Development of the CAPP
All members of the CAPP family are based on the same conceptual model. This model was derived in two steps. First, we considered the rich clinical descriptions of PPD symptomatology that have been provided by many scholars. We examined the literature to determine what are the core features of PPD; and equally important, what are not core features of PPD. The second step in the development process was the completion of detailed interviews with clinicians of diverse theoretical backgrounds - in Europe and North America - who have experience of patients with PPD. This is an important step because some other measures of psychopathy have been developed for non-clinical purposes and there is a danger that they lack clinical relevance - and clinical credibility. We asked clinicians to determine which symptoms they considered salient. These procedures revealed - both from theoretical and clinical perspectives - that existing measures of PPD traditionally suffered from significant construct under-representation; that is, they did not include important symptoms of the disorder. Through this process we identified six distinct domains (i.e., Attachment, Behavioural, Cognitive, Dominance, Emotional and Self) using rationale criteria (see “The CAPP Model” section for details).
It is noteworthy that all pathological expressions of the six CAPP domains can be linked theoretically to violence risk, to responsiveness to treatment, and to risk management more broadly. All pathological expressions of the domains have forensic relevance - in particular they have relevance for violence risk.
Having identified the domains of interest, we attempted to identify and define the key symptoms within each domain, ensuring that both the symptom and the domains were conceptually distinct. We used a lexical approach: essentially this approach is predicated on the view that the most important differences in human transactions will be encoded within natural language. Large numbers of symptom descriptive terms have been identified and subjected to detailed structural analysis; indeed, the Big Five view of human personality has its roots in this lexical approach. Being based on common language adjectival trait descriptors means that translation into other languages is facilitated and cross-cultural generalisability is likely to be enhanced.
After “deconstructing” each symptom into a relatively small number of trait-descriptive adjectives (usually 3), we then identified a number of behaviours that seemed to be closely linked to these traits. For example, for the first version we identified the behaviours - illustrative indicators - thought by a number of experts to be relevant to institutional adjustment and conduct.