The Patient’s Perspective: Exploring Factors that Contribute to Recovery from Eating Disorders The Patient’s Perspective: Exploring Factors That Contribute to Recovery from Eating Disorders

Eating Disorders (EDs) are quite distinct and difficult to treat mental disorders. Remarkably, when reviewing studies on the recovery process of ED’s, the patients’ experience of the therapeutic process has rarely been taken into account. To address the issue of the patients perspective on their recovery a study was conducted among ED patients. The main aims of the study were to investigate treatment factors, according to the patients view, that contributed to their recovery. Also, if there was any significant improvement within the first six months of the treatment and whether subjective improvement was in accordance with objective ED improvement and finally possible, if any, differences between different diagnostic groups. Overall, four major conclusions were produced. First, most of the patients appreciate a structured therapy in a specialized unit for EDs with a caring therapist that can form a strong therapeutic alliance with them. Second, most of the patients regard as recovery a change that goes beyond ED symptomatology to a more “holistic” improvement. Third, the patients opinion on therapy and recovery is not influenced by the diagnosis of the ED. Finally, the opinion of the patient and the objective evaluation of recovery present more similarities than differences. recovery: regular


Introduction
Eating disorders (EDs) are a group of mental disorders that are quite distinct and difficult to treat, as they manifest through a number of unique characteristics: a. Although they are not considered to be a type of psychosis, there are a number of patients suffering from anorexia nervosa (AN) who lack insight on and express delusional beliefs about primarily their body image [1].
b. The diagnosis of ED often does not remain stable over the years. It is not uncommon to observe a patient suffering from AN during adolescence, bulimia nervosa (BN) or other specified feeding and eating disorders (OSFEDs) during early adulthood and binge eating disorder (BED) during middle age. This has led a number of researchers to suggest a "transdiagnostic" approach to ED classification and therapy [2,3].
c. The first line of treatment for EDs is psychotherapy [4]. The difficulty with the application of psychotherapy, however, is that when treating patients whose symptomatology is highly reinforced, as is the case with EDs, then its effectiveness is substantially reduced. Moreover, a considerable number of patients, especially those suffering from AN, drop out of therapy for a variety of reasons [5].
d. The course of EDs is long; in many cases, patients need several years to recover. Moreover, a substantial number of patients do not recover fully or even partially and run the risk of developing a chronic disorder [6,7].
Overall, these four points stress the necessity of improving the available therapeutic interventions. It is alarming that there are published studies on AN reporting that specialized psychotherapeutic interventions are only equally or less effective than nonspecific supportive clinical management for the disorder [8]. Remarkably, when reviewing studies on the recovery process of EDs, the patient's experience of the therapeutic process has rarely been taken into account. The patient's perspective can and must play an important role on the development of new or the improvement of existing therapeutic interventions.

The patient's perspective of recovery from eating disorders
Over the last 20 years, patient evaluation of treatment has been a topic of a small number of studies. The earlier studies, which were published between 1990 and 2005, showed that most of the patients who were interviewed preferred individual therapy, group therapy and selfhelp groups than other forms of treatment, such as inpatient programs [9,10]. Interestingly, the more "specialized" therapies, such as behavioral and family therapy, were not regarded positively by the patients [9,10]. The majority of the patients were more satisfied when they were treated in a specialized unit for ED's or by a therapist experienced specifically in ED therapy [10,11]. Moreover, in clinical practice, there are patients who report that their need to be understood was often not met when they were treated by therapists who were not specialized in ED. The initial focus of treatment in weight gain and nutritional restoration seemed to be regarded positively by some patients and negatively by others [11][12][13]. A qualitative study showed that patients suffering from an ED preferred psychological types of treatment whereas medical interventions that focused exclusively on weight were not regarded as useful [14].
A more recent study, conducted in the Netherlands, evaluated the treatment of EDs from the patients' perspective in a large community-based sample. This study showed that treatment in specialized ED centers, self-help groups and treatments with a partner were reported to be most helpful [15]. The patients reported that the communication skills of mental health professionals, the therapist/patient working alliance, contact with peers and focus of treatment on ED symptoms as well as underlying issues were the most beneficial elements of their treatment [15].

Subjective vs. objective recovery
Most of the research on the outcome of EDs has been focused on the reduction in associated symptomatology, primarily nutrition restoration, normalization of body weight and absence of purging behaviors. While the above changes are sina qua non for achieving remission, it has been reported that when only these somatic criteria were considered, 79% of AN patients were considered recovered; when psychological criteria were taken into account, however, the recovery rate fell to 49% [16]. Moreover, a common observation in clinical practice is that when some patients "recover" from ED symptomatology, they become extremely anxious, distressed or even depressed. It becomes obvious that a gap might exist between the "official" recovery as measured by the researchers and the subjective evaluation of being well and free of the ED's "golden cage." Noordenbos and Seubring created a list of 52 possible criteria for recovery. This list represents the domains of eating behavior, body experience, physical and psychological well-being, and emotional and social functioning. They asked therapists and expatients to select the criteria that they viewed as important for recovery from EDs [17]. Interestingly, they found that the two groups agreed on most of the criteria for recovery. Considering the patients' experience at the end of their treatment, only four criteria had been realized by more than 80% of the patients: "I do not vomit," "I do not use laxatives," "I do not use diuretics" and "I do not use slimming pills." On the other hand, it is striking that three of the five criteria for body experience had been realized by less than 50% of the patients at the end of their treatment. Finally, three criteria were evaluated by both expatients and therapists as not important for recovery: regular menstruation, dental health and intimate relationships [17].

The therapist's characteristics
According to the patients' view, the therapist and especially a feeling of trust toward the therapist were the most important elements of what they considered high-quality therapy [18,19]. The patients believed that a good therapist needs to have "good communication skills, knowledge of and experience with EDs, the ability to facilitate engaging in a relationship with the patient, to listen to the patient, to stand beside the patient and work together, and to focus on the person and not the disorder" [18]. Therapists having specific expertise in EDs seem to be very important elements for a positive outcome of the therapy, both for the patients and their families [19]. Furthermore, the way that the therapist interacts with the patient plays a major role in the treatment's success, according to patient opinion. A positive encounter with the therapist can make the patient feel understandable, less alone, safe, valuable and in a better mood; a negative encounter can increase self-blame, loss of feeling of identity and desire to prove that she/he is "really sick," as well as decreased willingness to disclose, which can eventually lead to dropping out of therapy [20]. The "good" therapist characteristics can be summarized in four factors: "acceptance" of the individual's personality, preferences and difficulties; "vitality" in showing active interest in the patient's ideas, thoughts and problems; "challenging" the situation created by the ED symptomatology, focusing on the patient's resources, and offering active support while not pampering him/her and finally, "expertise" in the area of ED treatment so that he/she can offer security and guidance during therapy [19].

The importance of therapeutic alliance
The patient/therapist alliance has been acknowledged as a key element of the therapeutic process and a successful outcome in numerous studies across a range of treatment modalities and research settings [21]. It is not uncommon for the ED patient and therapist to have different opinions on therapeutic goals and the definition of their alliance and its dimensions. As one might expect, converging perspectives have been associated with positive outcome and vice versa [22]. Research data on the therapeutic alliance components have shown that patients view alliance in six basic components: collaborative work relationship, productive work, active commitment, bond, nondisagreement on goals/tasks and confident progress. Meanwhile, the therapists view alliance in four basic components: collaborative work relationship, therapist confidence and dedication, client commitment and confidence and client working ability [22]. Although there are similarities between the patients' and the therapists' view of therapeutic alliance, there are also differences. Patients appear to place greater emphasis on helpfulness, joint participation in the work of therapy and negative signs of the alliance [22].
Although the patient/therapist alliance has not been specifically investigated in the area of ED's research, the studies that compare the patients' and the therapists' opinions on the elements that are important for recovery from EDs have shown considerable converging perspectives [23]. Vanderlinden et al. have reported that "improving self-esteem," "improving body experience" and "learning problem solving skills" were considered core elements of the treatment by both the patients and their therapists [23].

Quality of life
The quality of life (QoL) of ED patients has been reported to be poor. Severity of ED symptomatology, especially low body mass index (BMI), but not duration of illness has been found to be predictive of low QoL [24]. Moreover, in terms of differences across diagnostic groups, individuals with a diagnosis of AN were found to have lower psychological and physical/ cognitive QoL than those with an OSFED or BN diagnosis [24]. Even after long-term treatment or recovery from EDs, the expatients reported only a slight improvement in their QoL [18]. In most cases, the QoL remained poorer than controls [25,26]. Self-image and well-being are often reported by current ED patients as the most affected areas of their QoL [18]. The patients mention that a sense of belonging, work or education, good physical health and a general sense of well-being are the most important elements of good QoL [18].

Self-efficacy
Another construct that can be explored through the patient's perspective of their ED recovery is self-efficacy. Self-efficacy has been defined as the individual's perceived ability to perform a particular behavior [27]. When considering EDs, self-efficacy can be thought of as the individual's subjective positive evaluation of his/her ability to eat without engaging in eating disordered behavior and to maintain a realistic body image [28]. There are ED studies showing that patients' self-efficacy around eating increases with treatment and is directly related with improvement in objective measurements, such as duration of hospitalization, drive for thinness and body dissatisfaction [29].

The study on the patients' perspective of therapy and recovery
The study was conducted among patients suffering from an ED who were treated at the Eating Disorders Unit of National and Kapodistrian University of Athens, 1st Psychiatric Department.

Aims
The main aims of the study were to investigate the following: 1. treatment factors, according to the patients' view, which contributed to their recovery; 2. the recovery criteria that had been identified by the patients during the first 6 months of their treatment; 3. if there was any significant improvement within the first 6 months of the treatment and whether subjective improvement was in accordance with objective improvement of ED symptomatology; and 4. possible differences, if any, between diagnostic groups regarding their perception of recovery and therapy.

Procedure
Each patient was approached during the initial assessment interview. The diagnosis of ED was confirmed by a psychiatrist specialized in EDs according to DSM-5 criteria. It should be noted, however, that due to the fact that the Eating Disorders Unit does not treat patients suffering from BED, the diagnosis of ED was either AN, BN or OSFED. The latter group consisted mainly of anorectic-type or bulimic-type patients who did not meet the full criteria for the diagnosis of AN or BN.
During the initial interview, the body mass index (BMI) of each patient was calculated following the measurement of her/his height and body weight. Also, the questionnaires for the measurement of ED symptomatology, quality of life and demographic data were admin-istered. After the initial interview, every patient who was eligible for treatment was assigned to a psychologist or psychiatrist specialized in cognitive behavioral therapy (CBT) for EDs. The inclusion criteria were diagnosis of an ED; age >17 years; adequate knowledge of the Greek language; and lack of psychosis, neurological disorders and substance misuse. Each therapist had weekly group supervision meetings with one of the authors (Fragiskos Gonidakis). All participants were contacted again by the main researcher (Dafni-Alexandra Karapavlou) 6 months after their first session. During the 6 months, the questionnaires on the subjective experience of therapy and recovery were administered alongside measurements of ED symptomatology and QoL Written informed consent, according to the Declaration of Helsinki, was obtained from all participants.

Instruments
For the objective measurement of improvement and recovery, the following questionnaires were used: 1. Eating Disorders Examination-Questionnaire (EDE-Q). The EDE-Q is a self-report questionnaire developed by Fairburn and Wilson [30]. It consists of 36 questions on eating behavior, clustered in four subscales: restraint, eating concern, shape concern and weight concern. Each question is rated on a 6-point Likert type scale and addresses the patient's last 28 days. When appropriate, respondents are requested to provide a frequency count. In a recent study by Giovazolias et al. [31], the validity of the Greek version of EDE-Q was investigated. The authors concluded that the results supported both the internal consistency, as well as the concurrent, convergent and discriminant validity of the EDE-Q global scale and its subscales. For the subjective patient experience of therapy and recovery, the following questionnaires were used: 1. Criteria on Eating Disorders Treatment Questionnaire (CEDT): To assess the quality of treatment from the patients' perspective, the CEDT was developed by the authors. The CEDT was based on the third part of the "Questionnaire for Eating Problems and Treatment" developed by de la Rie et al. [18]. Participants were asked to rate 70 items on a 5-point Likert type scale to assess the importance of each item for their treatment. The 70 items covered three major domains: treatment content, professionals involved in the treatment and the mental health facility that provides the treatment. Lastly, of the 70 items on the list, participants were asked to rank the 10 most important to them.
An exploratory principal component analysis (PCA) with varimax rotation was carried out to identify relevant factors of the 70 items regarding treatment. The results indicated that CEDT consists of a 6-factor structure: acceptance and bond with therapist, mastery and eating behaviors, treatment modalities, inpatient treatment, therapy structure and focus on underlying problems. Higher scores indicate higher importance of the factor. Cronbach's alphas were calculated to determine the reliability of the six identified factors, and the psychometric properties of the six scales were satisfactory (Cronbach's αs ranging from 0.782 to 0.848).
2. Criteria for Recovery from Eating Disorders Questionnaire. Noordenbos and Seubring [17] developed the Criteria for Recovery from Eating Disorders questionnaire. For the purpose of measuring the patients' experience of recovery, they created a list of the core characteristics and consequences of EDs based on the literature and on the criteria for recovery mentioned in effect and follow-up studies. They categorized these characteristics and consequences into groups representing somatic, behavioral, psychological, emotional and social factors. This procedure resulted in a list of 52 recovery criteria: 9 items related to eating behavior, 5 items concerning body attitude, 16 items on physical recovery, 8 items on psychological well-being, 9 items on emotional state and 5 items on social adjustment. Each item is rated on a 5-point Likert type scale, with higher scores indicating better feeling of recovery. Psychometric properties of the Greek version of the questionnaire with 6 subscales were satisfactory (Cronbach's αs ranging from 0.613 to 0.885).
Finally, additional information on sociodemographic characteristics, treatment history, weight history, hospitalization and medication history was obtained through a relevant questionnaire.

Participants
Of the 65 patients who were approached and were eligible to participate in the study, one refused to participate, possibly due to paranoid personality traits that made her oversuspicious about the "underlying true aim" of the study. From the 64 remaining patients, 42 completed the 6-month period of therapy (dropout rate: 34.4%). Forty of the completers were female and two were male. Regarding diagnosis, 17 were suffering from anorexia nervosa (AN), 18 from bulimia nervosa (BN) and seven from OSFED. The mean age of the participants was 26.7 (ranging from 17 to 47 years old). The mean age of ED onset was 18.5 (ranging from 7 to 31 years old). The mean duration of previous treatments was 3 years and 1 month (ranging from 1 month to 16 years). Half of the participants were university students (50%).

Statistical analysis
The mean scores of the answers to the CEDT and the Criteria for Recovery from Eating Disorders questionnaire were used as a way of ranking the most important factors according to the patients. The EDE-Q 6-month and WHOQOL-BREF initial and 6-month measurements followed a normal distribution while the EDE-Q initial measurement did not. For this reason, a paired sample t-test was used only for the comparison of the WHOQOL-BREF measurements; the Wilcoxon signed-rank test was selected to assess improvement in ED symptomatology measured by EDE-Q before treatment and after 6 months.
A one-way ANOVA was performed to assess possible differences between the three ED diagnostic groups. Finally, Pearson's bivariate correlations were performed to assess possible correlations between objective and subjective perceptions of ED symptomatology improvement. Table 1 summarizes the standardized mean scores of all CEDT factors for each of the three ED diagnostic groups. All the factors were found to contribute to the quality of treatment, having high mean scores, except for the "inpatient treatment" subscale. This was probably due to the fact that most of the participants have never been hospitalized in a mental health facility. "Acceptance and bond with therapist," "therapy structure and information" and "mastery and eating behaviors" had the highest mean scores. This result indicated that participants believed that therapeutic alliance, structured psychotherapy, and focus on how to gain control over ED symptomatology were vastly contributing to their therapy successful outcome. These three subscales were followed by "treatment modalities" and "focus on underlying problems" in terms of highest mean scores. As mentioned previously, the least identified factor was "inpatient treatment." A one-way ANOVA to compare differences between AN, BN and OSFED groups did not show any significant differences between diagnostic groups with the CEDT subscales ( Table 1).

Patients' view on the quality of treatment
A great variety were found in the ranking of individual CEDT treatment factors. The most often mentioned factor was "focus on the transition back to normal life." This was mentioned by 35.9% of the participants. Also, "focus on self-esteem" was mentioned by 31.1%, "focus on underlying problems" by 21.5% and "trust in therapist" by 21.4%. The factors that were not selected at all as important by any of the participants were the following: "being able to talk about religion," "being able to tell your story," "therapist with enough time," "therapist with a sense of humor," "the location for treatment should be easily reached," "explanation about somatic complaints and consequences of the eating disorder," "receiving standardized treatment," "companion as tutor/counselor," "role-playing as part of treatment" and " talking in groups." A Pearson's bivariate correlation analysis between CEDT and the 6-month EDE-Q subscales showed that "acceptance and bond with T = therapist" was positively correlated only with patients' BMI (r = 0.4, p = 0.09), "therapy structure" with EDE-Q's weight (r = 0.45, p = 0.03) and shape concerns (r = 0.4, p = 0.01) and "inpatient treatment" with EDE-Q's eating (r = 0.34, p = 0.03), weight (r = 0.4, p = 0.01) and shape concerns (r = 0.45, p = 0.03).

Objective and subjective recovery
A Wilcoxon signed-rank test indicated that there were significant differences in the ED symptomatology scores in both the overall and subscale scores of the EDE-Q questionnaire, showing important improvement over the 6 months of treatment ( Table 2). A paired sample t-test was conducted to compare QoL before and after 6 months of treatment. Significant improvement of QoL was found only in the general health scores (t = 3.7, p = 0.001) and the psychological domain scores (t = 4.5, p = 0.0001). Regarding the patients' view on the criteria of recovery, "physical recovery" had the highest mean score (mean = 48.1), followed by "emotional recovery" (mean = 31.3), "psychological recovery" (mean = 30.1) and "social recovery" (mean = 29.5). The least recognized areas of improvement were "eating behavior recovery" (mean = 28.6) and "body image recovery" (mean = 14.6). Again, a one-way ANOVA to compare differences between AN, BN and OSFED groups did not show any significant differences between diagnostic groups in their perception of recovery ( Pearson's bivariate correlation coefficient was computed to assess the relationship between the objective symptomatology improvement, as was measured by the EDE-Q, and the subjective improvement, as was rated by the patients (Table 4). Overall, there was a strong correlation between the ED symptomatology and the patients' subjective experience of psychological recovery. The feelings of improvement in self-esteem and body image, as well as reduction in self-judgment and negative feelings, were correlated with ED psychopathology. Finally, Pearson's bivariate correlation coefficient was also computed to assess the relationship between the WHOQOL-BREF QoL improvement and the subjective improvement as was rated by the patients ( Table 5). The stronger correlations were between objective social recovery and the WHOQL-BREF social relationships and overall recovery constructs and WHOQOL-BREF's psychological domain. Social recovery, emotional recovery, psychological recovery and overall recovery showed higher correlations with the improvement in the patients' QoL.

EDE-Q
Eating disorder recovery criteria Eating concern Restraint Shape concern Weight concern Total EDE-Q

Discussion
The first main aim of this study was to explore factors that are regarded by ED's patients as important elements of successful treatment. ED's patients stress the importance of feeling accepted and having a good relationship with their therapist. The importance of the therapeutic alliance in relation to therapy outcome is supported also by other studies [18]. Having a good relationship with the therapist and a feeling of acceptance in therapy were also found to correlate with normal BMI. Moreover, therapy structure and focus on ED recovery were considered very important by the patients. This could refect the fact that when applying for therapy, adult ED patients have often reached a point when ED symptomatology has become highly egodystonic and disruptive for their everyday life; as such, the patients are eager to experience a reduction in the severity of their symptomatology as soon as possible [34].
The second aim was to investigate the factors that the patients regard as important for their recovery. The recovery criteria that were mostly recognized by the patients were physical, emotional and psychological. From the patients' view, it seems that the improvement in ED symptomatology is mostly attributed to their perceptions of psychological recovery. This finding is supported by previous studies, which have shown that a medical, iatrogenic approach is not found helpful by patients [35].
The third aim of the study was to explore whether patients' subjective experiences of recovery were in accordance with objective improvements in ED symptomatology, as well as improvement in their QoL. More specifically, their perception of improvement in the psychological, emotional and social domains was highly correlated with improvement in the corresponding fields in their QoL, as well as in their overall QoL. This finding might suggest that when ED therapy focuses on improving the individual's psychological, emotional and social condition, this could result in reduction in ED symptomatology and QoL improvement or vice versa.
The results indicate that while there was an overall improvement in ED symptomatology during the first 6 months of treatment, some of the QoL domains did not show significant improvement. This could be explained by the short-time period between the two measurements, especially when considering that the mean duration of illness was 8.2 years.
Another point that needs to be taken into consideration is the discrepancy that might exist between the therapist's opinion and the psychometric measures that are used during the course of ED treatment [36]. It has been demonstrated that the "opinion" of the therapeutic team about the patients' treatment course has to be considered along different psychological, psychopathological and eating-related variables. Although for about 90% of the patients, there is a convergent between the therapist's clinical assessment of the therapy's outcome and the assessment by questionnaire, 10% of patients can be missclassified [36].
Finally, it was interesting that there was not any difference between the diagnostic subgroups of ED patients in regard to factors that were important to them for their therapy and recovery. This finding is in line with the notion that in the area of EDs, we could apply a transdiagnostic treatment model, as AN, BN and atypical EDs share common psychopathological mechanisms and can respond to a "universal" therapeutic process [23,37].

Conclusions
"Listen to what your patient is trying to tell you" is a phrase that is very often repeated in psychotherapy training and supervision meetings. Although to listen to your patient is common ground in psychotherapy, it is a novel area of research in the field of EDs. Overall, four major conclusions came out of the literature review and the study's results. First, most of the patients appreciate a structured therapy in a specialized unit for EDs with a caring therapist who can form a strong therapeutic alliance with them. Second, most of the patients regard recovery as a change that goes beyond ED symptomatology and into a more "holistic" improvement in their psychological, physical and social well-being. Third, the patients' opinion on therapy and recovery is not influenced by the diagnosis of the ED that they are suffering from. Finally, when compared, the opinion of the patient and the evaluation from the therapist or the appropriate clinical research measurements have more similarities than differences.