this finalother variables then were added to this final

model to check that no significant corre-model to check that no significant corre-

lates were missed. The final model waslates were missed. The final model was

adjusted for age and gender. All analysesadjusted for age and gender. All analyses

were conducted using STATA 6.0 (Stata-were conducted using STATA 6.0 (Stata-

Corp, 1999).Corp, 1999).

RESULTSRESULTS

RecruitmentRecruitment

Of those approached, 80% (708/892)Of those approached, 80% (708/892)

agreed to participate, 13% refused and aagreed to participate, 13% refused and a

further 7% were not interviewed for afurther 7% were not interviewed for a

variety of reasons, including inability tovariety of reasons, including inability to

establish contact. No significant differencesestablish contact. No significant differences

were found between participants and non-were found between participants and non-

participants in terms of basic demographicparticipants in terms of basic demographic

and clinical characteristics. However,and clinical characteristics. However,

compared with non-participants, patientscompared with non-participants, patients

who entered the trial had been ill for longerwho entered the trial had been ill for longer

(median duration 120 months(median duration 120 months v.v. 9696

months;months; UU¼51899.0;51899.0; PP¼0.04).0.04).

The demographic and clinical charac-The demographic and clinical charac-

teristics of the UK700 study profile haveteristics of the UK700 study profile have

been described elsewhere (Burnsbeen described elsewhere (Burns et alet al,,

1999). Tables 1 and 2 list these according1999). Tables 1 and 2 list these according

to the victim profile. In the sample, moreto the victim profile. In the sample, more

than half of the patients were young menthan half of the patients were young men

with long histories of illness (median ofwith long histories of illness (median of

10 years; median of 2 months in hospital10 years; median of 2 months in hospital

in the preceding 2 years). Most were diag-in the preceding 2 years). Most were diag-

nosed with schizoaffective disorder andnosed with schizoaffective disorder and

schizophrenia. Nearly one-third ofschizophrenia. Nearly one-third of

patients were African–Caribbean. Meanpatients were African–Caribbean. Mean

CPRS and DAS scores indicated thatCPRS and DAS scores indicated that

patients were moderately to severely ill.patients were moderately to severely ill.

2 3 42 3 4

VIOLENT VICTIMISATION IN SEVERE MENTAL ILLNES SVIOLENT VICTIMISATION IN SEVERE MENTAL ILLNES S

Prevalence of violent victimisationPrevalence of violent victimisation Information on victimisation was availableInformation on victimisation was available

for 691 (98%) of the subjects at interview;for 691 (98%) of the subjects at interview;

111 (16%) reported being a victim of vio-111 (16%) reported being a victim of vio-

lence in the previous year. With regard tolence in the previous year. With regard to

perceived threat, 269/678 (40%) were dis-perceived threat, 269/678 (40%) were dis-

satisfied with their personal safety andsatisfied with their personal safety and

301/677 (44%) were dissatisfied with the301/677 (44%) were dissatisfied with the

safety of their neighbourhood. Victimssafety of their neighbourhood. Victims

were significantly more likely to report feel-were significantly more likely to report feel-

ing personally unsafe (ing personally unsafe (nn¼65, 58%;65, 58%;

PP550.001) and unsafe in their neighbour-0.001) and unsafe in their neighbour-

hood (hood (nn¼66, 59%;66, 59%; PP550.001) than non-0.001) than non-

victims. Our interviews were conductedvictims. Our interviews were conducted

between 1994 and 1996. For comparison,between 1994 and 1996. For comparison,

crime figures collected at that time for thecrime figures collected at that time for the

British Crime Survey reveal an annualBritish Crime Survey reveal an annual

percentage of victimisation for contactpercentage of victimisation for contact

crime of 6.7% in London and 7.1% forcrime of 6.7% in London and 7.1% for

all inner cities. The figure for non-inner-cityall inner cities. The figure for non-inner-city

areas was 4.9% (Murless-Blackareas was 4.9% (Murless-Black et alet al,,

1996).1996).

Characteristics of victims:Characteristics of victims: univariate analysisunivariate analysis

The socio-demographic characteristics ofThe socio-demographic characteristics of

subjects reporting victimisation are pre-subjects reporting victimisation are pre-

sented in Table 1. Compared with non-sented in Table 1. Compared with non-

victims, victims were significantly morevictims, victims were significantly more

likely to be male, under 40 years and withlikely to be male, under 40 years and with

transient living conditions, including home-transient living conditions, including home-

lessness. Victims were less likely to havelessness. Victims were less likely to have

daily contact with their families and spentdaily contact with their families and spent

less time in independent accommodationless time in independent accommodation

in the community compared with non-in the community compared with non-

victims. Victims were more likely to havevictims. Victims were more likely to have

had contact with the law, with significantlyhad contact with the law, with significantly

more criminal convictions for violent andmore criminal convictions for violent and

non-violent crime and more recent self-non-violent crime and more recent self-

reported violent behaviour. There was noreported violent behaviour. There was no

significant association between being asignificant association between being a

victim of violence and being a member ofvictim of violence and being a member of

an ethnic minority group, recent employ-an ethnic minority group, recent employ-

ment or degree of deprivation of area ofment or degree of deprivation of area of

residence.residence.

The clinical characteristics of subjectsThe clinical characteristics of subjects

reporting victimisation are presented inreporting victimisation are presented in

Table 2. Although those with early illnessTable 2. Although those with early illness

onset, higher scores on general psycho-onset, higher scores on general psycho-

pathology and more unmet needs for carepathology and more unmet needs for care

were more likely to be victims, the lengthwere more likely to be victims, the length

of illness, level of negative symptoms andof illness, level of negative symptoms and

disability were not associated with victimdisability were not associated with victim

status. Compared with non-victims, victimsstatus. Compared with non-victims, victims

also were more likely to have a comorbidalso were more likely to have a comorbid

personality disorder. With regard to sub-personality disorder. With regard to sub-

stance misuse, victims used significantlystance misuse, victims used significantly

more illegal drugs but were not more likelymore illegal drugs but were not more likely

to misuse alcohol. Those using one illegalto misuse alcohol. Those using one illegal

drug were almost two and a half timesdrug were almost two and a half times

more likely to be victimised and those usingmore likely to be victimised and those using

two or more such drugs were over fourtwo or more such drugs were over four

times more likely to be victims of violencetimes more likely to be victims of violence

than those denying any use.than those denying any use.

Assessment of independent effectsAssessment of independent effects using multivariate analysisusing multivariate analysis

Table 3 presents the final multivariateTable 3 presents the final multivariate

model identifying the associations betweenmodel identifying the associations between

each variable and victim status, adjustedeach variable and victim status, adjusted

for age, gender and each other. Beingfor age, gender and each other. Being

homeless (homeless (PP¼0.01), using illegal drugs0.01), using illegal drugs

((PP550.001), being the perpetrator of an0.001), being the perpetrator of an

assault (assault (PP¼0.01), having greater current0.01), having greater current

symptomatology (symptomatology (PP¼0.02) and a comorbid0.02) and a comorbid

2 3 52 3 5

Table1Table1 Socio-demographic characteristics of the UK700 sample, by victim statusSocio-demographic characteristics of the UK700 sample, by victim status

VariableVariable VictimVictim Odds ratio (95%Odds ratio (95%

YesYes

((nn¼111)111)

NoNo

((nn¼580)580)

CI, unadjusted)CI, unadjusted)

Age,Age, nn (%)(%)

18^39 years18^39 years 84 (20%)84 (20%) 338 (80%)338 (80%) 2.32 (1.45^3.71)***2.32 (1.45^3.71)***

40^64 years40^64 years 26 (10%)26 (10%) 243 (90%)243 (90%) 11

Gender,Gender, nn (%)(%)

FemaleFemale 37 (12%)37 (12%) 257 (88%)257 (88%) 11

MaleMale 73 (18%)73 (18%) 324 (82%)324 (82%) 1.56 (1.02^2.40)*1.56 (1.02^2.40)*

Ethnicity,Ethnicity, nn (%)(%)

WhiteWhite 61 (17%)61 (17%) 297 (83%)297 (83%) 11

African^CaribbeanAfrican^Caribbean 30 (16%)30 (16%) 161 (84%)161 (84%) 0.68 (0.56^1.46)0.68 (0.56^1.46)

OtherOther 19 (13%)19 (13%) 123 (13%)123 (13%) 0.31 (0.43^1.31)0.31 (0.43^1.31)

Marital status,Marital status, nn (%)(%)

Ever marriedEvermarried 32 (14%)32 (14%) 205 (86%)205 (86%) 11

SingleSingle 78 (17%)78 (17%) 376 (83%)376 (83%) 1.32 (0.85^2.07)1.32 (0.85^2.07)

Occupation of father at birth,Occupation of father at birth, nn (%)(%)

Non-manualNon-manual 15 (18%)15 (18%) 69 (82%)69 (82%) 11

ManualManual 52 (16%)52 (16%) 282 (84%)282 (84%) 0.84 (0.45^1.59)0.84 (0.45^1.59)

UnemployedUnemployed 22 (15%)22 (15%) 127 (85%)127 (85%) 0.79 (0.38^1.63)0.79 (0.38^1.63)

Employed (past 2 years),Employed (past 2 years), nn (%)(%)

NoNo 88 (16%)88 (16%) 456 (84%)456 (84%) 11

YesYes 22 (15%)22 (15%) 125 (85%)125 (85%) 0.72 (0.54^1.51)0.72 (0.54^1.51)

Homeless (past 2 years),Homeless (past 2 years), nn (%)(%)

NoNo 96 (15%)96 (15%) 555 (82%)555 (82%) 11

YesYes 14 (35%)14 (35%) 26 (65%)26 (65%) 3.12 (1.56^6.17)***3.12 (1.56^6.17)***

Family contact (past year),Family contact (past year), nn (%)(%)

DailyDaily 31 (12%)31 (12%) 234 (88%)234 (88%) 11

Less than dailyLess than daily 76 (18%)76 (18%) 342 (82%)342 (82%) 1.67 (1.07^2.62)*1.67 (1.07^2.62)*

Independent living, mean (s.d.)Independent living, mean (s.d.) 15 (8)15 (8) 17 (8)17 (8) 0.97 (0.95^0.99)*0.97 (0.95^0.99)*

Assault (past 2 years),Assault (past 2 years), nn (%)(%)

NoNo 76 (13%)76 (13%) 493 (87%)493 (87%) 11

YesYes 33 (28%)33 (28%) 84 (72%)84 (72%) 2.54 (1.59^4.07)***2.54 (1.59^4.07)***

Violent conviction,Violent conviction, nn (%)(%)

NoNo 78 (14%)78 (14%) 487 (86%)487 (86%) 11

YesYes 32 (25%)32 (25%) 94 (75%)94 (75%) 2.13 (1.33^3.39)**2.13 (1.33^3.39)**

Non-violent conviction,Non-violent conviction, nn (%)(%)

NoNo 62 (13%)62 (13%) 401 (87%)401 (87%) 11

YesYes 48 (21%)48 (21%) 180 (79%)180 (79%) 1.72 (1.13^2.61)**1.72 (1.13^2.61)**

MINI score, mean (s.d.)MINI score, mean (s.d.) 442 (72)442 (72) 440 (63)440 (63) 1.00 (0.99^1.00)1.00 (0.99^1.00)

MINI,Mental Illness Needs Index.MINI,Mental Illness Needs Index. **PP550.05; **0.05; **PP550.01; ***0.01; ***PP550.001.0.001.

WALSH ET ALWALSH ET AL

personality disorder (personality disorder (PP¼0.006) were all0.006) were all

independently associated with being aindependently associated with being a

victim of violence.victim of violence.

DISCUSSIONDISCUSSION

Sixteen per cent of 691 patients living in theSixteen per cent of 691 patients living in the

community reported being the victim ofcommunity reported being the victim of

violence over 1 year, a figure that is moreviolence over 1 year, a figure that is more

than twice that recorded from generalthan twice that recorded from general

population figures in the UK during thepopulation figures in the UK during the

same period. Compared with non-victims,same period. Compared with non-victims,

victims of violence were significantly morevictims of violence were significantly more

likely to report more severe psychopatho-likely to report more severe psychopatho-

logical symptoms, to have been homeless,logical symptoms, to have been homeless,

to have misused drugs, to admit to havingto have misused drugs, to admit to having

assaulted another person and to suffer fromassaulted another person and to suffer from

a comorbid personality disorder.a comorbid personality disorder.

Strengths and weaknessesStrengths and weaknesses of the studyof the study This is the largest study to date to examineThis is the largest study to date to examine

the prevalence and correlates of violentthe prevalence and correlates of violent

victimisation in severe mental illness. Thevictimisation in severe mental illness. The

validity of our findings is increased by thevalidity of our findings is increased by the

use of operational definitions of psychosisuse of operational definitions of psychosis

and well-validated instruments based onand well-validated instruments based on

interview rather than records, compre-interview rather than records, compre-

hensive staff training and the availabilityhensive staff training and the availability

of additional sources of information, whichof additional sources of information, which

included case notes, information fromincluded case notes, information from

carers and clinical staff and officialcarers and clinical staff and official

criminal records. The participants werecriminal records. The participants were

recruited from four clinical centres andrecruited from four clinical centres and

were chosen to be representative of thosewere chosen to be representative of those

patients with chronic psychosis dwellingpatients with chronic psychosis dwelling

in the community and receiving care fromin the community and receiving care from

community mental health teams. Thecommunity mental health teams. The

choice of inner-city areas, with all theirchoice of inner-city areas, with all their

attendant problems, no doubt will haveattendant problems, no doubt will have

increased the prevalence of victimisationincreased the prevalence of victimisation

compared with rural samples (Hidaycompared with rural samples (Hiday et alet al,,

1999) and our results refer to urban rather1999) and our results refer to urban rather

than other areas. Owing to the cross-than other areas. Owing to the cross-

sectional nature of our data, we have beensectional nature of our data, we have been

able to examine only associations of violentable to examine only associations of violent

victimisation rather than predictive factors.victimisation rather than predictive factors.

We are therefore cautious about drawingWe are therefore cautious about drawing

inferences concerning causation based oninferences concerning causation based on

these data.these data.

The UK700 study did not employ aThe UK700 study did not employ a

general population or non-psychotic con-general population or non-psychotic con-

trol sample with whom we could comparetrol sample with whom we could compare

the prevalence of victimisation. We thusthe prevalence of victimisation. We thus

chose to rely on official records for compar-chose to rely on official records for compar-

ison, which were collected in a differentison, which were collected in a different

way and for different purposes. There isway and for different purposes. There is

evidence to suggest that individuals withevidence to suggest that individuals with

mental illnesses are more likely to be as-mental illnesses are more likely to be as-

saulted by people with whom they have asaulted by people with whom they have a

close relationship (Cascardiclose relationship (Cascardi et alet al, 1996). It, 1996). It

is therefore likely that victimisation willis therefore likely that victimisation will

be underreported for various reasons,be underreported for various reasons,

including protection of the perpetrator,including protection of the perpetrator,

shame and guilt, reluctance to discuss un-shame and guilt, reluctance to discuss un-

pleasant memories and fear of future vio-pleasant memories and fear of future vio-

lence. The comparative figures for thelence. The comparative figures for the

general population derive from anonymousgeneral population derive from anonymous

interviews with members of the public andinterviews with members of the public and

are therefore less susceptible to under-are therefore less susceptible to under-

reporting. Despite this, the difference inreporting. Despite this, the difference in

the prevalence of violent victimisation isthe prevalence of violent victimisation is

still impressive. We did not include non-still impressive. We did not include non-

violent victimisation, emotional abuse orviolent victimisation, emotional abuse or

social exploitation in our definition.social exploitation in our definition.

Prevalence of violent victimisationPrevalence of violent victimisation

Sixteen per cent of our subjects reportedSixteen per cent of our subjects reported

having been the victims of violence in thehaving been the victims of violence in the

previous year. Because information wasprevious year. Because information was

missing for 17 patients, the highest possiblemissing for 17 patients, the highest possible

2 3 62 3 6

Table 2Table 2 Clinical characteristics of the UK700 sample, by victim statusClinical characteristics of the UK700 sample, by victim status

Baseline risk factorBaseline risk factor VictimVictim Odds ratioOdds ratio

YesYes

((nn¼111)111)

NoNo

((nn¼580)580)

(95% CI)(95% CI)

Diagnosis,Diagnosis, nn (%)(%)

Schizoaffective disorderSchizoaffective disorder 54 (16%)54 (16%) 283 (84%)283 (84%) 11

SchizophreniaSchizophrenia 44 (17%)44 (17%) 221 (83%)221 (83%) 1.04 (0.67^1.61)1.04 (0.67^1.61)

Affective psychosisAffective psychosis 8 (17%)8 (17%) 40 (83%)40 (83%) 1.04 (0.46^2.36)1.04 (0.46^2.36)

Other psychosesOther psychoses 4 (10%)4 (10%) 37 (90%)37 (90%) 0.56 (0.19^1.65)0.56 (0.19^1.65)

Comorbid personality disorder,Comorbid personality disorder, nn (%)(%)

NoNo 57 (12%)57 (12%) 416 (88%)416 (88%) 11

YesYes 46 (25%)46 (25%) 137 (75%)137 (75%) 2.45 (1.58^3.78)2.45 (1.58^3.78)

Age at onset (years), mean (s.d.)Age at onset (years), mean (s.d.) 23 (7)23 (7) 26 (8)26 (8) 0.96 (0.93^0.99)**0.96 (0.93^0.99)**

Time ill (months), mean (s.d.)Time ill (months), mean (s.d.) 136 (110)136 (110) 151 (117)151 (117) 0.99 (0.99^1.00)0.99 (0.99^1.00)

CPRS total, mean (s.d.)CPRS total, mean (s.d.) 23 (14)23 (14) 18 (12)18 (12) 1.02 (1.01^1.04)**1.02 (1.01^1.04)**

SANS, mean (s.d.)SANS, mean (s.d.) 22 (16)22 (16) 21 (16)21 (16) 1.02 (0.99^1.01)1.02 (0.99^1.01)

DAS total, mean (s.d.)DAS total, mean (s.d.) 1.22 (0.78)1.22 (0.78) 1.13 (0.86)1.13 (0.86) 1.12 (0.88^1.41)1.12 (0.88^1.41)

Hospital admissions,Hospital admissions, nn (%)(%)

Less than twoLess than two 37 (12%)37 (12%) 416 (88%)416 (88%) 11

Two ormoreTwo ormore 73 (19%)73 (19%) 137 (75%)137 (75%) 1.87 (1.21^2.85)**1.87 (1.21^2.85)**

Unmet needs, mean (s.d.)Unmet needs, mean (s.d.) 3.15 (2.7)3.15 (2.7) 2.5 (2.3)2.5 (2.3) 1.11 (1.02^1.21)**1.11 (1.02^1.21)**

Drug use/misuse (past year),Drug use/misuse (past year), nn (%)(%)

NoneNone 65 (12%)65 (12%) 468 (88%)468 (88%) 11

OneOne 27 (24%)27 (24%) 83 (75%)83 (75%) 2.34 (1.41^3.88)2.34 (1.41^3.88)

Two ormoreTwo ormore 18 (38%)18 (38%) 30 (62%)30 (62%) 4.32 (2.27^8.18)**4.32 (2.27^8.18)**

Alcohol,Alcohol, nn (%)(%)

552^3units/day2^3units/day 93 (15%)93 (15%) 535 (85%)535 (85%) 11

442^3 units/day2^3 units/day 11 (24%)11 (24%) 35 (76%)35 (76%) 1.81 (0.88^3.68)1.81 (0.88^3.68)

CPRS,Comprehensive Psychopathological Rating Scale; DAS, Disability Assessment Scale; SANS, Scale for theCPRS,Comprehensive Psychopathological Rating Scale; DAS,Disability Assessment Scale; SANS, Scale for the Assessment of Negative Ssymptoms.Assessment of Negative Ssymptoms. ****PP550.01.0.01.

Table 3Table 3 Socio-demographic and clinical correlatesSocio-demographic and clinical correlates

of violent victimisationof violent victimisation

FactorFactor Odds ratioOdds ratio

(95% CI)(95% CI)11

Homeless (past 2 years)Homeless (past 2 years)

YesYes 2.67 (1.23^5.77)**2.67 (1.23^5.77)**

Assault (past 2 years)Assault (past 2 years)

YesYes 2.08 (1.18^3.43)**2.08 (1.18^3.43)**

Comorbid personality disorderComorbid personality disorder

YesYes 1.93 (1.20^3.10)**1.93 (1.20^3.10)**

CPRS total (mean)CPRS total (mean)

VictimVictim 1.02 (1.00^1.04)*1.02 (1.00^1.04)*

Drug use/misuse (past year)Drug use/misuse (past year)

NoneNone 11

OneOne 1.76 (1.01^3.09)1.76 (1.01^3.09)

Two ormoreTwo ormore 3.81 (1.87^7.77)***3.81 (1.87^7.77)***

CPRS,Comprehensive Psychopathological Rating Scale.CPRS,Comprehensive Psychopathological Rating Scale. 1. Adjusted for age, gender and all other variables in1. Adjusted for age, gender and all other variables in table.table. **PP550.05; **0.05; **PP550.01; ***0.01; ***PP550.001.0.001.

VIOLENT VICTIMISATION IN SEVERE MENTAL ILLNES SVIOLENT VICTIMISATION IN SEVERE MENTAL ILLNES S

prevalence for victimisation in the sampleprevalence for victimisation in the sample

was 18%, assuming that all those missingwas 18%, assuming that all those missing

had been victimised, and the lowest preva-had been victimised, and the lowest preva-

lence was 16%, assuming that they hadlence was 16%, assuming that they had

not. This gives a prevalence range of 16–not. This gives a prevalence range of 16–

18%, a figure more than twice that re-18%, a figure more than twice that re-

corded in the general population at thatcorded in the general population at that

time, according to the British Crime Survey.time, according to the British Crime Survey.

This prevalence figure is higher than thatThis prevalence figure is higher than that

reported in the USA. Hidayreported in the USA. Hiday et alet al (2002)(2002)

reported that 10% of persons with severereported that 10% of persons with severe

mental illness who had been deemed suit-mental illness who had been deemed suit-

able for enforced community treatmentable for enforced community treatment

post-discharge were victimised in the firstpost-discharge were victimised in the first

year. The comparative national rate wasyear. The comparative national rate was

3.1%. In an earlier study of the same3.1%. In an earlier study of the same

patients the 4-month period prevalence ofpatients the 4-month period prevalence of

victimisation was 8.2%, suggesting thatvictimisation was 8.2%, suggesting that

the annual prevalence rate would be some-the annual prevalence rate would be some-

what higher. Brekkewhat higher. Brekke et alet al (2001) followed(2001) followed

172 patients in the community for 3 years172 patients in the community for 3 years

to assess their vulnerability to risk andto assess their vulnerability to risk and

reported that 34% of their sample were vic-reported that 34% of their sample were vic-

tims of violence over this period, presentingtims of violence over this period, presenting

an annual risk closer to ours. Silver (2002),an annual risk closer to ours. Silver (2002),

in a case–control study, compared the pre-in a case–control study, compared the pre-

valence of violent victimisation amongvalence of violent victimisation among

270 recently discharged people with severe270 recently discharged people with severe

mental illness over 10 weeks post-dischargemental illness over 10 weeks post-discharge

with 477 neighbourhood controls. Usingwith 477 neighbourhood controls. Using

data from the Pittsburgh site of thedata from the Pittsburgh site of the

McArthur Risk Assessment Study, he foundMcArthur Risk Assessment Study, he found

that significantly more patients (15%) thanthat significantly more patients (15%) than

neighbourhood controls (7%) reportedneighbourhood controls (7%) reported

violent victimisation. Certain factors haveviolent victimisation. Certain factors have

been found to increase the risk of victimisa-been found to increase the risk of victimisa-

tion in the general population, includingtion in the general population, including

male gender, younger age, unemploymentmale gender, younger age, unemployment

and ethnic minority status. Despite controlsand ethnic minority status. Despite controls

being derived from the same neighbour-being derived from the same neighbour-

hood, patients still possessed more of thesehood, patients still possessed more of these

factors. Following statistical adjustment forfactors. Following statistical adjustment for

these and for the individual’s own violencethese and for the individual’s own violence

perpetration, patients were still nearlyperpetration, patients were still nearly

twice as likely to be violently victimisedtwice as likely to be violently victimised

than controls.than controls.

Factors associated with violentFactors associated with violent victimisationvictimisation

Our finding that victims of violence displayOur finding that victims of violence display

more severe clinical symptoms is consistentmore severe clinical symptoms is consistent

with previously published literature on thewith previously published literature on the

subject (Lehman & Linn, 1984; Brekkesubject (Lehman & Linn, 1984; Brekke etet

alal, 2001; Hiday, 2001; Hiday et alet al, 2002). Homelessness, 2002). Homelessness

(Hiday(Hiday et alet al, 1999), substance misuse, 1999), substance misuse

(Hiday(Hiday et alet al, 1999; Brekke, 1999; Brekke et alet al, 2001), 2001)

and a history of violence (Lehman & Linn,and a history of violence (Lehman & Linn,

1984) were also identified as significantly1984) were also identified as significantly

related to victimisation, as in previousrelated to victimisation, as in previous

work. However, it is difficult to make validwork. However, it is difficult to make valid

comparisons with other studies becausecomparisons with other studies because

researchers have either grouped non-violentresearchers have either grouped non-violent

victimisation together as a single outcomevictimisation together as a single outcome

(Hiday(Hiday et alet al, 1999, 2002) or they have used, 1999, 2002) or they have used

highly heterogeneous samples of patients.highly heterogeneous samples of patients.

Our results show that victims wereOur results show that victims were

more likely to misuse illegal substances, tomore likely to misuse illegal substances, to

have a recent history of assaulting othershave a recent history of assaulting others

and to be diagnosed with a comorbid per-and to be diagnosed with a comorbid per-

sonality disorder, all of which have beensonality disorder, all of which have been

shown previously to increase the risk forshown previously to increase the risk for

violent behaviour in the sample (Walshviolent behaviour in the sample (Walsh etet

alal, 2001; Moran, 2001; Moran et alet al, 2003). Victimisation, 2003). Victimisation

also has been found independently toalso has been found independently to

predict violence in the sample (Walshpredict violence in the sample (Walsh etet

alal, 2001). Childhood abuse and neglect, 2001). Childhood abuse and neglect

are risk factors for adult mental illnessare risk factors for adult mental illness

and have been shown to have a significantand have been shown to have a significant

impact on the likelihood of delinquency,impact on the likelihood of delinquency,

adult criminality and violence (Maxfieldadult criminality and violence (Maxfield

& Widom, 1996; Hiday& Widom, 1996; Hiday et alet al, 2001). Those, 2001). Those

with psychosis are more likely to be born inwith psychosis are more likely to be born in

cities (Marceliscities (Marcelis et alet al, 1998), and social drift, 1998), and social drift

(Goldberg & Morrison, 1963) makes them(Goldberg & Morrison, 1963) makes them

more likely to live in socially disorganisedmore likely to live in socially disorganised

and crime-ridden neighbourhoods and beand crime-ridden neighbourhoods and be

subjected to violence (Hidaysubjected to violence (Hiday et alet al, 2001)., 2001).

Our results show that those who have beenOur results show that those who have been

victimised were significantly more likely tovictimised were significantly more likely to

feel threatened and unsafe than others andfeel threatened and unsafe than others and

consequently it is more likely that they willconsequently it is more likely that they will

engage in violence themselves. It is there-engage in violence themselves. It is there-

fore conceivable that victimisation andfore conceivable that victimisation and

violence in severe mental illness share aviolence in severe mental illness share a

common pathway and that the occurrencecommon pathway and that the occurrence

of one or both outcomes will be determinedof one or both outcomes will be determined

by complex interactions between theseby complex interactions between these

factors across the life cycle. It should befactors across the life cycle. It should be

noted, however, that less than half ofnoted, however, that less than half of

victims reported committing an assault invictims reported committing an assault in

the 2 years before interview, indicating thatthe 2 years before interview, indicating that

an individual’s own violence may only ex-an individual’s own violence may only ex-

plain a proportion of violent victimisationplain a proportion of violent victimisation

in the sample. Furthermore, the linkin the sample. Furthermore, the link

between severe mental illness and violentbetween severe mental illness and violent

victimisation has been shown recently tovictimisation has been shown recently to

be independent of an individual’s ownbe independent of an individual’s own

tendency towards violence (Silver, 2002).tendency towards violence (Silver, 2002).

Compliance with treatment was notCompliance with treatment was not

measured in this study, but all subjects weremeasured in this study, but all subjects were

in contact with services, suggesting thatin contact with services, suggesting that

patients at particular risk of victimisationpatients at particular risk of victimisation

could be targeted for more assertivecould be targeted for more assertive

follow-up. One such assertive approach,follow-up. One such assertive approach,

called out-patient commitment, iscalled out-patient commitment, is practisedpractised

in certain states in North America,in certain states in North America, where itwhere it

has been shown to reduce significantlyhas been shown to reduce significantly

criminal victimisation in people with severecriminal victimisation in people with severe

mental illnesses (Hidaymental illnesses (Hiday et alet al, 2002). Within, 2002). Within

this approach, patients are ordered by lawthis approach, patients are ordered by law

to receive treatment and supervision by ato receive treatment and supervision by a

named treatment provider.named treatment provider.

Implications of the studyImplications of the study

It is becoming increasingly clear that thereIt is becoming increasingly clear that there

is a need to refocus the issue of communityis a need to refocus the issue of community

risk away from the danger posed by men-risk away from the danger posed by men-

tally ill individuals to the danger posed totally ill individuals to the danger posed to

them from other members of society (Walshthem from other members of society (Walsh

& Fahy, 2002). This has been highlighted& Fahy, 2002). This has been highlighted

by an American finding that patients withby an American finding that patients with

psychosis living in the community are 14psychosis living in the community are 14

times more likely to be the victims of a vio-times more likely to be the victims of a vio-

lent crime than to be arrested for such alent crime than to be arrested for such a

crime (Brekkecrime (Brekke et alet al, 2001). Further longitu-, 2001). Further longitu-

dinal work is needed to clarify the predic-dinal work is needed to clarify the predic-

tors of victimisation, which may be usedtors of victimisation, which may be used

to target vulnerable subgroups with addi-to target vulnerable subgroups with addi-

tional care. Enquiry about victimisation ex-tional care. Enquiry about victimisation ex-

periences does not form part of the routineperiences does not form part of the routine

psychiatric interpsychiatric interview. In light of our find-view. In light of our find-

ings and other emerging evidence on the sizeings and other emerging evidence on the size

of the victimisation problem among peopleof the victimisation problem among people

with mental illnesses, we suggest that suchwith mental illnesses, we suggest that such

enquiry be incorporated as standard.enquiry be incorporated as standard.

ACKNOWLEDGEMENTSACKNOWLEDGEMENTS

E.W. was funded by a WellcomeTraining Fellowship.E.W. was funded by a Wellcome Training Fellowship. The UK700 trial was funded by grants from the UKThe UK700 trial was funded by grants from the UK Department of Health and an NHS research andDepartment of Health and an NHS research and development programme.development programme.

The UK700 Group is a collaborative study teamThe UK700 Group is a collaborative study team involving four clinical and two non-clinical centres:involving four clinical and two non-clinical centres: Manchester Royal InfirmaryManchester Royal Infirmary: Tom Butler, Francis: Tom Butler, Francis Creed, Janelle Fraser, Peter Huxley, NicholasTarrier,Creed, Janelle Fraser, Peter Huxley, NicholasTarrier, Theresa Tattan.Theresa Tattan. King’s/Maudsley Hospital, LondonKing’s/Maudsley Hospital, London:: Tom Fahy, Catherine Gilvarry, Kwame McKenzie,Tom Fahy, Catherine Gilvarry, Kwame McKenzie, Robin Murray, Jim van Os, ElizabethWalsh.Robin Murray, Jim van Os, ElizabethWalsh. St Mary’sSt Mary’s Hospital/St Charles’ Hospital, London:Hospital/St Charles’ Hospital, London: John Green,John Green, Anna Higgitt, Elizabeth van Horn, Donal Leddy,Anna Higgitt, Elizabeth van Horn, Donal Leddy, Catherine Manley, Patricia Thornton, Peter Tyrer.Catherine Manley, Patricia Thornton, Peter Tyrer. St George’s Hospital, LondonSt George’s Hospital, London: Robert Bale,Tom Burns,:Robert Bale,Tom Burns, Matthew Fiander, Kate Harvey, Andy Kent, ChiaraMatthew Fiander, Kate Harvey, Andy Kent, Chiara Samele.Samele. Centre for Health Economics, YorkCentre for Health Economics, York: Sarah: Sarah Byford,David Torgerson,KenWright.Byford,David Torgerson,KenWright. London (Statis-London (Statis- tics)tics): Simon Thompson (: Simon Thompson (Royal Postgraduate MedicalRoyal Postgraduate Medical SchoolSchool) and IanWhite () and IanWhite (London School of Hygiene andLondon School of Hygiene and Tropical MedicineTropical Medicine).).

REFERENCESREFERENCES

—sberg, M., Montgomery, S. A., Perris,C.,—sberg, M., Montgomery, S. A., Perris, C., et alet al (1978)(1978) A comprehensive psychopathological rating scale.A comprehensive psychopathological rating scale. Acta Psychiatrica Scandinavica SupplementumActa Psychiatrica Scandinavica Supplementum,, 271271, 5^27., 5^27.

Andreasen,N. C.Andreasen, N. C. (1984)(1984) Modified Scale for theModified Scale for the Assessment of Negative Symptoms.Assessment of Negative Symptoms. Iowa City, IA:Iowa City, IA: University of Iowa.University of Iowa.

Brekke, J. S., Prindle,C.,Woo Bae, S.,Brekke, J. S., Prindle,C.,Woo Bae, S., et alet al (2001)(2001) Risks for individuals with schizophrenia who are living inRisks for individuals with schizophrenia who are living in the community.the community. Psychiatric ServicesPsychiatric Services,, 5252, 1358^1366., 1358^1366.

2 3 72 3 7

WALSH ET ALWALSH ET AL

Burns,T.,Creed, F., Fahy,T.,Burns,T., Creed, F., Fahy,T., et alet al (1999)(1999) IntensiveIntensive versus standard case management for severe psychoticversus standard case management for severe psychotic illness: a randomised trial.UK700 Group.illness: a randomised trial.UK700 Group. LancetLancet,, 353353,, 2185^2189.2185^2189.

Cascardi, M., Mueser, K.T., DeGiralomo, J.,Cascardi, M., Mueser, K.T., DeGiralomo, J., et alet al (1996)(1996) Physical aggression against psychiatric inpatientsPhysical aggression against psychiatric inpatients by family members and partners.by family members and partners. Psychiatric ServicesPsychiatric Services,, 4747,, 531^533.531^533.

Glover,G. R., Robin, E., Emami, J.,Glover,G. R., Robin, E., Emami, J., et alet al (1998)(1998) AA needs index for mental health care.needs index for mental health care. Social Psychiatry andSocial Psychiatry and Psychiatric EpidemiologyPsychiatric Epidemiology,, 3333, 89^96., 89^96.

Goldberg, E. M. & Morrison, S. L.Goldberg, E. M. & Morrison, S. L. (1963)(1963) Schizophrenia and social class.Schizophrenia and social class. British Journal ofBritish Journal of PsychiatryPsychiatry,, 109109, 785^802., 785^802.

Hiday,V. A., Swanson, J.W., Swartz, M. S.,Hiday,V. A., Swanson, J.W., Swartz, M. S., et alet al (1999)(1999) Criminal victimisation of persons with severeCriminal victimisation of persons with severe mental illness.mental illness. Psychiatric ServicesPsychiatric Services,, 5050, 62^68., 62^68.

__ ,, __ ,, __ ,, et alet al (2001)(2001) Victimization: a link betweenVictimization: a link between mental illness and violence?mental illness and violence? International Journal of LawInternational Journal of Law and Psychiatryand Psychiatry,, 2424, 559^572., 559^572.

__ , Swartz,M. S., Swanson, J.W.,, Swartz,M. S., Swanson, J.W., et alet al (2002)(2002) ImpactImpact of outpatient commitment on victimisation of peopleof outpatient commitment on victimisation of people with severe mental illness.with severe mental illness. American Journal of PsychiatryAmerican Journal of Psychiatry,, 88, 1403^1411., 1403^1411.

Jablensky, A., Schwartz, R. & Tomov,T.Jablensky, A., Schwartz, R. & Tomov,T. (1980)(1980) WHOWHO collaborative study of impairments and disabilitiescollaborative study of impairments and disabilities associated with schizophrenic disorders: a preliminaryassociated with schizophrenic disorders: a preliminary communication ^ objectives and methods.communication ^ objectives and methods. ActaActa Psychiatrica ScandinavicaPsychiatrica Scandinavica,, 6262, 152^163., 152^163.

Lehman, A. F. & Linn, L. S.Lehman, A. F. & Linn, L. S. (1984)(1984) Crimes againstCrimes against discharged mental patients in board-and-care homes.discharged mental patients in board-and-care homes. American Journal of PsychiatryAmerican Journal of Psychiatry,, 141141, 271^274., 271^274.

Marcelis, M., Navarro-Mateu, F., Murray, R.,Marcelis, M., Navarro-Mateu, F., Murray, R., et alet al (1998)(1998) Urbanization and psychosis: a study of1942^1978Urbanization and psychosis: a study of1942^1978 birth cohorts inThe Netherlands.birth cohorts inThe Netherlands. Psychological MedicinePsychological Medicine,, 44, 871^879., 871^879.

Maxfield, M. G. & Widom,C.Maxfield, M. G. & Widom,C. (1996)(1996) Cycle of violenceCycle of violence revisited 6 years later.revisited 6 years later. Archives of Pediatrics andArchives of Pediatrics and Adolescent MedicineAdolescent Medicine,, 150150, 390^395., 390^395.

McGuffin, P., Farmer, A. & Harvey, I.McGuffin, P., Farmer, A. & Harvey, I. (1991)(1991) AA polydiagnostic application of operational criteria inpolydiagnostic application of operational criteria in studies of psychotic illness.Development and reliabilitystudies of psychotic illness.Development and reliability of the OPCRITsystem.of the OPCRITsystem. Archives of General PsychiatryArchives of General Psychiatry,, 4848,, 764^770.764^770.

Moran, P.,Walsh, E.,Tyrer, P.,Moran, P.,Walsh, E.,Tyrer, P., et alet al (2003)(2003) Impact ofImpact of comorbid personality disorder on violence in psychosis.comorbid personality disorder on violence in psychosis. Report from the UK700 trial.Report from the UK700 trial. British Journal ofBritish Journal of PsychiatryPsychiatry,, 182182, 129^134., 129^134.

Mullen, P. E., Burgess, P.,Wallace,C.,Mullen, P. E., Burgess, P.,Wallace,C., et alet al (2000)(2000) Community care and criminal offending in schizophrenia.Community care and criminal offending in schizophrenia. LancetLancet,, 355355, 614^617., 614^617.

Murless-Black,C., Mayhew, P. & Percy, A.Murless-Black,C., Mayhew, P. & Percy, A. (1996)(1996) British Crime Survey 1996British Crime Survey 1996. London: Home Office.. London: Home Office.

Oliver, J.Oliver, J. (1991)(1991) The social care directive: developmentThe social care directive: development of a quality of life profile for use in community servicesof a quality of life profile for use in community services for the mentally ill.for the mentally ill. Social Work and Social SciencesSocial Work and Social Sciences ReviewReview,, 33, 53^60., 53^60.

Phelan, M., Slade, M.,Thornicroft, G.,Phelan, M., Slade, M.,Thornicroft, G., et alet al (1995)(1995) The Camberwell Assessment of Need: the validity andThe Camberwell Assessment of Need: the validity and

reliability of an instrument to assess the needs of peoplereliability of an instrument to assess the needs of people with severe mental illness.with severe mental illness. British Journal of PsychiatryBritish Journal of Psychiatry,, 167167, 589^595., 589^595.

Silver, E.Silver, E. (2002)(2002) Mental disorder and violentMental disorder and violent victimization: the mediating role of involvement invictimization: the mediating role of involvement in conflicted social relationships.conflicted social relationships. CriminologyCriminology,, 4040, 191^211., 191^211.

Spitzer, R., Endicott, J. & Robins, E.Spitzer, R., Endicott, J. & Robins, E. (1978)(1978) ResearchResearch Diagnostic Criteria: rationale and reliability.Diagnostic Criteria: rationale and reliability. Archives ofArchives of General PsychiatryGeneral Psychiatry,, 3535, 773^782., 773^782.

StataCorpStataCorp (1999)(1999) Stata Statistical Software, Release 6.0Stata Statistical Software, Release 6.0.. College Station,TX: Stata Corporation.College Station,TX: Stata Corporation.

Tyrer, P., Alexander, M. S., Cicchetti, D,Tyrer, P., Alexander, M. S., Cicchetti, D, et alet al (1979)(1979) Reliability of a schedule for rating personality disorders.Reliability of a schedule for rating personality disorders. British Journal of PsychiatryBritish Journal of Psychiatry,, 135135, 168^174., 168^174.

UK700 GroupUK700 Group (1999)(1999) Comparison of intensive andComparison of intensive and standard case management for patients with psychosis.standard case management for patients with psychosis. Rationale of the trial.Rationale of the trial. British Journal of PsychiatryBritish Journal of Psychiatry,, 174174,, 74^78.74^78.

Van Horn, E., Manley,C., Leddy, D.,Van Horn, E., Manley,C., Leddy, D., et alet al (2000)(2000) Problems in developing an instrument for the rapidProblems in developing an instrument for the rapid assessment of personality status.assessment of personality status. European PsychiatryEuropean Psychiatry,, 1616 (suppl.1), 29^33.(suppl. 1), 29^33.

Walsh, E. & Fahy,T.Walsh, E. & Fahy,T. (2002)(2002) Violence in society ^Violence in society ^ contribution of mental illness.contribution of mental illness. BMJBMJ,, 325325, 507^508., 507^508.

__ ,Gilvarry,C., Samele,C.,, Gilvarry,C., Samele,C., et alet al (2001)(2001) ReducingReducing violence in severe mental illness: randomised controlledviolence in severe mental illness: randomised controlled trial of intensive case management compared withtrial of intensive case management compared with standard care.standard care. BMJBMJ,, 323323, 1093^1096., 1093^1096.

__ , Buchanan, A. & Fahy,T. (2002), Buchanan, A. & Fahy,T. (2002) Violence andViolence and schizophrenia: examining the evidence.schizophrenia: examining the evidence. British Journal ofBritish Journal of PsychiatryPsychiatry,, 181181, 490^495., 490^495.

World Health Organization (1992World Health Organization (1992aa)) The Life ChartThe Life Chart.. Geneva:WHO.Geneva:WHO.

__ (1992(1992bb)) The ICD^10 Classification of Mental andThe ICD^10 Classification of Mental and Behavioural DisordersBehavioural Disorders.Geneva:WHO..Geneva:WHO.

2 3 82 3 8

CLINICAL IMPLICATIONSCLINICAL IMPLICATIONS

&& Violent victimisation of patients with psychosis is a significant problem: 16% wereViolent victimisation of patients with psychosis is a significant problem: 16% were violently victimised over1year.violently victimised over1year.

&& Those at particular risk of victimisation havemore severe symptoms and poorerThose at particular risk of victimisation havemore severe symptoms and poorer social circumstances.social circumstances.

&& Assertive enforced community treatment, which has been shown to reduceAssertive enforced community treatment, which has been shown to reduce victimisation in the USA, needs further evaluation in other settings.victimisation in the USA, needs further evaluation in other settings.

LIMITATIONSLIMITATIONS

&& The study was conducted in inner cities, so the findings are not generalisable toThe study was conducted in inner cities, so the findings are not generalisable to other settings.other settings.

&& It is a cross-sectional study that identified correlates rather than predictors ofIt is a cross-sectional study that identified correlates rather than predictors of victimisation.victimisation.

&& General population figures are used for comparison, rather than other controls.General population figures are used for comparison, rather than other controls.

ELIZABETHWALSH,MD, PAULMORAN,MD, Institute of Psychiatry, London; CHARLES SCOTT,MD,ELIZABETHWALSH,MD, PAULMORAN,MD, Institute of Psychiatry, London; CHARLES SCOTT,MD, University of California,Davis,CA,USA;KWAME McKENZIE,MRCPsych,Royal Free School of Medicine,University of California,Davis,CA,USA;KWAME McKENZIE,MRCPsych,Royal Free School of Medicine, London;TOMBURNS,MD, St George’s Hospital Medical School, London; FRANCIS CREED,MD,UniversityLondon;TOMBURNS,MD, St George’s Hospital Medical School, London; FRANCIS CREED,MD,University Department of Psychiatry,Manchester Royal Infirmary,Manchester; PETERTYRER,MD, St Mary’s HospitalDepartment of Psychiatry,Manchester Royal Infirmary,Manchester; PETERTYRER,MD, St Mary’s Hospital Medical School, London; ROBINM.MURRAY,MD,TOM FAHY,MD, Institute of Psychiatry, LondonMedical School, London; ROBINM.MURRAY,MD,TOM FAHY,MD, Institute of Psychiatry, London

Correspondence:Dr ElizabethWalsh, Section of Forensic Mental Health,The Institute of Psychiatry,DeCorrespondence:Dr ElizabethWalsh, Section of Forensic Mental Health,The Institute of Psychiatry,De Crespigny Park,Denmark Hill, London SE5 8AF,UK.E-mail: sppmemwCrespigny Park,Denmark Hill, London SE5 8AF,UK.E-mail: sppmemw@@iop.kcl.ac.ukiop.kcl.ac.uk

(First received 12 February 2003, final revision 19 May 2003, accepted 21May 2003)(First received 12 February 2003, final revision 19 May 2003, accepted 21May 2003)