An evaluation of the absolute and relative stability of alexithymia over 11 years in a Finnish general population

Abstract Objective We investigated if alexithymia, a personality construct with difficulties in emotional processing, is stable in the general population. Methods Altogether 3 083 unselected subjects aged 30 and older in Finland completed the 20-item Toronto Alexithymia Scale (TAS-20) in the longitudinal Health 2000 and Health 2011 general population surveys (BRIF8901). The stability of alexithymia at the 11-year follow-up was assessed with t -tests, correlations, and separate linear regression models with base-line and follow-up age, gender, marital status, education, and 12-month depressive and anxiety disorders as confounders. Results The mean score (SD) of the TAS-20 for the whole sample was 44.2 (10.4) in 2000 and 44.2 (10.9) in 2011 ( p = 0.731). The mean score of the TAS-20 subscale Difficulty Identifying Feelings increased by 0.3 points, Difficulty Describing Feelings decreased by 0.6 points and Externally Oriented Thinking increased by 0.3 points. The effect sizes of the changes varied from negligible to small. Age had little effect except for the group of the oldest subjects (75 – 97 years): the TAS-20 mean (SD) score was 49.1 (10.1) in 2000 and 53.1 (10.3) in 2011 ( p < 0.001), the effect size for the increase was medium. TAS-20 score in 2000 explained a significant proportion of variance in TAS-20 score in 2011. Controlling for all baseline confounders improved the model incrementally; the same applied to controlling for confounders at follow-up. Baseline depression or anxiety disorders were not associated with the TAS-20 scores in 2011, whereas current diagnoses were. Conclusions According to our large longitudinal study both the absolute and relative stability of alexithymia assessed with the TAS-20 are high in the adult general population.