Psychoform and Somatoform Dissociation Among Individuals with Eating Disorders

Nordic Journal of Psychiatry

ISSN: 0803-9488 (Print) 1502-4725 (Online) Journal homepage: https://www.tandfonline.com/loi/ipsc20

Psychoform and somatoform dissociation among individuals with eating disorders

Doris Nilsson, Annika Lejonclou & Rolf Holmqvist

To cite this article: Doris Nilsson, Annika Lejonclou & Rolf Holmqvist (2019): Psychoform and somatoform dissociation among individuals with eating disorders, Nordic Journal of Psychiatry, DOI: 10.1080/08039488.2019.1664631 To link to this article: https://doi.org/10.1080/08039488.2019.1664631

Published online: 11 Sep 2019.

Submit your article to this journal

Article views: 190

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=ipsc20 NORDIC JOURNAL OF PSYCHIATRY https://doi.org/10.1080/08039488.2019.1664631

ARTICLE Psychoform and somatoform dissociation among individuals with eating disorders

Doris Nilssona, Annika Lejoncloub and Rolf Holmqvistc aSection for Clinical Psychology, Department for Behavioural Sciences and Learning, Linkoping€ University, Linkoping,€ Sweden; bLinkoping€ University Hospital , Region Osterg€ otland,€ Sweden; cDepartment of Psychology, Institution for Behavioral Sciences and Learning, Linkoping€ University, Linkoping,€ Sweden

ABSTRACT ARTICLE HISTORY Objective: This study analyzed the prevalence of psychoform and somatoform dissociation among Received 21 May 2019 individuals with the whole spectrum of eating disorder diagnoses and compared it with ratings from a Revised 26 August 2019 non-clinical group. The relationship between dissociation and severity of eating disturbance was exam- Accepted 2 September 2019 ined as well as differences between the eating disorder diagnosis groups in extent of dissociation. The KEYWORDS validity of a new structural dissociation interview suitable for eating disorder patients was analyzed. Dissociation; psychoform; Method: Sixty individuals with eating disorder completed three self-report questionnaires: Dissociation somatoform; eating Questionnaire Sweden, Somatoform Dissociation Questionnaire and Eating Disorder Examination disorder; young adults Questionnaire. The ratings were compared with the scores in a female non-clinical group (N ¼ 245). Twenty patients with eating disorder diagnoses were interviewed with the Interview for Dissociative Disorders and Trauma Related Symptoms. The validity of the interview was tested by comparing the ratings on the interview subscales with the scores on the Dissociation questionnaires and the Somatoform Dissociation Questionnaire. Results: Participants with eating disorders reported a higher extent of both psychoform and somatoform dissociation compared with the non-clinical individuals. Analyses also showed a correlation between degree of dissociation and severity of eating disorder symptoms. No differences in dissociation were found between the ED subgroups. Participants reporting more dissociation got higher ratings on the interview, indicating convergent validity. Discussion: Eating disorders seem to be associated with presence and severity of dissociative symptoms. The extent of dissociation needs to be assessed for these individuals as treatment may benefit from a focus on such symptoms in order to increase its effect.

Introduction anesthesia [5,6]. Several of these psychosomatic symptoms as well as conversion symptoms are found under somatic distur- Dissociation is defined as a disruption in the normal integra- bances in DSM-5 (1). Psychoform and somatoform dissociation tive functions of consciousness, memory, emotion, identity, are phenomena that affect the whole dynamic bio psycho- perception, body representation, motor control and behavior. logical system [3,5,7,8]. It includes experiences such as amnesia, identity disturbances, Several origins of dissociation have been suggested, such depersonalization and derealization [1]. Dissociation may be as exposure to potentially traumatizing events, problems in manifested in both mind and body, as psychoform and soma- the attachment history and neurobiological factors [7,9–11]. toform dissociation. The concept of psychoform dissociation Dissociation has been seen as a defense mechanism [2,4] suggests that elements of the mind such as thoughts, sensa- and as a natural reaction of protection against overwhelming tions and emotions are kept apart [2,3]. Psychoform dissoci- traumatic experiences [12]. Dissociation may occur on a con- ation includes symptoms related to the mental functions such tinuum that ranges from” normal” dissociation, like day- as memory, consciousness and identity [2,4]. Somatoform dis- dreaming, to the most severe form of dissociation, sociation refers to dissociative symptoms that are manifested Dissociative Identity Disorder [3,4]. Dissociative disturbances physically, and which are not possible to explain by medical may be associated with psychiatric disorders including post- causes and where the symptoms are thought to have devel- traumatic stress disorder, borderline personality disorder and oped due to a failure to integrate a somatic aspect of an eating disorders [8,9,13,14]. experience [3,5]. The somatoform dissociation includes a range During the two last decades there has been a growing of physical problems, like motor disturbances, analgesia and interest in the possible relationship between dissociation and

CONTACT Doris Nilsson [email protected] Section for Clinical Psychology, Department for Behavioural Sciences and Learning, Linkoping€ University,Linkoping€ S-581 85, Sweden This article has been republished with minor changes. These changes do not impact the academic content of the article. ß 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way. 2 D. NILSSON ET AL. eating disturbances. In general, persons suffering from an patients with bulimic behavior [8,17]. Problems in the early eating disorder seem to have difficulties to distinguish attachment relationship and experiences of adverse child- between outer and inner experiences and between physical hood circumstances have also have been found to increase and emotional reactions [15,16]. Several studies have found the risk of psychopathology, including dissociation and eat- that dissociative symptoms are more common in individuals ing disorder, later in life [3,9,11]. However, some researchers with an eating disorder [17–19]. Some researchers argue that argue that there is probably not a causal link between eating high levels of dissociation seem to be linked to more severe disturbance and dissociation. Instead both problems may be eating symptoms [16,20]. The majority of empirical studies of related to a general vulnerability [26,27]. In sum, the relations associations between eating disorders and dissociation have between dissociation and variants of eating disorders may focused on bulimic behavior, where the association is well be multi-facetted and there is an obvious need to further established [17,18,21]. There are, however, few studies that analyze these relations. analyze extent of dissociation among patients with a restrict- Many studies have found associations between dissoci- ive anorexic diagnosis [20]. As an exception, Farrington et al. ation and extent of bulimic eating disorders. The specific aim [20] found that individuals with restrictive anorexia showed of this study was to analyze the prevalence of psychoform as relatively low levels of both psychoform and somatoform well as somatoform dissociation in all the different sub- dissociation. groups of eating disorder patients. A second aim was to ana- Persons with bulimia and anorexia of the binge/purge lyze differences between the eating disorder diagnosis subtype, in contrast, have been found to score high on both subgroups, also those with restrictive anorexia, concerning forms of dissociation and particularly on somatoform dissoci- psychoform and somatoform dissociation. A third aim was to ation [16,19]. Several ways of interpreting the associations study the relationship between extent of dissociation and between bulimia and dissociative symptoms have been sug- severity of eating disturbance since this relationship has not gested [17,22]. Bulimic behavior may be seen as a way to been clearly described in previous research. produce a dissociative state creating a possibility to escape In addition to these aims, a preliminary assessment was from unwanted experiences. Dissociation seems to fluctuate made of the usefulness and the validity of the Interview for during the binge eating and purging cycle and may be inter- dissociative disorders and trauma-related symptoms preted as episodically linked to the eating disorder rather (IDDTS) [8]. than to the general psychopathology [17]. Some studies sug- gest that negative affects and dissociative experiences – for an example caused by an interpersonal situation – may act as triggers to the binge episodes [22]. The eating behavior Materials and methods provides the person with a concrete and bodily experience of reality. The dissociation and the binge eating would thus Participants have a function for affect regulation [21,22]. These persons The clinical group consisted of 60 females aged 14 to seem to have difficulties to distinguish between outer and 30 years (M age ¼ 20.0, median ¼ 20.0, SD ¼ 3.26) who met inner experiences and between physical and emotional reac- ICD-10 [28] criteria for eating disorder diagnoses. The partici- tions. The purging process may be a way of getting back to pants were recruited from a Child and Adolescent Psychiatric reality from a dissociative state [15,16]. outpatient unit specializing in eating disorders and were Somatoform dissociation has been found to be an import- diagnosed by experienced clinicians, physicians and thera- ant problem in eating disorder patients [5,16,19,23]. The pists. Several professionals participated in the specialized degree of somatic dissociation seems to be related to the degree of dissatisfaction about the individual’s own body. It evaluation team. The team diagnosed eating disorder syn- may be based on a sensitivity about the body image and dromes but no other psychiatric diagnoses. When diagnos- seems especially connected with bulimic behavior [24,25]. ing, a number of evaluation instruments were used, e.g. the – The relationship has been found so often that it has been Structured Eating Disorder Interview SEDI, Eating Disorders suggested that this form of dissociation may distinguish indi- Examination Questionnaire (EDEQ), Clinical Impairment viduals using purging and other compensatory behaviors Assessment (CIA), Psychiatric Status Rating scale for anorexia/ from those who do not. Somatic dissociation has been inter- bulimia (PSR). preted as a defense strategy against emotional overload- The group included patients with anorexia nervosa ing [19]. (n ¼ 20), atypical anorexia nervosa (n ¼ 12), bulimia nervosa Somatic dissociation may also be seen as linked to a his- (n ¼ 5), atypical bulimia nervosa (n ¼ 4), overeating associated tory of trauma and studies have found associations between with other psychological disturbances (n ¼ 1), other eating dis- bodily expressions of dissociation and posttraumatic stress orders (n ¼ 1) and eating disorders, unspecified (n ¼ 17), disorder in persons with eating disorders [5,13]. High extent In order to simplify analyses, three diagnosis groups were of dissociative symptoms as well as eating problems have formed as follows: been found among persons who have been traumatized from sexual abuse during childhood and/or in adulthood Eating disorder, anorectic (including anorexia nervosa and [17,26]. The dissociation seems to be correlated with the atypical anorexia nervosa), n ¼ 32, aged 15 to 30 years (M severity of the abuse [26] and seems to be more common in age ¼ 19.8, median ¼ 19.0, SD ¼ 3.4) NORDIC JOURNAL OF PSYCHIATRY 3

Eating disorder, bulimic (including bulimia nervosa and vomiting, laxative use, use of diuretics and hard exercising. atypical bulimia nervosa) n ¼ 9 aged 17 to 29 years (M age The EDE-Q consists of four subscales: restraint, eating con- ¼ 21.4, median ¼ 20.0, SD ¼ 3.6 cern, shape concern and weight concern as well as a global Eating disorder, Unspecified (including overeating associ- score which is the average of the four subscales. The instru- ated with other psychological disturbances and other eating ment has good psychometric properties [38] concerning disorders) n ¼ 19, aged 14 to 23 years (M age ¼ 19.6, internal consistency [39] and temporal stability [40]. median ¼ 20.0, SD ¼ 2.78) Convergent validity with the EDE interview supports the use of the instrument as a screening tool for eating disorder psy- In the non-clinical group data was gathered from 245 chopathology [41]. Normative data for EDE-Q have been young women aged 15 to 19 years (M age ¼ 16.7, median ¼ published in e.g. in England [42] in the USA [43] and in 17, SD ¼ 1.04). The sample was recruited from secondary Sweden [44]. In a recent Norwegian study cutoff was sug- schools and high schools representing different socioeco- gested at 2.5, based on data across different levels of BMI nomic areas in two cities in the south of Sweden. and age and different diagnostic groups [45]. This cutoff is used in the present study. Measures Interview for dissociative disorders and trauma- Eating disorder diagnoses were assessed according to the related symptoms Classification of Mental and Behavioral Disorders, [28] by expe- In order to more thoroughly assess dissociative symptoms rienced clinicians. among for instance eating disorder patients, the Interview The Dissociation Questionnaire (DIS-Q) [29] is a self-report for Dissociative Disorders and Trauma-related Symptoms questionnaire with 63 items designed to measure dissoci- (IDDTS) [8] has been created. The IDDTS is a semi-structured ation. The instrument gives a total scale score and scores on interview for examining and determining dissociative symp- four subscales intended to assess different aspects of dissociative experiences: 1) identity confusion/fragmentation, 2) toms and disorders. For persons with eating disorders the loss of control 3) amnesia and 4) absorption. The items have IDDTS interview has several advantages compared to the five answer alternatives from 1¼ “not at all” to 5¼ commonly used interview for assessing dissociation, SCID-D “extremely”. The subscale scores are obtained by summing [46]. The IDDTS contains several questions about eating con- up the total scores and dividing the number of items in each cerns and somatoform dissociative symptoms and questions subscale. Reliability and validity are satisfactory [29]. The around complex comorbidity and trauma-related symptoms Swedish translation of the scale (DIS-Q-Sweden) has been [8]. The IDDTS is based on the structural dissociation theory used in several studies and has been found to have good [3]. This is an integrative theory that takes into account psy- psychometric properties such as reliability, internal consist- chological as well as somatoform symptoms. According to ency and stability over time [12,30]. The cutoff between clin- this theory, dissociation occurs through a structural separ- ical and non-clinical groups has been set to 2.5 [12,29]. ation of the self as different parts within the same insuffi- The Somatoform Dissociation Questionnaire (SDQ- 20) [31] ciently integrated personality. is a 20-item self-administered instrument measuring somato- The interview, which consists of several sections, starts form manifestations of dissociation. The answer alternatives with evaluating physical complaints and somatoform dis- concerning different physical complaints range from 1¼ “not sociative symptoms. The main part of the interview contains true at all for me” to 5¼ “extremely true”. Each question is questions about psychiatric symptoms that may be con- followed by a supplementary question asking whether the nected with traumatization such as substance abuse, eating physical symptom is possible to explain by medical causes or problems, self-destructive behavior, sleeping and mood prob- not. The total score is obtained by summarizing the answers lems, anxiety and panic symptoms. The interview ends with of the 20 items. The instrument has shown good reliability examining psychoform dissociative symptoms. The frequency and validity [31]. The Swedish translation of the scale [32] and severity of the symptoms are evaluated by the inter- has been used in two studies in Sweden and was found to viewer. Based on the content and manner of presentation of have satisfactory psychometric properties [33,34]. The cutoff the examples the clinician decides to what extent dissocia- between clinical and non-clinical groups has been set to tive or other trauma-related symptoms may be present. 1.5 [34,35] Instructions concerning criteria for the examples are included Eating Disorder Examination Questionnaire, version 6.0 in the text for the interviewer. The interview was formed (EDE-Q) was administered to assess the extent of eating dis- through many years of clinical and research-related experien- order. The EDE-Q v.6.0 is a self-report questionnaire [36] ces [8] but its validity has not yet been assessed. This is the derived from the Eating disorder examination [37] that uses first time IDDTS is used in Sweden. a 7-point Likert type scale assessing the frequencies of eating problems over the past 28 days from no days ¼ 0 to every- Procedure day ¼ 6. This 36-item instrument examines the attitudinal, emotional and behavioral symptoms of disordered eating For inclusion in the clinical group, ninety-eight potential par- and assesses the symptoms of purging and binge eating ticipants were asked to take part in the study. Thirty-eight over the same period. The symptoms of purging include persons decided not to participate before completing the 4 D. NILSSON ET AL. questionnaires. The attrition was due to the patients’ not Cronbach’s alpha for the whole DIS-Q scale was a ¼ 0.96. feeling well enough to participate or that they had left treat- For Identity confusion, it was a¼ 0.96, for Loss of control ment or in some cases that they did not submit their a ¼ 0.88, for Amnesia a ¼ 0.85, and for Absorption a ¼ 0.67. answers or declined to participate. The self-report question- For SDQ-20, Cronbach’s alpha was a ¼ 0.83. There were large naires were administered by trained and experienced thera- differences between the clinical and non-clinical groups on pists at the eating disorder outpatient clinic after information all subscales on DIS-Q-Sweden and on SDQ-20 as shown in about the study and a request for participation had been Table 1. In the clinical group, the mean score 2.56 on DIS-Q- given to the participants. Information was posted to the Sweden was just above the cutoff limit for clinical dissoci- parents and caretakers of individuals below 18 years. The ation 2.5 [12,29]. The mean for the non-clinical group was patients were assured of the confidentiality of their answers below the cutoff. and then gave informed consent and completed the ques- The difference was somewhat smaller on SDQ-20. The tionnaires in the waiting room while visiting the clinic. Some mean score for both the clinical and the non-clinical group participants needed help from their therapists to understand was under the cutoff for clinical significance which is 1.5. the questions and some asked to take the questionnaires In Table 2, the mean scores and standard deviations for home for completion. It was considered more important to the clinical group on EDE-Q are presented. receive responses than to keep to a uniform procedure. The The results in Table 2 show that the mean total score on participants had reached different stages in their treatment EDE-Q was above the cutoff (2.5). Cronbach’s alpha for EDE- process. Some were new at the clinic while others were at Q was a ¼ 0.91. the end of their treatment. In Table 3, correlations between DIS-Q-Sweden and EDE-Q Among these sixty eating disorder patients, ten patients are presented. with the highest sum-scale scores on Dis-Q-Sweden and ten There were moderate to strong correlations between sev- participants with the lowest sum-scale scores were inter- eral of the subscales of EDE-Q and DIS-Q-Sweden. The sub- viewed with the Interview for Dissociative Disorders and scale Eating concern had particularly strong associations with Trauma Related Symptoms (IDDTS) without the interviewer the DIS-Q-Sweden scales. The correlations with SDQ-20 were knowing the Dis-Q-Sweden scores. Information was given to significant but somewhat weaker. the participants and a request about participation. They were assured of the confidentiality of their answers. Differences between diagnosis groups The differences in psychoform and somatoform dissociation Statistical analyses between the eating disorder diagnosis groups were tested – Statistical analyses were made with parametric tests (t-tests, with the Kruskal Wallis test. The results are presented in analyses of variance and Pearson correlations) and with non- Table 4. parametric tests (Kruskal–Wallis and Spearman correlations) There were no differences between the diagnosis groups when group sizes were less than 30. Effect sizes, using on either psychoform or somatoform dissociation. Cohen’s d, were assessed when differences between means were significant. Interview for dissociative disorders and trauma- related symptoms Ethical considerations In Table 5, the mean scores on the subscales in IDDTS The study was approved by the Human Research Ethics are presented. Committee, Faculty of Health Sciences, Linkoping€ University (Dnr. 02-196). Informed consent was given by all participants. Table 2. Means scores and standard deviations for EDE-Q (n ¼ 60). Mean (std) Median EDE-Q total 3.66 (1.53) 3.98 Results Restraint 3.26 (1.87) 3.30 Eating concern 2.93 (1.43) 3.20 In Table 1 mean differences between the clinical groups and Shape concern 4.46 (1.61) 5.13 Weight concern 3.56 (1.81) 3.90 control group on Dis-Q and SDQ-20 are presented.

Table 1. Mean differences between the clinical group (n ¼ 60) and the control group (n ¼ 245) on DIS-Q –Sweden and SDQ-20. Control group Mean (std) Clinical group Mean (std) Mean difference p-value Cohen’s d DIS-Q Total 1.91 (0.52) 2.56 (0.69) 0.65 <.001 1.08 Identity confusion 1.66 (0.62) 2.73 (0.91) 1.07 <.001 1.39 Loss of control 2.13 (0.65) 2.74 (0.86) 0.61 <.001 .81 Amnesia 1.71 (0.55) 1.96 (0.67) 0.24 .004 .39 Absorption 2.13 (0.69) 2.81 (0.80) 0.69 <.001 .92 SDQ-20 1.21 (0.26) 1.50 (0.46) 0.29 <.001 .81 control group n ¼ 242. NORDIC JOURNAL OF PSYCHIATRY 5

Table 3. Correlations between DIS-Q-Sweden and EDE-Q (n ¼ 60). Restraint Eating concern Shape concern Weight concern EDE-Q total Identity confusion 0.59 0.66 0.58 0.50 0.63 Loss of control 0.37 0.61 0.43 0.35 0.47 Amnesia 0.48 0.54 0.50 0.42 0.53 Absorption 0.39 0.46 0.44 0.38 0.46 DIS-Q total 0.54 0.67 0.57 0.48 0.62 SDQ-20 0.37 0.36 0.39 0.38 0.41 p < .05 p < .01 p < .001.

Table 4. Tests of differences (Kruskal-Wallis) on DIS-Q-Sweden and SDQ-20 between the eating disorder diagnoses. Anorexia (n ¼ 32) Bulimia (n ¼ 9) Eating disorder Unspecified (n ¼ 19) Dissociation scale Mean (std) Mean (std) Mean (std) p DIS-Q total 2.48 (0.68) 2.57 (0.64) 2.69 (0.73) .279 Identity confusion 2.67 (0.92) 2.56 (0.80) 2.91 (0.94) .819 Loss of control 2.58 (0.73) 3.15 (1.20) 2.80 (0.85) .138 Amnesia 1.90 (0.67) 1.71 (0.55) 2.16 (0.68) .382 Absorption 2.75 (0.86) 2.85 (0.61) 2.90 (0.80) .202 SDQ-20 1.51 (0.38) 1.39 (0.31) 1.53 (0.60) .725

Table 5. Ratings on the subscales in IDDTS (n ¼ 20). Interview subscales Mean Median Standard deviation Somatoform symptoms 1.55 1.00 1.47 Amnesia 1.00 0.50 1.21 Depersonalization 2.15 2.00 1.14 Derealization 1.10 1.00 1.07 Identity problems 1.65 1.50 1.31 Indication of dissociated parts of the personality 0.45 0.00 0.89 Pseudo hallucinations 0.65 0.00 0.93 Schneider’s symptom 0.90 1.00 1.02 Dissociation during the interview 0.50 0.00 0.76

Table 6. Correlations (Spearman’s rho) between DIS-Q-Sweden and SDQ and subscales in IDDTS (n ¼ 20). DIS-Q SDQ Interview subscales Identity confusion Loss of control Amnesia Absorption Total DIS-Q Somatoform symptoms 0.82 0.66 0.84 0.63 0.75 0.86 Amnesia 0.78 0.55 0.82 0.58 0.66 0.77 Depersonalization 0.82 0.74 0.80 0.66 0.76 0.78 Derealization 0.57 0.44 0.65 0.55 0.58 0.74 Identity problems 0.75 0.66 0.69 0.59 0.72 0.72 Indications of dissociated parts 0.54 0.41 0.56 0.36 0.41 0.54 Pseudo hallucinations 0.64 0.55 0.71 0.51 0.55 0.70 Schneider’s symptom 0.63 0.56 0.82 0.38 0.51 0.69 Dissociation during the interview 0.39 0.17 0.41 0.16 0.21 0.33 p < .05 p < .01 p < .001.

The results in Table 5 show that depersonalization, iden- different (1.57 vs. 2.99, t(18)¼70.85, p<.001). In Table 7, the tity problems and somatoform problems were most common scores in the two groups and significance of the differences in the interview. are shown. In Table 6, correlations (Spearman’s rho) between ratings The results in Table 7 show that participants with high on DIS-Q-Sweden and IDDTS are presented. and low scores on DIS-Q-Sweden differed significantly in The IDDTS scales were generally strongly correlated with their answers on the interview on all dimensions except dis- DIS-Q-Sweden and particularly with SDQ-20. Strong correla- sociating during the interview. The strongest differences tions were found between identity confusion and amnesia were found for depersonalization, somatoform symptoms, and most of the IDDTS subscales. The correlations for SDQ- amnesia, and identity problems. 20 were strong for many of the IDDTS scales and particularly for somatoform symptoms. Discussion In a final analysis, differences on the IDDTS subscales between the group with highest scores on the DIS-Q- This study investigated the presence and extent of both psy- Sweden sum scale and the group with lowest scores on this choform and somatoform dissociation in individuals suffering scale were compared with Mann–Whitney’s U-test. The scores from eating disorder compared with a non-clinical group. on DIS-Q- Sweden in these two groups were significantly The results showed that the persons with eating disorder 6 D. NILSSON ET AL.

Table 7. Ratings on IDDTS in the groups with the highest (n ¼ 10) and the group with the lowest (n ¼ 10) ratings on DIS-Q-Sweden and the significance of the difference between these groups (Mann–Whitney). Interview subscales Low rating Mean (std) High ratings Mean (std) Mean Difference p-value Cohen’sd Somatoform symptoms 0.60 (0.97) 2.50 (1.27) 1.90 .005 1.70 Amnesia 0.20 (0.42) 1.80 (1.23) 1.60 .004 1.95 Depersonalization 1.30 (.82) 3.00 (.67) 1.80 .001 2.40 Derealization 0.50 (0.71) 1.70 (1.06) 1.30 .015 1.48 Identity problems 0.80 (1.03) 2.50 (.97) 1.80 .003 1.80 Indication of dissociated parts 0 (0) .90 (1.10) 0.50 .063 0.91 Pseudo hallucinations 0.10 (0.32) 1.20 (1.03) 0.70 .019 1.04 Schneider’s symptom 0.30 (0.48) 1.50 (1.08) 0.90 .023 1.15 Dissociation during the interview 0.30 (0.67) .70 (.82) 0.40 .247 0.54

reported substantially and significantly more psychoform and disgust towards their own body [15,21–25]. It is known that somatoform dissociation symptoms than the non-clinical dissociation sometimes is a way to deal with bodily feelings group. The mean score for the clinical group was over the related to previous trauma [5,13,22,23,25,49]. As patients cutoff limit for psychoform dissociation. The results confirm with eating disorders often have been exposed to traumatic previous studies that have found associations between dis- experiences, the high rate of dissociation might be caused sociation symptoms like absorption, identity confusion and by such experiences. loss of control and eating disorder symptoms [14,15,19,21]. The perceived severity of eating disorder was associated The associations between binge eating and vomiting on with the degree of dissociation. The correlation between dif- one hand and dissociation on the other may vary. For some ficulties of eating problems and somatoform dissociation was individuals binge eating and vomiting can be perceived as significant but slightly weaker than for psychoform an opportunity to escape from negative feelings [20,22]. dissociation. Such behavior may also be experienced as a loss of control We found no differences between the eating disorder that may evoke additional negative affects like shame subgroups concerning dissociation. The results could be due whichinturnleadstodissociation.Forotherindividuals, to inadequate power to detect differences between the eat- dissociation may precede the bulimic symptoms and ing disorder diagnoses. Recent studies, however, have found vomiting may be a means to come out of the dissoci- similar results, implying that the eating disorder sub-diagno- ation [14,17]. ses may not differ with regard to predisposing factors or Persons with eating disorders seem to be unsure in their comorbidity [13,25,50,51]. own identity and often tend to withdraw in interpersonal To get a deeper understanding of the symptoms of dis- relations [11,15,18,20,21]. In a longitudinal Finish study, asso- sociation a newly developed interview, the Interview for ciations between social phobia, depression, eating disorder, Dissociative Disorders and Trauma Related Symptoms (IDDTS) and a reluctance to seek help for the eating disorder were was used. This is the first time this interview instrument is found [47]. used in research. The purpose of IDDTS is to identify both The absorption may be a way to deal with difficult affects somatoform and psychoform dissociation. We chose to use and to withdraw into oneself. Problems may be experienced the sub-scales relevant for persons with eating dis- through bodily symptoms using the body as a representation order problems. of mental states [21]. Abbate-Daga et al. [48] found a high The ratings on the interview correlated significantly with percentage of psychosomatic symptoms and conversion the ratings on DIS-Q-Sweden as well as with the ratings on symptoms in different eating disorder groups. SDQ-20, suggesting adequate convergent validity with these The ratings on SDQ-20 in our study showed that the eat- established measures. ing disorder group scored significantly higher on somato- The clinical value of the interview is that it may give a form dissociation compared to the non-clinical group which more clear understanding of the patient’s personal experien- is consistent with previous research [13,16,19,22,23,25]. ces of the dissociations symptoms. Several individuals The level of somatoform dissociation was not, however, as described a relief from this opportunity to talk about dis- high as in other studies and was below the clinical cutoff. sociative experiences. In the interviews a number of serious One reason could be the low proportion of patients with dissociative symptoms were described such as somatic prob- bulimia. Persons with this diagnosis have been found to lems and trauma-related symptoms. During the interviews score high on somatoform dissociation in previous studies some participants described experiences indicative of deper- [16,19,23–25]. This study also included restrictive diagnoses sonalization as linked to strongly affective and stressful eat- in the clinical group which may have influenced the results. ing situations or associated to interpersonal relationships. Another reason could be that some of the participants were These symptoms often seem to emerge in a bodily way. at the end of their treatment. Successful treatment may have Sometimes the depersonalization was described as a positive decreased the dissatisfaction with the individual’s body and way of escaping and sometimes as an unpleasant experience the tendency to dissociate. outside the individual’s control. Difficulties with fragmenta- Individuals with eating problems often describe a dis- tion and identity confusion were also described as very turbed sense of body self containing feelings of shame and stressful experiences. NORDIC JOURNAL OF PSYCHIATRY 7

Research has shown that dissociative symptoms often are Notes on contributors not noticed or may be mistaken as neuropsychiatric symp- Doris Nilsson is Assistant professor in Clinical psychology at Linkopings€ toms or other somatic or psychiatric symptoms [4,8,11] .This University and has done quite a lot of research concerning Children study shows that it is important to inform clinicians about and adolescents with dissociation and has been working clinically with the dissociative type of psychopathology and develop instru- children and adolecensts for many years. ments for assessment to reduce such inaccuracies. Annika Lejonclou is Licensed Psychotherapist and have been working with adolescents with eating disorder for many years.

Limitations Rolf Holmqvist is Professor of Clinical psychology at Linkoping€ University and has been doing a lot of research in Clinical psychology A problem with studies based on self-report questionnaires is and work in clinical practice. that answers may be inaccurate because they are subject to memory failure and to the respondent’s current mood. The ORCID problem of common-method variance, implying that persons tend to give similar answers to all questions about mental Rolf Holmqvist http://orcid.org/0000-0003-2093-2510 health and well-being, must also be considered. Some of the References participants, when interviewed, indicated that they had more problems than they had rated on the questionnaires. [1] American Psychiatric Association. Diagnostic and statistical man- Eating disorders are often co-morbid with mood disorders. ual of mental disorders, 5th edition. Washington, DC: APA; 2013. [2] Dell PF, O’Neil JA. editors. Dissociation and Dissociative Disorders, It is possible that the dissociative symptoms were associated DSM-V and beyond. New York: Routledge; 2009. with mood problems. The present study could have benefit- [3] Van der Hart O, Nijenhuis ERS, Steele K. The haunted self. ted from a comparison with a clinical group with Structural dissociation and the treatment of chronic traumatiza- mood disorders. tion. New York: W.W. Norton & Company; 2006. Another major limitation is that the clinical group com- [4] Diseth TH. Dissociation in children and adolescents as reaction to trauma-an overview of conceptual issues and neurobiological fac- prised only 60 patients, and thus the diagnostic subgroups tors. Nord J Psychiatry. 2005;59(2):79–91. were small. The sample in the clinical group may therefore [5] Nijenhuis E. Somatoform dissociation and somatoform dissocia- not have been representative of an outpatient eating dis- tive disorders In: Dell PF, ONeil JA, editors. Dissociation and the order group, considering especially the low number of per- dissociative disorders: DSM-V and beyond. New York: Routledge, sons with bulimia. The individuals in the clinical group were 2009. p. 259–285. [6] Brown RJ. Dissociation and somatoform disorders In: Kennedy F, on the average some years older than the non-clinical group, Kennerley H, Pearson D, editors. Cognitive behavioural a fact that may be considered to make the comparisons less approaches to the understanding and treatment of dissociation. appropriate. So for future research it is a recommendation to New York: Routledge/Taylor & Francis Group, 2013. p. 133–147. make a case control study with matched groups concerning [7] Schore AN. Attachment trauma and the developing right brain: age and gender. origins of pathological dissociation In: Dell PF, ONeil JA, editors. Dissociation and the dissociative disorders: DSM-V and beyond. New York: Routledge, 2009. p. 107–141. Conclusions [8] Boon S, Draijer N, Matthess H. Interview for Dissociative Disorders and Trauma-related Symptoms/IDDTS; 2007. Available from: Our findings indicate that both psychoform and somatoform https://www.fbanken.se/form/156/interview-for-dissociative-disor- dissociation need to be measured in the assessment of indi- ders-and-trauma-related-symptoms [accessed 6 Sept 2019] viduals with eating disorders. Both screening instruments [9] Lyons-Ruth K. Contributions of the mother-infant relationship to dissociative, borderline, and conduct symptoms in young adult- and the structural interview may be used when there is a hood. Infant Ment Health J. 2008;29(3):203–218. need to better understand the symptoms. Since the dissocia- [10] Jepsen EKK, Langeland W, Heir T. Early traumatized inpatients tive symptoms seem to be associated with the severity of high in psychoform and somatoform dissociation: characteristics the eating disorder a focus on such problems may lead to and treatment response. J Trauma Dissoc. 2014;15(5):572–587. better treatment and may potentially reduce the eating dis- [11] Petrucelli J. My body is a cage’: interfacing interpersonal neuro- biology, attachment, affect regulation, self-regulation, and the turbance. An important task is to find interventions that in regulation of relatedness in treatment with patients with eating addition to focusing on the eating disturbance problems also disorders In: Petrucelli J, editor. Body- states: interpersonal and target dissociative problems. relational perspectives on the treatment of eating disorders. New York: Routledge/Taylor & Francis Group, 2015. p. 35–56. [12] Nilsson D, Svedin CG. Evaluation of the Swedish version of dis- Disclosure statement sociation questionnaire (Dis-Q), Dis-Q-Sweden, among adolescents. JTrauma Dissoc. 2006;7(3):65–89. There is no conflict of interest in the present study for any of the authors [13] Tagay S, Schlegl S, Senf W. Traumatic events, post traumatic stress symptomatology and somatoform symptoms in eating dis- Funding order patients. Eur Eat Disorders Rev. 2010;18(2):124–132. [14] La Mela C, Maglietta M, Lucarelli S, et al. Pretreatment outcome This study was supported by The Department of Research and indicators in an eating disorderoutpatient group: the effects of Development, Region Osterg€ otland€ and by Gyllenstiernska self-esteem, personality disorders and dissociation. Compr Krapperup stiftelsen. Psychiatry. 2013;54(7):933–942. 8 D. NILSSON ET AL.

[15] Skårderud F. Eating ones words, part 2: The embodied mind and properties of the Swedish versions of the SDQ-20 and SDQ-5. reflective function in anorexia nervosa – theory. Eur Eat Disorders Nord J Psychiatry. 2015;69(2):152–160. Rev. 2007;15(4):243–252. [35] Simeon D, Smith RJ, Knutelska M, et al. Somatoform dissociation [16] Tibon S, Rothschild L. Dissociative states in eating disorders: an in depersonalization disorder. J Trauma Dissoc. 2008;9(3):335–348. empirical Rorschach study. Psychoanal Psychol. 2009;26(1):69–82. [36] Fairburn CG, Beglin SJ. Eating disorder examination questionnaire [17] McShane J, Zirkel S. Dissociation in the binge–purge cycle of buli- In: Fairburn CG, editor. Cognitive behavior therapy and eating mia nervosa. J Trauma Dissociation. 2008;9(4):463–479. disorders. New York: Guilford Press, 2008. p. 309–313. [18] La Mela C, Maglietta M, Castellini G, et al. Dissociation in eating [37] Fairburn CG, Cooper Z. The eating disorders examination. In: disorders: relationship between dissociative experiences and Fairburn CG, Wilson GT, editors. Binge eating: nature, assessment, binge-eating episodes. Compr Psychiatry. 2010;5:393–400. and treatment (12th ed). New York: The Guilford Press, 1993. p. [19] Waller G, Babbs M, Wright F, et al. Somatoform dissociation in 317–331. eating-disordered patients. Behav Res Ther. 2003;41(5):619–627. [38] Berg KC, Peterson CB, Frazier P, et al. Psychometric evaluation of [20] Farrington A, Waller G, Neiderman M, et al. Dissociation in ado- the eating disorder examination and eating disorder examin- lescent girls with anorexia: relationship to comorbid psychopath- ation-questionnaire: a systematic review of the literature. Int J Eat – ology. J Nerv Mental Dis. 2002;190(11):746–751. Disord. 2012;45(3):428 438. [21] Hallings-Pott C, Waller G, Watson D, et al. State dissociation in [39] Peterson CB, Crosby RD, Wonderlich SA, et al. Psychometric prop- bulimic eating disorders: an experimental study. Int J Eat Disord. erties of the Eating Disorder Examination-Questionnaire: factor 2005;38(1):37–41. structure and internal consistency. Int J Eat Disord. 2007;40(4): – [22] Engelberg M, Steiger H, Gauvin L, et al. Binge antecedents in 386 389. bulimic syndromes: an examination of dissociation and negative [40] Mond JM, Hay PJ, Rodgers B, et al. Temporal stability of the affect. Int J Eat Disord. 2007;40(6):531–536. Eating Disorder Examination Questionnaire. Int J Eat Disord. 2004; – [23] Mountford VA. The eating disorders and dissociation In: Kennedy 36(2):195 203. F, Kennerley H, Pearson D, editors. Cognitive behavioural [41] Mond JM, Hay PJ, Rodgers B, et al. Validity of the eating disorder examination questionnaire (EDE-Q) in screening for eating disor- approaches to the understanding and treatment of dissociation. ders in community samples. Behav Res Ther. 2004;42(5):551–567. New York: Routledge/Taylor & Francis Group, 2013. p. 104–118. [42] Carter JC, Aime AA, Mills JS. Assessment of Bulimia nervosa: a [24] Beato L, Rodriguez Cano T, Belmonte A. Relationship of dissocia- comparison of interview and self-report questionnaire methods. tive experiences to body shape concerns in eating disorders. Eur Int J Eat Disord. 2001;30(2):187–192. Eat Disorders Rev. 2003;11(1):38–45. [43] Luce KH, Crowther JH. The reliability of the eating disorder exam- [25] Fuller-Tyszkiewicz M, Mussap A. The relationship between dissoci- ination – self-report questionnaire version (EDE-Q). Int J Eat ation and binge eating. J Trauma Dissoc. 2008;9(4):445–462. Disord. 1999;25(3):349–351. [26] Mercado R, Martinez-Taboas A, Pedrosa O. Childhood sexual [44] Welch E, Birgegard A, Parling T, et al. Eating Disorder abuse, eating disturbance and dissociation: a study in Puerto Examination Questionnaire and Clinical Impairment Assessment Rico. J of Psych Trauma. 2008;7(4):298–309. Questionnaire: general population and clinical norms for young [27] Gerke C, Mazzeo S, Kliewer W. The role of depression and dissoci- adult women in Sweden. Behav Res Ther. 2011;49(2):85–91. ation in the relationship between childhood trauma and bulimic [45] Rø Ø, Reas DL, Stedal K. Eating Disorder Examination symptoms among ethnically diverse female undergraduates. Questionnaire (EDE-Q) in Norwegian adults: discrimination – Child Abuse Neglect. 2006;30(10):1161 1172. between female controls and eating disorder patients. Eur Eat [28] World Health Organization. International classification of diseases, Disorders Rev. 2015;23(5):408–412. 10th edition (ICD-10). Geneva: WHO; 1992. [46] Steinberg M. Structured Clinical Interview for DSM-IV Dissociative [29] Vanderlinden J, Van Dyck R, Vandereycken W, et al. The dissoci- Disorders (SCID-D). Washington DC: American Psychiatric Press; ation questionnaire (DIS-Q): development and characteristics of a 1994. new self-report questionnaire. Clin Psychol Psychother. 1993;1(1): [47] Ranta K, V€a€an€anen J, Frojd€ S, et al. Social phobia, depression and 21–27. eating disorders during middle adolescence: longitudinal associa- [30] Svedin CG, Nilsson D, Lindell C. Traumatic experiences and dis- tions and treatment seeking. Nord J Psychiatry. 2017;71(8): sociative symptoms among Swedish adolescents. A pilot study 605–613. using Dis-Q-Sweden. Nord J Psychiatry. 2004;58(5):349–355. [48] Abbate-Daga G, Delsimdime N, Nicotra B, et al. Psychosomatic [31] Nijenhuis ERS, Spinhoven P, van Dyck R, et al. The development syndromes and anorexia nervosa. BMC Psychiatry. 2013;13(1):14. and psychometric properties of the somatoform dissociation [49] Moulton SJ, Newman E, Power K, et al. Childhood trauma and questionnaire (SDQ-20). J Nerv Mental Dis. 1996;184(11):688–694. eating psychopathology: a mediating role for dissociation and [32] Nilsson D, Nordfelt S, Wadsby M. SDQ-20 and SDQ-5. Translation emotion dysregulation? Child Abuse Neglect. 2015;39:167–174. to Swedish. BUP-Elefanten; 2006. [50] Lech B, Holmqvist R, Andersson G. Affect consciousness and eat- [33] H€aggstrom€ M. SDQ-20 Sweden: Validation of the Swedish version ing disorders. Short term stability and subgroup characteristics. of the Somatoform Dissociation Questionnaire. Master thesis, Eur Eat Disord Rev. 2011;18:1–6. Linkoping€ University; 2007. [51] Lejonclou A, Nilsson D, Holmqvist R. Variants of potentially trau- [34] Nilsson D, Lejonclou A, Svedin CG, et al. Somatoform dissociation matizing life events in eating disorder patients. Psychol Trauma. among Swedish adolescents and young adults: the psychometric 2014;6(6):661–667.