Investigation of the psychometric properties of the Toronto Alexithymia Scale in men with and without compulsive sexual behavior.

Abstract

Background: Previous studies have provided initial evidence supporting the association between compulsive sexual behavior (CSB) and alexithymia, but these studies did not investigate the psychometric property of a measure of alexithymia in individuals with CSB, which is necessary.

Aim: This study investigated the factor analysis, reliability, and construct validity of the Toronto Alexithymia Scale (TAS) in men with CSB and control individuals.

Methods: This cross-sectional study included 418 participants (304 individuals with CSB and 114 control individuals) who underwent a semi-structured psychiatric interview and completed the following instruments: the Sexual Compulsivity Scale, Hypersexual Disorder Screening Inventory, Beck Anxiety Inventory, Beck Depression Inventory, TAS, and Barratt Impulsiveness Scale. The eligible participants were men 18 years of age or older who were literate and residing in Brazil. Individuals who met the diagnostic criteria for Goodman's criteria for sex addiction were subsequently assessed for the excessive sexual drive (International Classification of Diseases-Tenth Revision F52.7) criteria. Those who met this second criteria were considered individuals with CSB. Participants who did not reach this point were considered control individuals. We conducted factor analysis, reliability analysis (internal consistency and temporal stability), and discriminant and construct validity analyses.

Outcomes: The outcomes included the TAS total score and scores on TAS factors 1, 2, 3, and 4.

Results: The extracted factors explained 44% of the variance in the TAS. Factor 1 explained 21%, but 3 items (items 2, 9, and 21) did not load onto this factor. The Cronbach's alpha was 0.83, and the reproducibility (intraclass correlation coefficient) was 0.70. The TAS can differentiate between individuals with CSB and control individuals. The different forms of validity were demonstrated through correlations between factors 1 to 4 and the total score, as well as with impulsivity, hypersexuality, sexual compulsivity, and depression. Surprisingly, anxiety was only weakly correlated with factors 1 and 2. Moreover, the TAS-4 score was not correlated with impulsivity.

Clinical implications: The TAS can be used in clinical practice to identify men with difficulties in recognizing subjective experiences, and proper interventions can subsequently be provided to these patients to increase their treatment efficacy.

Strengths and limitations: Various dimensions of alexithymia covary with other key psychopathological symptoms of CSB. This study examined a convenience sample. The results cannot be generalized to the broader population. Factors 3 and 4 presented low internal consistency (0.50).

Conclusion: In general, TAS presented good psychometric properties in a sample mainly composed of individuals with CSB.