Background: The study investigated the following hypotheses: (1) Alexithymi
a is significantly associated with dissociation. (2) Pathological versus no
npathological dissociation is associated with alexithymic traits. Methods:
Psychiatric in- and outpatients (n = 173) and nonclinical subjects (n = 38)
were investigated with the Toronto Alexithymia Scale (TAS-20), the FDS (Ge
rman version of Dissociation Experience Scale) and the Symptom Check List (
SCL-90-R; GSI). Correlation analyses followed by MANOVA and logistic regres
sion were performed. Results: Significant correlations and partial correlat
ions, controlling for GSI, were observed between dissociation and alexithym
ia, The MANOVA demonstrated significantly higher scores for the two TAS-20
dimensions 'difficulty identifying feelings' and 'difficulty expressing fee
lings' in the group with pathological dissociation. On the basis of the TAS
-20 subscores, logistic regression analysis correctly classified 72.5% of t
he cases into the pathological and the nonpathological dissociation group.
Conclusions: These results support our hypothesis that pathological traits
of dissociation are highly associated with alexithymia. A model is discusse
d in which alexithymic characteristics may contribute to the development of
pathological dissociation and stress-related disorders such as posttraumat
ic stress disorder. Copyright (C) 2000 S. Karger AG, Basel.