Aggressiveness and suicidal behavior: A cluster analytic approach in clinical and non-clinical subjects

Aims: An attempt was made to examine the relationship between different dim ensions of aggression and suicidal behavior in clinical and non-clinical su bjects. Method: A clinical sample of 68 suicide attempters (39 females, 29 males) and a matched sample of 70 non-clinical subjects (38 females, 32 mal es) were subgrouped by cluster analytic techniques. Classification variable s included the scales of the Aggressive Factors Questionnaire (AFQ). Discri minant function analyses were performed on the clusters to determine the qu ality of group separation and to assess the power of each input variable. E xternal variables such as the amount of hopelessness, suicidal ideation and the number of suicide attempts were used to assess the clinical relevance of the identified clusters. Results: Two subgroups were identified in each sample. Variables that contributed most to group separation were Auto-aggre ssion/Depression and Spontaneous Aggression in the non-clinical sample and Excitability in the clinical sample, respectively. The clusters identified within the non-clinical sample differentiated between ideators and non-idea tors. Cluster allocation within the clinical sample was indicative of the n umber of suicide attempts. Conclusions: Clusters defined on the basis of di fferent dimensions of aggressiveness turned out to be a classification with significant implications for the assessment of suicidal behavior.