alexithymia, body image and eating disorders

In: Body Image: Perceptions, Interpretations and Attitudes ISBN 978-1-61761-992-2 Editor: Sophia B. Greene © 2010 Nova Science Publishers, Inc.

Chapter 7

ALEXITHYMIA, BODY IMAGE AND EATING DISORDERS Domenico De Berardis1,2, *, Viviana Marasco1, Daniela Campanella1, Nicola Serroni1, Mario Caltabiano1, Luigi Olivieri1, Carla Ranalli1, Alessandro Carano2, Tiziano Acciavatti2, Giuseppe Di Iorio2, Marilde Cavuto3, Francesco Saverio Moschetta1, and Massimo Di Giannantonio2 1

NHS, Department of Mental Health, Psychiatric Service of Diagnosis and Treatment, Hospital “G. Mazzini” Teramo, Italy 2 Department of Neurosciences and Imaging, Chair of Psychiatry, University “G. d’Annunzio” of Chieti, Italy 3 IASM, L’Aquila, Italy

ABSTRACT It is widely recognized that the body dissatisfaction and an excessive concern about body weight and shape are core characteristic of Eating Disorders (EDs) and are used to determine self-worth. Recently, there was an increased interest about the body image as a multidimensional issue that involves perceptual, attitudinal and behavioral characteristics. Many researchers have focused their attention mainly to the perceptual and attitudinal aspects of body image whereas only few studies have investigated the behavioral consequences related to a negative body image. Moreover, it is known that alexithymia may play an important role in EDs: specifically alexithymics patients may show a higher psychological distress than nonalexithymics and the presence of an alexithymic trait may be related to a higher severity of EDs themselves. Some core aspects of alexithymic construct, as a difficulty in distinguishing emotional states from bodily sensations, may be more characterized in patients with EDs and a possible explanation might be that ED * Correspondence: Domenico De Berardis, MD, PhD. NHS Dipartimento di Salute Mentale, Servizio Psichiatrico Diagnosi e Cura, Ospedale Civile “G. Mazzini” Teramo, p.zza Italia 1, 64100 Teramo (Italy) • Tel. +39 0861429708 • Fax +39 0861429706 • E-mail: [email protected]

2

Domenico De Berardis, Viviana Marasco, Daniela Campanella, et al. patients may appear dramatically and deeply incapable of being in touch with their inner emotive world. As consequence, these subjects may focus their attention on negative perceptual aspects of body bypassing emotional experiences. Taken together, these findings may suggest that alexithymia and body image disturbances may be strongly correlated in EDs and, therefore, the aim of this paper will be to elucidate these relationships along with the presentation of a clinical study on 64 patients with a DSM-IV diagnosis of anorexia nervosa.

INTRODUCTION: THE ALEXITHYMIA CONSTRUCT Coined by Peter Sifneos, the term “alexithymia” was introduced to designate a cluster of cognitive and affective characteristics that were observed among patients with psychosomatic diseases [1]. The alexithymia construct, formulated from clinical investigations, is multifaceted and includes four distinct characteristics: (a) difficulty in identifying and describing feelings, (b) difficulty in distinguishing feelings from the bodily sensations, (c) diminution of fantasy, and (d) concrete and poorly introspective thinking [2]. Alexithymic individuals have affective dysregulation, the inability to self soothe and manage emotions because of a lack of awareness of emotions [3]. Thus, the adaptive informational value of emotions that is important for emotion regulation, often eludes these individuals. These cognitive characteristics have been attributed to an impaired capacity to elevate emotions from a sensorimotor level of experience to a representational level, where they can be used as signaling responses to internal or external events and modulated by psychological mechanisms [4]. The alexithymic patients show significantly higher levels of anxiety, depression, and general psychological distress [5]. Alexithymic individuals are prone to both “functional’’ somatic symptoms and symptoms of emotional turmoil because they are psychologically poorly equipped [6]. The characteristic attributes of alexithymic behaviour are particularly evident in social relationships with high emotional relevance. A persistent affect-avoiding interpersonal behaviour may be maladaptive and can cause disturbances and conflicts in such important relationships, finally contributing to an increased risk of symptoms such as depression or anxiety [7]. On the basis of the recent knowledge from neurobiology, Bermond [8] distinguishes two main forms of alexithymia (Types I and II). Type I alexithymia is characterized by the absence of the emotional experience and, consequently, by the absence of the cognition accompanying the emotion. Type II alexithymia is characterized by a selective deficit of emotional cognition with sparing of emotional experience. In non-clinical samples, the prevalence of alexithymia ranges from 0% to 28% [9]. An increasing body of research indicates that alexithymia features exist not only in classic psychosomatic disorders but also in other severe and chronic somatic diseases and psychiatric disorders such as Eating Disorders (EDs), Somatoform Disorders, Major Depression and other Axis I disorders [10, 11, 12, 13, 14]. Several researchers have pointed out that alexithymic patients may respond poorly both to pharmacotherapy and psychotherapy [15, 16]. Moreover, it has been suggested that, among alexithymics, emotional experiences may not reach full conscious symbolic and verbal elaboration during psychodynamic psychotherapy [17]. Furthermore, patients with

Alexithymia, Body Image and Eating Disorders

3

alexithymia were described as having little interest in introspective and analytical cognitive activity: this may prevent alexithymics from gain benefits from psychotherapeutic interventions that require such activities [18, 19]. It was also reported that the negative reaction of the therapist to an alexithymic patient might be a mechanism through which alexithymia negatively affects the outcome of psychotherapy. Under this view, psychotherapeutic methods, which primarily focus on the verbalization of inner emotional states are probably not the first choice for alexithymic patients [Errore. Il segnalibro non è definito.]. Subjects with alexithymia could undermine these therapeutic strategies in a formal and superficial manner, which leads to pseudotherapeutic effects, based in fact on social desirability [Errore. Il segnalibro non è definito.]. Subjects with alexithymia are not blind to emotional information, but they probably avoid the processing and expression of their own affective states [20]. This could suggest, particularly in the beginning of a psychotherapeutic treatment, that one should be cautious about the demonstrative expression of emotional signals or focusing on emotional conflicts to ensure the therapeutic attachment of high alexithymic patients [21].

MEASUREMENT OF ALEXITHYMIA Many of the older studies on alexithymia raised concerns about their scientific validity, as they were conducted with measures that were shown to lack consistence, reliability and validity such as the Schalling Sifneos Personality Scale (SSPS) and the MMPI alexithymia scale [22]. In 1985, the Toronto Alexithymia Scale (TAS-26) [23] was introduced as a reliable and valid measure of the construct. After several revisions, the actual 20-item version (TAS-20) was developed and, to date, it is the most widely used instrument to measure alexithymia [24, 25]. The TAS-20 is a selfreport scale comprised of 20 items that are rated on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree); five items are negatively keyed. It requires respondents to indicate to what extent they agree with each item. Items 4, 5, 10, 18 and 19 in the TAS-20 are reverse scored. Factor analyses have suggested that TAS-20 consists of three subfactors: difficulty in identifying feelings (DIF), difficulty in describing feelings (DDF), and externally oriented thinking (EOT) [Errore. Il segnalibro non è definito.]. The first factor (DIF) consists of seven items assessing the ability to identify feelings and to distinguish them from the somatic sensations that accompany emotional arousal (e.g., #2: “I have physical sensations that even doctors don't understand” and #6: “When I am upset, I don't know if I am sad, frightened, or angry”). Factor 2 (DDF) consists of five items assessing the ability to describe feelings to other people (e.g., #2: “It is difficult for me to find the right words for my feelings” and #11: “I find it hard to describe how I feel about people”). Factor 3 (EOT) consists of eight items assessing externally oriented thinking (e.g., #8: “I prefer to just let things happen rather than to understand why they turn out that way” and #15: “I prefer talking to people about their daily activities rather than their feelings”). Cut-off scores for the TAS-20 were provided by Taylor et al. [Errore. Il segnalibro non è definito.] and a score ≥ 61 is considered to be within the alexithymic range.

4

Domenico De Berardis, Viviana Marasco, Daniela Campanella, et al.

However, despite the usefulness of this scale, researchers raised some concerns about self-report measures of alexithymia, as the TAS-20. It is reasonable to think that some subjects may be unaware of their difficulties in identifying and describing their feelings, thereby limiting their capacity to report reliably and accurately such deficits on self-report measures [26, 27] . Moreover, as reported by Koiman et al., the TAS-20 has been criticized for not including items that directly assess the reduced fantasy and imaginal thinking facet of the alexithymia construct [28]. Recently, the authors of TAS-20 have developed the Toronto Structured Interview for Alexithymia (TSIA) which demonstrated adequate item characteristics, inter-rater, internal, and retest reliability, and evidence of concurrent and factorial validity, but, to date, results of its use in clinical researches are limited [29]. The TSIA has a hierarchical factor structure consisting of two higher-order factors (domain scales), each composed of two lower-order factors (facet scales). The domain scales are “operative thinking” related to the lower-order externally oriented thinking and imaginal processes and “affect awareness” related to the lower-order factors difficulty identifying feelings and difficulty describing feelings. As stated by author themselves, the most important difference between the TSIA and the TAS-20 is the presence of an imaginal processing scale on the TSIA, that was not retained in the development of the TAS-20, as there were indications that these items had low corrected item-total correlations and were associated with social desirability response bias [Errore. Il segnalibro non è definito.]. The most common rating scales employed to measure alexithymia are reported in the Table 1. Table 1. Most common rating scales employed to measure alexithymia Self–Rating Scales

Diagnostic Interviews

Projective Techniques

Q-sort Methodology

Schalling-Sifneos Personality Scales (SSPS)

Structured Interview for Alexithymia (TSAI)

Rorschach Alexithymia Scale (RAS)

California Q-Set Alexithymia Prototype (CAQ-AP)

MMPI Alexithymia Scale (MMPI-A)

Diagnostic Criteria for Psychosomatic Research (DRPR)

SAT-9 (Archetypal 9 test)

Bermond-Vorst Alexithymia Questionnaire (BVAQ)

Beth Israel Hospital Psychosomatic Questionnaire (BIQ)

Toronto Alexithymia Scale (TAS)

Observer Alexthymia Scale (OAS)

TAS-R (TASRevised)

Levels of Emotional Awareness Scale (LEAS) *

TAS-20 (TAS-


5

Revised)

* The LEAS is often employed to measure alexithymia but it wasn’t specifically designed to evaluate the alexithymic construct.

MAY ALEXITHYMIA PLAY A ROLE IN THE PSYCHOPATHOLOGY OF EATING DISORDERS? Concerning EDs, Bruch [30] suggested that the difficulty to distinguish and describe feelings is one of the main problems in ED patients, related to a sense of general inadequacy and a lack of control over one's life. Moreover, Bruch [31, 32] proposed that in anorexia nervosa an approach that helps patients to become aware of and identify inner states, including emotions, would be useful. It is known that alexithymia may play an important role in EDs: specifically alexithymics patients may show a higher psychological distress than nonalexithymics [33] and the presence of an alexithymic trait may be related to a higher severity of EDs themselves [34, 35]. Furthermore, there are evidences that ED patients are considerably more alexithymic than apparently healthy controls [36, 37] and some studies have specified that alexithymia is more related to the psychological characteristics of patients with EDs than to the eating behavior itself [Errore. Il segnalibro non è definito., 38 39]. Following Bruch’s suggestions, Taylor, Bagby, and Parker [Errore. Il segnalibro non è definito.] conceptualized ED as affect regulation disorders. Using the TAS [Errore. Il segnalibro non è definito., Errore. Il segnalibro non è definito.] in its different versions, empirical studies reported alexithymia rates ranging from 22.9 to 77.1% for patients with anorexia nervosa and from 32.3 to 56% for patients with bulimia nervosa. It is reported that emotional expression may be inversely related to body dissatisfaction [40]. However, concerning non-clinical samples, in an interesting and well conducted study, Quinton and Wagner [41] found that alexithymia did not predict neither total EAT-26 score, nor two EDI-2 subscales measuring aspects of eating psychopathology. They concluded that, although disturbed emotional functioning is a feature of eating disorders, it did not relate

6


directly to the core psychopathology. Some core aspects of alexithymic construct, as a difficulty in distinguishing emotional states from bodily sensations, may be more characterized in patients with EDs and a possible explanation might be that ED patients may appear dramatically and deeply incapable of being in touch with their inner emotive world [42]. As consequence, these subjects may focus their attention on negative perceptual aspects of body bypassing emotional experiences [43]. Taylor et al. [Errore. Il segnalibro non è definito.] stated that in eating disorders a deficit on the cognitive processing of emotions appears, but not on the operational cognitive style, suggesting that the lack of close relationships of anorexic patients could be due to alexithymia. Alexithymia is associated with interpersonal distrust, ineffectiveness, and lack of interoceptive awareness in ED, but it is not related to drive for thinness and body dissatisfaction. On the other hand, starvation, hyperactivity, bingeing and vomiting, could be attempts to regulate distressing and undifferentiated emotional states in these patients [Errore. Il segnalibro non è definito.]. Moreover, there are several reasons to believe that alexithymia construct could play a major role in the illness course of eating disorders: due to their cognitive limitations in emotion regulation, alexithymic individuals with eating disorders may resort to maladaptive self-stimulatory behaviors such as starving, bingeing, or drug misuse to self-regulate disruptive emotions [44]. Speranza et al. [45] found that one of the facets of the alexithymia construct, the difficulty in identifying feelings, was a negative prognostic factor for the longterm outcome of patients with eating disorders. Patients with the greatest difficulties at identifying emotions at baseline were more often symptomatic at follow-up and showed a less favorable clinical improvement. Moreover, the relative stability shown alexithymia over time legitimates its use as a potential prognostic factor in eating disorders. However, when alexithymia is evaluated in patients with EDs, the concomitant presence of anxiety and depression should be controlled in order to find an answer to the unsolved question of alexithymia as a state or a trait in ED. De Groot et al. [Errore. Il segnalibro non è definito.] controlling for depression, found differences in total TAS-26 and in the factor Difficulty in Identifying Feelings, when comparing bulimic patients and controls. Sexton et al. [46] also controlled for depression and used the TAS-26, and they found that the factor Difficulty in Identifying Feelings was more associated with the clinical state of depression in ED, as already suggested by Parker et al. [47] and De Groot et al. [Errore. Il segnalibro non è definito.]. The factor Difficulty in Expressing Feelings did not change when there was a decrease of depression in restrictive anorexics, and it was also associated with personality disorders. They concluded that difficulty in describing feelings was a trait in these patients, and that difficulty in identifying feelings was a state. Corcos et al. [48] confirmed that alexithymia had an increased prevalence in eating disorders, but its occurrence could not be interpreted without taking depression into account. They also reported that increased rates of alexithymia in anorexic patients, compared to bulimic patients, seemed to be more closely related to depression than to an increased alexithymic way of functioning itself. Jimerson et al. [49], using the TAS-26, and controlling anxiety and depression, compared controls and bulimics free of major depression and they found differences between both groups in their difficulty to identify feelings. They suggested that alexithymia, as it is associated with low self-esteem and insecurity, could be enhancing anxiety and depression in bulimic patients, and that it might be secondary to concurrent depression in certain patients. Moreover, Eizaguirre et al. [50] evaluated alexithymia and its relationship with anxiety and depression in 151 females with an ED (25 with anorexia nervosa, restricting subtype, 44 with


7

anorexia nervosa, bulimic subtype, 82 with bulimia nervosa and a control group of 43 females). They showed that patients with ED presented higher rates of alexithymia than controls, but after controlling for anxiety and depression the differences among groups disappeared. Depression and anxiety predicted and correlated positively with alexithymia. Therefore they suggested that alexithymia was closely related to anxiety and depression, and could be considered as a trait or a state in patients with ED.

ALEXITHYMIA AND BODY IMAGE: MAY THE BODY CHECKING BE THE LINK BETWEEN ALEXITHYMIA AND EATING DISORDERS? It is widely recognized that the body dissatisfaction and an excessive concern about body weight and shape are core characteristic of Eating Disorders (ED) and are used to determine self-worth [51, 52]. Recently, there was an increased interest about the body image as a multidimensional issue that involves perceptual, attitudinal and behavioral characteristics [53, 54, 55, 56]. Many researchers have focused their attention mainly to the perceptual and attitudinal aspects of body image whereas only few studies have investigated the behavioral consequences related to a negative body image [57]. Patients with or at risk of ED often have a negative perception of several body parts; in some cases, they avoid social situations that may point out their physical appearance and exert a ritualistic checking on their body weight and shape [54, 58]. The body checking could be considered somewhat similar to compulsive behaviors; through this checking patients are often able to avoid the anxiety that derives from negative concerns about their body weight and shape [59, Errore. Il segnalibro non è definito.]. Examples of body checking are the repetitive measure of body weight, the frequent exposures at mirror in order to verify possible body shape changes, the use of particular clothes that can “measure” the fatness or the thinness, the pinching of several body parts to verify their consistence, the comparison with other people about the own body weight and shape, the checking to see if thighs rub together and many more [Errore. Il segnalibro non è definito.,Errore. Il segnalibro non è definito.]. Examining a sample of 260 candidates (44 men and 216 women) for gastric bypass surgery at a medical center, Grilo et al. [60] reported that checking and avoidance behaviors were significantly associated with overevaluation of weight and shape (most patients in their study reported they “always” or “usually” avoided clothing that made them particularly aware of their body shape). This result was also confirmed in patients with Binge Eating Disorder [61]. On the other hand, in ED patients, paradoxically, the body checking may reinforce the body dissatisfaction focusing further attention on concerns related to a negative body image [62]. Fairburn, Shafran, and Cooper [63] have pointed out the rule of the “body control” in the clinical evolution of anorexia. The body control and checking are used to monitor the body weight and shape changes, but increase the perceived imperfections and may lead to a higher body weight control. As consequence, a hypervigilant body control preserves the negative beliefs about presumed abnormal body shape. Moreover, they have hypothesized that the normal variations of body weight may be directly related to mood swings in ED patients. Since the body control and checking may play a role in the development and maintenance of an ED, the evaluation of the body checking behaviors may be useful in

8


therapeutic programs aimed to help patients with ED [64, 65]. In fact, as clinical observations indicate that body checking increase both the patient's preoccupation with body shape and weight and the motivation to maintain dietary restraint, a therapeutic program aimed to reducing body checking may contribute to reduce body dissatisfaction and, consequently, ED symptoms [66]. Furthermore, it is reasonable to think that individuals with anorexia nervosa and bulimia nervosa develop a highly organized cognitive schema concerning body- and weight-related information [67]. Body checking behaviors, like social scanning, may be both positively and negatively reinforced and strengthen the concerns about body size and shape as well as food and eating. Body checking rituals are used to regulate emotions through confirmation or attenuation of fears (and are therefore negatively reinforced), whereas in many circumstances the checking behaviors reinforce disordered patterns of behavior when the ritual results in an unfavorable or negative perception [68]. Analyzing clinical samples, Calugi et al. [69] have compared the body checking of different diagnostic groups of ED patients and have showed that bulimia nervosa patients check more than those with anorexia nervosa, while an Eating Disorder Not Otherwise Specified (EDNOS) group had results that were intermediate between these two more defined diagnostic groups. More recently, Mountford et al. [] have showed that patients with anorexia nervosa and binge eating disorder had lower levels of body checking cognitions and behaviours than patients with bulimia nervosa. Interestingly, the EDNOS patients had the highest level body checking cognitions and behaviours. It is widely accepted that alexithymic traits may negatively influence the perception of own body image with a presence of a body dissatisfaction, even in absence of a clinically defined ED. Our research group evaluated alexithymia and body image in 64 women with Premenstrual Dysphoric Disorder (PMDD) testing whether alexithymic traits may influence severity of PMDD or body distress [70]. We found that alexithymia was associated with more severe PMDD symptoms and alexithymics exhibited significantly poorer appearance evaluation and body satisfaction than nonalexithymics. On the basis of these data, we subsequently aimed to evaluate in a non-clinical sample of undergraduate women, the relationships between alexithymia, body checking and body image, identifying predictive factors associated with the possible risk of developing an ED [71]. To do this, TAS-20, Body Checking Questionnaire (BCQ), Eating Attitudes Test (EAT26), Body Shape Questionnaire (BSQ), Interaction Anxiousness Scale (IAS), Rosenberg SelfEsteem Scale (RSES) and the Beck Depression Inventory (BDI) were administered to 254 undergraduate females. We found that alexithymics had more consistent body checking behaviors and higher body dissatisfaction than nonalexithymics. In addition, alexithymics also reported a higher potential risk for ED (higher scores on EAT-26) when compared to nonalexithymics. Difficulty in identifying and describing feelings subscales of TAS-20, Overall appearance and Specific Body Parts subscales of BCQ as well as lower self-esteem was associated with higher ED risk in a linear regression analysis. Thus, a combination of alexithymia, low self-esteem, body checking behaviors and body dissatisfaction may be a risk factor for symptoms of ED at least in a non-clinical sample of university women. On the basis of our results, we hypothesized that the presence of alexithymia could play an indirect role in pathogenesis and maintenance of abnormal eating behaviors facilitating the presence of depressive symptoms and lower self-esteem. Depressive symptoms and lower self-esteem may directly exacerbate abnormal eating behaviors or have repercussions on body checking and body dissatisfaction with a worsening of abnormal eating behaviors themselves. On its


9

own, abnormal eating behaviors may worsen self-esteem and body dissatisfaction. In fact, it is reported that abnormal eating attitudes and lower self-esteem may be linked to higher bodyimage dissatisfaction [72, 73, 55]. In addition, difficulty in identifying and describing feelings may let misinterpret perceptual and behavioral aspects of body image. This may cause a higher body checking that, on its own, may lower self-esteem and increase negative body-image perception. As consequence, it can be triggered a vicious circle that may conduct to a higher risk of developing and maintaining a possible ED. In accordance with Overton et al. [74], we suggested that women with EDs are proficient at using disordered eating behaviors to manipulate their experience of both positive and negative emotional states. This dynamic may be more pervasive especially in the presence of alexithymia and should be recognized as an important maintenance factor. ED patients under- and/or over-regulate emotions due to an impaired ability to use blends of emotion to coping with emotional experience [Errore. Il segnalibro non è definito.]. This hypothesis is consistent with de Groot and Rodin [75] who suggested that individuals with EDs may either have little access to their emotional life or feel dominated and overwhelmed by it. In conclusion, alexithymic individuals may have a more prominent body checking that generates, on the basis of cognitive biases, a negative body image, thus increasing the risk of developing an ED [65,76]. Negative body image is an important component of a variety of prevalent health problems in females such as depression, obesity, and the spectrum of disordered eating [77]. This spectrum encompasses varying combinations and degrees of binge-eating and unhealthy forms of weight management such as restrictive dieting and self-induced vomiting. At the extreme end of the spectrum are the well-known syndromes of anorexia nervosa, bulimia nervosa, and binge-eating disorder. Stice [78] conducted a meta-analytic review of longitudinal studies examining the ability of body dissatisfaction (“negative body image”) and other hypothesized risk factors to predict increases in eating pathology. In general, data from these studies confirm that negative body image is an independent predictor of disordered eating. Overall, Stice’s (2002) review supports a model linking pressures to be thin from family, friends, and media to internalization of the slender beauty ideal and to overvaluation of appearance as a feature of self-concept. These components set the stage for negative body image, which in turn increases the probability of four interlocking components of eating pathology: dietary restraint, binge-eating, body checking and negative affect.

CLINICAL STUDY The aim of the clinical study was to evaluate relationships between alexithymia, body checking and body dissatisfaction in a sample of adult patients with Anorexia Nervosa (AN).

1. Methods

10


1.1. Participants The participants consisted of 64 women with a diagnosis of AN (mean age 23.9 ± 4.15; mean BMI 16.1 ± 1.29), each of whom was referred to one of two specialist eating disorder services. They were recruited at the point of assessment and were diagnosed (using DSM-IV criteria [79]) by trained clinicians, using a semi-structured interview. Body mass index (BMI) was recorded for each participant. BMI was calculated by dividing weight in kilograms by height in meters squared. A BMI below 18.5 indicates unusual thinness. A BMI between 18.6 and 24.9 is considered normal. A BMI over 25 indicates overweight or obese status. 1.2. Measures 1.2.1. Body Checking Questionnaire (BCQ) The BCQ is a reliable and valid measure of body checking behaviors [Errore. Il segnalibro non è definito.]. It is a 23-item self-report questionnaire with higher scores associated with more intense body dissatisfaction, body-image avoidant behaviors, and general eating disturbances. The BCQ measures a high-order factor (body checking) with three subfactors that are highly correlated: overall appearance (OA), specific body parts (SBP) and idiosyncratic checking (IC). The Italian version has been validated by Calugi et al. [Errore. Il segnalibro non è definito.] and confirmatory factor analysis confirmed the same three-factor structure of the English version. In our study, Cronbach's α was 0.92 for the BCQ total score and 0.88, 0.90, 0.79, respectively, for the OA, SBP and IC subfactors. 1.2.2. Toronto Alexithymia Scale (TAS-20) Alexithymia was measured using the 20-item TAS-20, the most widely used measure of alexithymia [3]. The TAS-20 has a three-factor structure [80]. Factor l assesses the capacity to identify feelings and to distinguish between feelings and the bodily sensations of emotional arousal (Difficulty in Identifying Feelings, DIF); Factor 2 reflects the inability to communicate feelings to other people (Difficulty in Describing Feelings, DDF); Factor 3 assesses Externally Oriented Thinking (EOT). Cut-off scores for the TAS-20 are provided by Bagby et al. [Errore. Il segnalibro non è definito.] and a score of 61 and above is considered to be within the alexithymic range. The Italian version of the TAS-20 was used [81]. In our study, Cronbach's α was 0.85. 1.2.3. Body Shape Questionnaire (BSQ) The BSQ is a 34-item self-rating scale that estimates the participants' disturbed perceptions of body size and body shape [82]. Higher scores reflect greater body-image concerns. The BSQ is a widely used instrument in studies of eating and weight disorders. The Italian version was used [83]. In our study, Cronbach's α was 0.95. 1.2.4. Beck Depression Inventory (BDI) Depressed mood was assessed with the Italian version of BDI [84], a 21-item self-report scale with higher scores indicating more depressive symptoms. In our study, Cronbach's α was 0.90.


11

1.2.5. State-trait Anxiety Inventory (STAI) Anxiety was assessed with the STAI scale [85] that consists of two 20-item Likert selfreport scales. The State Anxiety Scale consists of items asking individuals to indicate how they feel ‘at this moment’ (STAI-S) on a 4-point scale ranging from 1 ‘not at all’ to 4 ‘very much so’. The Trait Anxiety Scale (STAI-T) consists of items asking individuals to indicate how they ‘generally’ feel on a 4-point scale ranging from 1 ‘almost never’ to 4 ‘almost always’. In our study, Cronbach's α was 0.90 for the STAI-S and 0.91 for the STAI-T.

2. Statistical Analysis Descriptive statistics (means and standard deviations as appropriate) and percentages were computed for the study sample on demographic variables and all psychometric scales. The differences between alexithymics and nonalexithymics were tested by using analyses of covariance (ANCOVA) with TAS-20 positivity/negativity as factor and age, BMI, BDI and STAI scores as covariates. A blockwise linear regression analysis was performed in order to find which variables were associated with body dissatisfaction (BSQ score as dependent variable). In the first block BMI, demographic variables and measures of psychological distress (such as BDI and STAI) were entered. In the second block body checking measure (BCQ subscales) was added to the model. DIF, DDF and EOT subscales of TAS-20 were entered in the last block. The quality of the regression model was also tested using the Durbin–Watson statistic (a value between 0 and 4 indicating the amount of autocorrelation within the model with an optimum of 2.0). P values ≤.05 were considered to be statistically significant. All statistical testing was two-sided. Statistical analyses were performed using SPSS for Windows release 10.0.0 (2000). All data are expressed, if otherwise specified, as mean ± standard deviation.

3. Results TAS-20 score was 53.6 ± 11.5. 39.1% (n=25) of 64 subjects scored 61 or more on TAS20 total score and therefore were categorized as alexithymics. BDI score was 16.7 ± 9.4. No differences between groups were found concerning age, marital status and occupation. The results of ANCOVA controlling for age, BMI, BDI and STAI scores showed that alexithymics had more consistent body checking behaviors (higher scores on BCQ and subscales OA, SBP, IC – for all measures p