Examining the utility of a temperament model for modelling non-melancholic depression


G. Parker School of Psychiatry, University of New South Wales, and the Mood Disorders Unit, Prince of Wales Hospital, Sydney, Australia


Objective: Depression is currently modelled dimensionally, along severity, duration and recurrency dimensions. An alternative model allows dimensional expressions of temperament and personality to influence risk to onset as well as persistence. Here, we examine the utility of a temperament model.

Method: A questionnaire assessing temperament dimensions and a number of depression variables was administered to a large routine general practice sample and with the temperament measure also completed by a small clinical sample.

Results: `Anxious worrying' and `irritable' dimensions were identified as internalizing and externalizing expressions of a trait anxiety dimension, three other `temperament' dimensions (i.e. `introversion', `self-centred' and `obsessive') were refined, while a `self-blame' dimension intruded into the factor analytic solution. High scores on the `anxious worrying' dimension were associated with all depression parameters. The `irritable', `introversion' and `self-blame' dimensions were less clearly linked with depression variables, while higher scores on the `self-centred' and `obsessional' dimensions did not appear to increase the chance of depression onset, persistence or recourse to treatment.

Conclusion: A temperament-based approach appears to have some conceptual utility in modelling depression, and particularly, the non-melancholic disorders. It is likely, however, to require complementing with refined at-risk personality dimensions.