Typology of mood assessment curves by means of cluster analysis of computerized mathematical descriptions of time-series mood data collected from depressive inpatients treated with antidepressants.

A recently developed method for the computerized description of mood curves was applied to self-evaluation mood data of a large sample of 136 depressive inpatients. The values of the mathematical parameters from this description were analysed by cluster analysis to determine different types of mood courses of patients undergoing therapy with antidepressants. In general, the most unfavorable type of mood course was overrepresented in the group of neurotic depressive patients, the most favorable one in the group of endogenous depressive patients. But in this latter diagnostic group, a remarkable proportion of patients with the unfavorable mood course was observed, indicating the well-known phenomenon of non-response to antidepressants. The results can be interpreted as a validation of the new method of computerized description of mood curves. This method might lead to interesting possibilities in drug treatment evaluation.