Mora‑Maltas et al. Journal of Eating Disorders (2023) 11:155 Journal of Eating Disorders https://doi.org/10.1186/s40337-023-00875-z RESEARCH Open Access Cognitive flexibility and DSM‑5 severity criteria for eating disorders: assessing drive for thinness and duration of illness as alternative severity variables Bernat Mora‑Maltas1,2†, Ignacio Lucas1,2,3†, Roser Granero2,3,4, Cristina Vintró‑Alcaraz1,2,3, Romina Miranda‑Olivos1,2,3, Isabel Baenas1,2,3, Isabel Sánchez1,2,3, Jessica Jiménez‑del Toro1,2, Jéssica Sánchez‑González1,2, Isabel Krug5, Javier Tapia1,6, Susana Jiménez‑Murcia1,2,3,7 and Fernando Fernández‑Aranda1,2,3,7* Abstract Background The severity criteria for eating disorders (EDs) proposed in the DSM-5 have been established with‑ out sufficient empirical support. Drive for thinness (DT) and duration of illness have been proposed as two alter‑ native severity measures, however their empirical evidence is also limited. To date, no research has assessed the validity of current eating disorder (ED) severity criteria regarding cognitive flexibility factors. Cognitive flexibility is often impaired in EDs, becoming a possible severity symptom. The current study assessed for the first time (1) whether the severity indexes for EDs proposed in the DSM-5 were associated with deficits in cognitive flexibility and, (2) whether drive for thinness and illness duration, acted as an alternative, more meaningful severity indices for defi‑ ciencies in cognitive flexibility. Methods Participants were 161 patients diagnosed with an ED, who were categorized according to DSM-5 severity categories, DT and duration of illness. Discriminative capacity of each classification was assessed for cognitive flexibil‑ ity measured by Wisconsin card sorting test (WCST). Results The findings for the DSM-5 classification comprised: (a) In the anorexia nervosa (AN) group, patients with moderate severity showed better scores in WCST than patients with mild and severe/extreme severity. Also, patients with moderate severity showed lower percentage of cognitive flexibility deficits than the other two sever‑ ity categories; (b) For the binge spectrum disorders (BSD) group, the patients with mild severity showed a higher percentage of cognitive flexibility deficits than did the moderate and severe/extreme categories. When assessing the alternative severity index of DT, no differences were found in cognitive flexibility in any of the groups. Regard‑ ing illness duration, in the AN group the task performance of the patients with longer illness duration was worse than the performance of the short duration group and, in the BSD group, patients with longer duration also showed more deficits in cognitive flexibility than the patients with shorter duration of illness. † Bernat Mora-Maltas and Ignacio Lucas shared first authorship. *Correspondence: Fernando Fernández‑Aranda ffernandez@bellvitgehospital.cat Full list of author information is available at the end of the article © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecom‑ mons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Mora‑Maltas et al. Journal of Eating Disorders (2023) 11:155 Page 2 of 12 Conclusions Our findings point out the limitations of the DSM-5 severity criteria to categorize cognitive flexibility in EDs and support illness duration as an alternative severity approach for EDs. Plain English summary The current severity criteria for eating disorders were defined without enough evidence and present several limita‑ tions. Therefore, some researchers proposed alternative criteria like the drive for thinness or the duration of the disor‑ der. Eating disorders are characterized by cognitive impairments such as deficits in cognitive flexibility. Cognitive flex‑ ibility is the ability to adapt our behaviour to the needs of the environment. According to our results, current severity criteria do not classify correctly in terms of cognitive flexibility. Moreover, this study shows that patients with longer illness duration show less cognitive flexibility. Therefore, this study highlights the limitations of current severity criteria for eating disorder to classify according to cognitive flexibility. Our findings also show the importance of taking into account the illness duration in order to assess the severity of the disorder. Keywords Eating disorders, Neuropsychology, DSM-5-TR, Illness duration, Severity ratings Background For BSD, incongruous findings regarding the utility of The fifth edition of the Diagnostic and Statistical Man- the DSM-5 severity index have been obtained. Studies ual of Mental Disorders (DSM-5) [1] and its text revised that support the proposed DSM-5 criteria found that version (DSM-5-TR) [2] propose four different sever- patients with severe and extreme BN had more psychi- ity gradients for eating disorders (EDs) (‘mild’, ‘moder- atric comorbidities, functional impairments, perfec- ate’ ‘severe’ and ‘extreme’). The severity of the disorder tionism, and ED body-related attitudes and behaviours is determined by the body mass index (BMI = kg/m2) for than those in the mild and moderate categories [17, 25– anorexia nervosa (AN). Regarding binge spectrum dis- 28]. Conversely, a study by Zayas et al. [18] supported orders (BSD), a classification previously used in the lit- the utility of the BN severity index with respect to ED erature [3] that includes bulimia nervosa (BN) and binge psychopathology in females, but not in males, and other eating disorder (BED), the severity is determined by the studies found rarely or no differences for these variants weekly frequency of episodes of inappropriate compen- across the BN severity groups [5, 29, 30]. Furthermore, satory behaviours for BN, and the number of weekly in some other studies few patients diagnosed with binge episodes for BED. However, a recent systematic BN fell into the severe or extreme categories, raising review and meta-analysis presented a compound of sci- doubts about the cut-off points of the BN classification entific evidence that acknowledged the limitations of the [17, 31, 32]. The literature has also been contradictory current severity classification for EDs [4]. According to with regard to BED. Some studies reported differences the literature, the criteria and cut-offs used for defining between severity groups in relation to BMI, ED charac- the severity of EDs are controversial and lack sufficient teristics, comorbidity (personality disorders, biases and empirical support [5–10]. emotional difficulties), and other factors of distress and With respect to AN, inconsistent results have been impairment [5, 11, 27, 33, 34]. However, various studies found, maybe because using only BMI does not take found no differences between severity categories with into account important factors such as physical (e.g. respect to psychiatric comorbidity, prognostic predic- weight history), psychological or cognitive factors that tion or body attitude [6, 8, 11, 17, 30, 34, 35]. In addi- may reflect different kinds of impairment [11–13]. In tion, as seen in BN, few individuals with BED fall into this line, several studies assessing individuals with AN the severe or extreme BED categories [4–6, 8, 17]. did not find significant variation among the different Due to the uncertainty that exists regarding the func- DSM-5 severity groups regarding distress, psychiatric tionality of the DSM-5 severity index, researchers have comorbidity or further attitudinal eating disorder (ED) introduced transdiagnostic indices to find alternative symptoms (i.e., shape concerns/weight) [5, 6, 9, 14–18]. severity classifications for each ED subtype [4]. In this Moreover, Dakanalis et al. [19] found that patients with line, previous research has proposed other alternative less severe AN showed more bingeing and purging measures for ED severity to overcome the limitations behaviours than the ones with more severity, and the presented by the DSM-5 classification. For example, presence of binge-purge behaviours in AN have been Krug et al. [6] proposed an alternative transdiagnos- associated with more psychopathology [20–22], more tic indicator of ED severity, the drive for thinness (DT) relapses [23] and poorer treatment outcomes [24]. dimension of the eating disorders inventory-2 (EDI-2) Mora‑Maltas et al. Journal of Eating Disorders (2023) 11:155 Page 3 of 12 [36]. DT is defined as an extreme fear of weight gain, in cognitive flexibility are associated with the inability to which reinforces disordered eating patterns (especially stop compulsive overeating [50]. Moreover, poor perfor- restrictive eating) [37–39]. However, DT is just one fac- mance of BSD patients in the WCST have proven to be tor involved in the complex process of EDs [40]. a predictor of bad treatment outcome [62]. However, in Duration of illness has also been reported as an alter- BSD there are inconsistent results, with studies reporting native indicator of severity for EDs. Specifically, the EDs deficits in cognitive flexibility [50, 63], and others finding literature has repeatedly demonstrated that a shorter no significant differences [64, 65] compared to controls. duration of illness is related to a more favourable out- Until now, the literature has taken little account of cog- come of treatment for EDs [41–44]. For instance, a study nitive factors as an effective measure to assess the sever- by Fernández-Aranda et al. [45] has shown that dura- ity of EDs. In fact, to our knowledge, there are no studies tion of illness was linked with poor response to treat- that evaluate cognitive deficits of patients with an ED ment, suggesting that the duration of illness could be a based on the DSM-5 severity indices. good marker of severity. In this study, the duration cut- off points from which there would be a greater risk of The current study having poor results were: 12 years for patients diagnosed The current study assessed, for the first time: (1) Whether with AN, 13 years for patients with BN, and 21 years for the severity indices for EDs proposed in the DSM-5 were patients with BED. In addition, other empirical studies associated with deficits in cognitive flexibility and, (2) have reported an association between the duration of ill- Whether DT and illness duration, acted as an alternative, ness and deficits in cognitive flexibility in AN [46] and more meaningful severity indices for deficiencies in cog- BSD [47]. nitive flexibility. Cognitive flexibility is defined as the ability to adjust individual’s beliefs or behaviour in response to new situ- Methods ations, which is essential for behaviour self-regulation Participants [48]. A common neuropsychological task to measure The participants were 161 adults (130 females and cognitive flexibility is the Wisconsin Card Sorting Test 31 males) who met the DSM-5 criteria for an ED. All (WCST; [49]). Although the role of cognitive flexibility patients were diagnosed by experienced psychologists. in EDs is still being studied, there is increasing support Those who were diagnosed according to DSM-IV-TR cri- that this cognitive malfunction may be one of the fac- teria [66] were reanalysed and recoded post-hoc using tors that influences the development and maintenance DSM-5 criteria [2]. Most patients were treatment naïve of EDs. A meta-analysis carried out by Wu et al. [50] (n = 116, 72%). The number of patients with one previous reported cognitive flexibility impairments in patients treatment was n = 24 (14.9%), with two previous treat- with restrictive subtype of AN, BN and BED. In addition, ments was n = 10 (6.2%), and with three or more previous a recent study associated AN and BSD with impairments treatments n = 11 (6.8%). There was a positive correla- in different executive domains, including cognitive flex- tion between the number of previous treatments and ill- ibility [51]. In that study, poor cognitive performance ness duration (non-parametric correlation = 0.333) in the correlated with anxious, depressive, and ED symptoms. whole sample. The distribution of the ED diagnoses was: Other results also associated cognitive flexibility impair- 72 AN-Restrictive (AN-R), 28 AN-Binge-eating/Purging ments in EDs with comorbid symptomatology, such as type (AN-BP), 34 BN, and 27 BED. Patients with a diag- depression or anxiety [52]. In patients with AN, lack of nosis of AN-R and AN-BP were categorized in the AN cognitive flexibility has been considered a neurocogni- group, and patients diagnosed with BN and BED were tive endophenotype that may contribute to compulsive categorized in the BSD group. This classification was and rigid behaviour [53–55]. Specific treatments, such based on the common physiological and psychological as the cognitive remediation therapy, are aimed at the factors that differentiate ED subtypes [3]. The AN group cognitive impairments shown by these patients [56–58]. had a mean age of 27.28 years (SD = 8.99). The BSD group Complement usual treatment with cognitive remediation had a mean age of 35.62 years (SD = 10.64). The age range therapy have proven to produce a significant improve- of the total sample is between 17 and 58 years. Table 1 ment in eating disorder-specific health-related quality of displays a detailed sociodemographic description of each life and a greater reduction of eating disorder psychopa- group, and the result of the statistical comparison. Exclu- thology [59]. However, the efficacy of cognitive remedia- sion criteria were having an intellectual disability, the tion therapy remains unknown due to a lack of conclusive presence of an organic mental disorder or an active psy- data from other studies, which have found no evidence chotic disorder. that this treatment improves eating disorder symptoms Data were collected between November 2007 and Janu- [60, 61]. For BSD, a meta-analysis reported that deficits ary 2020 at the Eating Disorders Unit of the Bellvitge Mora‑Maltas et al. Journal of Eating Disorders (2023) 11:155 Page 4 of 12 Table 1 Sample description AN group (n = 100) BSD group (n = 61) p n % n % Sex Females 92 92.0 38 62.3 < 0.001* Males 8 8.0 23 37.7 Marital status Single 81 81.0 36 59.0 0.006* Married 15 15.0 16 26.2 Divorced 4 4.0 9 14.8 Education Primary 33 33.0 26 42.6 0.324 Secondary 38 38.0 23 37.7 University 29 29.0 12 19.7 Employment Unemployed 43 43.0 22 36.1 0.384 Employed/student 57 57.0 39 63.9 Social position High 12 12.0 4 6.6 0.028* Mean-high 21 21.0 20 32.8 Mean 24 24.0 23 37.7 Mean-low 34 34.0 9 14.8 Low 9 9.0 5 8.2 Mean SD Mean SD p Age (years old) 27.28 8.99 35.62 10.64 < 0.001* Onset ED (years old) 21.06 8.58 25.21 11.38 0.009* Duration ED (years) 6.22 6.30 10.41 8.12 < 0.001* EDI-2 drive for thinness 9.72 7.31 14.38 5.12 < 0.001* Body mass index (kg/m2) 16.21 1.50 30.94 8.93 < 0.001* Inappropriate compensatory behaviours 2.67 7.84 4.69 6.57 0.094 AN anorexia nervosa group (AN-R anorexia nervosa restrictive, AN-BP anorexia nervosa binge-purging), BSD binge spectrum disorders symptoms group (BN bulimia nervosa, BED binge eating disorder), ED eating disorder, SD standard deviation, EDI-2 eating disorders inventory-2; Inappropriate compensatory behaviours: number of vomits, laxatives and diuretics per week. *Bold: significant comparison University Hospital (Barcelona, Spain). All participants Psychological assessment received information about the procedure and signed an Participants were evaluated using the DT subscale of the informed consent form. All procedures were approved by Eating Disorders Inventory-2 (EDI-2) [36]. This ques- the Ethical Committee of the Bellvitge University Hospi- tionnaire evaluates cognitive and behavioural features tal in accordance with the Helsinki Declaration of 1975 as related to the ED. DT factor is defined as the extreme fear revised in 1983 (Refs. 34/05, 307/06). of weight gain and over-preoccupation about diet and weight. In this study, Cronbach’s alpha for the EDI-2 DT subscale was 0.884 (indicating good internal consistency). Sociodemographic and clinical information Sociodemographic data were collected from each par- Neuropsychological assessment ticipant. These data included age, education level, mari- The computerized version of the Wisconsin Card Sorting tal status and employment, as well as ED onset and Test (WCST) [49, 69] was used to evaluate cognitive flex- duration. Sex data is also reported, showing a higher ibility. The WCST includes 128 cards that comprise three proportion of females, but in accordance with preva- available categories: number (N), colour (C) and shape lence estimates of EDs [67]. Social position was calcu- (S). For a right pair, participants must identify the sorting lated using the Hollinghead method [68]. rule, receiving the feedback of “Right” or “Wrong” after each sort. By trial and error, the participant must learn Mora‑Maltas et al. Journal of Eating Disorders (2023) 11:155 Page 5 of 12 to change the sorting categories according to the given EDs [45]. This previous study calculated the duration feedback. Initially, C is the correct sorting category, and cut-off points from when there would be a higher risk positive feedback is given only if the card is placed in of having poor treatment outcomes in each subtype of the pile with the same colour. After 10 consecutive right ED. Hence, the cut-off points used in this study were pairs, the rule is changed, and then, another sorting rule 12 years for patients diagnosed with AN-R and AN-BP, must be identified. There are up to six attempts to detect 13 years for patients with BN, and 21 years for patients the sorting rule and five rule shifts during the task. Each with BED. rule attainment is referred to as “category completed”. The task ends when all 128 cards are sorted or after the six full categories are completed. The number of com- Procedure pleted categories is recorded, as well as the percentages Evaluations were conducted in two separate sessions of errors, perseverative responses, perseverative errors, prior to the psychological treatment. In the first one, we non-perseverative errors, and conceptual level responses. collected sociodemographic data and conducted the psy- The presence of deficits in cognitive flexibility was deter- chological assessment. And, in the second one, partici- mined by scores below the ­16th percentile in any of the pants completed a computerized version of the WCST WCST scales perseverative errors, non-perseverative [49, 69]. errors, and number of completed categories, based on normative data published in the manual of the task and in accordance with previous literature about cognitive func- Statistical analysis tions [70, 71]. The method to calculate presence of defi- The statistical analysis was performed with Stata17 (Stata cits in cognitive flexibility is not described in the manual, Press, 2021) for Windows. Chi-squared tests (χ2) were but have been previously used in eating disorders [47]. done for the comparison of categorical variables between the groups (e.g. cognitive flexibility deficits), and analy- Severity index categorization sis of variance (ANOVA) was done for the comparison of DSM‑5 quantitative measures (e.g. neuro-psychological scores). The DSM-5 severity classifications were carried out fol- The effect size of the proportion and mean comparisons lowing the criteria proposed in the manual [1]. For the was estimated through Cramer’s-V coefficient for cat- AN group, four severity categories were defined accord- egorical variables [the thresholds 0.06, 0.15 and 0.30 were ing to patients’ BMI: mild (> 17.0 kg/m2), moderate considered for low/poor effect size, moderate/medium, (16–16.99 kg/m2), severe (15–15.99 kg/m2) o extreme and high/large (Cohen, 1998)] and partial eta-squared (< 15 kg/m2). The severity of the ED in patients diagnosed coefficient (η2) for quantitative measures [the thresholds with BN was defined by the number of inappropriate for low/poor, moderate/medium and high/large were compensatory behaviours per week (vomits, laxatives and 0.06, 0.10, and 0.25 (Levine and Hullett, 2002)]. Increase diuretics): mild (1–3 episodes/week), moderate (4–7 epi- of Type-I error due to multiple comparisons was con- sodes/week), severe (8–13 episodes/week) and extreme trolled using the Finner Method (Finner, 1993), a family- (> 14 episodes/week). In patients with a BED diagnosis, wise procedure that has proved to be more powerful than the severity categories were defined by the same catego- the classical Bonferroni correction. rization as for BN but taking into account binge eating episodes per week instead of compensatory behaviours. Results Drive for thinness Severity distribution Krug et al. [6] used an alternative categorization for ED Figure 1 displays bar-charts for the severity levels accord- severity based on DT symptomatology using the EDI-2 ing to the three classification methods of the study DT subscale. We used the same cut-off point for clas- (DSM-5, DT and illness duration). Among patients sifying low DT (≤ 14) participants and high DT (> 14) in the AN group, the more prevalent categories were: participants, based on the recommendations by [72] for severe-extreme severity based on the DSM-5 severity screening purposes. This cut-off point has also been used levels (grouping 40.0% of patients), low score based on in other previous studies [73]. the EDI-2 DT scale (66.0%), and short duration (78.0%) based on the illness duration. Among patients of the BSD Illness duration group, the more prevalent categories were mild severity Illness duration cut-off points were based on a previ- for the DSM-5 classification system (36.1%), high score ous recent study that highlighted the importance of the based on the EDI-2 DT scale (59.0%), and short duration duration of the disorder in the treatment outcome of based on the illness duration (75.4%). Mora‑Maltas et al. Journal of Eating Disorders (2023) 11:155 Page 6 of 12 Fig. 1 Severity prevalence estimates according to the three classification methods of the study (DSM-5, drive for thinness and illness duration) Cognitive flexibility based on the DSM‑5 severity Discussion classification A first aim of the present study was to determine if the Table 2 shows the discriminative capacity of the classifi- DSM-5 severity criteria for EDs were able to assess the cation methods for cognitive flexibility based on DSM-5 presence of deficits in cognitive flexibility, considering severity criteria. Among patients in the AN group, the that it could be a core symptom of severity in EDs. Our moderate severity group classified by the DSM-5 criteria second aim was to evaluate whether other alternatives showed better performance in WCST (PctErrors, Pct- variables such as DT or illness duration could be associ- PersRsps, PctPersErrors, PctCLRsps) than the mild and ated with poorer cognitive flexibility. Using the DSM-5 severe/extreme groups. This classification also reported severity criteria for AN, we observed that the moderate that patients grouped in the mild and severe/extreme severity group performed better in the WCST than the categories showed a higher percentage of deficits in cog- mild and severe/extreme groups, which presented a simi- nitive flexibility than the moderate group. Among BSD lar performance. In the same line, we found that patients group, patients classified in the mild severity category classified in the mild and severe/extreme groups accord- showed higher percentage of deficits in cognitive flexibil- ing to DSM-5 criteria, presented a higher percentage of ity than the moderate and severe/extreme categories. deficits in cognitive flexibility than the moderate severity group. Considering the BSD group, our study did not find significant differences in cognitive flexibility performance Cognitive flexibility based on alternative severity between the DSM-5 severity groups. Additionally, the classifications less severe group showed more cognitive flexibility defi- Table 3 shows the discriminative capacity of alternative cits than the other two groups. These results suggest that severity classification methods for cognitive flexibility. DSM-5 severity criteria were not able to discriminate No differences in cognitive flexibility were obtained for between cognitive flexibility levels of patients diagnosed the DT classification in any group. However, regarding with a BSD. Regarding both clinical groups, the present duration of the disorder, patients of the AN group and findings showed that DT did not discriminate poor cog- a short duration of the disorder showed a better perfor- nitive flexibility. However, duration of illness did present mance in WCST (PctErrors, PctPersRsps, PctPersErrors, discriminative capacity to assess poor cognitive flexibil- PctNonPersError, PctCLRsps) than the patients with a ity, resulting in an alternative severity classification for longer duration. In the patients of the BSD group classi- EDs. In the AN group, the WCST performance of the fied by duration of illness, no differences were found in long duration group was worse than the performance of the task scores. Nevertheless, the duration of the illness the short duration group. Similarly, in the BSD group, the differentiated the presence of deficits in cognitive flex- long duration group included a higher percentage of peo- ibility in patients with BSD, with a higher percentage of ple with cognitive flexibility deficits. deficits in cognitive flexibility for the patients with longer These results illustrate that the DSM-5 severity rat- duration. ings for AN, that are exclusively based on BMI, do not Mora‑Maltas et al. Journal of Eating Disorders (2023) 11:155 Page 7 of 12 Table 2 Discriminative capacity on cognitive flexibility for the DSM-5 severity classification AN group (n = 100) WCST T-scores DSM-5 criteria Mild (G1) n = 33 Mod(G2) n = 27 S/E (G3) n = 40 Factor group Pairwise comparisons G1–G2 G1–G3 G2–G3 Mean SD Mean SD Mean SD p ES p p p WCST: PctErrors 45.36 11.37 52.44 10.31 46.30 11.24 0.033* 0.068 0.015* 0.719 0.028* WCST: PctPersRsps 46.91 11.91 55.11 10.71 46.33 10.59 0.004* 0.108† 0.005* 0.823 0.002* WCST: PctPersErrors 46.91 11.91 53.74 10.45 46.10 10.63 0.016* 0.082 0.019* 0.756 0.006* WCST: PctNonPersErrors 45.67 10.69 51.70 10.41 47.23 9.86 0.072 0.053 0.026* 0.521 0.084 WCST: PctCLRsps 44.79 11.75 52.37 10.15 46.05 11.62 0.026* 0.072 0.011* 0.636 0.027* Deficit n % n % n % p ES G1–G2 G1–G3 G2–G3 Cognitive flexibility 13 40.6% 2 7.4% 11 27.5% 0.015* 0.291† 0.004* 0.240 0.041* BSD group (n = 61) WCST T-scores DSM-5 criteria Mild (G1) n = 22 Mod(G2) n = 21 S/E (G3) n = 18 Factor group Pairwise comparisons G1–G2 G1–G3 G2–G3 Mean SD Mean SD Mean SD p ES p p p WCST: PctErrors 41.68 13.12 46.05 9.62 47.94 11.22 0.209 0.052 0.216 0.091 0.608 WCST: PctPersRsps 44.18 14.71 47.05 11.13 50.06 16.48 0.431 0.029 0.510 0.197 0.511 WCST: PctPersErrors 42.95 14.60 46.81 10.98 49.28 16.76 0.367 0.034 0.377 0.166 0.590 WCST: PctNonPersErrors 43.00 11.56 46.86 7.95 48.78 9.06 0.163 0.061 0.198 0.066 0.540 WCST: PctCLRsps 41.73 12.71 45.81 9.71 47.89 11.27 0.219 0.051 0.242 0.092 0.570 Deficit n % n % n % p ES G1–G2 G1–G3 G2–G3 Cognitive flexibility 11 50.0% 5 23.8% 2 11.1% 0.021* 0.355† 0.046* 0.009* 0.303 AN group anorexia nervosa group, BSD group binge spectrum disorders group, WCST Wisconsin card sorting test, PctErrors percentage of errors, PctPersRsps percentage of perseverative responses, PctPersErrors percentage of perseverative errors, PctNonPersErrors percentage of non-perseverative errors, PctCLRsps percentage of conceptual level responses, Mod moderate, S/E severe/extreme, SD standard deviation, ES effect size [partial eta-squared for ANOVA (η2) and Cramer-V for chi-square test (C-V)]; *Bold: significant comparison; †Bold: effect size into the mild/moderate to the high/large range (η2 > 0.10 or C-V > 0.15). correspond to the deficits in cognitive flexibility. In important domains, such as cognitive flexibility, that do patients diagnosed with AN, poor cognitive flexibility not map onto the current, linear severity criteria. In this may be associated with the perseveration of maladaptive particular sample, cognitive flexibility seems to be associ- cognitive and behavioural patterns [74–76]. Therefore, ated with other factors, such as the duration of illness. it may contribute to the maintenance of the fixation on The present results suggest that duration of illness weight loss, weight control and calorie counting, exces- could be a better variable than DSM-5 severity criteria sive exercise routines, or instilled body image distortion to identify poor cognitive flexibility in patients diagnosed [77, 78], all of which are clinical symptoms in AN [30, 79]. with an ED. The poor cognitive flexibility presented by Our findings are consistent with previous research that the two clinical groups with longer duration of illness highlighted the limited clinical utility of DSM-5 severity could be suggested as a common EDs feature. Some stud- specifiers for AN [4, 5, 9, 18]. Considering the BSD group, ies have associated difficulties in cognitive flexibility with taking into consideration that poor cognitive flexibility is a fixed idea and rigid eating style based on idiosyncratic frequently linked to the inability to cut off compulsive rules and with a greater resistance to be modified by ther- overeating [50] and difficulties set shifting attention away apy [78, 82]. Therefore, it seems that in EDs, the difficulty from ED-related stimuli [80], these findings are consist- to adapt to new behaviour and rules in a changing envi- ent with other studies that highlight the limited clinical ronment could be associated with longer duration of the support for DSM-5 severity criteria for BN [5, 30, 81] disorder. In addition, cognitive rigidity is a variable that is and BED [8, 11, 34, 35]. The results highlight the limita- at odds with the notion of change and is likely to present tions of the DSM-5 severity criteria for EDs, as there are difficulties to therapy and, therefore, is considered one of Mora‑Maltas et al. Journal of Eating Disorders (2023) 11:155 Page 8 of 12 Table 3 Discriminative capacity on cognitive flexibility for alternative severity classifications AN group (n = 100) WCST T-scores EDI-2 drive for thinness Duration of the disorder Low n = 66 High n = 34 Factor group Short n = 78 Long n = 22 Factor group Mean SD Mean SD p ES Mean SD Mean SD p ES WCST: PctErrors 46.94 11.36 49.03 11.29 0.385 0.008 49.13 10.92 42.41 11.44 0.013* 0.061 WCST: PctPersRsps 48.62 11.43 49.41 12.09 0.749 0.001 50.54 11.44 43.05 10.43 0.007* 0.072 WCST: PctPersErrors 48.08 11.25 49.12 11.79 0.667 0.002 50.01 11.15 42.82 10.64 0.008* 0.069 WCST: PctNonPersError 47.14 10.45 49.44 10.47 0.299 0.011 49.01 10.38 44.05 10.06 0.049* 0.039 WCST: PctCLRsps 46.59 11.65 48.79 11.54 0.371 0.008 48.78 11.21 42.23 11.76 0.018* 0.055 Deficit n % n % p ES n % n % p ES Cognitive flexibility 19 29.2% 7 20.6% 0.353 0.093 18 23.4% 8 36.4% 0.222 0.123 BSD group (n = 61) WCST T-scores EDI-2 drive for thinness Duration of the disorder Low n = 25 High n = 36 Factor group Short n = 46 Long n = 15 Factor group Mean SD Mean SD p ES Mean SD Mean SD p ES WCST: PctErrors 46.32 11.98 44.14 11.36 0.474 0.009 46.43 11.74 40.73 10.21 0.098 0.046 WCST: PctPersRsps 48.40 13.11 45.86 14.89 0.495 0.008 48.80 14.95 41.07 9.42 0.065 0.057 WCST: PctPersErrors 47.68 12.99 45.08 15.05 0.487 0.008 48.13 14.99 40.07 9.33 0.055 0 WCST: PctNonPersErrors 46.40 10.16 45.78 9.75 0.810 0.001 47.09 9.97 42.80 9.01 0.144 0.036 WCST: PctCLRsps 46.68 11.90 43.75 11.10 0.329 0.016 46.35 11.57 40.67 10.16 0.095 0.047 Deficit n % n % p ES n % n % p ES Cognitive flexibility 9 36.0% 9 25.0% 0.354 0.119 10 21.7% 8 53.3% 0.020* 0.298 † AN group anorexia nervosa group, BSD group binge spectrum disorders group, WCST Wisconsin card sorting test, PctErrors percentage of errors, PctPersRsps percentage of perseverative responses, PctPersErrors percentage of perseverative errors, PctNonPersErrors percentage of non-perseverative errors, PctCLRsps percentage of conceptual level responses, Mod moderate, S/E severe/extreme, SD standard deviation, ES effect size [partial eta-squared for ANOVA (η2) and Cramer-V for chi-square test (C-V)]; *Bold: significant comparison; †Bold: effect size into the mild/moderate to the high/large range (η2 > 0.10 or C-V > 0.15). the factors associated with a worse prognosis [74]. In fact, variable in terms of cognitive deficits. Although cogni- in EDs, the lack of response to treatment has been linked tive features are not yet considered an important severity to the illness duration and therefore, with chronicity [45]. measure in EDs, several studies have focused interven- Regarding duration and cognition, while some studies tion on improving executive functions as alternative have observed that longer illness duration and severity treatment with promising results [57, 88]. Therefore, the in ED symptomatology were associated with executive improvement of executive functions, such as cognitive dysfunctions [47, 83, 84], other studies did not find sig- flexibility, could influence a greater ability to adapt and nificant associations between longer illness duration and guide problematic behaviours. However, a wide range of cognitive deficits [85, 86]. Therefore, the literature on the commonly shared features (i.e. affective, cognitive, bio- impact of illness duration on executive functions requires logical, or personality) that can occur across all ED diag- further investigation. noses have to be taken into account as well. The preceding leads us to suggest that the DSM sever- ity classification may not adequately reflect the severity Limitations and strengths from a neurocognitive point of view. Consequently, it The results of this study must be interpreted in light of its may be beneficial to consider alternative factors in order limitations. First, our sample size was limited to test the to define severity classifications for EDs. Our results do discriminative power of DSM-5 severity criteria regard- not support DT as a transdiagnostic measure of severity ing cognitive flexibility across all ED subtypes. Second, for EDs. This variable may present some limitations as a related to the previous one, due to the heterogeneity of transdiagnostic measure to assess eating disorders sever- each group it is not possible to reach conclusions about ity as, for example, some individuals could have differ- differences between all the ED subtypes included in the ent concerns, such as muscularity [87]. According to our study (e.g., BN, BED). Future studies should include results, duration of illness seems to be a better severity a sufficient sample of each subtype to better identify Mora‑Maltas et al. Journal of Eating Disorders (2023) 11:155 Page 9 of 12 the characteristics of each. Third, people diagnosed ED Eating disorder EDs Eating disorders with an ED usually present deficits in different cogni- EDI-2 Eating disorders inventory-2 tive domains, hence, other cognitive domains could also ES Effect size have been explored, such as decision making or working Mod Moderate severity PctErrors Percentage of errors in the Wisconsin Card Sorting Test memory. Fourth, although the study used for establishing PctPersRsps Percentage of perseverative responses in the Wisconsin the duration thresholds is representative of the popula- Card Sorting Test tion included in this study, using the thresholds proposed PctPersErrors Percentage of perseverative errors in the Wisconsin Card Sorting Test in a single study could represent a limitation. Fifth, it is PctNonPersErrors Percentage of non-perseverative errors in the Wisconsin important to highlight that this study included only a Card Sorting Test treatment seeking adult ED population. Consequently, PctCLRsps Percentage of conceptual level responses in the Wiscon‑ sin Card Sorting Test our findings might not generalize to other populations, S/E Severe/extreme severity such as adolescents with an ED, for which some studies SD Standard deviation [89] revealed that cognitive impairment was not linked WCST Wisconsin Card Sorting Test to AN. Future studies need to verify the DSM-5 severity Acknowledgements index for EDs and other transdiagnostic severity indica- Not applicable. tors across a range of different treatment seeking and Author contributions community samples. BM-M: Conceptualization, Writing—Original draft; IL: Conceptualization, However, our study also presents some remarkable Writing—Original draft; RG: Data analysis, Writing- Original draft; CV-A: Data strengths. First, our results are in line with previous stud- acquisition; RM-O: Data acquisition; IB: Data acquisition; IS: Data acquisition; JJ-T: Data acquisition; JS-G: Data acquisition; IK: Writing—review and editing; ies that reported the limitations of the DSM-5 severity JT: Writing—review and editing; SJ-M: Writing—review and editing, Funding criteria. Moreover, this study includes further evidence acquisition; FF-A: Conceptualization, Writing—review and editing, Funding that strengthens the conclusions derived from those acquisition. All authors approved the final version of the manuscript. studies, because these limitations seem to extrapolate to Funding cognitive domains such as cognitive flexibility. Second, We thank CERCA Programme/Generalitat de Catalunya for institutional these results encourage the application of a transdiagnos- support. This manuscript and research were supported by grants from the Instituto de Salud Carlos III (ISCIII) [PI20/00132, PI17/01167], by the Delegación tic severity indicator based on illness duration. del Gobierno para el Plan Nacional Sobre Drogas [2017I067 and 2021I031], by the Department of Health of the Generalitat de Catalunya by the call ‘Acció instrumental de programes de recerca orientats en l’àmbit de la recerca i la Conclusions innovació en salut’ [SLT006/17/00246] and cofounded by FEDER funds/Euro‑ The present study makes noteworthy contributions to pean Regional Development Fund (ERDF), a way to build Europe. Additional evidence the limitations of the DSM-5 and DSM-5-TR support was received from EU-H2020 Grants [Eat2beNICE/ H2020-SFS-2016-2; Ref: 728018; and PRIME/ H2020-SC1-BHC-2018-2020; Ref: 847879] and by severity criteria for EDs. The proposed severity classi- AGAUR-Generalitat de Catalunya (2021 SGR 824). RG is supported by the fication does not demonstrate good discrimination in Catalan Institution for Research and Advanced Studies (ICREA-Academia, terms of cognitive flexibility levels, a core significant fea- 2021-Programme). CIBER Fisiopatología de la Obesidad y Nutrición (CIBERobn) is an initiative of ISCIII. IL was supported by the Ministerio de Ciencia e Inno‑ ture of EDs. Furthermore, our findings show that the ED vación (Juan de la Cierva-Formación Program, FJC2021-046494-I). duration is associated with cognitive flexibility deficits, confirming that illness duration can be a good marker Availability of data and materials The datasets generated during and/or analyzed during the current study are of severity in EDs. Future studies should aim to further not publicly available due to ethical restrictions in order to protect the confi‑ demonstrate all the limitations of the DSM-5 and DSM- dentiality of the participants, but are available from the corresponding author 5-TR severity classification for EDs and also to propose on reasonable request. alternative severity variables. Declarations Abbreviations Ethics approval and consent to participate AN Anorexia nervosa All participants received information about the procedure and signed an ANOVA Analysis of variance informed consent form. All procedures were approved by the Ethical Com‑ AN-BP Anorexia nervosa, binge-purgative subtype mittee of the Bellvitge University Hospital in accordance with the Helsinki AN-R Anorexia nervosa, restrictive subtype Declaration of 1975 as revised in 1983 (Refs. 34/05, 307/06). BED Binge eating disorder BMI Body mass index Consent for publication BN Bulimia nervosa Not applicable. BSD Binge spectrum disorders DT  Drive for thinness dimension of the eating disorders Competing interests inventory-2 FFA and SJM received consultancy honoraria from Novo Nordisk and FF-A edi‑ DSM-5 Diagnostic and statistical manual of mental disorders, fifth torial honoraria as EIC from Wiley. The rest of the authors declare no conflict of edition interest. The funders had no role in the design of the study; in the collection, DSM-5-TR Diagnostic and statistical manual of mental disorders, fifth analyses, or interpretation of data; in the writing of the manuscript, or in the edition, text revision decision to publish the results. Mora‑Maltas et al. Journal of Eating Disorders (2023) 11:155 Page 10 of 12 Author details 16. Smink FRE, van Hoeken D, Oldehinkel AJ, Hoek HW. Prevalence and sever‑ 1 Clinical Psychology Unit, University Hospital Bellvitge and CIBERobn, Feixa Lla‑ ity of DSM-5 eating disorders in a community cohort of adolescents. Int J rga s/n 08907 L’Hospitalet del Llobregat, Barcelona, Spain. 2 Psychoneurobiol‑ Eat Disord. 2014;47:610–9. https://​doi.​org/​10.​1002/​eat.​22316. ogy of Eating and Addictive Behaviours Group, Bellvitge Biomedical Research 17. Smith KE, Ellison JM, Crosby RD, Engel SG, Mitchell JE, Crow SJ, et al. The Institute (IDIBELL), Barcelona, Spain. 3 CIBER de Fisiopatología de la Obesidad y validity of DSM-5 severity specifiers for anorexia nervosa, bulimia nervosa, Nutrición (CIBERobn), Instituto de Salud Carlos III, Barcelona, Spain. 4 Departa‑ and binge-eating disorder. Int J Eat Disord. 2017;50:1109–13. https://​doi.​ ment de Psicobiologia i Metodologia, Universitat Autònoma de Barcelona, org/​10.​1002/​eat.​22739. Barcelona, Spain. 5 Melbourne School of Psychological Sciences, The University 18. Zayas LV, Wang SB, Coniglio K, Becker K, Murray HB, Klosterman E, et al. of Melbourne, Melbourne, Australia. 6 Gerencia Territorial Metropolitana Sud. Gender differences in eating disorder psychopathology across DSM-5 Hospital Universitari de Bellvitge, Barcelona, Spain. 7 Department of Clinical severity categories of anorexia nervosa and bulimia nervosa. Int J Eat Sciences, School of Medicine and Health Sciences, University of Barcelona, Disord. 2018;51:1098–102. https://​doi.​org/​10.​1002/​eat.​22941. Barcelona, Spain. 19. Dakanalis A, Alix Timko C, Colmegna F, Riva G, Clerici M. Evaluation of the DSM-5 severity ratings for anorexia nervosa in a clinical sample. Psychia‑ Received: 27 April 2023 Accepted: 22 August 2023 try Res. 2018;262:124–8. https://​doi.​org/​10.​1016/j.​psych​res.​2018.​02.​009. 20. Devoe DJ, Dimitropoulos G, Anderson A, Bahji A, Flanagan J, Soumbasis A, et al. The prevalence of substance use disorders and substance use in anorexia nervosa: a systematic review and meta-analysis. J Eat Disord. 2021;9:161. https://​doi.​org/​10.​1186/​s40337-​021-​00516-3. 21. Peat C, Mitchell JE, Hoek H, Wonderlich S. Validity and utility of subtyping References anorexia nervosa. Int J Eat Disord. 2009;42:590–4. https://​doi.​org/​10.​1002/​ 1. American Psychiatric Association. Diagnostic and statistical manual of eat.​20717. mental disorders. Washington, D.C.: American Psychiatric Association; 22. Sanchez I, Lucas I, Munguía L, Camacho-Barcia L, Giménez M, Sánchez- 2013. https://​doi.​org/​10.​1176/​appi.​books.​97808​90425​596. González J, et al. Food addiction in anorexia nervosa: Implications for the 2. American Psychiatric Association. Diagnostic and statistical manual of understanding of crossover diagnosis. Eur Eat Disord Rev. 2022;30:278– mental disorders. Washington, D.C.: American Psychiatric Association 88. https://​doi.​org/​10.​1002/​erv.​2897. Publishing; 2022. https://​doi.​org/​10.​1176/​appi.​books.​97808​90425​787. 23. Carter JC, Mercer-Lynn KB, Norwood SJ, Bewell-Weiss CV, Crosby RD, 3. Treasure J, Duarte TA, Schmidt U. Eating disorders. Lancet. 2020;395:899– Woodside DB, et al. A prospective study of predictors of relapse in 911. https://​doi.​org/​10.​1016/​S0140-​6736(20)​30059-3. anorexia nervosa: Implications for relapse prevention. Psychiatry Res. 4. Dang AB, Giles S, Fuller-Tyszkiewicz M, Kiropoulos L, Krug I. A system‑ 2012;200:518–23. https://​doi.​org/​10.​1016/j.​psych​res.​2012.​04.​037. atic review and meta-analysis on the DSM–5 severity ratings for eating 24. Kelly AC, Carter JC. Eating disorder subtypes differ in their rates of psy‑ disorders. Clin Psychol Sci Pract. 2022;29:325–44. https://​doi.​org/​10.​1037/​ chosocial improvement over treatment. J Eat Disord. 2014;2:2. https://​doi.​ cps00​00078. org/​10.​1186/​2050-​2974-2-2. 5. Gianini L, Roberto CA, Attia E, Walsh BT, Thomas JJ, Eddy KT, et al. Mild, 25. Dakanalis A, Bartoli F, Caslini M, Crocamo C, Zanetti MA, Riva G, et al. Valid‑ moderate, meaningful? Examining the psychological and functioning ity and clinical utility of the DSM-5 severity specifier for bulimia nervosa: correlates of DSM-5 eating disorder severity specifiers. Int J Eat Disord. results from a multisite sample of patients who received evidence-based 2017;50:906–16. https://​doi.​org/​10.​1002/​eat.​22728. treatment. Eur Arch Psychiatry Clin Neurosci. 2017;267:823–9. https://​doi.​ 6. Krug I, Binh Dang A, Granero R, Agüera Z, Sánchez I, Riesco N, et al. Drive org/​10.​1007/​s00406-​016-​0712-7. for thinness provides an alternative, more meaningful, severity indicator 26. Dakanalis A, Clerici M, Riva G, Carrà G. Testing the DSM-5 severity indica‑ than the DSM-5 severity indices for eating disorders. Eur Eat Disord Rev. tor for bulimia nervosa in a treatment-seeking sample. Eat Weight Disord. 2021;29:482–98. https://​doi.​org/​10.​1002/​erv.​2776. 2017;22:161–7. https://​doi.​org/​10.​1007/​s40519-​016-​0324-2. 7. Lucas I, Jiménez-Murcia S, Fernández-Aranda F. DSM-5 severity ratings 27. Dakanalis A, Colmegna F, Zanetti MA, Di Giacomo E, Riva G, Clerici for eating disorders: some clinical implications and future directions. Clin M. Evaluation of the DSM-5 severity specifier for bulimia nervosa in Psychol Sci Pract. 2022. https://​doi.​org/​10.​1037/​cps00​00096. treatment-seeking youth. Child Psychiatry Hum Dev. 2018;49:137–45. 8. Lydecker JA, Ivezaj V, Grilo CM. Testing the validity and clinical utility of https://​doi.​org/​10.​1007/​s10578-​017-​0735-y. the severity specifiers for binge-eating disorder for predicting treatment 28. Shroff H, Reba L, Thornton LM, Tozzi F, Klump KL, Berrettini WH, et al. Fea‑ outcomes. J Consult Clin Psychol. 2020;88:172–8. https://​doi.​org/​10.​1037/​ tures associated with excessive exercise in women with eating disorders. ccp00​00464. Int J Eat Disord. 2006;39:454–61. https://​doi.​org/​10.​1002/​eat.​20247. 9. Machado PPP, Grilo CM, Crosby RD. Evaluation of the DSM-5 severity 29. Gorrell S, Hail L, Kinasz K, Bruett L, Forsberg S, Delucchi K, et al. A test of indicator for anorexia nervosa. Eur Eat Disord Rev. 2017;25:221–3. https://​ the DSM-5 severity specifier for bulimia nervosa in adolescents: can we doi.​org/​10.​1002/​erv.​2508. anticipate clinical treatment outcomes? Int J Eat Disord. 2019;52:586–90. 10. Regier DA, Kuhl EA, Kupfer DJ. The DSM-5: classification and criteria https://​doi.​org/​10.​1002/​eat.​23034. changes. World Psychiatry. 2013;12:92–8. https://​doi.​org/​10.​1002/​wps.​ 30. Nakai Y, Nin K, Noma S, Teramukai S, Fujikawa K, Wonderlich SA. The 20050. impact of DSM-5 on the diagnosis and severity indicator of eating dis‑ 11. Grilo CM, Ivezaj V, White MA. Evaluation of the DSM-5 severity indica‑ orders in a treatment-seeking sample. Int J Eat Disord. 2017;50:1247–54. tor for binge eating disorder in a community sample. Behav Res Ther. https://​doi.​org/​10.​1002/​eat.​22777. 2015;66:72–6. https://​doi.​org/​10.​1016/j.​brat.​2015.​01.​004. 31. Grilo CM, Ivezaj V, White MA. Evaluation of the DSM-5 severity indicator 12. Hirst RB, Beard CL, Colby KA, Quittner Z, Mills BM, Lavender JM. Anorexia for bulimia nervosa. Behav Res Ther. 2015;67:41–4. https://​doi.​org/​10.​ nervosa and bulimia nervosa: a meta-analysis of executive functioning. 1016/j.​brat.​2015.​02.​002. Neurosci Biobehav Rev. 2017;83:678–90. https://​doi.​org/​10.​1016/j.​neubi​ 32. Jenkins PE, Luck A, Cardy J, Staniford J. How useful is the DSM-5 severity orev.​2017.​08.​011. indicator in bulimia nervosa? A clinical study including a measure of 13. Marucci S, Ragione LD, De Iaco G, Mococci T, Vicini M, Guastamacchia E, impairment. Psychiatry Res. 2016;246:366–9. https://​doi.​org/​10.​1016/j.​ et al. Anorexia nervosa and comorbid psychopathology. Endocr Metab psych​res.​2016.​10.​011. Immune Disord Drug Targets. 2018;18:316–24. https://​doi.​org/​10.​2174/​ 33. Dakanalis A, Riva G, Serino S, Colmegna F, Clerici M. Classifying adults 18715​30318​66618​02131​11637. with binge eating disorder based on severity levels. Eur Eat Disord Rev. 14. Dalle Grave R, Sartirana M, El Ghoch M, Calugi S. DSM-5 severity specifiers 2017;25:268–74. https://​doi.​org/​10.​1002/​erv.​2518. for anorexia nervosa and treatment outcomes in adult females. Eat Behav. 34. Grilo CM, Ivezaj V, White MA. Evaluation of the DSM-5 severity indi‑ 2018;31:18–23. https://​doi.​org/​10.​1016/j.​eatbeh.​2018.​07.​006. cator for binge eating disorder in a clinical sample. Behav Res Ther. 15. Reas DL, Rø Ø. Investigating the DSM-5 severity specifiers based on 2015;71:110–4. https://​doi.​org/​10.​1016/j.​brat.​2015.​05.​003. thinness for adults with anorexia nervosa. Int J Eat Disord. 2017;50:990–4. 35. Forrest LN, Jacobucci RC, Grilo CM. Empirically determined severity lev‑ https://​doi.​org/​10.​1002/​eat.​22729. els for binge-eating disorder outperform existing severity classification Mora‑Maltas et al. Journal of Eating Disorders (2023) 11:155 Page 11 of 12 schemes. Psychol Med. 2022;52:685–95. https://​doi.​org/​10.​1017/​S0033​ 56. Tchanturia K, Lloyd S, Lang K. Cognitive remediation therapy for anorexia 29172​00022​87. nervosa: current evidence and future research directions. Int J Eat Disord. 36. Garner DM. Eating disorder inventory-2: professional manual. Odessa: 2013;46:492–5. https://​doi.​org/​10.​1002/​eat.​22106. Psychological Assessment Resources; 1991. 57. Tchanturia K, Giombini L, Leppanen J, Kinnaird E. Evidence for cogni‑ 37. Chernyak Y, Lowe MR. Motivations for dieting: drive for thinness tive remediation therapy in young people with anorexia nervosa: is different from drive for objective thinness. J Abnorm Psychol. systematic review and meta-analysis of the literature. Eur Eat Disord Rev. 2010;119:276–81. https://​doi.​org/​10.​1037/​a0018​398. 2017;25:227–36. https://​doi.​org/​10.​1002/​erv.​2522. 38. De Young KP, Lavender JM, Steffen K, Wonderlich SA, Engel SG, Mitchell 58. Tchanturia K, Davies H, Lopez C, Schmidt U, Treasure J, Wykes T. Letter to JE, et al. Restrictive eating behaviors are a nonweight-based marker of the Editor: neuropsychological task performance before and after cogni‑ severity in anorexia nervosa. Int J Eat Disord. 2013;46:849–54. https://​ tive remediation in anorexia nervosa: a pilot case-series. Psychol Med. doi.​org/​10.​1002/​eat.​22163. 2008;38:1371–3. https://​doi.​org/​10.​1017/​S0033​29170​80037​96. 39. Peñas-Lledó E, Bulik CM, Lichtenstein P, Larsson H, Baker JH. Risk for 59. Dingemans AE, Danner UN, Donker JM, Aardoom JJ, Van Meer F, Tobias self-reported anorexia or bulimia nervosa based on drive for thin‑ K, et al. The effectiveness of cognitive remediation therapy in patients ness and negative affect clusters/dimensions during adolescence: a with a severe or enduring eating disorder: a randomized controlled trial. three-year prospective study of the TChAD cohort. Int J Eat Disord. Psychother Psychosom. 2013;83:29–36. https://​doi.​org/​10.​1159/​00035​ 2015;48:692–9. https://​doi.​org/​10.​1002/​eat.​22431. 5240. 40. Davidsen AH, Hoyt WT, Poulsen S, Waaddegaard M, Lau M. Eating 60. Hagan KE, Christensen KA, Forbush KT. A preliminary systematic review disorder severity and functional impairment: moderating effects of ill‑ and meta-analysis of randomized-controlled trials of cognitive remedia‑ ness duration in a clinical sample. Eat Weight Disord. 2017;22:499–507. tion therapy for anorexia nervosa. Eat Behav. 2020;37:101391. https://​doi.​ https://​doi.​org/​10.​1007/​s40519-​016-​0319-z. org/​10.​1016/j.​eatbeh.​2020.​101391. 41. Errichiello L, Iodice D, Bruzzese D, Gherghi M, Senatore I. Prognostic 61. Brockmeyer T, Schmidt H, Leiteritz-Rausch A, Zimmermann J, Wünsch- factors and outcome in anorexia nervosa: a follow-up study. Eat Weight Leiteritz W, Leiteritz A, et al. Cognitive remediation therapy in anorexia Disord. 2016;21:73–82. https://​doi.​org/​10.​1007/​s40519-​015-​0211-2. nervosa—a randomized clinical trial. J Consult Clin Psychol. 2021;89:805– 42. Reas DL, Williamson DA, Martin CK, Zucker NL. Duration of illness 15. https://​doi.​org/​10.​1037/​ccp00​00675. predicts outcome for bulimia nervosa: a long-term follow-up study. 62. Mallorquí-Bagué N, Lozano-Madrid M, Granero R, Mestre-Bach G, Int J Eat Disord. 2000;27:428–34. https://​doi.​org/​10.​1002/​(sici)​1098-​ Vintró-Alcaraz C, Sánchez I, et al. Cognitive and clinical gender-related 108x(200005)​27:4%​3c428::​aid-​eat7%​3e3.0.​co;2-y. differences among binge-spectrum eating disorders: analysis of therapy 43. Steinhausen H-C, Weber S. The outcome of bulimia nervosa: findings response predictors. Eur Eat Disord Rev. 2022. https://​doi.​org/​10.​1002/​erv.​ from one-quarter century of research. Am J Psychiatry. 2009;166:1331– 2961. 41. https://​doi.​org/​10.​1176/​appi.​ajp.​2009.​09040​582. 63. Voon V. Cognitive biases in binge eating disorder: the hijacking of deci‑ 44. Vall E, Wade TD. Predictors of treatment outcome in individuals with sion making. CNS Spectr. 2015;20:566–73. https://​doi.​org/​10.​1017/​S1092​ eating disorders: a systematic review and meta-analysis. Int J Eat 85291​50006​81. Disord. 2015;48:946–71. https://​doi.​org/​10.​1002/​eat.​22411. 64. Zakzanis KK, Campbell Z, Polsinelli A. Quantitative evidence for distinct 45. Fernández-Aranda F, Treasure J, Paslakis G, Agüera Z, Giménez M, Gran‑ cognitive impairment in anorexia nervosa and bulimia nervosa. J Neu‑ ero R, et al. The impact of duration of illness on treatment nonresponse ropsychol. 2010;4:89–106. https://​doi.​org/​10.​1348/​17486​6409X​459674. and drop-out: exploring the relevance of enduring eating disorder 65. Lang K, Stahl D, Espie J, Treasure J, Tchanturia K. Set shifting in children concept. Eur Eat Disord Rev. 2021;29:499–513. https://​doi.​org/​10.​1002/​ and adolescents with anorexia nervosa: an exploratory systematic review erv.​2822. and meta-analysis. Int J Eat Disord. 2014;47:394–9. https://​doi.​org/​10.​ 46. Reville M-C, O’Connor L, Frampton I. Literature review of cognitive 1002/​eat.​22235. neuroscience and anorexia nervosa. Curr Psychiatry Rep. 2016;18:18. 66. American Psychiatric Association. Diagnostic and Statistical Manual of https://​doi.​org/​10.​1007/​s11920-​015-​0651-4. Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR). vol. 1. Arling‑ 47. Miranda-Olivos R, Testa G, Lucas I, Sánchez I, Sánchez-González J, ton: American Psychiatric Association; 2000. https://​doi.​org/​10.​1176/​appi.​ Granero R, et al. Clinical factors predicting impaired executive func‑ books.​97808​90423​349. tions in eating disorders: the role of illness duration. J Psychiatr Res. 67. Preti A, De Girolamo G, Vilagut G, Alonso J, De Graaf R, Bruffaerts R, et al. 2021;144:87–95. https://​doi.​org/​10.​1016/j.​jpsyc​hires.​2021.​09.​042. The epidemiology of eating disorders in six European countries: results of 48. Lezak M. Neuropsychological assessment. 1995. the ESEMeD-WMH project. J Psychiatr Res. 2009;43:1125–32. https://​doi.​ 49. Heaton RK. Wisconsin card sorting test computer version 2.0. Odessa. org/​10.​1016/j.​jpsyc​hires.​2009.​04.​003. 1993. 68. Hollingshead AB. Four factor index of social status. Yale J Sociol. 50. Wu M, Brockmeyer T, Hartmann M, Skunde M, Herzog W, Friederich HC. 2011;8:21–51. Set-shifting ability across the spectrum of eating disorders and in over‑ 69. Grant DA, Berg EA. A behavioral analysis of degree of reinforcement and weight and obesity: a systematic review and meta-analysis. Psychol ease of shifting to new responses in a Weigl-type card-sorting problem. J Med. 2014;44:3365–85. https://​doi.​org/​10.​1017/​S0033​29171​40002​94. Exp Psychol. 1948;38:404–11. https://​doi.​org/​10.​1037/​h0059​831. 51. Diaz-Marsa M, Pemau A, de la Torre-Luque A, Vaz-Leal F, Rojo-Moreno 70. Munguía L, Lucas I, Jiménez-Murcia S, Mora-Maltas B, Granero R, Miranda- L, Beato-Fernandez L, et al. Executive dysfunction in eating disorders: Olivos R, et al. Executive functions in binge spectrum eating disorders relationship with clinical features. Prog Neuro-Psychopharmacol Biol with comorbid compulsive buying. Eur Eat Disord Rev. 2021;29:854–67. Psychiatry. 2023;120:110649. https://​doi.​org/​10.​1016/j.​pnpbp.​2022.​ https://​doi.​org/​10.​1002/​erv.​2855. 110649. 71. Tanner-Eggen C, Balzer C, Perrig WJ, Gutbrod K. The neuropsychological 52. Perpiñá C, Segura M, Sánchez-Reales S. Cognitive flexibility and decision- assessment of cognitive deficits considering measures of performance making in eating disorders and obesity. Eat Weight Disord. 2017;22:435– variability. Arch Clin Neuropsychol. 2015;30:217–27. https://​doi.​org/​10.​ 44. https://​doi.​org/​10.​1007/​s40519-​016-​0331-3. 1093/​arclin/​acv008. 53. Abbate-Daga G, Buzzichelli S, Marzola E, Aloi M, Amianto F, Fassino S. 72. Peñas-Lledó E, Fernández-Aranda F, Jiménez-Murcia S, Granero R, Penelo Does depression matter in neuropsychological performances in anorexia E, Soto A, et al. Subtyping eating disordered patients along drive for thin‑ nervosa? A descriptive review. Int J Eat Disord. 2015;48:736–45. https://​ ness and depression. Behav Res Ther. 2009;47:513–9. https://​doi.​org/​10.​ doi.​org/​10.​1002/​eat.​22399. 1016/j.​brat.​2009.​03.​003. 54. Holliday J, Tchanturia K, Landau S, Collier D, Treasure J. Is impaired 73. Davenport E, Rushford N, Soon S, McDermott C. Dysfunctional metacog‑ set-shifting an endophenotype of anorexia nervosa? Am J Psychiatry. nition and drive for thinness in typical and atypical anorexia nervosa. J Eat 2005;162:2269–75. https://​doi.​org/​10.​1176/​appi.​ajp.​162.​12.​2269. Disord. 2015;3:24. https://​doi.​org/​10.​1186/​s40337-​015-​0060-4. 55. Westwood H, Stahl D, Mandy W, Tchanturia K. The set-shifting profiles 74. Danner UN, Sanders N, Smeets PAM, van Meer F, Adan RAH, Hoek HW, of anorexia nervosa and autism spectrum disorder using the Wisconsin et al. Neuropsychological weaknesses in anorexia nervosa: set-shifting, Card Sorting Test: a systematic review and meta-analysis. Psychol Med. central coherence, and decision making in currently ill and recovered 2016;46:1809–27. https://​doi.​org/​10.​1017/​S0033​29171​60005​81. Mora‑Maltas et al. Journal of Eating Disorders (2023) 11:155 Page 12 of 12 women. Int J Eat Disord. 2012;45:685–94. https://​doi.​org/​10.​1002/​eat.​ 22007. 75. Roberts ME, Tchanturia K, Stahl D, Southgate L, Treasure J. A systematic review and meta-analysis of set-shifting ability in eating disorders. Psy‑ chol Med. 2007;37:1075–84. https://​doi.​org/​10.​1017/​S0033​29170​70098​ 77. 76. Stedal K, Rose M, Frampton I, Landrø NI, Lask B. The neuropsychological profile of children, adolescents, and young adults with anorexia nervosa. Arch Clin Neuropsychol. 2012;27:329–37. https://​doi.​org/​10.​1093/​arclin/​ acs032. 77. Aloi M, Rania M, Caroleo M, Bruni A, Palmieri A, Cauteruccio MA, et al. Decision making, central coherence and set-shifting: a compari‑ son between binge eating disorder, anorexia nervosa and healthy controls. BMC Psychiatry. 2015;15:1–10. https://​doi.​org/​10.​1186/​ s12888-​015-​0395-z. 78. Kanakam N, Treasure J. A review of cognitive neuropsychiatry in the tax‑ onomy of eating disorders: state, trait, or genetic? Cogn Neuropsychiatry. 2013;18:83–114. https://​doi.​org/​10.​1080/​13546​805.​2012.​682362. 79. Steinglass JE, Walsh BT, Stern Y. Set shifting deficit in anorexia nervosa. J Int Neuropsychol Soc. 2006;12:431–5. https://​doi.​org/​10.​1017/​s1355​ 61770​60605​28. 80. Diamond A. Executive functions. Annu Rev Psychol. 2013;64:135–68. https://​doi.​org/​10.​1146/​annur​ev-​psych-​113011-​143750. 81. Smith KE, Mason TB, Johnson JS, Lavender JM, Wonderlich SA. A systematic review of reviews of neurocognitive functioning in eating disorders: the state-of-the-literature and future directions. Int J Eat Disord. 2018;51:798–821. https://​doi.​org/​10.​1002/​eat.​22929. 82. Kaye WH, Fudge JL, Paulus M. New insights into symptoms and neuro‑ circuit function of anorexia nervosa. Nat Rev Neurosci. 2009;10:573–84. https://​doi.​org/​10.​1038/​nrn26​82. 83. Grau A, Magallón-Neri E, Faus G, Feixas G. Cognitive impairment in eating disorder patients of short and long-term duration: a case-control study. Neuropsychiatr Dis Treat. 2019;15:1329–41. https://​doi.​org/​10.​2147/​NDT.​ S1999​27. 84. Roberts ME, Tchanturia K, Treasure JL. Exploring the neurocognitive signature of poor set-shifting in anorexia and bulimia nervosa. J Psychiatr Res. 2010;44:964–70. https://​doi.​org/​10.​1016/j.​jpsyc​hires.​2010.​03.​001. 85. Cavedini P, Bassi T, Ubbiali A, Casolari A, Giordani S, Zorzi C, et al. Neu‑ ropsychological investigation of decision-making in anorexia nervosa. Psychiatry Res. 2004;127:259–66. https://​doi.​org/​10.​1016/j.​psych​res.​2004.​ 03.​012. 86. Galimberti E, Martoni RM, Cavallini MC, Erzegovesi S, Bellodi L. Motor inhibition and cognitive flexibility in eating disorder subtypes. Prog Neuropsychopharmacol Biol Psychiatry. 2012;36:307–12. https://​doi.​org/​ 10.​1016/j.​pnpbp.​2011.​10.​017. 87. Lavender JM, Brown TA, Murray SB. Men, muscles, and eating disor‑ ders: an overview of traditional and muscularity-oriented disordered eating. Curr Psychiatry Rep. 2017;19:1–7. https://​doi.​org/​10.​1007/​ s11920-​017-​0787-5. 88. Roberts ME. Feasibility of group cognitive remediation therapy in an adult eating disorder day program in New Zealand. Eat Behav. 2018;30:1– 4. https://​doi.​org/​10.​1016/j.​eatbeh.​2018.​04.​004. 89. Miles S, Gnatt I, Phillipou A, Nedeljkovic M. Cognitive flexibility in acute anorexia nervosa and after recovery: a systematic review. Clin Psychol Rev. 2020;81:101905. https://​doi.​org/​10.​1016/j.​cpr.​2020.​101905. Publisher’s Note Ready to submit your research ? Choose BMC and benefit from: Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions