To identify empirical clusters of CBD employing personality traits, as well as patients' sex, age and the age of CBD onset as indicators.
An agglomerative hierarchical clustering method defining a combination of the Schwarz Bayesian Information Criterion and log-likelihood was used.
Three clusters were identified in a sample of n = 110 patients attending a specialized CBD unit a) “male compulsive buyers” reported the highest prevalence of comorbid gambling disorder and the lowest levels of reward dependence; b) “female low-dysfunctional” mainly included employed women, with the highest level of education, the oldest age of onset, the lowest scores in harm avoidance and the highest levels of persistence, self-directedness and cooperativeness; and c) “female highly-dysfunctional” with the youngest age of onset, the highest levels of comorbid psychopathology and harm avoidance, and the lowest score in self-directedness.
Sociodemographic characteristics and personality traits can be used to determine CBD clusters which represent different clinical subtypes. These subtypes should be considered when developing assessment instruments, preventive programs and treatment interventions.