Is there evidence for a latent class called ‘hypomanic temperament’?

Background: Affective disorders belong to the most common psychiatric disorders. Several risk factors have been postulated and empirically investigated. Researchers like Akiskal [Interpersonal Factors in the Origin and Course of Affective Disorders, Gaskell, London, 1996] have pointed out the associations between sub-affective temperaments and affective disorders. However, no study has dealt with the issue whether there is a latent class of such sub-affective temperaments or if such temperaments are best conceptualized as fully dimensional. We investigated whether the Hypomanic Personality Scale [J. Abnorm. Psychol. 121 (1986) 214–222] as an indicator of hyperthymia is taxonic in structure. Methods: We chose two different samples to address this issue: A sample of young adults (n=1966) and another sample of adolescents (n=4045). We ran MAXCOV-HITMAX analyses based on identical subsets of items in both samples. Results: Neither in the sample of young adults nor in the sample of adolescents there was evidence for a latent class called ‘hypomanic temperament’. Limitation: Only one indicator for vulnerability and one procedure to test for latent classes was used. Furthermore, we do not know how many of our sample had a life-time history or current affective disorders. Conclusions: The hypomanic–hyperthymic temperament is best conceptualized as a dimension in the general population. However, before drawing final conclusions about the taxonicity of the risk for affective disorders, more research is needed using different measures, samples and methods to resolve this question of the dimensionality of vulnerability. Additionally, the question remains open how to conceptualize mania itself.