Editor:

We read with interest the report by Dr. Grant¹ regarding the co-occurrence of personality disorders in persons with kleptomania. In particular, we noted that of 12 subjects with kleptomania who had personality disorders, one-fourth had Schizoid Personality Disorder as diagnosed by SCID-II. This combination of compulsive behavior (stealing) and schizoid traits suggests to us that such subjects may, in fact, have a Pervasive Developmental Disorder (PDD).

The neurodevelopmental history is crucial in diagnosing PDDs such as Asperger’s Disorder and PDD NOS (not otherwise specified). In our experience, and in the literature, affected persons are not reliable historians in characterizing their levels of social adjustment.^2,3 Absent developmental history, adults with PDDs may appear to have Schizoid Personality Disorder. In a factor analysis of the Personality Disorders, Shedler and Westen⁴ found these traits to represent most prototypically Schizoid Personality Disorder: lacks close friends and relationships; lacks social skills and tends to be socially awkward, with inappropriate social behavior; appears to have a limited or constricted range of emotions; and feels like an outcast. These traits can also be seen in PDDs: affected persons demonstrate impaired reciprocal social behavior and impaired social communication. The third axis of impairment in PDDs is that of obsessive narrow interests. Kleptomania could represent such an interest.

Chen et al.⁵ reported the case of a 21-year-old man with Asperger’s Disorder whose previous diagnoses included schizophrenia and kleptomania. He compulsively stole and also hoarded items such as cups and bags. He was expelled from high school for obsessive stealing. Siponmaa et al.⁶ performed comprehensive diagnostic assessments on a subgroup of juvenile offenders. Using DSM-III and International Classification of Diseases (ICD)-9, 53 percent of subjects had a diagnosis of Mixed Personality Disorders. Using DSM-IV and ICD-10 and the PDD diagnostic criteria of Gilberg, they found that 30 percent of subjects met criteria for PDDs (confirmed and probable cases). We hypothesize that as more studies use expanded diagnostic systems with collateral information, psychiatrists may find an increasing prevalence of PDD in groups previously characterized as having Cluster A Personality Disorders.

In Dr. Grant’s study, kleptomaniacs with Personality Disorders had an average age of stealing onset of 13.4 years, whereas non-personality disordered subjects’ mean age at onset was 27.4 years (p < .006). The age discrepancy is broad and consistent with the childhood onset of obsessive behavior in PDDs.

Dr. Grant’s series of 12 subjects with kleptomania who sought treatment at a university center includes 9 females and 3 males. The ratio of males to females in PDD samples typically is two to four males to one female.⁷ There are two possible reasons for this significant gender discrepancy (female overrepresentation). Generally, females are higher utilizers of health care services than are males. This suggests males with kleptomania may be less treatment-seeking than are females. Also, Wolfe et al.⁸ suggest that the phenotype of females with PDDs is poorly understood and may not closely track the male phenotype. We think an expanded developmental history with collateral information studying Dr. Grant’s cohort could be very informative. This should include systematic screening for PDDs, particularly since DSM-IV-TR guidelines for diagnosing Schizoid and Schizotypal Personality Disorders require that PDDs be ruled out.

We thank Dr. Grant for this interesting study, and we look forward to further exploratory comparisons of Compulsive Behaviors, Cluster A Disorders, and the adult manifestations of PDDs.