Somatoform disorders as disorders of affect regulation

Abstract

Objective

To determine the role of undifferentiated and dysregulated affects in somatoform disorders by using a multimethod assessment approach of alexithymia.

Methods

Forty patients with ICD-10 somatoform disorders (SoD) and 20 healthy controls, matched for age, education and sex, were included in the study. Alexithymia was assessed using the Toronto Alexithymia Scale (TAS-20), the Affect Consciousness Interview (ACI), and the Levels of Emotional Awareness Scale (LEAS). All classifications were made blinded with regard to clinical status.

Results

Scores of the ACI and the TAS-20 showed that alexithymia is higher in SoD than in healthy controls. No differences were found on the LEAS. In terms of the multidimensionality of the alexithymia construct, our results indicate a specific positive association between SoD and a proneness to experience undifferentiated affects. The three subfactors of the TAS-20 were differentially related to non-self-report measures of alexithymia and to negative affectivity (NA). Only the cognitive facet of the TAS-20 (externally oriented thinking [EOT]) was related to the LEAS and the ACI. In contrast, the affective facets of the TAS-20—difficulties identifying feelings (DIF) and difficulties describing feelings (DDF)—were substantially related to NA.

Conclusion

The findings highlight the important role of impaired affect regulation and NA in the process of somatization.

Introduction

Somatizing patients are often characterized by a tendency to experience and communicate psychological distress in form of somatic symptoms and to seek medical help for them [1]. The idea that a diminished capacity to consciously experience and differentiate affects and express them in an adequate or healthy way is an underlying factor of SoD is discussed here.

One of the most elaborated and well-researched constructs for describing personality-related difficulties in the processing and regulation of emotion is alexithymia [2]. Several empirical studies have explored a possible relationship between alexithymia and somatoform disorders (SoD). In two earlier studies using alexithymia only as a dichotomous construct, a high prevalence of alexithymia was found in patients with chronic pain [3], [4]. Other studies reported increased levels of alexithymia in SoD as compared to healthy controls [5], [6], [7]. Patients with SoD were also found to show elevated alexithymia scores, when compared with medically ill patients [8], [9], [10]. Two further studies found no differences in alexithymia between somatizing patients and other clinical control groups [11], [12].

Despite the evidence by the abovementioned studies of a link between alexithymia and somatization, the empirical findings remain controversial [13]. Primarily methodological limitations accounted for the difficulties in the interpretation of data. The first concerns the measurement of alexithymia. In the past decade, findings on alexithymia in patients with SoD were mostly based on self-report measures. Although in current research the Toronto Alexithymia Scale (TAS-20) [14], [15], [16] is the best validated instrument to measure alexithymia, the exclusive use of self-report measures for assessing alexithymia remains subject to criticism. It has been argued that it is to some extent paradoxical to ask alexithymic persons who are characterized by a diminished affective insight to give an accurate estimation of their affective disturbances. Yet, as Lumely [17] commented, although plausible, this must be tested against data. The author as well as the creators of the TAS therefore recommended that studies be conducted using multiple alexithymia measures. To date, studies comparing the TAS-20 with non-self-report measures of alexithymia are few in number.

The interpretation of existing studies concerning the link between alexithymia and somatization is further complicated by the insufficient attention that has been given to the overlap of alexithymia with negative emotional distress. Several studies have found that alexithymia correlated with depression and anxiety [18], [19], [20]. Even though alexithymia is separate from the construct of depression [21], it must be tested whether the association between alexithymia and somatization is mediated by depression or anxiety.

In the present study, we assessed alexithymia using the TAS in combination with non-self-report instruments of alexithymia. We also included a measure of negative affectivity (NA) to control for the effect of NA. In addition to these established instruments, new and promising measures of alexithymia-related constructs that avoid self-ratings have recently been introduced. Two of them, an interview-based measure (Affect Consciousness Interview [ACI]) [22] and a performance-based measure of alexithymia (Levels of Emotional Awareness Scale [LEAS]) [23], were used in this study. The ACI is theoretically grounded in Tomkin's affect and script theory [24], [25] and in contemporary self-psychology [26], [27]. Affect consciousness is considered to reflect a stable pattern of affect (schema) organization. It is operationalized in degrees of awareness, tolerance, emotional and conceptual expression across nine basic affect categories. The LEAS assesses the structural level of affect representation according to a cognitive–developmental model of emotional awareness. The hierarchical model of affect development is based on Piaget's theory of cognitive development [28] and Werner and Kaplan's theories of symbolization and language development [29]. The LEA model postulates five levels of emotional organization ranging from globally organized somatic and action dominated levels to increasingly differentiated organized symbolic levels.

The primary purpose of the present study was to further clarify the associations between alexithymia and somatization. Based on the abovementioned research strategy, we therefore sought to determine whether patients with SoD are more alexithymic then healthy controls. The second aim of the study was to further investigate the convergent and discriminant validity of the TAS-20. We therefore addressed the question of how the TAS-20 and non-self-report measures of alexithymia (ACI, LEAS) and a measure of NA were related to each other. The sample used in this study was already described in an article on attachment representation in SoD, which has been submitted for publication [30].