Temporal stability of antisocial personality disorder: Blind follow-up study at 8 years

The study objective was to examine the temporal stability of the antisocial personality disorder (ASPD) diagnosis based on whether specific antisocial symptoms were considered to be related to substance abuse. A total of 407 adults who were initially part of a family study of alcoholism and sociopathy were blindly reassessed an average of 8 years later, using the Home Environment and Lifetime Psychiatric Evaluation Record (HELPER) and basing diagnoses on the clinician's best final estimate using all sources of data. 'Narrow' and 'broad' ASPD diagnoses were made at both times based on whether individual symptoms were counted toward diagnosis if they occurred in the setting of significant substance abuse. κ values varied from 0.31 to 0.68, with more restrictive methods of diagnosis being less stable. After deriving estimates of sensitivity and specificity of diagnosis, the probability of being a 'case' could be assigned based on the reported number of conduct problems occurring before age 15 as a clinical covariate for diagnosis. We conclude that diagnosing ASPD without attempting to attribute the cause of individual symptoms to substance abuse results in substantially greater temporal stability. Using a broader definition, the diagnosis of ASPD is highly sensitive (P = .97) and specific (q = 0.93). These findings may allow more accurate diagnosis of ASPD in drug-abusing individuals.