Alexithymia and somatisation: A quantitative review of the literature

https://doi.org/10.1016/S0022-3999(02)00467-1Get rights and content

Abstract

Objective: To present a quantitative review of the empirical literature on somatisation and alexithymia. Methods: Medline and PsycLIT searches for relevant studies were conducted. Meta-analytical techniques were applied to quantify the strength of the associations that were found. Results: A small to moderate relationship was found between general alexithymia and somatic symptom reporting. The alexithymia dimension measuring difficulty in identifying feelings showed the strongest association with symptom reports. The alexithymia dimension measuring externally oriented thinking was virtually unrelated to somatic symptom reports. Compared to healthy control populations, subjects suffering from a somatoform condition were significantly more alexithymic, with effect sizes ranging from moderate to large. The studies comparing somatoform to medical or psychiatric conditions yielded inconclusive results. Conclusions: By means of quantitative procedures, an association between general alexithymia and somatic symptom reporting was established. Due to the use of questionnaires that can only check for symptoms, not whether these symptoms are medically explained or not, it is however not possible to draw conclusions on somatisation properly defined. The inconsistent results found when comparing somatoform conditions to medical and psychiatric controls may be attributed to confounding variables. In future studies, these variables should be statistically controlled to establish a more consistent pattern of associations between somatoform conditions and alexithymia. It is, however, equally feasible that this inconsistency reflects the nonspecific character of the association between alexithymia and somatisation. The presence of only one prospective study does not allow to draw conclusions on alexithymia as a predisposing factor for somatisation.

Introduction

Somatisation has been defined in different ways. In spite of their differences, these definitions have one element in common, namely the presence of somatic symptoms that can not be (adequately) explained by organic findings [1]. Epidemiological studies have demonstrated a high prevalence of such symptoms in the general population [2] and in all medical settings [3], [4]. Somatisation has been operationalised in distinct ways. It can be measured along a continuum, in which case it is defined as the number of medically unexplained symptoms reported, or as diagnostic categories, grouped together under the heading of somatoform disorders in the Diagnostic and Statistical Manual of Mental Disorders (DSM) [5], [6], [7]. In addition, a number of discrete somatoform disorders or functional somatic syndromes have been described in the literature [8].

A large number of factors have been studied to explain the phenomenon of somatisation, including a number of personality traits, hypothesised as risk factors for the development or persistence of medically unexplained symptoms. This review focuses on one of them, alexithymia. Sifneos [9] was the first to use the word alexithymia, which literally means “no words for feelings” in order to capture what he considered to be a core characteristic of patients suffering from psychosomatic diseases. Although the alexithymia concept has originated within psychoanalysis, the wider scientific arena has subsequently taken an interest in it. This has resulted in an accumulating body of research supporting the view that alexithymia is reflecting a deficit in the cognitive processing and regulation of emotions [10], [11]. Although alexithymia has been originally conceptualised as a dimensional construct, distributed normally in the general population [10], cut-off scores have been established empirically, enabling researchers to distinguish alexithymic from non-alexithymic subjects [12], [13].

A large number of scales have been developed in an attempt at quantifying alexithymic characteristics. Most of the early measures have, however, been rather hastily constructed and have subsequently been shown to lack reliability and validity. The self-report Toronto Alexithymia Scale (TAS) was the first measure that was developed with reasonable concern for test development procedures and psychometric quality. In an attempt at further improving its psychometric properties, three subsequent versions of the TAS have been developed: the TAS, the TAS-R and the TAS-20. These three versions differ from each other with respect to total number of items and factor structure. The following dimensions are however part of all three versions: (1) difficulty identifying feelings and distinguishing between feelings and bodily sensations (DIF), (2) difficulty describing feelings (DDF) and (3) externally oriented thinking (EOT) (for a detailed review of measurement issues, see Ref. [14]). More recently, there has been some discussion regarding the factor structure of the TAS-20, with some studies showing a two-factor structure instead of a three-factor structure [15], [16]. The internal consistency of the EOT subscale has been demonstrated to be considerably lower than that of the other subscales [12], [17], which could be attributed to the presence of a large number of negatively keyed items in this subscale [17]. In addition, because one of the subscales of the TAS (“reduced daydreaming”) had little theoretical coherency with the other subscales, this dimension was dropped from the subsequent versions of the scale. From a theoretical point of view, this implies that alexithymia as measured by the TAS-R and the TAS-20 is not completely congruent with the alexithymia construct as defined by Bagby and Taylor [14].

Next to the TAS, two new questionnaires for measuring alexithymia have been developed recently, the psychometric qualities of which look promising: (1) the Amsterdam Alexithymia Scale (AAS) [18], a self-report measure of alexithymia consisting of five factors, and (2) the Observer Alexithymia Scale (OAS) [19], [20], an observer-rated questionnaire intended for use by lay people and clinicians. The major problem with the OAS is that the connection between somatisation and alexithymia is considered as an a priori fact as “somatizing” is one of the five subscales of the OAS. Whereas the AAS does not suffer from this problem, so far no studies have been published on the relationship between somatisation and alexithymia, as measured by the AAS.

Despite the criticism that may be passed on the TAS, it remains at present the most widely used instrument for assessing alexithymia. It is also the only scale that has been cross-validated in different languages. For that reason, this review will be confined to studies that have used one of the three versions of the TAS, in an attempt at quantifying the association between alexithymia and somatisation. Subsequently, these review data will be discussed, not only from the perspective of potentially confounding factors, but also in the light of the question whether the association between alexithymia and somatisation is specific or not.

The review addresses three questions: (1) Is there an association between alexithymia and somatisation and, if so, what is the magnitude of this association? (2) Is there evidence supporting the fact that alexithymia is a predisposing factor for somatisation? Finally, (3) Do the different dimensions of alexithymia as measured by the TAS (DIF, DDF and EOT) relate differently to somatisation?

Section snippets

Search strategy

Studies on the relationship between alexithymia and somatisation were identified by means of systematic searches of the databases Medline and PsycLIT with somatisation, medically unexplained somatic symptoms, functional somatic symptoms, somatoform disorder(s), functional somatic syndrome(s), chronic fatigue syndrome, fibromyalgia, functional gastrointestinal disorders, functional dyspepsia, irritable bowel syndrome, premenstrual syndrome, interstitial cystitis, temporomandibular disorders,

Relationship between number of symptoms reported and alexithymia

A total number of 16 publications were retrieved, yielding 18 study samples. Table 1 summarizes the findings.

With respect to total alexithymia score, results were quite consistent. With the exception of two studies (Study 3 [27]) [33], all studies have found a significant positive correlation with different measures of somatic symptom reporting. The mean correlation coefficient, all study samples combined and weighted by sample size, is .23, indicating a small to moderate effect size. Regarding

Relationship between somatisation and alexithymia

The findings summarised in Table 1, Table 3 establish the existence of a small to moderate relationship between somatisation and alexithymia. Two remarks are nevertheless in place. A first remark relates to the measurement of somatisation. Whereas even a broad definition of somatisation requires the presence of physical complaints that cannot be accounted for by organic pathology [1], most studies have used questionnaires that can only check for symptoms, not whether these symptoms are

Acknowledgements

We would like to thank Stan Maes, PhD (Leiden University) and Benjamin Fischler, MD, PhD (University of Leuven) for reading and commenting upon an earlier version of the manuscript.

References (69)

  • M Bach et al.

    Independency of alexithymia and somatisation

    Psychosomatics

    (1996)
  • BJ Cox et al.

    Alexithymia in panic disorder and social phobia

    Compr Psychiatry

    (1995)
  • IJ Deary et al.

    Neuroticism, alexithymia and medically unexplained symptoms

    Pers Individ Differ

    (1997)
  • M Bach et al.

    Alexithymia and somatisation: relationship to DSM-III-R diagnoses

    J Psychosom Res

    (1994)
  • H Miyaoka et al.

    A psychiatric appraisal of “glossodynia”

    Psychosomatics

    (1996)
  • MA Lumley et al.

    The relationship of alexithymia to subjective and biomedical measures of disease

    Psychosomatics

    (1997)
  • TM Tojek et al.

    Stress and other psychosocial characteristics of patients with psychogenic nonepileptic seizures

    Psychosomatics

    (2000)
  • B Wood et al.

    Personality and social attitudes in chronic fatigue syndrome

    J Psychosom Res

    (1999)
  • KR Cohen et al.

    Is alexithymia related to psychosomatic disorder and somatizing?

    J Psychosom Res

    (1994)
  • GJ Taylor et al.

    Alexithymia and somatic complaints in psychiatric out-patients

    J Psychosom Res

    (1992)
  • M Joukamaa et al.

    Alexithymia in primary health care patients

    J Psychosom Res

    (1995)
  • J Kauhanen et al.

    Social factors in alexithymia

    Compr Psychiatry

    (1993)
  • RD Lane et al.

    Sociodemographic correlates of alexithymia

    Compr Psychiatry

    (1998)
  • S Berthoz et al.

    Alexithymia and anxiety: compounded relationships? A psychometric study

    Eur Psychiatry

    (1999)
  • K Honkalampi et al.

    Depression is strongly associated with alexithymia in the general population

    J Psychosom Res

    (2000)
  • JDA Parker et al.

    Alexithymia and depression: distinct or overlapping constructs?

    Compr Psychiatry

    (1991)
  • MA Lumley

    Alexithymia and negative emotional conditions

    J Psychosom Res

    (2000)
  • TN Wise et al.

    The relationship between somatosensory amplification, alexithymia, and neuroticism

    J Psychosom Res

    (1994)
  • P Spinhoven et al.

    Somatization and somatosensory amplification in psychiatric outpatients: an explorative study

    Compr Psychiatry

    (1997)
  • R Kellner

    Somatisation. Theories and research

    J Nerv Ment Dis

    (1990)
  • K Kroenke et al.

    Symptoms in the community. Prevalence, classification, and psychiatric comorbidity

    Arch Intern Med

    (1993)
  • JI Escobar et al.

    Abridged somatisation: a study in primary care

    Psychosom Med

    (1998)
  • P Fink

    The use of hospitalizations by persistent somatizing patients

    Psychol Med

    (1992)
  • Diagnostic and statistical manual of mental disorders

    (1980)
  • Cited by (252)

    View all citing articles on Scopus
    View full text