رابطه آلکسیتیمیا و اختلال شخصیت وابسته: تجزیه و تحلیل بعدی

The present study had two aims and used two different samples. The first aim was to determine if alexithymia and dependent personality disorder (DPD) are distinct or overlapping constructs. The second aim was to determine the specificity and the stability of the relationship between alexithymia and DPD. The first study used exploratory principal components analysis (PCA) in a sample of 477 non-clinical subjects who completed three questionnaires measuring alexithymia (Twenty item Toronto Alexithymia Scale, i.e. TAS-20), dependent personality disorder (Dependent Personality Questionnaire, i.e. DPQ) and depression (Beck Depression Inventory-II, i.e. BDI-II). The second study used a sample of 305 subjects consecutively admitted to an outpatient department of legal medicine. The subjects completed (at admission and 3 months later) the Structured Clinical Interview for DSM-IV, screen questionnaire (SCID-II-SQ), the TAS-20 and the BDI. Multiple regressions were done. For the first study, the PCA yielded a four-factor solution with no overlap of the significant factor loadings for the items from each scale and with the factors corresponding to their respective construct. For the second study, multiple regressions showed that only avoidant personality disorder was an independent predictor of the TAS-20 scores. Alexithymia is a construct that is distinct and separate from DPD and depression. Alexithymia is not a stable feature of DPD while it is a core feature of avoidant personality disorder.

Alexithymia is a disorder of affect regulation characterized by difficulties identifying and describing emotions, an impoverished fantasy life and externally oriented thinking (Taylor et al., 1997). Alexithymia is associated with depression, anxiety and addiction notably substance and alcohol abuse and has also been implicated in the pathogenesis of physical illness notably psychosomatic disease (Hintikka et al., 2001, De Rick and Vanheule, 2007 and Kojima, 2012). From a conceptual point of view, alexithymia can be considered either as a state or a trait. In 1980, the DSM-III (American Psychiatric Association, 1980) identified excessive interpersonal dependency as a separate disorder of personality, the dependent personality disorder (DPD), for the first time. Numerous studies have since found significant relationships between interpersonal dependency or DPD and anxiety or depression (Bornstein, 1993 and Disney, 2013). Moreover, one meta-analytic review reported that excessive interpersonal dependency was associated with increased risk for physical illness in prospective studies (Bornstein, 1998). Just like alexithymia, excessive interpersonal dependency can constitute either a state or a trait (DPD). The relationships between alexithymia or DPD and psychiatric or somatic illness suggest the existence of a potential conceptual common construct between the two dimensions. Firstly, Ruesch (1948) described the infantile personality as the core problem of psychosomatic disease. The infantile personality is characterized by dependency: “the infantile personality relies upon cues received from others, which necessitates permanent guidance and support by others” (p. 138). The infantile personality is also characterized by an excessive degree of conformance and several features indicating a disorder of affect regulation: “the infantile personality does not possess the necessary techniques for social interaction and communication… signs used for self-expression originate in the somatic sphere or are related to action, and interpersonal relations on the level of verbal symbolization are rudimentary or non-existent” (p. 142). Secondly, the pattern of correlations between DPD or alexithymia and the five-factor model of personality is similar. Luminet et al. (1999) and De Gucht et al. (2004) found that alexithymia was positively correlated with neuroticism and negatively correlated with extraversion and openness to experience. For DPD, Bornstein and Cecero (2000) conducted a meta-analysis of studies exploring the relationship between DPD and the Big Five and reported that dependency was positively correlated with neuroticism and negatively correlated with extraversion and openness to experience. Strong associations between alexithymia and several dimensions of the Cloninger׳s psychobiological model of personality have been reported. Grabe et al. (2001) in psychiatric patients and Lee et al. (2010) in healthy subjects found that low reward dependence and low self-directedness were predictors of alexithymia, rated by the Twenty item Toronto Alexithymia Scale (TAS-20). To our knowledge, the relationship between the Cloninger׳s model of personality and DPD has not been studied but one study (MacDonald et al., 2013) in psychiatric outpatients has explored the relationships between attachments dimensions and temperament and personality. Attachment anxiety dimension was significant correlated with self-directedness (negative correlation) and with DPD dimensional score (positive correlation) whereas attachment avoidance dimension was significant correlated with self-directedness or reward dependence (negative correlations) and with DPD dimensional score (positive correlation). This study suggested that DPD dimension and self-directedness or reward dependence could be correlated and that these associations could be mediated by attachment dimensions. Thirdly, the relationship between adult attachment style and alexithymia or DPD is similar. Several studies in university students reported firstly significant positive associations between alexithymia and insecure attachments styles notably preoccupied and avoidant/fearful styles and secondly negative significant associations between alexithymia and secure attachment style (Montebarocci et al., 2004, Wearden et al., 2005 and Besharat and Khajavi, 2013). Concerning DPD one study on 357 psychiatric patients (MacDonald et al., 2013) and one study on 1407 non-psychiatric subjects (Brennan and Shaver, 1998) found significant associations between insecure attachment styles, notably preoccupied and avoidant/fearful styles and DPD. Moreover Livesley et al. (1990) have proposed that the criteria of DPD could be more represented by two behavioral dimensions called dependent and attachment behaviors than a unitary dimension. Using confirmatory factorial analysis on the DSM-IV items of the DPD Gude et al. (2006) reported in a sample of 1078 psychiatric patients a two-factor structure of the DPD named perceived incompetence and dysfunctional attachment. Despite the potential link between the concepts of alexithymia and DPD few studies have explored the relationship between these two dimensions. To the best of our knowledge, only five studies have examined the link between DPD and alexithymia. The first study (Bach et al., 1994) reported significant associations in a sample of 182 overweight women between the score of the Toronto Alexithymia Scale (TAS) and several dimensional scores of personality disorders, following DSM-III-R criteria, rated by the Personality Diagnostic Questionnaire revised. 55.5% of the women had at least one DSM-III-R axis I lifetime diagnosis, essentially affective or anxiety disorders. DPD as well as avoidant and schizotypal personality disorders were significantly associated with the scores of the TAS. In this study depression was not controlled for. The second study De Rick and Vanheule, 2007 has examined the relationship between alexithymia rated by the Bermond Vorst Alexithymia Questionnaire and DSM-IV personality disorders traits in a sample of 101 alcoholic inpatients and in 101 healthy subjects of the general population. In this study depression was controlled using the Beck Depression Inventory. Personality disorder traits were assessed using the Assessment of DSM-IV Personality Disorders. In the alcoholic sample positive significant associations were found between alexithymia and schizoid, avoidant and antisocial personality traits. These results were not replicated in the non-clinical sample. The third study (Loas and Cormier, 2009) compared 18 DPD subjects, 108 nondependent personality disorder subjects and 179 subjects without personality disorders on the Twenty-item Toronto Alexithymia Scale (TAS-20) (Parker et al., 1993) and the abridged version of the Beck depression inventory (BDI) scores. The sample consisted of 305 subjects admitted to an outpatient department of legal medicine. The diagnoses of personality disorders following DSM-IV criteria were reached using the Structural Clinical Interview II (SCID-II) questionnaire. DPD subjects had significant higher scores on the TAS-20 than the two other groups even when depression was controlled for. The fourth study (Loas and Cormier, 2008) used the same sample as the second study for a follow-up study. The subjects completed at 3 months the TAS-20 and BDI. Contrary to the results at initial assessment, DPD subjects presented non-significant differences on the TAS-20 comparatively to the subjects with non-dependent personality disorders at follow-up. This last result suggested that alexithymia had a lower absolute stability in DPD than in other personality disorders. The fifth study (Nicolo et al., 2011) explored the relationship between alexithymia, rated by the TAS-20 and personality disorders evaluated by the SCID-II. The sample included 388 adult and adolescent outpatients. The authors used a categorical approach: the subjects were classified as having alexithymia, indeterminate alexithymia or no alexithymia using the cutoff scores of the TAS-20. Significant associations between alexithymia and personality disorders were found for four personality disorders: avoidant, dependent, depressive and passive-aggressive personality disorders. When depression was controlled for, using the depression subscale of the SCL-90-R, alexithymia remained associated only with avoidant personality disorder (Nicolo et al., 2012). One limitation on these four last studies was the use of categorical diagnosis of DPD although dimensional definition could have better represented the underlying constructs and presented higher reliability (Morgan et al., 2013). The purpose of the present study was to examine the relationship between alexithymia and dependent personality constructs, firstly using a dimensional definition of DPD and secondly controlling for the potential effect of depression. To explore this relationship two studies were conducted. Firstly, factor analysis of the correlation matrix comprising items from both questionnaires measuring alexithymia (TAS-20), dependent personality (Dependent Personality Questionnaire, DPQ) (Tyrer et al., 2004) and depression (Beck Depression Inventory-II, BDI-II) (Beck et al., 1996) was carried out. If alexithymia, dependent personality and depression are distinct, then factor analysis of a correlation matrix comprising items from both the TAS-20, DPQ and BDI-II should produce different factors corresponding to their respective construct. Secondly, the relationship between alexithymia and DPD was explored by examining the relationships between alexithymia, depression and each personality disorder evaluated at two periods of time. The specificity and the stability of the relationship between alexithymia and DPD were tested.