Psychopathy was the first personality disorder to be recognized in psychiatry. According to Schneider (1923), a German psychiatrist, the term psychopathy referred to a variety of personality disorders (psychopathic personalities [PDs]) as extreme variants of normal personality. It has been given many different labels (Hare, 1991) such as psychopathic inferiority, character deficiency, moral insanity, and manipulative personality. The current interest in the disorder is (at least partly) attributable to the development of the Hare Psychopathy Checklist-Revised (PCL-R; Hare, 1991 and Hare et al., 1990) and the abundance of empirical research it has generated over the past two decades. PCL-R items are personality traits and behaviors, which are scored on a 3-point scale (2=the item definitely applies to the participant, 1=the item applies to a certain extent, 0=the item does not apply to the participant), yielding a maximum total of 40. A score of 30 or more is recommended by Hare (1991) to identify the prototypical psychopath. PCL-R items define two correlated oblique factors, Factor 1 (callous and remorseless style of relating to other people), primarily at high levels of the construct, and Factor 2 (unstable, socially deviant lifestyle) at low levels of the construct (Cooke & Michie, 1997). Recently, however, Cooke and Michie (2001), using confirmatory factor analysis, identified distinct interpersonal, affective, and behavioral factors of which the measurement is uncontaminated by items reflecting antisocial behavior
Evidence gathered in the last decade demonstrates that the PCL-R scale is highly reliable when used with trained and experienced raters. Studies in a variety of countries have typically obtained intraclass correlations (ICCs) >.80 for a single rater. Internal consistency (alpha coefficients >.80; mean interitem correlations >.22) is also high. Considerable evidence has accrued attesting to the construct-related validity of the PCL-R. In several (mostly North American) studies Hart & Hare, 1989, Hemphill et al., 1994, Schroeder et al., 1983 and Smith & Newman, 1990, an expected pattern of relations with clinical assessments of DSM-III-R Axis I and II disorders American Psychiatric Association, 1980 and American Psychiatric Association, 1987 is reported, the interpretation of which is greatly clarified by an analysis of the two-factor structure of the PCL-R (Hart & Hare, 1997). In addition, there is increasing evidence that PCL-R scores are related, in appropriate ways, to so-called psychopathy-related self-report scales, as well as to a variety of behavioral variables Bodholt et al., 2000, Hare, 1991 and Hart & Hare, 1997.
The most common finding in studies that have examined the association between PCL-R psychopathy and DSM-III-R Axis I mental disorders is that a diagnosis of PCL-R psychopathy is rarely significantly associated with individual Axis I pathology other than substance-use disorders Hart & Hare, 1989, Nedopil et al., 1998, Rice & Harris, 1995 and Stålenheim & von Knorring, 1996. Hart and Hare (1989), for example, reported that patients with a diagnosis of PCL-R psychopathy (total score ≥30) were nine times less likely to receive any Axis I principal diagnosis than were other patients. However, moderate to strong associations between the PCL-R total and Factor 2 scores and substance-related disorders, and weak relationships between Factor 1 scores and substance abuse were found Hart & Hare, 1989 and Rutherford et al., 2000. Smith and Newman (1990) assessed substance-use disorder with a structured interview in 360 male prison inmates. Analyses revealed that PCL-R psychopathy was significantly associated with both alcohol and drug abuse/dependence disorders. Other studies Hart et al., 1992 and Hemphill et al., 1994 also found significant correlations between PCL/PCL-R scores and drug abuse/dependence diagnoses; however, correlations with alcohol abuse/dependence diagnoses were not significant. Similar associations were found in European samples of prisoners and forensic psychiatric patients Andersen et al., 1999 and Stålenheim & von Knorring, 1996.
With regard to the association with PDs, the majority of PCL-R psychopaths meet the criteria for antisocial PD, whereas a large proportion of participants with the antisocial PD diagnosis do not meet the PCL-R criteria for psychopathy Hart & Hare, 1989 and Stålenheim & von Knorring, 1996. The correlation between PCL-R scores and (dimensional) diagnoses of antisocial PD is usually quite high, that is, r=.55 to .65 (Hart & Hare, 1989). The prevalence rates of (PCL-R) psychopathy among samples of forensic participants (15–30%), however, are much lower than those for the DSM diagnosis of antisocial PD (50–80%; Hare, 1985 and Hart et al., 1994). Results further indicate that the PCL-R score correlates positively with DSM-III-R Axis II Cluster B disorders (“dramatic–erratic–emotional”) and negatively with Cluster C personality, the “anxious–fearful” cluster (Hart & Hare, 1985; Hart et al., 1994). Rutherford, Alterman, Cacciola, and McKay (1997), for example, found strong and significant correlations between the PCL-R total score and the number of symptoms of DSM-III-R APD, borderline, narcissistic, and histrionic PD in a sample of 250 male methadone patients. Hart and Hare (1989) reported positive correlations between PCL-R total scores and categorical diagnoses of DSM-III antisocial and histrionic PD in a sample of 80 North American men remanded by the courts for inpatient assessment of competency to stand trial. PCL-R Factor 1 scores were negatively correlated with the prototypicality ratings of avoidant and dependent PD. A diagnosis of psychopathy was significantly associated with only one DSM-III Axis II disorder, namely, APD (odds ratio=11.32). Finally, examining 61 Swedish male forensic psychiatric patients, Stålenheim and von Knorring (1996) found that PCL-R-defined psychopathy was strongly associated with the presence of Cluster B disorders (t=7.89, P<.0001) and a diagnosis of antisocial personality disorder (χ2=27.9, P<.001) according to DSM-III-R criteria.
To summarize, previous work has generally supported the construct validity of Hare's PCL-R in relation to assessments of DSM-III-R Axis I and II disorders, based on semistructured interviews. Most of this work, however, has involved North American criminal and forensic samples. We do not know whether the findings reported are generalizable to European forensic psychiatric samples. In addition, to the best of our knowledge, no study has been published that systematically examined the association between PCL-R psychopathy and DSM-IV Axis I and II disorders (American Psychiatric Association, 1994).
