The distribution of psychopathy among a household population: categorical or dimensional?

The study aimed to examine the distribution of psychopathic traits in a representative household population to identify whether a transition point is reached on a continuum of psychopathy to indicate a 'disease' or categorical entity.

Mixture Poisson distribution and epidemiological procedures were used to examine the distribution of the Psychopathy Checklist Screening Version (PCL:SV) score in a sample of 638 adults in households in Great Britain. Analysis aimed to identify a cut-off within the population using the distribution of continuous scores (mean and (1/2) SD) and validate a 'probable' psychopathy category using a scale of social and behavioural problems as an external validator.

The distribution of psychopathy within the population was quasicontinuous, represented by a mixture of three-Poisson distributions with differing demography and comorbid Axis I and II psychopathology. Independent calculation indicated a cut-score at 11.8 on the PCL:SV. There was an exponential rise of associated social and behavioural problems at a transition point of 11.3. The prevalence of 'probable' psychopathy was 3.6% (95% CI = 2.3-5.5%) above this level.

The findings suggest a transition from a non-clinical to clinical state of psychopathy which can be defined categorically using a cut-off on the PCL:SV. The cut-off approximates to that previously recommended for identification of a case using the instrument. Above this critical level, individuals are at exceptional risk of compulsory care or incarceration due to multiple social and behavioural problems. Psychopathy should be considered for future inclusion in DSM-V and successfully combines both categorical and dimensional approaches to diagnosis.