a1 Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands
a2 Virginia Institute for Psychiatric and Behavioral Genetics, Richmond, VA, USA
a3 Department of Psychiatry, Virginia Commonwealth University, Richmond, VA, USA
Background Previous research has shown that stressful life events (SLEs) influence the pattern of individual depressive symptoms. However, we do not know how these differences arise. Two theories about the nature of psychiatric disorders have different predictions about the source of these differences: (1) SLEs influence depressive symptoms and correlations between them indirectly, via an underlying acute liability to develop a dysphoric episode (DE; common cause hypothesis); and (2) SLEs influence depressive symptoms and correlations between them directly (network hypothesis). The present study investigates the predictions of these two theories.
Method We divided a population-based sample of 2096 Caucasian twins (49.9% female) who reported at least two aggregated depressive symptoms in the last year into four groups, based on the SLE they reported causing their symptoms. For these groups, we calculated tetrachoric correlations between the 14 disaggregated depressive symptoms and, subsequently, tested whether the resulting correlation patterns were significantly different and if those differences could be explained by underlying differences in a single acute liability to develop a DE.
Results The four SLE groups had markedly different correlation patterns between the depressive symptoms. These differences were significant and could not be explained by underlying differences in the acute liability to develop a DE.
Conclusions Our results are not compatible with the common cause perspective but are consistent with the predictions of the network hypothesis. We elaborate on the implications of a conceptual shift to the network perspective for our diagnostic and philosophical approach to the concept of what constitutes a psychiatric disorder.
(Received May 03 2011)
(Revised August 25 2011)
(Accepted August 31 2011)
(Online publication November 18 2011)