ArticlePDF Available

Psychopathy among prisoners in England and Wales

Authors:

Abstract and Figures

Most research into psychopathy among prisoners is based on selected samples. It remains unclear whether prevalences are lower among European populations. This study aimed to measure the prevalence of psychopathy, and the distribution and correlates of psychopathic traits in a representative national sample of prisoners. Psychopathy was measured using the revised Psychopathy Checklist (PCL-R) in a second stage, cross-sectional survey of prisoners in England and Wales in 1997 (n=496). Poisson regression analysis was carried out to examine independent associations between correlates and PCL-R total and factor scores. The prevalence of categorically diagnosed psychopathy at a cut off of 30 was 7.7% (95%CI 5.2-10.9) in men and 1.9% (95%CI 0.2-6.9) in women. Psychopathic traits were less prevalent among women. They were correlated with younger age, repeated imprisonment, detention in higher security, disciplinary infractions, antisocial, narcissistic, histrionic, and schizoid personality disorders, and substance misuse, but not neurotic disorders or schizophrenia. The study concluded that psychopathy and psychopathic traits are prevalent among male prisoners in England and Wales but lower than in most previous studies using selected samples. However, most correlates with psychopathic traits were similar to other studies. Psychopathy identifies the extreme of a spectrum of social and behavioral problems among prisoners.
Content may be subject to copyright.
Psychopathy among prisoners in England and Wales
Jeremy Coid
a,
, Min Yang
a
, Simone Ullrich
a
, Amanda Roberts
a
, Paul Moran
b
, Paul Bebbington
c
,
Traolach Brugha
d
, Rachel Jenkins
e
, Michael Farrell
b
, Glyn Lewis
f
, Nicola Singleton
g
, Robert Hare
h
a
Queen Mary University of London, UK
b
Institute of Psychiatry, London, UK
c
Royal Free and University College, UK
d
University of Leicester, UK
e
King's College London, UK
f
University of Bristol, UK
g
Drugs Analysis and Research Programme, London, UK
h
University of British Columbia, Canada
abstractarticle info
Keywords:
Psychopathy
Prisoners
Epidemiology
Sampling
Most research into psychopathy among prisoners is based on selected samples. It remains unclear whether
prevalences are lower among European populations. This study aimed to measure the prevalence of
psychopathy, and the distribution and correlates of psychopathic traits in a representative national sample of
prisoners. Psychopathy was measured using the revised Psychopathy Checklist (PCL-R) in a second stage,
cross-sectional survey of prisoners in England and Wales in 1997 (n=496). Poisson regression analysis was
carried out to examine independent associations between correlates and PCL-R total and factor scores. The
prevalence of categorically diagnosed psychopathy at a cut off of 30 was 7.7% (95%CI 5.210.9) in men and
1.9% (95%CI 0.26.9) in women. Psychopathic traits were less prevalent among women. They were correlated
with younger age, repeated imprisonment, detention in higher security, disciplinary infractions, antisocial,
narcissistic, histrionic, and schizoid personality disorders, and substance misuse, but not neurotic disorders
or schizophrenia. The study concluded that psychopathy and psychopathic traits are prevalent among male
prisoners in England and Wales but lower than in most previous studies using selected samples. However,
most correlates with psychopathic traits were similar to other studies. Psychopathy identies the extreme of
a spectrum of social and behavioral problems among prisoners.
© 2009 Elsevier Ltd. All rights reserved.
1. Introduction
Psychopathy is a personality disorder associated with multiple
social and behavioral problems (Cornell et al., 1996; Hill, Neumann, &
Rogers, 2004) and has an exceptionally poor prognosis among the
mental disorders (Andersen, Sestoft, Lillebaek, Mortensen, & Kramp,
1999; Hare, 2003). Although not currently included as a separate
diagnostic category in the ICD or DSM classications, interest in
psychopathy has grown and its measurement has become increasingly
important in risk assessment. It is a rare condition affecting less than
1% of the household population (Coid et al., 2009) but highly
prevalent among prisoners and associated with homelessness and
psychiatric hospitalization over the lifespan. However, there are
remarkable differences in reported prevalence rates of psychopathy
among samples of prisoners in different countries withina range of 3%
to 73%, (Assadi et al., 2006; Coid, 1998; Cooke, 1996; Hare, 2003;
Moran, 1999; Ullrich, Paelecke, Kahle, & Marneros, 2003).
Psychopathy, measured using the Psychopathy Checklist Revised
(PCL-R; Hare, 2003), incorporates aspects of antisocial behavior as well
as core personality traits. Studies of the factor structure of psychopathy
indicate the importance of different subcomponents (Cooke & Michie,
2001). They are now incorporated into a four-factormodel based on
conrmatory factor analyses (Hare & Neumann, 2006; Neumann,
Vitacco, Hare, & Wupperman, 2005; Vitacco, Neumann, & Jackson,
2005), although initially referred to as two factorfour facet modelin
the second edition of the PCL-R (Hare, 2003). This model allows for
ner descriptive analysis of individuals encountered in clinical practice
and facilitates empirical study of the subcomponents of psychopathy
(see Fig. 1). The components of psychopathy comprise deceitful
interpersonal style,affective deciency,impulsivenessor life-
style(depending on the assumption of three or four underlying
factors) and the antisocialcomponent. This differentiation of
psychopathy now includes possibly differing etiological factors
(Blonigen, Hicks, Krueger, Patrick, & Lacono, 2005; Viding, Blair,
International Journal of Law and Psychiatry 32 (20 09) 134141
Funding support: Department of Health, England and Wales.
Corresponding author. Barts and the London Medical School, Queen Mary University
of London, Forensic Psychiatry Research Unit, William Harvey House, 61 Bartholomew's
Close, London EC1A 7BE, UK. Tel.: +44 20 7601 8138; fax: +44 20 7601 7969.
E-mail address: j.w.coid@qmul.ac.uk (J. Coid).
0160-2527/$ see front matter © 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijlp.2009.02.008
Contents lists available at ScienceDirect
International Journal of Law and Psychiatry
Moftt, & Plomin, 2005), with evidence of differing neuro-cognitive
dysfunction associated with certain subcomponents (Blair, Mitchell, &
Blair, 2005; Hare, 2003). However, research into psychopathy has
almost exclusively focused on non-representative samples (including
samples of convenience) using primarily North American male
prisoners selected from high and medium secure institutions (Hare,
2003). Psychopathic individuals are likely to be concentrated in these
locations due to serious criminal behavior and behavioral disorder
whilst incarcerated (Coid, 1998) and the generalisability of ndings
from these studies is unclear. In contrast, a representative sample of
Scottish prisoners (Cooke, 1994) combined with selected samples of
English offenders (Hare, 2003, pp. 205210; Hare, Clark, Grann, &
Thornton, 2000) demonstrated lower PCL-R scores than North
American samples. Based on Item Response Theory analyses (IRT), it
was subsequently argued that, when making the diagnosis, the
standard PCL-R cut score for psychopathy should be lowered for UK
populations (Cooke & Michie, 1999; Cooke, Michie, Hart, & Clark,
2005). However, Bolt, Hare, and Neumann (2007), also using IRT
analyses but with a different anchor item selection method, have
proposed that the recommended PCL-R cut-off score of 30 reects
approximately the same level of psychopathy in the UK as in North
America. According to their analyses, lower scores in one country do
not necessarily mean lack of scalar equivalence, implying that the cut
score for determining prevalence should be similar in the UK and North
America.
Our aim was to determine whether certain correlates of psycho-
pathy observed in non-representative samples are found equally in a
more representative sample of an entire correctional jurisdiction. We
therefore estimated the prevalence of psychopathy using a cut-off
score of 30, examined the distribution of psychopathic traits, and
elucidated the correlates of total and factor scores of psychopathy
among a representative sample of the prisoner population in England
and Wales, aged 1664, assessed in 1997. We used the PCL-R to
examine the relationship between measures of psychopathy and
demography, verbal intelligence, DSM-IV Axis II personality disorder
traits, ICD-10 clinical syndromes, offending behavior, and behavioral
problems in the prison setting. We aimed to investigate the overall
and gender specic prevalence of categorically diagnosed psycho-
pathy in a representative sample of prisoners. Furthermore, we were
interested in the differential associations of the factors of psychopathy
with the above mentioned outcome variables. The conceptualization
of psychopathy as three or four factor model is a recent development.
Consequently, few studies have investigated whether the components
demonstrate similar or different correlations, and which may be
highly relevant for future understanding of psychopathy. However, as
the position regarding the cut-off in a European population, together
with the ongoing debate as to whether a three- or four-factor solution
best t data on psychopathy remains unresolved, we adhered to the
recommended manual cut-off of 30 and examined the four-factor
model of psychopathy.
2. Method
2.1. Sample
The sample comprised 496 participants in the second of a two-
stage survey of psychiatric morbidity among prisoners in England and
Wales, aged 1664 years, carried out by the Ofce for National
Statistics in 1997 (Singleton, Meltzer, Gatward, Coid, & Deasy, 1998).
All 131 penal establishments were included, then containing 61,944
prisoners, including 46,872 male sentenced,12,302 male remand, and
2770 women prisoners. Different sampling fractions were applied to
assure the requisite number of interviews for each group of prisoners.
This included 1 in 34 male sentenced prisoners, 1 in 8 male remand
prisoners, and 1 in 3 women prisoners, either remand or sentenced. In
the last four weeks of the survey, the sampling fraction changed to 1in
50 male sentenced, as a larger number of this group had been
interviewed. Samples were taken from all prison locations in the rst
phase to avoid over- or under-sampling those with mental health
problems in locations such as Health Care and to be representative of
the entire national prison estate. The survey therefore included all
prisons (115 male, 11 women, 5 mixed). This included a sample of
16.2% from Young Offender Institutions, 10.3% from open prisons or
lowest security category D, 5.1% closed prisons, 7.4% category B, 25%
category C, 31.6% local prisons including sentenced and remanded by
local courts, and 5.9% from dispersal prisons, the highest on the
security scale. Substitution of prisoners no longer available for
interview, including those transferred or released, with new prisoners
was performed for those on remand.
In the rst stage, 3563 prisoners were selected, of whom 3142
(88%) completed full interviews. 37 failed to complete a full interview,
198 (6%) refused, and 53 (1%) could not take part, mainly due to
language problems. Interviewers could not contact 118 (3%), and were
advised not to interview 15.
In the second stage, 661 prisoners, a 1 in 5 random subsample,
were then selected for clinical interview, of whom 505 (76%) were
interviewed, 105 (16%) could no longer be contacted, and 50 (8%)
refused.
Fig. 1. Items in the 4-factor model of psychopathy (Hare, 2003).
135J. Coid et al. / International Journal of Law and Psychiatry 32 (2009) 134141
2.2. Diagnostic measures
Self-report measures were taken in the rst stage using laptop
computers: the Clinical Interview Schedule (CIS-R; Lewis & Pelosi,
199 0) measures six ICD-10 syndromes in the week preceding
interview, including mixed anxiety and depressive disorder, general-
ized anxiety disorder, depressive episodes, phobias, obsessive
compulsive disorder, panic disorder; and a brief measure of
perceptualverbal intelligence, the Quick Test (Ammons & Ammons,
1962). We also analyzed a combined category of the ICD conditions.
Additional self-report measures included socio-demography and
behavior in the prison setting. Information on criminal convictions
was obtained from prison records.
Psychopathy was measured in the second phase using the
Psychopathy Checklist Revised (PCL-R; Hare,1991, 2003) consisting
of 20 items scored 0, 1, or 2 based upon clinical interview and review
of le information. This was administered after rst completing the
Schedules for Clinical Assessment in Neuropsychiatry (SCAN; Wing et
al., 1990; World Health Organisation, 1999) for ICD-10 clinical
syndromes, and the Structured Clinical Interview for Axis II disorders
(SCID-II; First, Gibbon, Spitzer, & Williams, 1997) measuring cate-
gories of DSM-IV personality disorder. The scoring of the PCL-R
allowed us to present our results using continuous scores as well as
categorical measures, including the recommended score of 30 as the
cut score, above which a diagnosis of psychopathy is attributed.
The second stage of the survey was conducted by six psychiatrists
at specialist registrar level and two clinical psychologists trained in a
group to use the PCL-R assessment procedure and scoring. This
involved the viewing of videotapes of assessment interviews to
establish norms for scoring individual items. Cronbach's Alpha
coefcients of total, male, and female PCL-R scores were within the
acceptable range (total 0.89, male= 0.88, female= 0.90) suggesting
good internal consistency. Inter-item correlations (M=0.29,
SD= 0.13, Md= 0.29) also indicated satisfactory homogeneity.
2.3. Statistical analysis
For PCL-R total and four factor scores, descriptive analyses of
demographic characteristics and social adversity measures were
derived using SPSS (v12). Pearson's correlation coefcients were
calculated for inter-item correlations, and Cronbach's alpha coef-
cients for overall internal consistency among the 20 items of PCL-R.
Partial correlation analysis was performed for the four factors of
psychopathy, controlling for gender. Poisson regression analysis was
applied to investigate the association between the PCL-R total scores
and demographic characteristics of respondents, Axis II disorders,
intelligence, SCAN diagnosis, CIS-R neurotic syndromes, index
offences, adult behavior problems and life events experienced by
respondents. These were adjusted for confounders or co-morbid
disorders, applicable to each variable of interest. As the four factors are
strongly associated with each other, the analysis took into account
their inter-correlations when investigating associations between the
variables of interest and factors scores. We used multivariate Poisson
regression analysis, which treats the four factors scores as repeated
measures within each respondent. The full variancecovariance
structure of the four factor scores was then captured by the model.
The effect of each variable on the PCL-R scores was tested by the
standard Z-score statistic. All regression analyses were performed in
MLwiN (V2.0). Extra Poisson variation was allowed in the Poisson
regression model to reect the excessive variation of the PCL-R scores
due to extreme values.
3. Results
3.1. Prevalence
The prevalence of psychopathy using a PCL-R cut-off of 30 was 7.7%
(95% CI: 5.210.9) in men and 1.9% (0.26.9) in women. Remanded
men had a higher prevalence (9.4%, 5.515.0) than sentenced men
(6.2%, 3.310.6). Sentenced women demonstrated the same preva-
lence as the entire sample of sentenced and remanded women
prisoners (1.9%, 0.28.1). The gender ratio for psychopathy at this cut
score was 4:1. Mean total PCL-R scores were 15.03 (SD 9.14) for men
and 8.31 (SD 8.59) for women.
Fig. 2 shows the distribution of PCL-R total scores among men and
women in the prison population. A relatively larger proportion of
Fig. 2. Prevalence of psychopathy.
Table 1
Inter-factor correlation.
Pearson's simple correlation Partial correlation
(adjusted for gender)
Factor 1 Factor 2 Factor 3 Factor 1 Factor 2 Factor 3
Factor 2 0.44*** 0.42***
Factor 3 0.55*** 0.54*** 0.52*** 0.48***
Factor 4 0.47*** 0.48*** 0.73*** 0.43*** 0.44*** 0.75***
***pb0.0001 (two tailed).
136 J. Coid et al. / International Journal of Law and Psychiatry 32 (2009) 134141
women had none, or very few psychopathic traits (44.8% scored less
than 5).
3.2. Demography, intelligence and prison location
Partial correlation coefcients between the four factors, controlling
for gender, demonstrated that all were signicantly correlated
(Table 1). The strongest correlations were between the antisocial
(F4) and lifestyle (F3) factors.
Lower mean PCL-R total scores were observed among prisoners
aged 3564 compared to those 1634 years, accounted for by the
affective, lifestyle, and antisocial, but not interpersonal, factors
(Table 2). Men scored signicantly higher than women for total
scores and all factors. Non-UK born prisoners demonstrated lower
total, lifestyle, and antisocial scores. However, black prisoners scored
higher on interpersonal and lifestyle factors.
Single and unmarried cohabiting prisoners had signicantly higher
total and antisocial factor scores. Married/widowed prisoners had
lower lifestyle scores. Educational qualications were associated with
lower total, affective, lifestyle, and antisocial scores. However, there
were no associations between social class and total PCL-R scores.
Prisoners from social classes I and II scored higher than those from
IIINM to VI, and those unemployed before imprisonment. Unem-
ployed individuals scored higher on the lifestyle factor.
There was no association between prisoner status (remanded or
sentenced) and psychopathy scores after controlling for other
demographic variables.
Regression analyses of verbal intelligence and PCL-R scores (after
controlling for gender and factor inter-correlations) demonstrated
negative associations with total scores [β(SE)= 0.012 (0.005),
pb0.05], lifestyle scores [β(SE) =0.011 (0.005), pb0.05], and
Table 2
Sociodemographic and socioeconomic characteristics of the total sample (n= 496).
Demographic characteristic Category group Respondents Total Factor 1 Factor 2 Factor 3 Factor 4
Interpersonal Affective Lifestyle Antisocial
N(%) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD)
Age group 1634 379 (76.4) 15.30 (9.20) 1.84 (2.02) 2.62 (2.50) 4.92 (3.03) 4.46 (3.23)
3554 109 (22.0) 8.71 (7.98)** 1.34 (2.00) 1.56 (2.05)* 2.4 4 (2.49)** 2.35 (2.80)**
5574 8 (1.6) 1.50 (2.50)** 0.00 (0.00) 1.13 (2.23) 0.13 (0.35)* 0.00 (0.00)
Gender Male 391 (78.8) 15.00 (9.14) 1.82 (2.07) 2.65 (2.44) 4.67 (3.02) 4.41 (3.23)
Female 105 (21.2) 8.31 (8.59)*** 1.27 (1.74)* 1.29 (2.16)*** 2.93 (3.09)** 2.12 (2.74)***
Non-UK born UK born 442 (89.1) 14.20 (9.37) 1.74 (2.04) 2.47 (2.46) 4.48 (3.07) 4.17 (3.28)
Non UK born 54 (10.9) 8.57 (8.40)* 1.35 (1.80) 1.44 (2.15) 2.80 (3.05)* 1.94 (2.37)***
Ethnic origin White 412 (83.1) 14.10 (9.23) 1.65 (1.99) 2.49 (2.46) 4.52 (3.06) 4.14 (3.23)
Black 59 (11.9) 11.50 (10.6) 2.15 (2.34)** 1.86 (2.37) 3.29 (3.12)* 2.97 (3.47)
Asian 10 (2.0) 8.90 (7.45) 1.00 (0.94) 1.60 (2.27) 3.20 (3.71) 2.30 (1.89)
Others 15 (3.0) 10.60 (9.48) 1.80 (1.70) 1.33 (2.13) 3.00 (3.16) 2.80 (3.36)
Marital status before prison Single 177 (35.7) 15.20 (9.23) 1.85 (2.00) 2.62 (2.54) 4.77 (3.02) 4.55 (3.18)
Divorced/separated 49 (9.9) 7.98 (6.06)* 1.16 (1.25) 1.45 (1.88) 2.59 (2.49) 1.76 (2.24)**
Married/widowed 72 (14.5) 8.57 (9.34)* 1.26 (1.98) 1.69 (2.21) 2.43 (3.16)* 2.28 (2.85)*
Cohabiting 198 (39.9) 15.40 (9.18) 1.86 (2.16) 2.60 (2.48) 4.98 (2.90) 4.49 (3.29)
Education qualication None 215 (43.3) 15.21 (9.43) 1.73 (2.01) 2.77 (2.63) 4.83 (3.05) 4.49 (3.25)
Any 281 (56.7) 12.37 (9.26)** 1.68 (2.03) 2.05 (2.25)* 3.89 (3.11)* 3.49 (3.23)*
Social class I and II 52 (10.5) 10.19 (9.07) 2.27 (2.61) 1.73 (2.20) 2.69 (2.77) 2.48 (3.01)
IIINM 48 (9.7) 10.73 (9.71) 1.73 (2.10)* 1.83 (2.44) 3.50 (3.22) 2.44 (2.74)
IIIM 136 (27.4) 12.81 (8.95) 1.35 (1.79)*** 2.35 (2.50) 4.13 (3.15) 3.66 (3.16)
IV 133 (26.8) 13.67 (9.61) 1.71 (2.02)** 2.21 (2.19) 4.28 (3.06) 4.11 (3.34)
V and VI 64 (12.9) 15.56 (9.24) 1.70 (1.91)** 2.48 (2.49) 4.86 (2.93) 5.09 (3.32)
Missing label 63 (12.7) 18.20 (8.37) 1.95 (1.88)* 3.51 (2.69) 6.08 (2.61)* 5.24 (2.92)
Remanded No 299 (60.3) 12.05 (9.32) 1.56 (2.07) 2.07 (2.36) 3.85 (3.11) 3.38 (3.15)
Yes 197 (39.7) 15.96 (9.12) 1.91 (1.91) 2.81 (2.52) 4.98 (3.01) 4.75 (3.28)
When comparing the mean scores between the category levels within each variable, other demographic variables and inter-factor correlations are adjusted for. The group missing
labelin social class consists of mostly young men, white, no qualication, single or cohabiting with antisocial personality disorder.
*p0.05, **p0.01, ***p0.001.
Table 3
Association between PCL-R scores and prison location.
Prison Location n(%) PCL-R
Total Factor 1 Factor 2 Factor 3 Factor 4
Interpersonal Affective Lifestyle Antisocial
Open
(reference)
42 (8.5) 6.3
(7.6)
0.8 (1.5) 1.7 (2.1) 1.9 (2.5) 1.5 (2.3)
Local
and dispersal
253 (51.0) 14.6
(9.6)*
1.9 (2.1) 2.5 (2.5) 4.6 (3.1)** 4.1 (3.3)*
Other closed 134 (27.0) 12.1
(8.7)*
1.5 (2.0) 2.0 (2.5) 3.8 (2.9)** 3.4 (3.0)*
Young offenders
institution
67 (13.5) 17.3
(8.5)
1.8 (2.0) 2.8 (2.3) 5.5 (2.8)* 5.8 (2.9)**
Adjustments: age, sex, ethnicity, schizophrenia, drug disorder, and alcohol disorder.
*p0.05, **p0.01. Numbers are Mean (SD).
Table 4
Associations between dimensional scores of Axis II personality disorder criteria and
PCL-R scores.
Axis II disorder Total Factor 1 Factor 2 Factor 3 Factor 4
Interpersonal Affective Lifestyle Antisocial
Avoidant 0.43 1.25 0.01 1.89 0.43
Dependent 0.36 0.24 0.45 0.01 0.33
Obsessivecompulsive 1.59 0.74 1.51 1.29 0.84
Paranoid 0.29 0.23 0.53 0.75 1.11
Schizotypal 1.17 1.68 0.22 1.35 0.80
Schizoid 3.33** 0.49 5.62*** 0.99 1.88
Histrionic 2.67** 2.04* 0.55 2.64** 0.63
Narcissistic 2.47* 6.96*** 2.22* 0.85 0.0 0
Borderline 0.24 0.11 1.40 0.09 0.34
Conduct disorder 4.50*** 1.92 0.76 1.41 7.56***
Adult antisocial 15.0*** 6.10*** 9.48*** 12.4*** 9.86***
Adjustments: age, sex, ethnicity, marital status before prison, drug disorder, alcohol
disorder, affective/anxiety disorder (SCAN diagnosis), psychosis, intercorrelations of
four factors, and comorbid PD scores (factor 4 was not controlled for conduct disorder
and adult antisocial for factor 1, factor 2, and factor 3). The association is presented by
z-score as the partial regression coefcient over its standard error.
*p0.05, **p0.01, ***p0.001.
137J. Coid et al. / International Journal of Law and Psychiatry 32 (2009) 134141
antisocial scores [β(SE)=0.013 (0.006), pb0.05]. There were no
signicant correlations with factors 1 and 2.
Prisoners housed in open prisons had the lowest mean PCL-R
scores (Table 3). Total scores were signicantly higher in local and
dispersal and other closed locations, but not Young Offenders
institutions. This was accounted for by factors 3 and 4, but not factors
1 and 2.
3.3. Correlates with personality disorder and Axis I clinical syndromes
Correlations between criteria scores of individual DSM-IV Axis II
personality disorders and PCL-R scores are demonstrated in Table 4.
Total scores in the combined male and female sample were
signicantly correlated with adult antisocial, conduct disorder,
schizoid, histrionic, and narcissistic scores. Factor 1 (interpersonal)
scores were positively correlated with narcissistic, adult antisocial,
and histrionic scores. The affective factor (F2) correlated positively
with adult antisocial, schizoid and narcissistic scores. Lifestyle (F3)
scores were signicantly correlated with adult antisocial and
histrionic and the antisocial factor (F4) with adult antisocial and
conduct disorder scores.
Table 5 demonstrates that there were no signicant associations
between total PCL-R scores and ICD-10 clinical syndromes measured
using the SCAN. Associations between anxiety and phobic disorders
were conned to factor 3. An association was observed between
schizophrenia and factor 1 but not total scores.
We examined associations between six clinical affective and
anxiety disorders derived from the self-report, CIS-R. No signicant
associations were found between these measures and total scores or
factor scores.
3.4. Substance misuse
Associations between total PCL-R scores and disorders due to
substance use are demonstrated in Table 6. All categories of substance
use disorders except multiple drug use were signicantly associated
with lifestyle and antisocial scores. There were additional associations
between opioid, sedative, and cocaine use disorders and interperso-
nal, and sedative, cocaine, and solvent use disorders and affective
scores.
3.5. Criminal behavior
The age at which prisoners rst appeared in court demonstrated a
strong negative correlation with total PCL-R scores (pb0.001). After
additional adjustments for the other factor scores, age was also
signicantly negatively correlated with factors 1 (pb0.05), 2 (pb0.01),
3(pb0.001), and 4 (pb0.001). Scores were also positively correlated
Table 5
Association between SCAN categories of Axis I affective/anxiety disorder (SCAN diagnosis) and PCL-R scores.
SCAN category Present/
absent
Respondents
N(%)
Total Factor 1 Factor 2 Factor 3 Factor 4
Interpersonal Affective Lifestyle Antisocial
Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD)
Depressive disorder No 407 (82.1) 13.2 (9.47) 1.62 (2.00) 2.35 (2.47) 4.13 (3.11) 3.78 (3.26)
Yes 89 (17.9) 15.7 (8.99) 2.08 (2.05) 2.43 (2.33) 5.10 (3.03) 4.56 (3.24)
Anxiety disorder No 441 (88.9) 13.3 (9.43) 1.65 (1.98) 2.32 (2.43) 4.17 (3.11) 3.92 (3.27)
Yes 55 (11.1) 16.2 (9.03) 2.13 (2.29) 2.69 (2.55) 5.33 (2.94)* 4.64 (3.32)
Phobias No 463 (93.3) 13.5 (9.53) 1.66 (2.01) 2.38 (2.47) 4.20 (3.13) 3.89 (3.27)
Yes 33 (6.7) 15.7 (7.69) 2.27 (2.04) 2.06 (2.15) 5.76 (2.49)* 4.36 (3.28)
Obsessivecompulsive
disorder
No 488 (98.4) 13.4 (9.37) 1.68 (2.02) 2.32 (2.42) 4.25 (3.11) 3.88 (3.27)
Yes 8 (1.6) 23.9 (7.08) 2.88 (1.81) 5.13 (2.53) 7.38 (1.60) 6.38 (2.45)
Schizophrenia No 458 (92.3) 13.2 (9.40) 1.60 (1.95) 2.32 (2.43) 4.20 (3.14) 3.79 (3.23)
Yes 38 (7.7) 18.1 (8.64) 2.92 (2.41)* 2.82 (2.6 6) 5.50 (2.59) 5.53 (3.31)
Brain dysfunction No 485 (97.8) 13.5 (9.42) 1.68 (2.01) 2.36 (2.46) 4.24 (3.11) 3.87 (3.25)
Yes 11 (2.21) 20.1 (7.73) 2.73 (2.10) 2.27 (2.05) 6.82 (2.23) 6.36 (3.14)
Adjustments: age, sex, ethnicity, psychosis, drug disorder, alcohol disorder, and intercorrelations of four factors.
*p0.05.
Table 6
Association between disorders due to substance use (SCAN) and PCL-R scores.
Substance Present/
absent
Respondents
N(%)
Total Factor 1 Factor 2 Factor 3 Factor 4
Interpersonal Affective Lifestyle Antisocial
Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD)
Opioids No 344 (69.4) 11.34 (9.17) 1.52 (1.99) 2.11 (2.42) 3.49 (3.01) 3.11 (3.07)
Yes 152 (30.6) 18.73 (7.90)*** 2.11 (2.03)* 2.93 (2.43) 6.14 (2.49)*** 5.77 (2.94)***
Cannabis No 377 (76.0) 11.86 (9.18) 1.54 (1.98) 2.15 (2.37) 3.66 (3.01) 3.35 (3.12)
Yes 119 (23.0) 19.14 (7.96)*** 2.23 (2.04) 3.03 (2.58) 6.33 (2.53)*** 5.75 (3.06)***
Sedatives/hypnotics No 410 (82.7) 12.46 (9.39) 1.60 (2.00) 2.22 (2.41) 3.84 (3.04) 3.58 (3.23)
Yes 86 (17.3) 19.08 (7.52)*** 2.17 (2.03)* 3.05 (2.54)* 6.49 (2.48)*** 5.58 (2.95)***
Cocaine No 375 (75.6) 11.55 (8.90) 1.43 (1.86) 2.14 (2.41) 3.60 (2.98) 3.26 (3.04)
Yes 121 (24.4) 19.98 (8.11)*** 2.55 (2.25)** 3.06 (2.45)* 6.47 (2.46)*** 5.98 (3.10)***
Stimulants No 374 (75.4) 11.74 (9.25) 1.58 (2.01) 2.14 (2.40) 3.59 (3.00) 3.29 (3.12)
Yes 122 (24.6) 19.31 (7.53)*** 2.07 (2.00) 3.03 (2.48) 6.48 (2.38)*** 5.88 (2.93)***
Solvents No 448 (90.3) 12.92 (9.34) 1.66 (2.04) 2.22 (2.41) 4.06 (3.05) 3.72 (3.24)
Yes 48 (9.7) 19.96 (7.83)*** 2.10 (1.72) 3.67 (2.45)* 6.54 (2.81)** 5.79 (2.94)*
Multiple drug use No 445 (89.7) 12.96 (9.40) 1.65 (2.03) 2.23 (2.42) 4.07 (3.07) 3.74 (3.26)
Yes 51 (10.3) 19.24 (7.68)** 2.16 (1.86) 3.53 (2.40) 6.29 (2.77)** 5.49 (2.90)
Alcohol No 223 (45.0) 10.10 (8.98) 1.38 (1.79) 1.86 (2.34) 3.17 (3.11) 2.77 (3.03)
Yes 273 (55.0) 16.47 (8.81)*** 1.97 (2.15) 2.77 (2.46) 5.23 (2.80)*** 4.87 (3.15)***
Adjustments: age, sex, ethnicity, neurotic, psychosis, alcohol disorder, and drug disorder (for alcohol only).
*p0.05, **p0.01, ***p0.001.
138 J. Coid et al. / International Journal of Law and Psychiatry 32 (2009) 134141
with the number of previous periods of imprisonment (total PCL-R
pb0.001, F1 pb0.01, F2 pb0.01, F3 pb0.001, F4 pb0.001). However,
few signicant associations emerged between PCL-R scores and
individual categories of offending behavior leading to current
imprisonment.
Total PCL-R scores were signicantly associated only with offences
of theft (pb0.05) and (negatively) with drug offences (pb0.01). There
were no independent associations between factor 1 and any category
of offending behavior. Factor 2 was signicantly associated with minor
offences of violence (pb0.05), theft (pb0.05), and criminal damage
(pb0.01). Factor 3 was independently associated only with theft
(pb0.05), and factor 4 with robbery (pb0.05). There were negative
associations between factors 1 (pb0.01), 2 (pb0.05), 3 (pb0.01), and
4(pb0.001) and drug offences.
3.6. Behavioral patterns in prison and the community
There were associations between living off crime prior to imprison-
ment and totalPCL-R (pb0.01), factor 2 (pb0.05),factor 3 (pb0.01), and
factor 4 (pb0.01), but not factor 1 scores. There were no signicant
associations observed with reports of having been the victim of violence
in the family home, sexual abuse, nancial problems, or having been
admitted to a psychiatric hospital and total or individual factor scores.
However, total and individual factor scores were signicantly (pb0.01)
associated with periods of homelessness. There was also an association
between reports of attempted suicide and factor 3 (lifestyle) (pb0.05),
but not with total or other factor scores.
In prison, PCL-R scores were associated with receiving additional
punishments as indicated by signicant associations between being
placed in solitary connement (total pb0.001, F1 pb0.05, F2 pb0.01,
F3 pb0.001, F4 pb0.001) and receiving added days to sentences (total
pb0.01, F1 NS, F2 pb0.05, F3 pb0.05, F4 pb0.01). There were no
associations between PCL-R scores and reports of being physically
assaulted by other prisoners, having belongings stolen, or physical
intimidation to hand over belongings. There was an association
between factor 4 (antisocial) scores and reports of being threatened
with violence (pb0.01), but no associations between this experience
and other factor scores. Total scores (pb0.01), and factors 1 (pb0.05),
2(pb0.05), 3 (pb0.05), and 4 (pb0.01) were associated with reports
of receiving unwanted sexual attentions from other prisoners. There
were no signicant associations with reporting being forced to have
sex.
4. Discussion
4.1. Prevalence of psychopathy among prisoners
Mean PCL-R scores of men in this survey were lower than the large
pooled experimental samples of North American male prison inmates
(M=22.1) described by Hare (2003), and substantively lower than
those for females (M=19.0). They were similar, however, to a pooled
UK male sample (M=16.1; Cooke et al., 2005). Nevertheless, only one
of these UK pooled samples was representative of an entire
correctional jurisdiction, and the North American samples were
more likely to come from high and medium security institutions.
Cooke and Michie (1999) argued that sample selection did not explain
the lower prevalence of psychopathy among Scottish prisoners. They
recommended an adjusted PCL-R cut score of 25 to receive the
diagnosis, but more recently made a further adjustment of approxi-
mately two points (28; Cooke et al., 2005). However, Bolt et al. (2007)
concluded that an adjustment of only half a point is justied.
Nevertheless, the ndings of this study indicate an even greater
discrepancy between England/Wales and Scotland than between
England/Wales and North America. In the only comparable survey of
an entire correctional jurisdiction, Cooke (1994) found the prevalence
of psychopathy at a cut-off of 30 among Scottish sentenced men to be
only half that of equivalent prisoners in England and Wales. Cooke and
Michie (1999) subsequently argued that Scottish prisoners required
higher levels of the underlying trait before certain characteristics
become apparent compared to North American prisoners and forensic
patients, together with the possibility that Scottish psychopaths were
more likely to migrate. This last point received only partial support
from a comparison between Scottish and English-born prisoners using
data obtained in the rst stage of our survey. Prisoners born in
Scotland scored signicantly higher on adult antisocial behavior, had
more previous imprisonments, and were signicantly more criminally
versatile. However, age at rst conviction, and scores on Axis II
personality disorders found to correlate with total PCL-R and factor
scores in this study, demonstrated no signicant differences. This
suggested that if migration was a factor among Scottish-born
individuals in English prisons, it was secondary to their extensive
criminal lifestyle rather than their personalities.
4.2. Distribution of psychopathic traits among prisoners
The continuous distribution of psychopathic traits in the male prison
population of England and Wales was similar tothat observed in pooled,
male experimental samples from North America, but very different for
women prisoners (Hare, 2003) and differed from that of men and
women in the household population of Britain (Coid et al., 2009). This
suggested either that North American female samples were more highly
selected and atypical, or that the criminal justice process leading to
imprisonment in Canada and the USA is more selective of women with
psychopathic traits. The latter is unlikely and representativeness of this
female prison sample from England and Wales, including subcategories
of non-UK born women illegally importing drugs, and those serving
sentences in open prisons, together with the high prevalence of women
with severe mental disorder (Singleton et al., 1998), may have
determined the distribution of psychopathy scores, similar to the half-
normal distribution observed in the general population (Coid et al.,
2009).
Within the prison population of England and Wales, psychopathic
prisoners are more likely to have extreme features along a spectrum,
including younger age, early onset of criminal behavior, prolic
offending, repeated imprisonment, living off crime in the community,
and periods of homelessness. Although prisoners with psychopathic
traits were found among most categories of offending behavior,
including serious offences, we found that they are more likely to be in
prison for common, petty crimes. This differs from the impression
conveyed by experimental samples selected from high security
settings for study of the association between psychopathy and
violence. However, they were less likely to be found in low security
settings, such as open prisons in England and Wales, due to behavioral
problems in prison.
Previously observed associations between assault and robbery
offences (Hart & Hare, 1997) were partly conrmed by specic
associations with individual factors, together with disruptive behavior
in prison, reected in punishments of solitary connement and
addition of days to sentences. Although prisoners with high levels of
psychopathic traits were more likely to report receiving threats of
violence and unwanted sexual attentions, they did not report being
physically victimized, in contrast to prisoners with affective, anxiety,
and psychotic disorders (Coid et al., 2002). Threatening behavior and
sexual overtures may have resulted from their own behavior, and
where psychopathy is more typically associated with intimidation,
victimization, and aggressive homosexual behavior towards other
prisoners (Coid, 1998).
4.3. Demography, intelligence and the 4-factor model
Differential correlations with a series of measures highlight the
multidimensional nature of the construct of psychopathy and suggest
139J. Coid et al. / International Journal of Law and Psychiatry 32 (2009) 134141
the likelihood of multiple etiological determinants. However, the
selected nature of prison populations means that rm conclusions
cannot be drawn regarding the apparent decline in psychopathy with
age in the general population. Nevertheless, we did not observe a
decline in facet 1 in the household survey (Coid et al., 2009), similar to
our observations among prisoners. Older prisoners included more
serving life sentences, including domestic homicides, and more sex
offenders against children. However, our ndings are in accordance
with previous studies investigating the effects of ageing on psycho-
pathy scores using a cross-sectional design. In a study of Harpur and
Hare (1994), it was demonstrated that the traditional factor 2 of the
PCL-R (social deviance) declined with age, whereas the psychopathic
traitsincluding the interpersonal style and affective deciency
(factor 1) remained stable across various age groups. This was
conrmed by Ullrich et al. (2003) who applied the three-factor
model of psychopathy in a sample of German prisoners and found an
age related decline only for factor 3 (impulsivity). These ndings were
not unexpected against the background of the assumption of
psychopathic traits(interpersonal and affective) representing the
core personality traits (which are supposed to demonstrate stability),
whereas traits constituting social deviance(impulsiveness and
antisocial) are more strongly behavior-related.
Non-UK born prisoners showed no increased tendency to psycho-
pathy, perhaps because a proportion had become involved in illegal
importation of drugs for economic reasons rather than as part of a
criminal lifestyle. Associations between interpersonal and lifestyle
factors and black ethnicity were of interest and correlated with certain
differences in specic Axis-II personality disorder categories between
white and black prisoners previously observed (Coid et al., 2002). In
general, psychopathic traits did not show strong associations with
social class, although interpersonal factor 1 scores were signicantly
higher among prisoners of higher social class. Finally, there was no
association between psychopathy and being a remanded prisoner
after adjusting for age.
Low intelligence is an important predictor of offending in the
general population (Farrington,1997), corresponding to our nding of
associations between low verbal intelligence and an antisocial and
impulsive/irresponsible lifestyle, but not with the interpersonal or
affective factors. Previous research in an adolescent forensic sample
demonstrated a positive association between verbal intelligence and
factor 1, and an inverse correlation with factor 2 (Salekin, Neumann,
Leistico, & Zalot, 2004). Similarly, a positive correlation with verbal
intelligence was found with factor 1 and an inverse correlation with
affective (factor 2) and lifestyle items (factor 3) in a study of
psychiatric patients (Vitacco et al., 2005).
4.4. Correlates with personality disorder and clinical syndromes
Consistent with research on the factor structure of the PCL-R (Hare,
2003; Neumann et al., 2005), we found that the four factors are highly
inter-correlated. We used partial correlations to adjust for potential
confounding which revealed their independent associations. We
therefore found fewer associations with Axis II personality scores
than in previous forensic samples (Hare, 2003; Hart, Cox, & Hare,
1995). However, our ndings were generally similar, in that adult
antisocial disorder, conduct disorder, narcissistic and histrionic scores
derived from diagnostic instruments correlated most strongly with
psychopathy (Hare, 2003; Hart et al., 1995). Schizoid personality
disorder was also associated. The glibness and supercial charm
characterizing factor 1 was, as expected, strongly associated with
narcissistic and histrionic personality disorder. The nding of an
association between factor 2 and schizoid personality disorder would
be explained by affective dysfunction characterizing this factor.
Impulsive and irresponsible lifestyle characteristics of factor 3
correlated with histrionic personality disorder in this prison sample.
Conduct disorder was associated only with factor 4 and total scores,
suggesting the importance of an early onset of behavioral disturbance
correlating with the antisocial factor.
In this sample, all factors were highly correlated with antisocial
behavior in adulthood as measured by antisocial personality disorder
scores after age 15. However, lack of a correlation with borderline
personality disorder was unexpected, particularly as it had previously
been found in the British household population (Coid et al., 2009).
Among prisoners, this association was confounded by adult antisocial
behavior.
According to Cleckley's (1941) original observations, psychopathic
individuals show neither psychoneurotic nor psychotic symptoms.
Our ndings only partly support this. There were relatively weak
associations observed between schizophrenia and factor 1, and
between phobic and anxiety disorders and factor 3. In contrast to a
household survey using the CIS-R (Coid et al., 2009), no associations
were found with obsessivecompulsive disorder. This suggests that at
lower levels of psychopathic traits there may be weak associations
with OCD at the general population level, but among populations with
higher trait levels and more individuals with psychopathy, these
associations are absent.
On the other hand, the observation of an association between
anxiety and phobic disorders is of some interest. The relationship
with psychopathy has been controversial. It is probable that the
associations we found were robust because the overall prevalence of
all anxiety d isorders in this prison sa mple was very high (Singleton et
al., 1998), imprisonment, and especially remand, being a highly
stressful si tuational experience fo r most prisoners. It has been argu ed
that low anxiousness should itself be included as a criterion for
psychopathy. But also that there should be a distinction between
primary(with low anxiety) and secondary(with high anxiety)
psychopaths (see Widiger, 2006). It may also be important to
distinguish between fearfulness and anxiousness. Persons high in
fearlessness engage in substantial risk taking, but then may
experience anxiety in relation to their producing and encountering
highly stressful events (Frick, Lilienfeld, Ellis, Loney, & Silverthorn,
1999; Lilienfeld,1994). However, it is not clear whethe r thrill-seeking
behavior is best understood as reecting fearlessness, as an
impulsive disposition, or both. The association with anxiety and
phobic disorders with factor 3 in this study suggests that this may be
associated through impulsiveness.
All previous studies examining substance misuse and psychopathy
have demonstrated associations (Hare, 2003; Schubert, Wolf, Patter-
son, Grande, & Pendleton, 1988). Psychopathic individuals in forensic
samples are also relatively more likely to have problems due to alcohol
misuse than other offenders, but in the British household population
the association was weak (Coid et al., 2009).
Few specic associations were found between substance misuse
and either factor 1 or 2, suggesting that the effects of these factors on
choice of substances were weak. Overwhelmingly, substance misuse
was part of an antisocial and irresponsible lifestyle.
4.5. Methodological limitations
Although this sample was representative of the prison population
of England and Wales, it is important to consider the highly selected
nature of any prison population when examining correlates of
psychopathy; most prisoners in any country are young, male, of low
social class or previously unemployed. A disproportionate number in
the UK are black (Coid et al., 2002), and characterized by very high
levels of psychiatric morbidity (Singleton et al., 1998). These factors
must be taken into account when comparing ndings with those in
the general population (Coid et al., 2009) and in making comparisons
with ndings from studies using experimental samples, often selected
from unrepresentative prison locations deliberately chosen in antici-
pation that there will be a raised prevalence of psychopathic
individuals for study.
140 J. Coid et al. / International Journal of Law and Psychiatry 32 (2009) 134141
The sampling frame ensured that the survey was highly repre-
sentative and attrition rate at both stages was low. Few prisoners
refused to participate, attrition being largely accounted for by rapid
movement of some prisoners between institutions or unexpected
release. However, this study examined correlates with psychopathic
traits in the prison population and not with psychopathy as
categorically dened. Furthermore, collateral information on previous
behavior was relatively limited and previous psychological and
psychiatric reports were rarely available to interviewers. This may
have biased PCL-R scores so that the true prevalence was higher than
we have recorded. The clinically trained raters had information about
previous criminal behavior, together with measures from the rst
stage of the survey. However, PCL-R ratings were taken in the context
of an interview that gave primacy to the diagnosis of ICD-10 clinical
syndromes. In the context of recommendations for administration of
the instrument (Hare, 2003), they must be considered limited.
5. Summary and conclusion
Our study demonstrated that the prevalence of psychopathy is
remarkably higher in male than in female prisoners. The same held
true for observations of dimensional scores of the interpersonal,
affective, impulsive and antisocial components. Although the pre-
valences and scores are higher than in a representative non-forensic
and non-psychiatric community sample, the pattern of gender
differences was identical (Coid et al., 2009). These ndings suggest
that psychopathy measured with the PCL-R and PCL: SV does not
reect a different construct depending on the population under study
but a more severe condition among individuals detained in prison.
The differentiation of psychopathy into its sub-components
revealed important associations about the differential associations of
the interpersonal, affective, lifestyle and antisocial factors and various
outcome measures. Every factor demonstrated a specic pattern of
relationship with co-morbid psychopathology and social and beha-
vioral problems relevant for a better understanding of the construct.
References
Ammons, R. B., & Ammons, C. H. (1962). The Quick Test: Provisional manual. Psycho-
logical Reports,11,111161.
Andersen, H. S., Sestoft, D., Lillebaek, T., Mortensen, E. L., & Kramp, P. (1999). Psychopathy
and psychopathological proles in prisoners on remand. Acta Psychiatrica Scandina-
vica,99(1), 3339.
Assadi,S.M.,Noroozian,M.,Pakravannejad,M.,Yahyazadeh,O.,Aghayan,S.,Shariat,S.V.,etal.
(2006). Psychiatric morbidity among sentences prisoners: Prevalence study in Iran.
British Journal of Psychiatry,188 ,159164.
Blair, J., Mitchell, D., & Blair, K. (2005). The psychopath, emotion, and the brain. Malden,
MA: Blackwell Publishing.
Blonigen, D. M., Hicks, B. M., Krueger, R. F., Patrick, C. K., & Lacono, W. G. (2005).
Psychopathic personality traits: Heritability and genetic overlap with internalizing
and externalizing psychopathology. Psychological Medicine,35,637648.
Bolt, D., Hare, R. D., & Neumann, C. S. (2007). Score metric equivalence of the PCL-R
across North American and UK criminal offenders: A critique of Cooke et al. (2005)
and new analysis. Assessment,14(1), 4456.
Cleckley, H. (1941). The mask of sanity. St Louis, MO: Mosby.
Coid, J. W. (1998). The management of dangerous psychopathy in prison. In T. Millon, E.
Simonsen, M. Birket-Smith, & R. D. Davis (Eds.), Psychopathy, Antisocial, Criminal
and Violent Behaviour (pp. 431457). New York: Guilford Press.
Coid, J., Petruckevitch, A., Bebbington, P., Brugha, T., Bhugra,D., Jenkins, R., et al. (2002).
Ethnic differences in prisoners. I: Criminality and psychiatric morbidity. British
Journal of Psychiatry,181,473480.
Coid, J., Ullrich, S., Yang, M., Ullrich, S., Roberts, A., & Hare, R. (2009). Prevalence and
correlates of psychopathic traits in a household population. International Journal of
Law and Psychiatry,32,6573.
Cooke, D. J. (1994). Psychological disturbance in the Scottish prison system: Prevalence,
precipitants, and policy. Edinburgh: Scottish Home and Health Department.
Cooke, D. J. (1996). Psychopathic personality in different cultures. What do we know?
What do we need to nd out? Journal of Personality Disorders,10,2340.
Cooke, D. J., & Michie, C. (1999). Psychopathy across cultures: Scotland and North
America compared. Journal of Abnormal Psychology,108,5868.
Cooke, D. J., & Michie, C. (2001). Rening the construct of psychopathy: Towards a
hierarchical model. Psychological Assessment,13,171188.
Cooke, D. J., Michie, C., Hart, D., & Clark, D. (2005). Assessing psychopathy in the UK:
Concerns about cross-sectional generalisability. British Journal of Psychiatry,186,
335341.
Cornell, D. J., Warren, J., Hawk, G., Stafford, E., Oram, G., & Pine, D. (1996). Psychopathy
in instrumental and reactive violent offenders. Journal of Consulting and Clinical
Psychology,64, 783790.
Farrington, D. (1997). Human development and criminal careers. In N. Maguire, R.
Morgan, & R. Reiner (Eds.), The Oxford Handbook of Criminology (pp. 361408).,
2nd edition Oxford: Oxford University Press.
First, M. B., Gibbon, M., Spitzer, R. L., & Williams, J. B. W. (1997). Structured clinical
interviews for DSM-IV Axis-II personality disorders. Washington: American Psychia-
tric Press.
Frick, P. J., Lilienfeld, S. O., Ellis, M., Loney, B., & Silverthorn, P. (1999). The association
between anxiety and psychopathy dimensions in children. Journal of Abnormal
Child Psychology,27, 383392.
Hare, R. D. (1991). The Hare Psychopathy Checklist Revised. Toronto, Ontario: Multi-
Health Systems.
Hare, R. D. (2003). Manual for the Revised Psychopathy Checklist, (2nd edition) Toronto,
ON, Canada: Multi-Health Systems.
Hare,R. D., & Neumann, C. N.(2006). The PCL-R assessment of psychopathy:Development,
structural properties, and new directions. In C. Patrick (Ed.), Handbook of Psychopathy
(pp. 5888). New York: Guilford.
Hare, R. D., Clark, D., Grann, M., & Thornton, D. (2000). Psychopathy and the predictive
validity of the PCL-R: An international perspective. Behavioral Sciences and the Law,
18, 623645.
Harpur, T. J., & Hare, R. D. (1994). The assessment of psychopathy as a function of age.
Journal of Abnormal Psychology,103,604609.
Hart, S. D., Cox, D. N., & Hare, R. D. (1995). The Hare Psychopathy Checklist: Screening
Version. New York: Multi-Health Systems Inc.
Hart, S. D., & Hare, R. D. (1997). Psychopathy: Assessment and association with criminal
conduct. In D. Stoff, J. Breiling, & J. Maser (Eds.), Handbook of Antisocial Behavior
(pp. 2235). New York, NY: Wiley & Sons.
Hill, C. D., Neumann, C. S., & Rogers, R. (2004). Conrmatory factor analysis of the
psychopathy checklist: Screening version in offenders with Axis I disorders. Psy-
chological Assessment,16,9095.
Lewis, G., & Pelosi, A. J. (1990). Manual of the Revised Clinical Interview Schedule (CIS-R).
London: Institute of Psychiatry.
Lilienfeld, S. O. (1994). Conceptual problems in the assessment of psychopathy. Clinical
Psychology Review,14,1738.
Moran, P., (1999). Antisocial personality disorder. An epidemiological perspective. London.
Royal College of Psychiatrists. Gaskell.
Neumann, C. S., Vitacco, M. J., Hare, R. D., & Wupperman, P. (2005). Deconstructing the
reconstructionof psychopathy: A comment on Cooke, Michie, Hart, & Clark.
Journal of Personality Disorders,19, 624640.
Salekin, R. T., Neumann, C. S., Leistico, A. -M. R., & Zalot, A. A. (2004). Psychopathy in
youth and intelligence: An investigation of Cleckley's hypothesis. Journal of Clinical
Child and Adolescent Psychology,333,731742.
Schubert, D. S., Wolf, A. W., Patterson, M. B., Grande, T. P., & Pendleton, L. (1988). A
statistical evaluation of the literature regarding the associations among alcoholism,
drug abuse, and antisocial personality disorder. International Journal of Addiction,
23,797808.
Singleton, N., Meltzer, H., Gatward, R., Coid, J., & Deasy, D. (1998). Psychiatric Morbidity
among prisoners in England and Wales. London: The Stationery Ofce.
Ullrich, S., Paelecke, M., Kahle, I., & Marneros, A. (2003). Categorical and dimensional
assessment of psychopathy in German offenders. Prevalence, gender differences
and aging. Der Nervenarzt,74,10021008.
Viding, E., Blair, R. J. R., Moftt, T. E., & Plomin, R. (2005). Evidence for substantial
genetic risk for psychopathy in 7-year-olds. Journal of Child Psychology and
Psychiatry,46, 592597.
Vitacco, M. J., Neumann, C. S., & Jackson, R. L. (2005). Testing a four-factor model of
psychopathy and its association with ethnicity, gender, intelligence, and violence.
Journal of Consulting and Clinical Psychology,73,466476.
Widiger, T. A. (2006). Psychopathy and DSM-IV psychopathology. In C. J. Patrick (Ed.),
Handbook of Psychopathy (pp. 156171). New York: Guilford Press.
Wing, J. K., Babor, T., Brugha, T., Burke, J., Cooper, J. E., Giel, R., et al. (1990). SCAN:
Schedules for Clinical Assessment in Neuropsychiatry. Archives of General Psychiatry,
47(6), 586593.
World Health Organisation Division of Mental Health (1999). SCAN: Schedules for
Clinical Assessment in Neuropsychiatry, Version 2.1. Geneva: World Health
Organisation.
141J. Coid et al. / International Journal of Law and Psychiatry 32 (2009) 134141
... Nevertheless, considerable differences (3-73%) in psychopathic prevalence rates among prisoners have been reported in different countries [14]. As documented in most relevant studies, prisoner populations are most frequently involved in personality and substance abuse disorders [8]. ...
... Several studies suggested the present applicability of certain male-related risk factors to females and the inability of violence committed by younger males and females [14,17]. Prisoners usually come from socioeconomically deprived families with criminal behavior and mostly psychiatric disorder backgrounds [18]. ...
... These findings are aligned with other studies [10,14,15,21]. Criminal behaviors are very complex and intertwined with political, legal, constabulary, medical, psychological, and social issues [12]. ...
Full-text available
Article
Personality disorders can lead to difficult social or occupational functional processes rooted in chronic maladaptive thoughts, feelings, and behaviors. This study aimed to investigate factors of personality disorder in prisoners from the central prison of Sanandaj, Iran. We conducted a cross-sectional study, which included all prisoners in the central prison of Sanandaj, Iran. The study sample includes 343 prisoners, of which 329 were male, and 14 were female, selected by randomized multistage sampling method. Participants filled in Millon's multi-axis clinical questionnaire. The data were analyzed using Chi-square, multiple logistic regression, and bootstrap analysis. There were 183 participants without personality disorders (53.4%) and 99 participants (28.9%) with cluster B personality disorders (narcissistic, histrionic, anti-social, and borderline). Male gender (OR=0.07) and elementary education level (OR=0.18) have a significant relationship with cluster A personality disorders (paranoid, schizoid, schizotypal). Cluster B has a significant relationship only with the elementary education level (OR=0.27). Cluster C (avoidant, dependent, obsessive-compulsive personality disorder) has a significant relationship with male gender (OR=0.20), elementary education level (OR=0.30), unemployment (OR=2.64), theft crime types (OR=0.38) and disputes and assaults (OR=0.18). Based on these results, psychological and psychiatric interventions in prisoners are suggested.
... Within this group, about one-third meet criteria for callous-unemotional (CU) traits in youth and psychopathy (ASPD+P) in adulthood, offending earlier, more widely, and more severely than those without psychopathy (ASPD-P) (Kosson et al., 2006). Despite the high prevalence of ASPD+/-P in forensic and penal settings (Fazel and Danesh, 2002;Coid et al., 2009b) and the large social and financial burden associated with ASPD+/-P (Heeks et al., 2018), evidence for successful treatments is lacking Khalifa et al., 2020), and understanding of causative mechanisms remains limited. ...
... It is, therefore, possible that the TriPM model may explain injurious behaviours in prisoners, amongst whom high levels of psychopathy are particularly common. 37 However, to our knowledge, there is no published research outside the US examining the association between the TriPM constructs and injurious and suicidal behaviours in incarcerated populations. ...
Article
Self-injurious behaviours and suicide attempts among incarcerated people are higher when compared with the general population. The current research is designed to examine the prevalence of self-injurious behaviours and suicide attempts among incarcerated males in Spain, and the factors associated with these behaviours. 201 men imprisoned in Spain completed anonymous self-report measures of demographic variables (e.g., age, marital status), the Triarchic Psychopathy Model (boldness, meanness, and disinhibition) and impulsive/premeditated aggression and aggressive behaviour (including physical, verbal, hostility, and anger). Official records were also consulted for information related to criminality (i.e., prior imprisonment and type of offence). Overall, 37.5% reported having injured themselves at a time of anger or despair and 24.9% reported attempting suicide at any point in their lives. Lower scores in meanness, higher scores on impulsive aggression, serious illnesses and previous imprisonment were associated with an increased risk of self-injurious and/or suicide attempting behaviours. Being married acted as a protective factor of attempting suicide. Trait variables related to psychopathy (lower meanness) and aggression (higher impulsive aggression) were associated with increased frequency of suicide attempts in prisoners even after controlling for personal and criminal variables. The prevention of high rates of suicide in prisons needs the treatment of those cognitive variables in addition to more contextual personal and criminal variables.
... Not surprisingly, these personality features are more prevalent in prison populations (e.g. Psychopathy, Coid et al., 2009). ...
Article
People deceive for different reasons, from avoiding interpersonal conflicts to preserving, protecting, and nurturing interpersonal relationships, and to obtaining social status and power. A growing body of research highlights the role of personality in both deception detection and production, with a particular focus on high Dark Triad (DT) traits (Narcissism, Machiavellianism and Psychopathy), for their shared tendency to engage in unethical self-benefitting behaviors, despite negative consequences for others. The main goal of the current scoping review was to bring together the studies investigating self-reported and performance-based deception production and detection performances, as presented in individuals characterized by high DT traits and point out the possible contribution of DT to deception research. To do so, we identified the relevant studies documenting the similarities and discrepancies between the three personality traits and presented their results, based on the procedure used for deception assessment: subjective or objective measurements for production / detection. Then, we discussed possible explanatory mechanisms for inter-individual differences in lie detection / production and argue for the contribution of DT to deception research beyond the typical personality models, particularly for the antisocial character of deception.
Article
Are psychopaths morally responsible? Should we argue with them? Remonstrate with them, blame them, sometimes even praise them? Is it worth trying to change them, or should we just try to prevent them from causing harm? In this book, Jim Baxter aims to find serious answers to these deep philosophical questions, drawing on contemporary insights from psychiatry, psychology, neuroscience and law. Moral Responsibility and the Psychopath is the first sustained, book-length philosophical work on this important and fascinating topic, and will be of deep interest and importance to researchers in these fields – not to mention anyone who has had to interact with a psychopath in their everyday life.
Article
Using recently published general population studies, this chapter updates an earlier Moran review on the epidemiology of ASPD. It examines the prevalence, comorbidity, psychosocial functioning, impact on health services and cultural differences of those with ASPD and, to a lesser extent, of those with psychopathy and on prisoners. It finds that the prevalence of ASPD ranges between 0.6% and 4.3% in men (which is significantly higher than in women). It has multiple comorbidities – both with other mental disorders (especially with substance misuse and affective disorder) together with other Cluster B PDs. Despite significant mental health and psychosocial impairment, those with ASPD rarely seek treatment for their disorder. There are marked cultural differences in the prevalence of ASPD with studies in the United States showing significantly higher rates than in Europe or in Asia. The prevalence of ASPD among prisoners is c. 55% among men and 31% among women with community prevalence rates for psychopathy ranging from 0.6% in the United Kingdom and 1.0% in the United States. Both ASPD and psychopathy have a positive association with both increased mortality and suicidality, suggesting that, together with their other impairments, they are a group that require mental health interventions.
Chapter
Psychopathy and psychosis are often misunderstood and incorrectly conflated. We examine the relationship between psychopathy and various psychotic and psychotic-spectrum disorders. The paucity of literature exploring the intersection of psychopathic traits and psychosis is surprising given that each domain is the subject of a rich clinical research literature. We first examine the relationship between psychopathy and psychotic-spectrum disorders, with emphases on schizophrenia, metacognition, violence, and patients within secure forensic settings. Most current literature focuses specifically on the relationship between psychopathy and schizophrenia. We also explore the relationship between psychopathy and personality disorders related to psychosis including Cluster A personality disorders. Lastly, we provide recommendations for future research. Specifically, more etiological factors should be considered and the possible link between psychopathy and a broader range of psychotic disorders should be examined. The relationship between psychopathy and schizotypal personality appears to be a particularly informative avenue for future research.KeywordsMetacognitionPsychopathyPsychosisSchizophreniaSchizotypalViolence
Full-text available
Article
Although psychopathy is widely conceptualised as a mental disorder, some researchers question the maladaptive nature of psychopathy, and argue that it might be advantageous from an evolutionary point of view. According to this view psychopathy can be seen as an evolutionary adaptative strategy that relies on deception and manipulation to gain short-term reproductive benefits. Psychopathy is also identified as a fast life strategy in response to early life stress and an adaptation to harsh environments. This paper investigates the evidence that psychopathic traits are adaptive, while also addressing the limitations of current evolutionary models of psychopathy based on frequency-dependent selection and life-history theory. We review recent studies on the fitness correlates of psychopathy and find that psychopathic traits present potential adaptive trade-offs between fertility and mortality, and offspring quantity and quality. On a proximate level, individual differences in stress reactivity and environmental risk factors in early development predispose to psychopathy through gene-environment interactions. We propose that environmental, developmental, social and cultural factors can mediate the relationship between psychopathic traits and fitness and therefore should be considered to make accurate predictions on the adaptive potential of psychopathy. We end by outlining gaps in the literature and making recommendations for future evolutionary research on psychopathy.
Full-text available
Article
Previous research has uncovered medical and psychological effects of anabolic–androgenic steroid (AAS) use, but the specific relationship between AAS use and risk-taking behaviors as well as between AAS use and psychopathic tendencies remains understudied. To explore these potential relationships, we anonymously recruited 492 biologically male, self-identified bodybuilders (median age 22; range 18–47 years) from online bodybuilding fora to complete an online survey on Appearance and Performance Enhancing Drug (APED) use, psychological traits, lifestyle choices, and health behaviors. We computed odds ratios and 95% confidence intervals using logistic regression, adjusting for age, race, education, exercise frequency, caloric intake, and lean BMI. Bodybuilders with a prior history of AAS use exhibited heightened odds of psychopathic traits, sexual and substance use risk-taking behaviors, anger problems, and physical problems compared to those with no prior history of AAS use. This study is among the first to directly assess psychopathy within AAS users. Our results on risk-taking, anger problems, and physical problems are consistent with prior AAS research as well as with existing frameworks of AAS use as a risk behavior. Future research should focus on ascertaining causality, specifically whether psychopathy is a risk associated with or a result of AAS use.
Article
In this paper, we build on a robust literature on push and pull factors to focus on the personality traits and values of individuals involved in organized crime. We distinguish organized crime from other kinds of criminal activity and recruit a unique sample of non‐incarcerated individuals verified by the Danish National Police to be involved in organized crime. We use comprehensive standardized psychological assessments of their big five personality traits, maladaptive dark traits and core values and drivers to compare them to an adult norm group. Danish individuals involved in organized crime are much less emotionally stable (d = 1.84), ambitious and self‐confident (d = 1.50), agreeable (d = 0.87) and conscientious (d = 0.65) than the norm group. At the same time, they have substantially higher scores on all but one of the 11 dark traits (Cohen's d ranging from 0.39 to 3.10). They are characterized by a high need for security (d = 1.14) as well as material (d = 0.96) and financial success (d = 0.81). While these patterns fit results previously found in the criminological literature, a latent class analysis reveals two separate groups. A subset of one third of our sample had somewhat less depressed scores on the big five and more moderate scores on the dark traits, indicating more adaptive personality structures. We consider this novel finding in terms of potential exits from a milieu of organized crime.
Article
A detailed account is given of the development, standardization, and evaluation of The Quick Test, a brief individual intelligence test based on perceptual-verbal performance. Three single forms, each consisting of 50 word-items based on one plate with four line drawings, cover abilities from the two-year level to the superior-adult level. Three to ten minutes are required to administer the QT to any person who can see the drawings, hear or read the word items, and give a yes-no signal. Standardization was based on the responses of 458 children and adults, a sample controlled on a simultaneous quota basis for age, sex, educational level and own or father's occupation. Single forms and combination forms (1+2, 2+3, 1+3, 1+2+3) were found to be highly reliable (.66 to .95). These estimates were corroborated in several independent studies ( N = 111), as were validity estimates (.77 to .96, correlations with the revised Stanford-Binet and the Wechslers). MA, IQ, and percentile norms are included as well as a discussion of practical problems in testing, suggestions for minimizing such problems, and ideas for further research. It is concluded that the QT is quite satisfactory for school, clinical, and research use, and since the QT shows strong correlation with the Full-Range Picture Vocabulary Test, the two tests can be used interchangeably.
Article
Psychopathy is characterized by diverse indicators. Clinical accounts have emphasized 3 distinct facets: interpersonal, affective, and behavioral, Research using the Psychopathy Checklist-Revised (PCL-R), however, has emphasized a 2-factor model, A review of the literature on the PCL-R and related measures of psychopathy, together with confirmatory factor analysis of PCL-R data from North American participants, indicates that the 2-factor model cannot be sustained. A 3-factor hierarchical model was developed in which a coherent superordinate factor, Psychopathy, is underpinned by 3 factors: Arrogant and Deceitful Interpersonal Style, Deficient Affective Experience, and Impulsive and Irresponsible Behavioral Style. The model was cross-validated on North American and Scottish PCL-R data, Psychopathy Screening Version data, and data derived from the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994) antisocial personality disorder field trial.
Article
Differences in the prevalence and presentation of psychopathic personality disorder between North America and Scotland were evaluated R D. Hare's (1991) Psychopathy Checklist-Revised ratings obtained from a sample of 2,067 North American male prisoners and forensic patients were compared with ratings obtained from 246 Scottish male prisoners. Item response theory methods were used to examine differences in the performance of items and to equate the scale across settings. The items had equal relevance to the description of psychopathic personality disorder in both settings; however, the Scottish prisoners had to have higher levels of the underlying latent trait before certain characteristics became apparent. The prevalence of the disorder appears to be lower in Scotland Explanations for the observed differences in terms of enculturation, socialization, and migration are explored.
Article
Das Persönlichkeitskonstrukt der "psychopathy" ist im Rahmen der Legalprognose von großer Evidenz. Eine Vielzahl empirischer Untersuchungen der letzten 20 Jahre belegt dessen prädiktive Valenz im Hinblick auf die Gefährlichkeit eines Straftäters. Untersuchungen im deutschsprachigen Raum gibt es jedoch vergleichsweise nur wenige. In der vorliegenden Arbeit wurde eine repräsentative, nichtselektierte Stichprobe von 416 Straftätern mittels der Screening-Version der Psychopathy-Checklist (PCL:SV) untersucht. "Psychopathy" wurde dabei sowohl kategorial als auch dimensional in Anlehnung an das Drei-Faktoren-Modell nach Cooke und Michie erfasst. Die Prävalenz von High-Scorern war deutlich geringer als in nordamerikanischen Untersuchungen, was auf interkulturelle Differenzen schließen lässt. Geschlechtseffekte zeigten sich dahingehend, dass männliche Straftäter auf allen drei Faktoren deutlich höhere Werte erreichten. Hinsichtlich verschiedener Altersgruppen zeichneten sich keine Unterschiede im betrügerisch-manipulativen interpersonellen Stil sowie dem defizitären affektiven Erleben ab, wohingegen der impulsive und unverantwortliche Lebensstil mit zunehmendem Alter geringer ausgeprägt war. Die Ergebnisse weisen somit auf schon berichtete interkulturelle Unterschiede in der Auftretenshäufigkeit von "psychopathy" und gehen konform mit empirischen Befunden zu sowohl Geschlechts- als auch Alterseffekten im Hinblick auf dieses Konstrukt.
Article
Psychopathic personality disorder is present in many cultures. A cross-cultural approach to the study of psychopathic personality disorder may advance our understanding of its etiology. Speculation concerning the importance of enculturation and socialization in the etiological process is presented. Data comparing differences in the prevalence of psychopathic personality disorder between Scotland and North America are reviewed: it is concluded that significant differences exist. Four empirical strategies, designed to enhance our understanding of cross-cultural differences, are proposed.