Criminal Behaviour and Mental Health (2014) Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/cbm.1923
Anti-social personality characteristics and psychotic symptoms: Two pathways associated with offending in schizophrenia
JOSANNE D. M. VAN DONGEN1,2, NICOLE M. L. BUCK1,3, MARKO BARENDREGT4,5, NICO M. VAN BEVEREN1,6, EDWIN DE BEURS4,5 AND HJALMAR J. C. VAN MARLE1, 1Department of Forensic Psychiatry, Erasmus University Medical Center, Rotterdam, The Netherlands; 2Institute of Psychology, Erasmus University Rotterdam, Rotterdam, The Netherlands; 3Forensic Psychiatric Center De Kijvelanden, Rhoon, The Netherlands; 4Department of Research & Development, Netherlands Institute for Forensic Psychiatry and Psychology (NIFP), Utrecht, The Netherlands; 5Foundation Benchmark GGZ, Bilthoven, The Netherlands; 6Delta Psychiatric Center, Poortugaal, The Netherlands ABSTRACT Background Several research groups have shown that people with schizophrenia who offend do not form a homogenous group. A three-group model claimed by Hodgins proposes distinguishing between people who start offending before the onset of psychosis (early starters), after psychosis onset but at age 34 years or under (late starters) and after psychosis onset but at age 35 years or older (late first offenders). Aims This study aimed to test the hypotheses (1) that the personality of early starters and non-psychotic offenders would be similar, but different from either late-starter group; (2) that the late-starter groups would be more likely to have positive psychotic symptoms than non-criminal patients with schizophrenia; and (3) that symptom types would differentiate the psychotic groups. Methods A retrospective file study was conducted on cases of 97 early starters, 100 late starters and 26 late first offenders all drawn from the Netherlands Institute of Forensic Psychiatry and Psychology (NIFP) archives 1993–2008, 115 non-psychotic offenders from 2005–2008 NIFP archives and 129 patients with schizophrenia and no criminal history from one general service in Rotterdam. Results Early starters closely resembled the non-psychotic offenders in their premorbid anti-social personality characteristics. The two late-onset offending psychosis groups
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Van Dongen et al.
were more likely to have persecutory and/or grandiose delusions than non-offenders with psychosis, but so were the early starters. Implications In a first study to compare subgroups of offenders with psychosis directly with non-psychotic offenders and non-offenders with psychosis, we found such additional support for a distinction between early and late starters with psychosis that different treatment strategies would seem indicated, focusing on personality and substance misuse for the former but psychotic symptoms for all. It remains to be seen whether the higher rate of alcohol misuse amongst late first offenders is a fundamental distinction or a function of age difference. Copyright © 2014 John Wiley & Sons, Ltd. Introduction Recent reviews have shown that there is a small but significant relationship between psychosis and (violent) criminal behaviour (e.g. Douglas et al., 2009; Fazel et al., 2009). The relationship between schizophrenia and violent crime, including homicide (Eronen et al., 1996; Schanda et al., 2004; Large et al., 2009), may be stronger than that between schizophrenia and non-violent crimes (e.g. Lindqvist and Allebeck, 1990; Coté and Hodgins, 1992; Belfrage, 1998). A number of researchers have suggested that there are distinct subgroups of offenders with schizophrenia, subdividing variously according to age of onset of offending and/or co-morbidity of personality disorder. Hodgins (2008, 2009) has suggested three types of offenders with schizophrenia: early starters, late starters and late first offenders. Early starters show anti-social personality characteristics early in life, which may be reflected in a childhood history of conduct problems (Hodgins and Coté, 1993; Hodgins, 1995; Hodgins et al., 1995). They may be similar to the ‘life course persistent’ delinquent group defined by Moffitt (1993), except for the fact that they also develop schizophrenia. Late starters, also called ‘adult starters’ (Hodgins, 1995; Kratzer and Hodgins, 1999), usually start offending in adulthood after the onset of a major mental disorder (e.g. schizophrenia). Their criminal behaviour is more likely to be attributable to cognitive and perceptual (positive) symptoms of the disorder (Hodgins, 1995; Hodgins et al., 1995). Hodgins (2009) also claims that once late starters have started their criminal behaviour, they will repeatedly engage in criminal behaviour and violence towards others, although others have shown that people with schizophrenia desist earlier than people in the general population (Lindqvist and Allebeck, 1990) or than other psychiatric patients (Wessely et al., 1994). A further subgroup of late first offenders, who suddenly commit a very serious offence after the onset of their schizophrenia, has also been suggested (Hodgins, 2008, 2009; Pedersen et al., 2010). We (Van Dongen et al., 2014) were the first independent group to study the late first offender group within this three-group typology and found no differences between the three offender groups regarding positive psychotic symptoms. This
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Pathways of offending in schizophrenia
seems to be inconsistent with the hypothesis that the offending of late starters and late first offenders is attributable to positive psychotic symptoms, whereas that of the early starters is not. The main aim of our present study, therefore, was to test the relative role of personality and positive symptoms of schizophrenia within one sample of patients and non-psychotic offenders. Our first hypothesis was that early starters would resemble offenders without schizophrenia with regard to having an anti-social personality trait, but differ in this from late starters and late first offenders with schizophrenia. We also retested what is, essentially, the alternative hypothesis that late starters and late first offenders would be more likely to have positive psychotic symptoms than the early-start offenders. Finally, we hypothesised that late starters and late first offenders would differ from non-offenders with schizophrenia on type of delusions and persecutory delusions in particular. Method Participants This records study drew on 367 clinical and/or penal records, representing the following groups: 97 people with a schizophrenia spectrum disorder and offending, the offending having started before the onset of even any prodrome of the illness; the latter was defined as onset of any disturbed behaviour, including loss of motivation, anhedonia, anger, irritability, attention problems and/or social withdrawal (see Yung and McGorry, 1996); the first offence was the date of the first conviction according to the official criminal record; 100 people with the two conditions who had started offending only after the onset of the illness but whose age was 34 years or under at the time of the first offence; 26 people similar to the 100 but whose offending had started at age 35 years or older; 115 repeat offenders with no psychotic illness; and 129 patients with schizophrenia and no criminal record. The aim had been to have a sample of 150 cases per group, but it became clear that the sample was limited by eligible cases available. Files for the offenderpatient groups were retrieved from the Observation Clinic of the Ministry of Justice (Pieter Baan Center; PBC) in Utrecht, the Netherlands, which is part of the Netherlands Institute of Forensic Psychiatry and Psychology (NIFP) for the years 1993–2008. First, the whole online database of the PBC was checked for eligible cases (with a diagnosis of schizophrenia spectrum disorder). The offender group without schizophrenia (first comparison group) was made up of
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Van Dongen et al.
all offenders referred for clinical evaluation by the NIFP in Rotterdam and Dordrecht in the years 2005 to 2008; we attempted to match for timescale of referral, but cases were not registered by that system before 2005. Files for the non-offender schizophrenia group (second comparison group) were drawn from all admissions to a general psychiatric ward at the Erasmus Medical Center, Rotterdam, the Netherlands, in the years 2002–2008; inclusion criteria were a diagnosis of schizophrenia spectrum disorder and no criminal record. Again, the earlier date was limited by the registration system. All eligible cases were included and data extracted from the full clinical record. Data quality may differ between the two systems because when a person accused of a serious offence is referred to the NIFP, she or he has to undergo a clinical pretrial assessment for the court, but when a person is referred to the Pieter Baan Center, she or he will be observed by behavioural experts for 7 weeks to determine whether the defendant has a mental disorder and his or her resultant likely responsibility for the act as charged (Van Marle, 2000). Nevertheless, pretrial assessments in both systems are standardised, and multidisciplinary in most cases, and therefore extensive and complete in their information. Measures All measures were scored from the clinical assessments, official records and court reports for the offender groups and the clinical record for the non-offender group. Variables were characterised simply as present or absent. Anti-social personality
Anti-social personality disorder was rated as present when the expert referred to an anti-social personality disorder, anti-social personality traits or psychopathic personality traits. Disruptive behaviour disorders were coded as present if it was indicated that the person was diagnosed with a conduct disorder, oppositional defiant disorder or attention-deficit hyperactivity disorder. In addition, two variables were scored regarding conduct problems before age 12 years, and between 12 and 18 years of age. Conduct problems were scored when there were problems such as running away, skipping school or criminal activities. Substance use
Similarly, when there was a file diagnosis of substance use disorder or substance dependence then the variable was accepted as evidence of the presence of a substance misuse disorder, rated separately for each of alcohol, cannabis and other/polysubstance use. In addition, we rated age of onset of substance use (up to age 14/15+ years). Positive symptoms
Visual and auditory hallucinations, and grandiose, influence, reference and persecutory delusions were rated as present or absent according to the clinical file.
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Pathways of offending in schizophrenia
Clinical diagnoses were standardised, using all five axes of the third or fourth edition of the Diagnostic and Statistical Manual (APA, 1980, 1987, 1994). The disorders included in the ‘schizophrenia spectrum’ were schizophrenia, schizoaffective, and delusional. The variables in the early-start and first offender group were scored by graduate-level research assistants before group allocation to the early-start and first offender group. Statistical analyses Chi-square test statistics were used to test differences between the groups in the domains of anti-social personality, substance use and positive symptoms. In cases where cells had counts less than 5, we used Fisher’s exact test. All analyses were conducted using a significance level of p < .05. Results The demographic and diagnostic characteristics of the five groups are summarised in Table 1. Is the offending of early starters associated with a premorbid anti-social personality disorder? Results for the bivariate analyses comparing early starters with offenders who had no psychosis are shown in Table 2, confirming that anti-social personality diagnosis or traits, ‘psychopathic’ traits, disruptive behaviour disorders and substance use disorders other than alcohol were more prevalent amongst early starters than non-psychotic offenders, as were conduct problems in youth. Early starters were, however, less likely to have used alcohol than non-psychotic offenders. A higher proportion of early than late starters had anti-social personality characteristics, substance use diagnoses and conduct problems in adolescence. No personality disorder diagnoses at all had been made in the late first offender group, but late first offenders were more likely to be using alcohol than early onset offenders. Is the offending of late starters and first offenders associated with positive symptoms of the schizophrenic disorder? Table 2 also shows the symptom comparisons between the psychotic groups. A higher proportion of early starters had persecutory and/or grandiose delusions than did the non-offenders with psychosis. The same pattern was apparent for the late starters and the late first offenders. The offender groups with psychosis did not differ from each other in the prevalence of positive symptoms of psychosis.
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Van Dongen et al.
Table 1: Demographic characteristics of the early starters, offenders, late starters, first offenders and non-offenders
Age (years) Male gender (%) Ethnicity Dutch (%) Other Western (%) Dutch Antilles (%) Morocco (%) Turkey (%) Surinam (%) Cape Verde (%) Refugee countries (%) Other (%) Psychotic disorder Schizophrenia (%) Schizoaffective (%) Delusional disorder (%)
Early starters (n = 97)
Offenders (n = 115)
Late starters (n = 126)
First offenders (n = 26)
Nonoffenders (n = 129)
31.95 (7.49) 97.9
32.75 (11.16) 87.0
35.70 (9.73) 91.3
43.41 (6.05) 65.4
30.05 (0.80) 76.7
40.2 3.1 5.2 15.5 3.1 17.5 0.0 4.1 11.3
60.0 2.6 9.6 2.6 3.5 8.7 3.5 3.5 6.1
45.2 1.6 0.8 5.6 4.0 11.1 6.3 0.8 24.6
73.0 3.8 3.8 0.0 3.8 0.0 3.8 0.0 11.4
36.4 3.1 0.0 10.9 4.7 20.2 4.7 7.0 13.2
83.5 3.1 13.4
N/A N/A N/A
78.6 9.5 11.9
57.7 23.1 19.2
92.2 4.7 3.1
Note: N/A, not applicable.
Discussion This is the first study that investigated the differences between early starters, late starters and late first offenders, all with psychosis, with two comparison groups variously eliminating the psychosis or the offending. Our first hypothesis was sustained, in that the early starters were similar to the non-psychotic offenders in most aspects of personality measured, although the early-starter psychosis group was more likely to have had specifically documented conduct disorder problems. Our hypothesis that late starters and late first offenders would be more likely to have positive psychotic symptoms was sustained to the extent that the two groups of offenders with psychosis who started offending after the diagnosis were more likely to have had psychotic symptoms recorded than the non-offenders with psychosis. This was especially true with respect to grandiose and persecutory delusions, but the early starters differed from the non-offenders with psychosis in this respect too. Although our study did not follow an experimental design, and therefore causal relations cannot be inferred, the results indicate that the hypotheses formulated by Hodgins are partly supported. Our results showed that early-start offenders are characterised by anti-social personality characteristics, such as conduct disorder
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Copyright © 2014 John Wiley & Sons, Ltd.
Positive symptoms Auditory hallucination
Substance use Substance misuse Alcohol Cannabis Other/poly substances Substance before 15 years
Anti-social personality APD diagnosis APD traits Psychopathic traits DBD disorder Conduct youth Conduct adolescence
20 (17.4)
24 (24.7)
N/A
17 (19.2) 30 (11.3) 39 (33.9)
2 (2.1) 20 (20.6) 37 (38.1)
61 (62.9)
69 (60.0)
41 (35.7) 28 (24.3) 72 (62.6)
30 (30.9) 40 (41.2) 83 (85.6)
59 (60.8)
30 (26.1) 78 (67.8) 4 (3.5)
OFN (%)
16 (16.50) 64 (66.0) 8 (8.2)
ESN (%)
LSN (%)
68 (68.0)
27 (27.0)
3 (3.0) 15 (15.0) 28 (28.0)
46 (46.0)
15 (15.0) 33 (33.0) 64 (64.0)
9 (9.0) 37 (37.0) 1 (1.0)
Percentages
14 (53.8)
2 (7.7)
5 (9.5) 1 (3.8) 1 (3.8)
7 (26.9)
0 (6.3) 5 (19.2) 10 (38.5)
0 (0.0) 0 (0.0) 0 (0.0)
FON (%)
85 (65.9)
6 (10.0)
7 (5.4) 29 (22.5) 33 (25.6)
69 (53.5)
N/A N/A N/A
N/A N/A N/A
SPN (%)
N/A
1.73
Fisher*** 3.47 0.41
0.02
0.53 6.89** 14.12***
2.85 0.08 Fisher
Valuea
ES versus OF
0.57
0.13
Fisher*** 1.06 2.29
4.35*
1.90 1.43 12.09***
2.50 16.55*** Fisher*
Valuea
ES versus LS
0.70
3.58
Fisher** Fisher* 11.30***
9.48**
10.65** 4.28* 24.67***
Fisher* 35.76*** Fisher
Valuea
ES versus FO
0.22
N/A
N/A N/A N/A
N/A
N/A N/A N/A
N/A N/A N/A
Valuea
ES versus SP
1.82
N/A
N/A N/A N/A
N/A
N/A N/A N/A
N/A N/A N/A
Valuea
LS versus FO
Table 2: Percentages and results of the univariate tests regarding anti-social personality, substance use and positive symptoms
0.11
N/A
N/A N/A N/A
N/A
N/A N/A N/A
N/A N/A N/A
Valuea
LS versus SP
(Continues)
1.36
N/A
N/A N/A N/A
N/A
N/A N/A N/A
N/A N/A N/A
Valuea
FO versus SP
Pathways of offending in schizophrenia
(2014) DOI: 10.1002/cbm
Copyright © 2014 John Wiley & Sons, Ltd.
N/A
N/A
N/A
N/A
N/A
36 (37.1)
22 (22.7)
35 (36.1)
69 (71.1)
OFN (%)
18 (18.6)
ESN (%)
LSN (%)
68 (68.0)
32 (32.0)
22 (22.0)
40 (40.0)
30 (30.0)
Percentages
22 (84.6)
13 (50.0)
4 (15.4)
8 (30.8)
7 (26.9)
FON (%)
65 (50.4)
43 (33.3)
28 (21.7)
17 (13.2)
32 (24.8)
SPN (%)
N/A
N/A
N/A
N/A
N/A
Valuea
ES versus OF
0.23
0.37
0.01
0.17
3.50
Valuea
ES versus LS
1.94
1.67
0.66
0.36
0.89
Valuea
ES versus FO
9.87**
0.19
0.03
17.67***
1.26
Valuea
ES versus SP
2.79
2.91
0.55
0.75
0.09
Valuea
LS versus FO
7.18**
0.05
0.01
21.68***
0.77
Valuea
LS versus SP
10.30***
2.61
0.53
Fisher*
0.05
Valuea
FO versus SP
Note: ES = early starter; OF = offender without major mental disorder; LS = late starter; FO = first offender; SP = patient with schizophrenia without criminal record; N/A, not applicable; DBD, disruptive behaviour disorders; APD, antisocial personality disorder. a degree of freedom = 1. *p < .05. **p < .01. ***p ≤ .001.
Visual hallucination Grandiose delusions Influence delusions Reference delusions Persecutory delusions
Table 2: (Continued)
Van Dongen et al.
(2014) DOI: 10.1002/cbm
Pathways of offending in schizophrenia
problems at an early age, substance use and anti-social personality disorder symptoms or signs. In this respect, these people resemble offenders without a schizophrenic disorder more than they resemble those people with schizophrenia who start offending after the clear onset of the illness and those whose first offence is many years after that. Presence of psychotic symptoms pertinent to the offending, by contrast, did not distinguish the groups. This may be explained in two ways. First, it may be that early starters start their offending because of early anti-social personality characteristics and subsequently develop a schizophrenic disorder, but the symptoms are co-incidental to their offending. Alternatively, early starters may start their offending because of anti-social personality characteristics, but subsequent psychotic symptoms are associated with development or maintenance of the offending. Hodgins’ (2008, 2009) unique contribution to the early-start and late-start offender typology was to add a third offender group, namely the first late offenders. Our findings here and elsewhere (Van Dongen et al., 2014) suggest that, although this third type may be warranted, first late starters and late starters are more similar to each other than different. Most importantly, it seems likely that in both late-onset offending groups, the offending is associated with positive symptoms of the schizophrenic disorder, especially delusions, but not anti-social personality traits. Therefore, our findings are more strongly supportive of the two pathway model – that there is one that may be associated with delusions but is as or more strongly associated with anti-social personality characteristics, and one that is associated more or less purely with delusional ideation (Taylor et al., 1998; Taylor, 2008). It has to be noted that a number of authors have argued that conduct problems may be a precursor of schizophrenia (e.g. Ferdinand and Verhulst, 1995; Kim-Cohen et al., 2003), but of all the psychosis groups, only the early-start offenders showed this problem. The results also showed that early starters were more likely than offenders to misuse substances other than alcohol, although numbers were too small to be able to discriminate between subgroups of illicit drug users. This was also true in the English high-security hospital study (Taylor et al., 1998). It may be that conduct problems lead to the misuse of cannabis at an early age, thereby increasing the risk of developing a psychotic disorder in those who already have this vulnerability (see Fergusson et al., 2006; Henquet et al., 2005), but this was far from being the only substance involved. Our study has two important limitations. First, one cannot rule out the possibility that people in the patient group without a criminal record will become offenders in the future. Thus, conclusions using a comparison with this group have to be interpreted with caution. The main limitation, however, is that, because of the need for a large sample, we had to rely on file data that, furthermore, were from different sources. File data for the offender groups were from forensic assessments that, by definition, had adopted a more structured and intensive approach to assessment. File data for the non-offenders were from standard
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Van Dongen et al.
clinical files. These reports contain much important information about the patient, but are not as extensive as the forensic assessments, and would be as likely to affect symptom reporting as any other aspect of the case. There was no attempt to measure personality traits on the non-offender group. A third limitation is that our study was cross-sectional, so, not only is it not possible to infer any causality but also the conduct and personality disorder diagnoses would inevitably have relied to some extent on retrospective recall. Prospective longitudinal studies in this field, however, tend to have such small sample sizes of the groups of interest that it is difficult to draw conclusions for these different reasons. In conclusion, our study does not rule out a three-group model for understanding the role of psychosis in offending pathways when these two problems come together, but there was very little difference between the two late-onset groups. This has important implications for treatment, which would have to be differently balanced in assessment and treatment strategies if it is to accommodate important differences in the needs of offender patients. Acknowledgement We thank Luc van Seggelen, the NIFP, Rotterdam/Dordrecht division, for his facilitation of data collection. References American Psychiatric Association (1980) Diagnostic and Statistical Manual of Mental Disorders (3rd ed.). Washington, DC: American Psychiatric Association. American Psychiatric Association (1987) Diagnostic and Statistical Manual of Mental Disorders (Revised 3rd ed.). Washington, DC: American Psychiatric Association. American Psychiatric Association (1994) Diagnostic and Statistical Manual of Mental Disorders: DSM-IV (4th ed.). Washington, DC: American Psychiatric Association. Belfrage H (1998) A ten-year follow-up of criminality in Stockholm mental patients. British Journal of Criminology 38: 145–155. Coté G, Hodgins S (1992) The prevalence of major mental disorders among homicide offenders. International Journal of Law and Psychiatry 15: 69–99. Douglas KS, Guy LS, Hart SD (2009) Psychosis as a risk factor for violence to others: A metaanalysis. Psychological Bulletin 135: 679–706. Eronen M, Tiihonen J, Hakola P (1996) Schizophrenia and homicidal behavior. Schizophrenia Bulletin 22: 83–89. Fazel S, Gulati G, Linsell L, Geddes JR, Grann M (2009) Schizophrenia and violence: Systematic review and meta-analysis. PLoS Medicine 6: e1000120. Ferdinand RF, Verhulst FC (1995) Psychopathology from adolescence into young adulthood: An 8-year follow-up study. American Journal of Psychiatry 152: 1586–1594. Fergusson D, Poulton R, Smith P, Boden J (2006) Cannabis and psychosis. British Medical Journal 332: 172–175. Henquet C, Murray R, Linszen D, van Os J (2005) The environment and schizophrenia: The role of cannabis use. Schizophrenia Bulletin 31: 608–612.
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Pathways of offending in schizophrenia
Hodgins S (1995) Major mental disorder and crime: An overview. Psychology, Crime & Law 2: 5–17. Hodgins S (2008) Violent behaviour among people with schizophrenia: A framework for investigations of causes, and effective treatment, and prevention. Philosophical Transactions of the Royal Society, B: Biological Sciences 363: 2505–2518. Hodgins S (2009) Violent behaviour among people with schizophrenia: A framework for investigations of causes, effective treatment, and prevention. In Hodgins S, Viding E, Plowdowski A (eds) The Neurobiological Basis of Violence: Science and Rehabilitation. Oxford: Oxford University Press; pp. 43–65. Hodgins S, Coté G (1993) The criminality of mentally disordered offenders. Criminal Justice and Behaviour 20: 115–129. Hodgins S, Coté G, Toupin J (1995) Major mental disorder and crime: An etiological hypothesis. Issues in Criminological and Legal Psychology 24: 74–80. Kim-Cohen J, Caspi A, Moffitt TE, Harrington HL, Milne BJ, Poulton R (2003) Prior juvenile diagnoses in adults with mental disorder: developmental follow-back of a prospective longitudinal cohort. Arch. Gen. Psychiatry 60: 709–717. DOI:10.1001/archpsyc.60.7.709 Kratzer L, Hodgins S (1999) A typology of offenders: A test of Moffitt’s theory among males and females from childhood to age 30. Criminal Behaviour and Mental Health 9: 59–73. Large M, Smith G, Nielssen O (2009) The relationship between the rate of homicide by those with schizophrenia and the overall homicide rate: A systematic review and meta-analysis. Schizophrenia Research 112: 123–129. Lindqvist P, Allebeck P (1990) Schizophrenia and crime: A longitudinal follow-up of 644 schizophrenics in Stockholm. British Journal of Psychiatry 157: 345–350. Moffitt TE (1993) “Life-course persistent” and “adolescence-limited” antisocial behaviour: A developmental taxonomy. Psychological Review 100: 674–701. Pedersen L, Rasmussen K, Elsass P, Hougaard H (2010) The importance of early anti-social behaviour among men with a schizophrenia spectrum disorder in a specialist forensic psychiatry hospital unit in Denmark. Criminal Behaviour and Mental Health 20: 295–304. Schanda H, Knecht G, Schreinzer D, Th S, Ortwein-Swoboda G, Th W (2004) Homicide and major mental disorders: A 25-year study. Acta Psychiatrica Scandinavica 110: 98–107. Taylor PJ (2008) Psychosis and violence: Stories, fears, and reality. Canadian Journal of Psychiatry 53: 647–659. Taylor PJ, Leese M, Williams D, Butwell M, Daly R, Larkin E (1998) Mental disorder and violence. A special(high security) hospital study. British Journal of Psychiatry 172: 218–226. Van Dongen JDM, Buck NML, Van Marle HJC (2014) First offenders with psychosis: Justification of a third type within the early/late start offender typology. Crime and Delinquency 60: 126–142. DOI:10.1177/0011128713505490 Van Marle HJC (2000) Forensic psychiatric services in the Netherlands. International Journal of Law and Psychiatry 23: 515–531. Wessely SC, Castle D, Douglas AJ, Taylor PJ (1994) The criminal careers of incident cases of schizophrenia. Psychological Medicine 24: 483–502. Yung AR, McGorry PD (1996) The prodromal phase of first-episode psychosis: Past and current conceptualizations. Schizophrenia Bulletin 22: 353–370.
Address correspondence to: Dr Josanne D. M. Van Dongen, Clinical Psychology, Faculty of Social Sciences, Institute of Psychology, Erasmus University Rotterdam, Room T13-26, PO Box 1738, 3000 DR Rotterdam, The Netherlands. Email: j.d.m. [email protected]
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Anti-social personality characteristics and psychotic symptoms: Two pathways associated with offending in schizophrenia
JOSANNE D. M. VAN DONGEN1,2, NICOLE M. L. BUCK1,3, MARKO BARENDREGT4,5, NICO M. VAN BEVEREN1,6, EDWIN DE BEURS4,5 AND HJALMAR J. C. VAN MARLE1, 1Department of Forensic Psychiatry, Erasmus University Medical Center, Rotterdam, The Netherlands; 2Institute of Psychology, Erasmus University Rotterdam, Rotterdam, The Netherlands; 3Forensic Psychiatric Center De Kijvelanden, Rhoon, The Netherlands; 4Department of Research & Development, Netherlands Institute for Forensic Psychiatry and Psychology (NIFP), Utrecht, The Netherlands; 5Foundation Benchmark GGZ, Bilthoven, The Netherlands; 6Delta Psychiatric Center, Poortugaal, The Netherlands ABSTRACT Background Several research groups have shown that people with schizophrenia who offend do not form a homogenous group. A three-group model claimed by Hodgins proposes distinguishing between people who start offending before the onset of psychosis (early starters), after psychosis onset but at age 34 years or under (late starters) and after psychosis onset but at age 35 years or older (late first offenders). Aims This study aimed to test the hypotheses (1) that the personality of early starters and non-psychotic offenders would be similar, but different from either late-starter group; (2) that the late-starter groups would be more likely to have positive psychotic symptoms than non-criminal patients with schizophrenia; and (3) that symptom types would differentiate the psychotic groups. Methods A retrospective file study was conducted on cases of 97 early starters, 100 late starters and 26 late first offenders all drawn from the Netherlands Institute of Forensic Psychiatry and Psychology (NIFP) archives 1993–2008, 115 non-psychotic offenders from 2005–2008 NIFP archives and 129 patients with schizophrenia and no criminal history from one general service in Rotterdam. Results Early starters closely resembled the non-psychotic offenders in their premorbid anti-social personality characteristics. The two late-onset offending psychosis groups
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Van Dongen et al.
were more likely to have persecutory and/or grandiose delusions than non-offenders with psychosis, but so were the early starters. Implications In a first study to compare subgroups of offenders with psychosis directly with non-psychotic offenders and non-offenders with psychosis, we found such additional support for a distinction between early and late starters with psychosis that different treatment strategies would seem indicated, focusing on personality and substance misuse for the former but psychotic symptoms for all. It remains to be seen whether the higher rate of alcohol misuse amongst late first offenders is a fundamental distinction or a function of age difference. Copyright © 2014 John Wiley & Sons, Ltd. Introduction Recent reviews have shown that there is a small but significant relationship between psychosis and (violent) criminal behaviour (e.g. Douglas et al., 2009; Fazel et al., 2009). The relationship between schizophrenia and violent crime, including homicide (Eronen et al., 1996; Schanda et al., 2004; Large et al., 2009), may be stronger than that between schizophrenia and non-violent crimes (e.g. Lindqvist and Allebeck, 1990; Coté and Hodgins, 1992; Belfrage, 1998). A number of researchers have suggested that there are distinct subgroups of offenders with schizophrenia, subdividing variously according to age of onset of offending and/or co-morbidity of personality disorder. Hodgins (2008, 2009) has suggested three types of offenders with schizophrenia: early starters, late starters and late first offenders. Early starters show anti-social personality characteristics early in life, which may be reflected in a childhood history of conduct problems (Hodgins and Coté, 1993; Hodgins, 1995; Hodgins et al., 1995). They may be similar to the ‘life course persistent’ delinquent group defined by Moffitt (1993), except for the fact that they also develop schizophrenia. Late starters, also called ‘adult starters’ (Hodgins, 1995; Kratzer and Hodgins, 1999), usually start offending in adulthood after the onset of a major mental disorder (e.g. schizophrenia). Their criminal behaviour is more likely to be attributable to cognitive and perceptual (positive) symptoms of the disorder (Hodgins, 1995; Hodgins et al., 1995). Hodgins (2009) also claims that once late starters have started their criminal behaviour, they will repeatedly engage in criminal behaviour and violence towards others, although others have shown that people with schizophrenia desist earlier than people in the general population (Lindqvist and Allebeck, 1990) or than other psychiatric patients (Wessely et al., 1994). A further subgroup of late first offenders, who suddenly commit a very serious offence after the onset of their schizophrenia, has also been suggested (Hodgins, 2008, 2009; Pedersen et al., 2010). We (Van Dongen et al., 2014) were the first independent group to study the late first offender group within this three-group typology and found no differences between the three offender groups regarding positive psychotic symptoms. This
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Pathways of offending in schizophrenia
seems to be inconsistent with the hypothesis that the offending of late starters and late first offenders is attributable to positive psychotic symptoms, whereas that of the early starters is not. The main aim of our present study, therefore, was to test the relative role of personality and positive symptoms of schizophrenia within one sample of patients and non-psychotic offenders. Our first hypothesis was that early starters would resemble offenders without schizophrenia with regard to having an anti-social personality trait, but differ in this from late starters and late first offenders with schizophrenia. We also retested what is, essentially, the alternative hypothesis that late starters and late first offenders would be more likely to have positive psychotic symptoms than the early-start offenders. Finally, we hypothesised that late starters and late first offenders would differ from non-offenders with schizophrenia on type of delusions and persecutory delusions in particular. Method Participants This records study drew on 367 clinical and/or penal records, representing the following groups: 97 people with a schizophrenia spectrum disorder and offending, the offending having started before the onset of even any prodrome of the illness; the latter was defined as onset of any disturbed behaviour, including loss of motivation, anhedonia, anger, irritability, attention problems and/or social withdrawal (see Yung and McGorry, 1996); the first offence was the date of the first conviction according to the official criminal record; 100 people with the two conditions who had started offending only after the onset of the illness but whose age was 34 years or under at the time of the first offence; 26 people similar to the 100 but whose offending had started at age 35 years or older; 115 repeat offenders with no psychotic illness; and 129 patients with schizophrenia and no criminal record. The aim had been to have a sample of 150 cases per group, but it became clear that the sample was limited by eligible cases available. Files for the offenderpatient groups were retrieved from the Observation Clinic of the Ministry of Justice (Pieter Baan Center; PBC) in Utrecht, the Netherlands, which is part of the Netherlands Institute of Forensic Psychiatry and Psychology (NIFP) for the years 1993–2008. First, the whole online database of the PBC was checked for eligible cases (with a diagnosis of schizophrenia spectrum disorder). The offender group without schizophrenia (first comparison group) was made up of
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Van Dongen et al.
all offenders referred for clinical evaluation by the NIFP in Rotterdam and Dordrecht in the years 2005 to 2008; we attempted to match for timescale of referral, but cases were not registered by that system before 2005. Files for the non-offender schizophrenia group (second comparison group) were drawn from all admissions to a general psychiatric ward at the Erasmus Medical Center, Rotterdam, the Netherlands, in the years 2002–2008; inclusion criteria were a diagnosis of schizophrenia spectrum disorder and no criminal record. Again, the earlier date was limited by the registration system. All eligible cases were included and data extracted from the full clinical record. Data quality may differ between the two systems because when a person accused of a serious offence is referred to the NIFP, she or he has to undergo a clinical pretrial assessment for the court, but when a person is referred to the Pieter Baan Center, she or he will be observed by behavioural experts for 7 weeks to determine whether the defendant has a mental disorder and his or her resultant likely responsibility for the act as charged (Van Marle, 2000). Nevertheless, pretrial assessments in both systems are standardised, and multidisciplinary in most cases, and therefore extensive and complete in their information. Measures All measures were scored from the clinical assessments, official records and court reports for the offender groups and the clinical record for the non-offender group. Variables were characterised simply as present or absent. Anti-social personality
Anti-social personality disorder was rated as present when the expert referred to an anti-social personality disorder, anti-social personality traits or psychopathic personality traits. Disruptive behaviour disorders were coded as present if it was indicated that the person was diagnosed with a conduct disorder, oppositional defiant disorder or attention-deficit hyperactivity disorder. In addition, two variables were scored regarding conduct problems before age 12 years, and between 12 and 18 years of age. Conduct problems were scored when there were problems such as running away, skipping school or criminal activities. Substance use
Similarly, when there was a file diagnosis of substance use disorder or substance dependence then the variable was accepted as evidence of the presence of a substance misuse disorder, rated separately for each of alcohol, cannabis and other/polysubstance use. In addition, we rated age of onset of substance use (up to age 14/15+ years). Positive symptoms
Visual and auditory hallucinations, and grandiose, influence, reference and persecutory delusions were rated as present or absent according to the clinical file.
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Pathways of offending in schizophrenia
Clinical diagnoses were standardised, using all five axes of the third or fourth edition of the Diagnostic and Statistical Manual (APA, 1980, 1987, 1994). The disorders included in the ‘schizophrenia spectrum’ were schizophrenia, schizoaffective, and delusional. The variables in the early-start and first offender group were scored by graduate-level research assistants before group allocation to the early-start and first offender group. Statistical analyses Chi-square test statistics were used to test differences between the groups in the domains of anti-social personality, substance use and positive symptoms. In cases where cells had counts less than 5, we used Fisher’s exact test. All analyses were conducted using a significance level of p < .05. Results The demographic and diagnostic characteristics of the five groups are summarised in Table 1. Is the offending of early starters associated with a premorbid anti-social personality disorder? Results for the bivariate analyses comparing early starters with offenders who had no psychosis are shown in Table 2, confirming that anti-social personality diagnosis or traits, ‘psychopathic’ traits, disruptive behaviour disorders and substance use disorders other than alcohol were more prevalent amongst early starters than non-psychotic offenders, as were conduct problems in youth. Early starters were, however, less likely to have used alcohol than non-psychotic offenders. A higher proportion of early than late starters had anti-social personality characteristics, substance use diagnoses and conduct problems in adolescence. No personality disorder diagnoses at all had been made in the late first offender group, but late first offenders were more likely to be using alcohol than early onset offenders. Is the offending of late starters and first offenders associated with positive symptoms of the schizophrenic disorder? Table 2 also shows the symptom comparisons between the psychotic groups. A higher proportion of early starters had persecutory and/or grandiose delusions than did the non-offenders with psychosis. The same pattern was apparent for the late starters and the late first offenders. The offender groups with psychosis did not differ from each other in the prevalence of positive symptoms of psychosis.
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Van Dongen et al.
Table 1: Demographic characteristics of the early starters, offenders, late starters, first offenders and non-offenders
Age (years) Male gender (%) Ethnicity Dutch (%) Other Western (%) Dutch Antilles (%) Morocco (%) Turkey (%) Surinam (%) Cape Verde (%) Refugee countries (%) Other (%) Psychotic disorder Schizophrenia (%) Schizoaffective (%) Delusional disorder (%)
Early starters (n = 97)
Offenders (n = 115)
Late starters (n = 126)
First offenders (n = 26)
Nonoffenders (n = 129)
31.95 (7.49) 97.9
32.75 (11.16) 87.0
35.70 (9.73) 91.3
43.41 (6.05) 65.4
30.05 (0.80) 76.7
40.2 3.1 5.2 15.5 3.1 17.5 0.0 4.1 11.3
60.0 2.6 9.6 2.6 3.5 8.7 3.5 3.5 6.1
45.2 1.6 0.8 5.6 4.0 11.1 6.3 0.8 24.6
73.0 3.8 3.8 0.0 3.8 0.0 3.8 0.0 11.4
36.4 3.1 0.0 10.9 4.7 20.2 4.7 7.0 13.2
83.5 3.1 13.4
N/A N/A N/A
78.6 9.5 11.9
57.7 23.1 19.2
92.2 4.7 3.1
Note: N/A, not applicable.
Discussion This is the first study that investigated the differences between early starters, late starters and late first offenders, all with psychosis, with two comparison groups variously eliminating the psychosis or the offending. Our first hypothesis was sustained, in that the early starters were similar to the non-psychotic offenders in most aspects of personality measured, although the early-starter psychosis group was more likely to have had specifically documented conduct disorder problems. Our hypothesis that late starters and late first offenders would be more likely to have positive psychotic symptoms was sustained to the extent that the two groups of offenders with psychosis who started offending after the diagnosis were more likely to have had psychotic symptoms recorded than the non-offenders with psychosis. This was especially true with respect to grandiose and persecutory delusions, but the early starters differed from the non-offenders with psychosis in this respect too. Although our study did not follow an experimental design, and therefore causal relations cannot be inferred, the results indicate that the hypotheses formulated by Hodgins are partly supported. Our results showed that early-start offenders are characterised by anti-social personality characteristics, such as conduct disorder
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Copyright © 2014 John Wiley & Sons, Ltd.
Positive symptoms Auditory hallucination
Substance use Substance misuse Alcohol Cannabis Other/poly substances Substance before 15 years
Anti-social personality APD diagnosis APD traits Psychopathic traits DBD disorder Conduct youth Conduct adolescence
20 (17.4)
24 (24.7)
N/A
17 (19.2) 30 (11.3) 39 (33.9)
2 (2.1) 20 (20.6) 37 (38.1)
61 (62.9)
69 (60.0)
41 (35.7) 28 (24.3) 72 (62.6)
30 (30.9) 40 (41.2) 83 (85.6)
59 (60.8)
30 (26.1) 78 (67.8) 4 (3.5)
OFN (%)
16 (16.50) 64 (66.0) 8 (8.2)
ESN (%)
LSN (%)
68 (68.0)
27 (27.0)
3 (3.0) 15 (15.0) 28 (28.0)
46 (46.0)
15 (15.0) 33 (33.0) 64 (64.0)
9 (9.0) 37 (37.0) 1 (1.0)
Percentages
14 (53.8)
2 (7.7)
5 (9.5) 1 (3.8) 1 (3.8)
7 (26.9)
0 (6.3) 5 (19.2) 10 (38.5)
0 (0.0) 0 (0.0) 0 (0.0)
FON (%)
85 (65.9)
6 (10.0)
7 (5.4) 29 (22.5) 33 (25.6)
69 (53.5)
N/A N/A N/A
N/A N/A N/A
SPN (%)
N/A
1.73
Fisher*** 3.47 0.41
0.02
0.53 6.89** 14.12***
2.85 0.08 Fisher
Valuea
ES versus OF
0.57
0.13
Fisher*** 1.06 2.29
4.35*
1.90 1.43 12.09***
2.50 16.55*** Fisher*
Valuea
ES versus LS
0.70
3.58
Fisher** Fisher* 11.30***
9.48**
10.65** 4.28* 24.67***
Fisher* 35.76*** Fisher
Valuea
ES versus FO
0.22
N/A
N/A N/A N/A
N/A
N/A N/A N/A
N/A N/A N/A
Valuea
ES versus SP
1.82
N/A
N/A N/A N/A
N/A
N/A N/A N/A
N/A N/A N/A
Valuea
LS versus FO
Table 2: Percentages and results of the univariate tests regarding anti-social personality, substance use and positive symptoms
0.11
N/A
N/A N/A N/A
N/A
N/A N/A N/A
N/A N/A N/A
Valuea
LS versus SP
(Continues)
1.36
N/A
N/A N/A N/A
N/A
N/A N/A N/A
N/A N/A N/A
Valuea
FO versus SP
Pathways of offending in schizophrenia
(2014) DOI: 10.1002/cbm
Copyright © 2014 John Wiley & Sons, Ltd.
N/A
N/A
N/A
N/A
N/A
36 (37.1)
22 (22.7)
35 (36.1)
69 (71.1)
OFN (%)
18 (18.6)
ESN (%)
LSN (%)
68 (68.0)
32 (32.0)
22 (22.0)
40 (40.0)
30 (30.0)
Percentages
22 (84.6)
13 (50.0)
4 (15.4)
8 (30.8)
7 (26.9)
FON (%)
65 (50.4)
43 (33.3)
28 (21.7)
17 (13.2)
32 (24.8)
SPN (%)
N/A
N/A
N/A
N/A
N/A
Valuea
ES versus OF
0.23
0.37
0.01
0.17
3.50
Valuea
ES versus LS
1.94
1.67
0.66
0.36
0.89
Valuea
ES versus FO
9.87**
0.19
0.03
17.67***
1.26
Valuea
ES versus SP
2.79
2.91
0.55
0.75
0.09
Valuea
LS versus FO
7.18**
0.05
0.01
21.68***
0.77
Valuea
LS versus SP
10.30***
2.61
0.53
Fisher*
0.05
Valuea
FO versus SP
Note: ES = early starter; OF = offender without major mental disorder; LS = late starter; FO = first offender; SP = patient with schizophrenia without criminal record; N/A, not applicable; DBD, disruptive behaviour disorders; APD, antisocial personality disorder. a degree of freedom = 1. *p < .05. **p < .01. ***p ≤ .001.
Visual hallucination Grandiose delusions Influence delusions Reference delusions Persecutory delusions
Table 2: (Continued)
Van Dongen et al.
(2014) DOI: 10.1002/cbm
Pathways of offending in schizophrenia
problems at an early age, substance use and anti-social personality disorder symptoms or signs. In this respect, these people resemble offenders without a schizophrenic disorder more than they resemble those people with schizophrenia who start offending after the clear onset of the illness and those whose first offence is many years after that. Presence of psychotic symptoms pertinent to the offending, by contrast, did not distinguish the groups. This may be explained in two ways. First, it may be that early starters start their offending because of early anti-social personality characteristics and subsequently develop a schizophrenic disorder, but the symptoms are co-incidental to their offending. Alternatively, early starters may start their offending because of anti-social personality characteristics, but subsequent psychotic symptoms are associated with development or maintenance of the offending. Hodgins’ (2008, 2009) unique contribution to the early-start and late-start offender typology was to add a third offender group, namely the first late offenders. Our findings here and elsewhere (Van Dongen et al., 2014) suggest that, although this third type may be warranted, first late starters and late starters are more similar to each other than different. Most importantly, it seems likely that in both late-onset offending groups, the offending is associated with positive symptoms of the schizophrenic disorder, especially delusions, but not anti-social personality traits. Therefore, our findings are more strongly supportive of the two pathway model – that there is one that may be associated with delusions but is as or more strongly associated with anti-social personality characteristics, and one that is associated more or less purely with delusional ideation (Taylor et al., 1998; Taylor, 2008). It has to be noted that a number of authors have argued that conduct problems may be a precursor of schizophrenia (e.g. Ferdinand and Verhulst, 1995; Kim-Cohen et al., 2003), but of all the psychosis groups, only the early-start offenders showed this problem. The results also showed that early starters were more likely than offenders to misuse substances other than alcohol, although numbers were too small to be able to discriminate between subgroups of illicit drug users. This was also true in the English high-security hospital study (Taylor et al., 1998). It may be that conduct problems lead to the misuse of cannabis at an early age, thereby increasing the risk of developing a psychotic disorder in those who already have this vulnerability (see Fergusson et al., 2006; Henquet et al., 2005), but this was far from being the only substance involved. Our study has two important limitations. First, one cannot rule out the possibility that people in the patient group without a criminal record will become offenders in the future. Thus, conclusions using a comparison with this group have to be interpreted with caution. The main limitation, however, is that, because of the need for a large sample, we had to rely on file data that, furthermore, were from different sources. File data for the offender groups were from forensic assessments that, by definition, had adopted a more structured and intensive approach to assessment. File data for the non-offenders were from standard
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Van Dongen et al.
clinical files. These reports contain much important information about the patient, but are not as extensive as the forensic assessments, and would be as likely to affect symptom reporting as any other aspect of the case. There was no attempt to measure personality traits on the non-offender group. A third limitation is that our study was cross-sectional, so, not only is it not possible to infer any causality but also the conduct and personality disorder diagnoses would inevitably have relied to some extent on retrospective recall. Prospective longitudinal studies in this field, however, tend to have such small sample sizes of the groups of interest that it is difficult to draw conclusions for these different reasons. In conclusion, our study does not rule out a three-group model for understanding the role of psychosis in offending pathways when these two problems come together, but there was very little difference between the two late-onset groups. This has important implications for treatment, which would have to be differently balanced in assessment and treatment strategies if it is to accommodate important differences in the needs of offender patients. Acknowledgement We thank Luc van Seggelen, the NIFP, Rotterdam/Dordrecht division, for his facilitation of data collection. References American Psychiatric Association (1980) Diagnostic and Statistical Manual of Mental Disorders (3rd ed.). Washington, DC: American Psychiatric Association. American Psychiatric Association (1987) Diagnostic and Statistical Manual of Mental Disorders (Revised 3rd ed.). Washington, DC: American Psychiatric Association. American Psychiatric Association (1994) Diagnostic and Statistical Manual of Mental Disorders: DSM-IV (4th ed.). Washington, DC: American Psychiatric Association. Belfrage H (1998) A ten-year follow-up of criminality in Stockholm mental patients. British Journal of Criminology 38: 145–155. Coté G, Hodgins S (1992) The prevalence of major mental disorders among homicide offenders. International Journal of Law and Psychiatry 15: 69–99. Douglas KS, Guy LS, Hart SD (2009) Psychosis as a risk factor for violence to others: A metaanalysis. Psychological Bulletin 135: 679–706. Eronen M, Tiihonen J, Hakola P (1996) Schizophrenia and homicidal behavior. Schizophrenia Bulletin 22: 83–89. Fazel S, Gulati G, Linsell L, Geddes JR, Grann M (2009) Schizophrenia and violence: Systematic review and meta-analysis. PLoS Medicine 6: e1000120. Ferdinand RF, Verhulst FC (1995) Psychopathology from adolescence into young adulthood: An 8-year follow-up study. American Journal of Psychiatry 152: 1586–1594. Fergusson D, Poulton R, Smith P, Boden J (2006) Cannabis and psychosis. British Medical Journal 332: 172–175. Henquet C, Murray R, Linszen D, van Os J (2005) The environment and schizophrenia: The role of cannabis use. Schizophrenia Bulletin 31: 608–612.
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
Pathways of offending in schizophrenia
Hodgins S (1995) Major mental disorder and crime: An overview. Psychology, Crime & Law 2: 5–17. Hodgins S (2008) Violent behaviour among people with schizophrenia: A framework for investigations of causes, and effective treatment, and prevention. Philosophical Transactions of the Royal Society, B: Biological Sciences 363: 2505–2518. Hodgins S (2009) Violent behaviour among people with schizophrenia: A framework for investigations of causes, effective treatment, and prevention. In Hodgins S, Viding E, Plowdowski A (eds) The Neurobiological Basis of Violence: Science and Rehabilitation. Oxford: Oxford University Press; pp. 43–65. Hodgins S, Coté G (1993) The criminality of mentally disordered offenders. Criminal Justice and Behaviour 20: 115–129. Hodgins S, Coté G, Toupin J (1995) Major mental disorder and crime: An etiological hypothesis. Issues in Criminological and Legal Psychology 24: 74–80. Kim-Cohen J, Caspi A, Moffitt TE, Harrington HL, Milne BJ, Poulton R (2003) Prior juvenile diagnoses in adults with mental disorder: developmental follow-back of a prospective longitudinal cohort. Arch. Gen. Psychiatry 60: 709–717. DOI:10.1001/archpsyc.60.7.709 Kratzer L, Hodgins S (1999) A typology of offenders: A test of Moffitt’s theory among males and females from childhood to age 30. Criminal Behaviour and Mental Health 9: 59–73. Large M, Smith G, Nielssen O (2009) The relationship between the rate of homicide by those with schizophrenia and the overall homicide rate: A systematic review and meta-analysis. Schizophrenia Research 112: 123–129. Lindqvist P, Allebeck P (1990) Schizophrenia and crime: A longitudinal follow-up of 644 schizophrenics in Stockholm. British Journal of Psychiatry 157: 345–350. Moffitt TE (1993) “Life-course persistent” and “adolescence-limited” antisocial behaviour: A developmental taxonomy. Psychological Review 100: 674–701. Pedersen L, Rasmussen K, Elsass P, Hougaard H (2010) The importance of early anti-social behaviour among men with a schizophrenia spectrum disorder in a specialist forensic psychiatry hospital unit in Denmark. Criminal Behaviour and Mental Health 20: 295–304. Schanda H, Knecht G, Schreinzer D, Th S, Ortwein-Swoboda G, Th W (2004) Homicide and major mental disorders: A 25-year study. Acta Psychiatrica Scandinavica 110: 98–107. Taylor PJ (2008) Psychosis and violence: Stories, fears, and reality. Canadian Journal of Psychiatry 53: 647–659. Taylor PJ, Leese M, Williams D, Butwell M, Daly R, Larkin E (1998) Mental disorder and violence. A special(high security) hospital study. British Journal of Psychiatry 172: 218–226. Van Dongen JDM, Buck NML, Van Marle HJC (2014) First offenders with psychosis: Justification of a third type within the early/late start offender typology. Crime and Delinquency 60: 126–142. DOI:10.1177/0011128713505490 Van Marle HJC (2000) Forensic psychiatric services in the Netherlands. International Journal of Law and Psychiatry 23: 515–531. Wessely SC, Castle D, Douglas AJ, Taylor PJ (1994) The criminal careers of incident cases of schizophrenia. Psychological Medicine 24: 483–502. Yung AR, McGorry PD (1996) The prodromal phase of first-episode psychosis: Past and current conceptualizations. Schizophrenia Bulletin 22: 353–370.
Address correspondence to: Dr Josanne D. M. Van Dongen, Clinical Psychology, Faculty of Social Sciences, Institute of Psychology, Erasmus University Rotterdam, Room T13-26, PO Box 1738, 3000 DR Rotterdam, The Netherlands. Email: j.d.m. [email protected]
Copyright © 2014 John Wiley & Sons, Ltd.
(2014) DOI: 10.1002/cbm
