Using an Ecological Momentary Assessment (EMA) Smartphone Application in the Treatment of Binge Eating Disorder: A Proof-of-Concept Study

1. Introduction

Binge eating disorder (BED) is the most common eating disorder, with detrimental effects on individual’s health and well-being, as it may lead to obesity, depressive symptoms and substance abuse [1]. Treatment of BED mostly consists of elements of cognitive behaviour therapy (CBT) [2]. Group-based or individual CBT is effective in alleviating BED symptoms, with abstinence rates after one year varying between 46-52% for psychotherapy [2]. However, several individuals only reach partial remission and may face relapses at the longer term. This may be due to a host of different factors, including underlying alexithymia and problems with emotion regulation [3,4], and treatments that more effectively target maintenance mechanisms of BED are needed [5,6].

Beyond other aetiological factors, mood changes and emotion regulation processes seem to play a key role in binge eating among most individuals with BED [3,7,8,9]. The affect regulation model of binge eating posits that binge eating plays a role in the regulation of one’s negative affect, suggesting that binge eating episodes are preceded by a deterioration in affect or stress, while mood tends to improve after such an episode [7]. Many studies have tested this model utilizing Ecological Momentary Assessments (EMA) to capture momentary ratings and repeated assessments of emotions and activities in daily life. A recent comprehensive meta-analysis integrated results from 59 EMA studies, generally supporting the notion that deteriorations in emotions precede and thus seem to trigger binge eating [9]. Yet, evidence for the emotional consequences of binge eating depended on the type of analyses: the level of negative affect was generally worse at the first post-binge rating, yet temporal trajectories – as based on multilevel modelling in more recent studies – suggest slight improvements after binge eating [9]. Although these findings highlight the role of (fluctuations in) emotions, only few studies examined the clinical utility of EMA in BED treatment.

Le Grange et al. (2002) conducted a pilot study in which CBT treatment with and without EMA was compared among 41 individuals with BED [10]. EMA was implemented in the first two weeks of treatment using a wristwatch that beeped several times per day and a pre-printed diary. After each week, participants discussed the pattern of their binge eating episodes with the therapist. In 77% of these individuals (compared to 47% of the control group), the binge eating frequency reduced by half, yet findings were not statistically significant [10]. Another pilot study among 16 individuals with BED or Bulimia Nervosa used an EMA smartphone application as an add-on during an emotion-focused CBT [11]. The app registered daily emotions as cues for disordered eating, and alongside, delivered customized coping strategies during moments of risk. Participants found the app acceptable as treatment augmentation and indicated that the app helped accomplish therapy goals, yet paradoxically, compliance (daily registrations) was too low to trigger intervention strategies [11]. A few other studies investigated the use of ecological momentary interventions as add-on to CBT treatment, encompassing direct delivery of therapeutic support via smartphone applications, sometimes based oneself-initiated self-monitoring [12,13]. For these studies, it is, however, difficult to determine the role of EMA itself.

Taken together, EMA is a promising augmentation to current eating disorder therapies, because monitoring of affect and binge eating may provide insight in individual symptom patterns that can be used to tailor treatment [14]. Yet, as initial evidence suggests possible issues with compliance, it is necessary to value individuals’ and therapists’ experiences to define feasibility and acceptability [11], before EMA is incorporated as a tool during treatment. The current proof-of-concept study evaluated the usefulness of an EMA app for BED treatment through qualitative interviews with participants and therapists.

2. Materials and Methods

2.1. Design and Participants

This study was conducted at the Eating Disorder outpatient clinic of the Parnassia Groep, a mental health institute in the Netherlands. All individuals diagnosed with BED who enrolled in CBT between August 1st, 2021 and August 1st, 2022 (n=35, five treatment groups) were asked to use the EMA smartphone application as part of treatment. Twenty individuals consented for their EMA data to be analysed in the study, of whom 16 also consented to participate in a qualitative, semi-structured interview about their experiences with the app. Of the 8 therapists and treatment coordinators (referred to as ‘practitioners’), 5 consented to participate in an interview.

2.2. Procedure

Participants enrolled in treatment as usual, a group-based CBT for BED consisting of 20 (bi-)weekly sessions [15]. In the first session, an involved researcher from the Parnassia Groep introduced the EMA app and the study, and handed out patient information and an informed consent form. Participants could consent for use of their EMA data in our study. Additionally, we asked individuals and practitioners for their consent to participate in an interview on the usability of the app. In the second session, consent forms were collected. Then, both consenting and non-consenting individuals were asked to download the app, as it was part of the treatment protocol for everyone. Participants received personal login codes and instructions to download the app. During weeks 2-7, participants received daily invitations to fill out micro-questionnaires on emotions and binge eating in daily life. The reports of each participant were automatically included in a personal graph in their app to offer insight into their pattern of emotions and binges. An example of a fictitious personal graph is shown in the Supplementary Materials (Figure S1). The graph was discussed in weekly sessions with the practitioners. After week 7, interviews with participants and practitioners took place. The semi-structured one-on-one interviews on user experiences (feasibility, usefulness and added value) lasted 30-45 minutes, were conducted by trained master students in Clinical Psychology and were audiotaped and transcribed by them afterwards.

2.3. Measures

2.3.1. General Characteristics

Information on sex and age was retrieved from patient records. At intake, highest obtained educational level was asked, and height and weight were measured. Symptom severity pre- and post-treatment was self-reported on the validated Eating Disorder Examination Questionnaire (EDE-Q)[16,17].

2.3.2. EMA Measures

A smartphone application for EMA was developed in cooperation with tech-company Innovattic. Both developers and researchers tested the app multiple times before deploying it in the intervention. Security requirements were tested (including a penetration test by an external company) and approved by the security officer of Erasmus University Rotterdam. The app, called a Daily Diary, was installed by participants on their smartphones and prompted participants, via text message or notification, for data entry for six consecutive weeks at five semi-random intervals each day: 7–8.30 am, 10–12 am, 2–4 pm, 5–7 pm and 8–9 pm on weekdays and 9–11 am, 12–2 pm, 3–5 pm, 6–7.30 pm and 8.30–10.00 pm during weekends. Each sampling point could be filled in for one hour, with participants receiving a reminder prompt after 30 minutes, if needed. In total, there were 210 sampling points. Each sampling contained questions with predefined answering categories to enable rapid responses. Two questions asked where the participant was at that moment (at home, work, school, someone else’s home, outside, public entertainment space, sports club, on the road, somewhere else), and with whom (alone, partner, child(ren), friend(s), colleagues/classmates, acquaintance, parents, brother/sister, strangers, others). These questions were categorized as being at home versus elsewhere, and being alone versus not alone. Then questions asked about current emotions (‘At the moment, I feel…’), including nine different emotions each to be answered on a 7-point Likert scale ('not at all’ to 'very strongly'), from which sum scores on negative affect (insecure, sad, angry, anxious, lonely, worried and ashamed) and positive affect (happy, relaxed) were calculated. These items were based on the PANAS scale [18] and in consultation with people with lived experiences. Finally, participants were asked whether they had eaten much more than a regular meal or snack since the last assessment, using a 5-point Likert scale ('not at all’ to 'very much'). If answered with somewhat/quite a lot/very much, they received a follow up question on whether they had experienced the feeling that it was hard to stop eating (LOC, Loss Of Control eating). Besides these fixed sampling points, participants were able to report a binge eating episode if one occurred, using the same questions on overeating and LOC. The occurrence of overeating in combination with the LOC was defined as a binge eating episode.

2.3.3. Semi-Structured Interview

A semi-structured interview was held with participants and practitioners to evaluate their experiences with the EMA application during BED treatment. The interview schedule was specifically designed for this study and covered the themes: burden of symptoms (participants only), experience of treatment as usual (practitioners only), impact of EMA application on change in symptom burden and treatment course, and recommendations. The interview schedule for participants and practitioners are presented in the supplement (Figure S2 and S3).

2.4. Analyses

2.4.1. Qualitative Analysis of the Interviews

Interviews were transcribed verbatim and analysed with thematic analyses [19] in Atlas.ti 22 for windows. Deductive data-driven thematic analysis was conducted as follows: first, during the open coding, relevant sentences from the transcripts were reduced to codes showing the sentence’s core. Next, axial coding was used to compare codes with each other. Similar codes were merged into sub-themes. In the last step (selective coding), sub-themes were assigned to broader predefined themes relating to feasibility (compliance, user-friendliness, burden, and readability and comprehensibility of the graph) and usefulness of the app (new insights based on graph and reporting procedure, as well as contribution to reduce BED symptoms). We conducted open data-driven coding to ensure a data-driven and inclusive account of experiences. The deductive fitting of the open/axial coding to literature-derived main categories ‘feasibility and usefulness’ was performed at the end of the analyses (selective coding). The transcribing and coding were conducted by two junior research assistants, and theme formation was discussed in recurrent sessions with the juniors, a senior qualitative researcher and a senior binge eating researchers.

2.4.2. Quantitative EMA Analyses

To further explore the usability and validity of the EMA app for BED treatment, the daily registration data was analysed quantitatively to estimate bi-directional associations of affect and situational factors with overeating using autoregressive models [20,21,22]. The quantitative analyses were performed among those who 1) reported overeating and 2) had a response rate >30% (n=8). Due to the low compliance rate, results should be interpreted with caution and are therefore presented in the supplement (see Supplementary Materials: Figure S4, Table S1-S3).

3. Results

3.1. Characteristics

Characteristics of the total sample (n=20) and sample for qualitative analyses (n=16) are provided in Table 1. The two samples were quite comparable. The total sample (n=20) consisted of mostly females (75%) with a mean age of 34.95 years (SD=10.32) and a mean body mass index of 40.13 kg/m² (SD=8.11). On average, participants scored lower on the EDE-Q post- vs. pre-treatment (2.56 versus 3.38). The mean EMA response rate for the total study sample was 31.1% and ranged from 0.9% to 80.0%. Nine participants had a response rate >30% (>65 of 210 EMA reports were filled in; mean response rate of 56.4%).

3.2. Qualitative Analysis of the Interviews

3.2.1. Feasibility

Feasibility is defined as the merits and viability of the EMA app, which was covered in five themes in the interviews (see Table 2). Below, themes are denoted in italics, and subthemes in bold text. Quotes are denoted as Px for patient with anonymized participant code x, as Ty for the practitioners (therapists and treatment coordinators).

3.2.2. Compliance with EMA App

Compliance regarding the procedure was noted by participants indicating a few times that they used the app as much as possible. However, participants more frequently noted that they often lacked the opportunity to fill in the app, e.g., because they were traveling (P12). Some participants mentioned that they stopped using the app early, e.g. ‘I think I really tried for 1.5 weeks’ (P5). Within the treatment groups, it was experienced as hindering that not all individuals used the app or quit prematurely which affected the value of the app as part of group therapy. The practitioners noticed that group/social dynamics affected individual experiences, e.g., ‘in a negative group, more people tended to decline’ (T4) and that ‘some people joined out of loyalty with the therapist’ (T4). Practitioners also indicated that age might have affected participation, with the used technology being more appealing for younger participants. However, this was not reflected in the low response rates among the young adults (26 or younger) who indicated that they had difficulty incorporating time for registration in their daily schedules or who indicated that emotions were not relevant for their problems.

Compliance also regarded the content of the reporting. Several participants filled in emotions truthfully and without much thought, yet they differed substantially in registering binge eating. Some registered their binge eating episodes, while others did not for various reasons: they doubted whether the eating should be called a binge, did not want to report the binge, or were too preoccupied with the binge.

3.2.3. User-Friendliness of the EMA App

Despite a high user-friendliness of the questionnaires (e.g., clear questions that can be answered quickly), compliance may have been jeopardized by practical issues and some technical problems. Practical issues prohibiting a response included not having the smartphone nearby the whole day, the short time window of questionnaire availability, and inconvenient timing of the questionnaires. Participants suggested to improve the app by allowing to set own start and end time, and to broaden the time window for response. Technical issues regarded mostly the notifications, with some participants indicating they did not receive all notifications. However, it could also be that participants sometimes missed notifications. Practitioners echoed these comments, but ‘advised clients not to give up too quickly’ (T3). Note that over the course of the study, we changed the notification system, resulting in a steadier flow of notifications. Practitioners additionally noted that the app makes registration ‘easier’ and ‘better readable’.

The relevance of the questions also appeared as a user-friendliness subtheme. Some participants indicated that sufficient emotions were assessed, while others suggested to include options to personalize emotions and circumstances. The latter was also noted by one participant who thought the app focused too much on emotions and should focus more on circumstances.

3.2.4. Burden of the EMA App

The practical burden of the assessment raised contrasting experiences. Some participants experienced the app as positive and did not regard the reminders as distracting, while most participants noted that they got too many reminders, that only for some were distracting. Also, most participants indicated that the assessment period was too long, with the app becoming annoying. Practitioners agreed with these comments.

The app also presented an emotional burden, as for some participants, the registration of feelings was confronting and evoked shame. Indeed, practitioners warned that ‘the graph can cause a downward spiral if it does not go well’ (T5), although ‘it also has potential to increase awareness of emotions, which is often difficult for this patient population and thus can be a good thing’ (T5). Indeed, some participants indicated that the registration of feelings was fine because it resulted in self-insight and sometimes even brought relief, e.g., ‘it was sometimes nice to express that you aren’t feeling happy. And then afterwards it’ll be okay again’ (P9).

Unexpectedly, the app appeared to have an unintended negative side effect: participants indicated it felt unpleasant or even like a failure when they missed an assessment round. This aligns with the practitioners’ thoughts that some participants ‘filled in the app out of perfectionism’ (T1), a personality trait that is linked with binge eating [23].

3.2.5. Readability and Comprehensibility of the Graph

Generally, participants were positive about the appearance and readability of the graph (e.g., ‘graphs were clear’ (P7), ‘because of the colours I thought it was clear’ (P8)). Further, they recognized the goal of the graph. Yet, both participants and practitioners indicated that reading and interpreting the graphical patterns of (many) different emotions over the course of multiple days or weeks was difficult. However, the practitioners emphasized that particularly ‘assessments over a longer trajectory would say more’ (T2).

Understanding the graph was a challenge for some, sometimes resulting in giving up (e.g., ‘So I noticed that I didn’t look at [it] anymore’ (P10)). For others, the pattern was understandable and recognizable, but they did appreciate help from the therapist to achieve this, which was confirmed by practitioners. However, practitioners also pointed at the limited time they had to discuss the app during the sessions. Participants recommended a better embedding of the app in the treatment sessions to allow for more discussion and explanation of the registration.

3.3. Usefulness

Usefulness is defined as the EMA app being advantageous and helpful. Usefulness was covered by three themes in the interviews (see table 2). Below, themes are denoted in italics, and subthemes in bold text.

3.3.1. Gained New Insight Based on the Graph and Reporting

Most participants frequently indicated that the reporting and graph increased awareness of and insight in (the role of) their emotions. It was described by participants as a ‘check-in moment’ (P15), that ‘the graph helped to realize that ‘I’m less happy than I thought I was’ (P4) and that ‘I noticed that [insecurity] came up often’ (P13). This is notable considering the low response of several participants: apparently, few registrations may already increase insight. Only few participants reported that they did not gain new insights from reporting emotions or reading the graph, e.g., because they deemed the tracking of emotions as ‘a bit irrelevant’ as the participant described him/herself as ‘not an emotional person’ (P2). Practitioners added that the app may ‘induce a moment for reflection’ (T5) and ‘unravels subconscious emotions’ (T4), since it asks questions in ‘a real-life setting’ (T4) where participants have to report in-the-moment instead of retrospectively.

3.3.2. Whether the EMA App Helped to Reduce BED Symptoms

Only few participants indicated that the app helped in reducing BED symptoms, with one participant stating that recognizing the role of emotions in the occurrence of binges helped him/her to prevent binge eating. However, others mostly attributed changes in symptoms to treatment rather than to the app. Despite this, the potential added value of the app was experienced as positively by participants and practitioners. Some suggested that the effect may have been larger when the app would be used later in or after treatment, rather than at the start. Practitioners also recommended to ‘use the app as aftercare’ (T4) although they would ‘advise against using app as stand-alone treatment’ (T5).

4. Discussion

This proof-of-concept study examined the value of an EMA smartphone application as tool during CBT treatment for BED. Participants and practitioners indicated mixed experiences. Although the app was received as a user-friendly tool that increased awareness and insight in emotions for some, participants also perceived the reporting procedure as burdensome because of the many registrations. Moreover, reporting emotions and binge eating episodes was considered confronting and shameful for some participants. Together, this may explain the low number of registrations in the EMA application, and as a result, the quantitative analyses were likely underpowered and should be interpreted with caution.

4.1. Interpretation of Findings

Because of a low number of registrations and very low occurrences of binge eating episodes, our quantitative findings may have limited relevance compared to our qualitative findings, yet in spirit of full disclosure we report them in the supplementary material. Our modelling of bidirectional associations suggested that negative affect was not linked with later overeating, while in turn, higher levels of overeating were associated with increased negative affect one timepoint later. The latter is in line with findings of a recent review of multiple EMA studies regarding negative affect directly following a binge eating episode [9]. Nevertheless, it should be noted that our design differed substantially from most of the studies included in the review. Previous studies mainly conducted EMA assessments in non-treatment samples or prior to the start of treatment, while our EMA assessment took place during treatment. CBT treatment for BED often has a high initial treatment response in which the number of binge episodes rapidly declines [24]. When overeating does occur during treatment, individuals may experience particularly heightened negative emotions – such as guilt or shame – stemming from increased awareness, a pattern that emerged in our data despite limited statistical power. Importantly, this rapid decline in binge eating episodes may also explain the very low number of reported binge eating episodes in our sample. Interestingly, several ecological momentary interventions (EMIs) have now been developed and implemented either as stand-alone treatments or as additions to standard care [12,13]. Delivered through smartphone apps, these interventions rely on self-monitoring – including assessments of affect, eating disorder symptoms and food intake – and activate built-in support modules when needed. The resulting self-monitoring data offers a valuable opportunity to build on our preliminary findings and assess whether and how the affect regulation model shifts over the course of treatment. To our knowledge, this has not been examined before.

Compliance rates in previous EMA research among individuals with an eating disorder are generally higher than the 31% in our study. Of 36 studies reviewed by Borm et al., 2025 [9] that relied on comparable signal-contingent reporting, 24 studies had compliance rates between 53-97%, while 12 studies did not report on compliance. Differences in design between our and previous studies may explain the diverging compliance rates: nearly all studies were conducted pre-treatment, lasted for 7-14 days and offered substantial monetary rewards [9]. Perhaps a more comparable design was an EMI study in which participants were asked, among other tasks, to self-monitor their affect four times per day. The moderate initial compliance of 58% in the first two weeks deteriorated over the course of 19 weeks of treatment to 11% responding at least once a day in the last two weeks [11]. This may reflect tapering-off self-monitoring as treatment neared its end or the growing familiarity with emotions and contexts linked to binge eating that reduces the perceived need for continued monitoring but may also reflect the possibility that only a few weeks of reporting – rather than extended monitoring periods – are sufficient.

Low compliance among participants with BED was previously shown to be linked with individual and contextual factors such as lower positive affect, but not with participants’ demographic characteristics [25]. This was also reflected in our qualitative findings, which suggested that the low compliance may be due to the practical (e.g., high number of registrations) and emotional burden of the EMA app (e.g., due to increased emotional confrontation). Indeed, individuals with BED can experience difficulties in their emotional awareness (e.g., alexithymia) alongside difficulties in emotion regulation [3,4], but the confrontation seemed to help some increase their emotional awareness. Considering the low number of registrations, this increased awareness may emerge with far fewer registrations than we initially expected, suggesting a relatively straightforward strategy to enhance the overall treatment effect. Yet, an unintended side-effect from the use of the EMA app was a feeling of self-disapproval when missing a registration. Individuals with BED often show higher levels of perfectionism, which may drive fears about negative social evaluations and being afraid to make mistakes [23]. Signs of this measurement loyalty may help therapists to recognize perfectionism in their patients, which could be targeted during treatment. In conclusion, although the low compliance rates were detrimental for conducting robust analyses, they may not have undermined our primary goal: supporting the treatment process.

4.2. Strengths and Limitations

The strength of this study was the in-depth, qualitative overview on the applicability of a smartphone application during treatment, which helped to interpret the quantitative findings and highlighted the contrasting experiences between users. However, the study’s findings should also be interpreted considering some limitations. A first limitation was that, although the designed application was tested many times before the start of the study, notifications were not always received by all participants in the first two treatment groups due to technical issues. These issues influenced compliance rates and reduced generalizability. Indeed, the mean response rate in the first two treatment groups was 21.0% and after changing the method of notifications for the last three treatment groups, the mean response rate was 33.8%. However, it is not possible to determine to what extent this difference was due to technical problems or whether the app became more embedded in treatment. Secondly, our findings might also be influenced by underreporting of binge eating in the EMA app. Although this could reflect an initial treatment response, it may also be the result of socially desirable reporting, as some participants indicated in the interview that they did not report all binge eating episodes. Because of the low occurrence of binge eating episodes, and as LOC was registered conditional on the registration of overeating, we could only analyse overeating. Thirdly, over the course of the study, there was a high turnover of therapists because of the COVID-19 aftermath, illness, or transfers. This might have influenced the engagement of therapists in discussing the app during sessions, but at the same time also reflects commonly occurring treatment circumstances.

4.3. Recommendations

The added value of this proof-of-concept study lies in the qualitative evaluation of the implementation of an EMA smartphone application during BED treatment. Overall, the app was considered a user-friendly tool by most participants, while several participants also indicated the registrations were helpful in increasing awareness of (fluctuations in) emotions. This pattern was also observed in other qualitative EMA studies among BED patients [11], but also in adolescents in the general population [26], highlighting the broader potential of EMA to enhance awareness and serve as a valuable tool both within and beyond the clinical treatment context. Drawing on our real-world example of implementing an EMA application, we outline several recommendations to enhance the feasibility and added value of EMA during treatment for BED, as well as potentially for other psychiatric conditions.

Firstly, the assessment period could be shortened. Several participants indicated that the reporting enhanced awareness of and insight in their emotions, which apparently, considering the general low response of participants, may already be reached with a few registrations. So, in contrast to EMA research, daily registrations for multiple weeks and strict compliance may be less important for individualized treatment goals.

Secondly, the implementation may benefit from better alignment of the assessment period with the content and course of treatment. For example, a one-week assessment at the start of treatment may help patients focus on real-time emotional experiences and increase emotional awareness, whereas a one-week assessment later during treatment aligns with the timing of the emotion-regulation module of CBT, which targets the connection between emotions and binge eating symptoms [27]. This way, the practical burden is reduced while still obtaining valuable insights in emotions and the link with binge eating.

Thirdly, the EMA application should be well-embedded in treatment. To effectively and meaningfully incorporate EMA or any other tool into clinical practice (research), it is recommended to offer therapists the relevant comprehensive knowledge on the tool’s theoretical foundations and scientific evidence regarding therapeutic advantages. While scientific evidence is still developing, real-time monitoring of emotions, contexts and behaviours aligns well with the transdiagnostic model of eating disorders [27] and can be meaningfully integrated within this framework making use of modern technology. Next to that, therapists should have sufficient allocated time to discuss registrations during treatment sessions. Such prerequisites will help to avoid uneven utilization of treatment tools by practitioners.

Fourthly, given the rapid advancements in digital health technologies over recent years, it is further recommended that clinicians and researchers leverage existing m-health platforms, such as Avicenna and M-path, to implement daily self-monitoring with greater ease and efficiency in their target groups. These platforms were still in their infancy when we started developing our application, but provide sound digital infrastructures nowadays that enhance the feasibility, scalability, and ecological validity of EMA protocols, while also facilitating data management and integration into therapeutic workflows. Indeed, clinician dashboards allow therapists to easily create and adjust EMA schedules, including the ability to tailor items to individual patients – an important feature considering evidence that the emotions that are most relevant for binge eating vary from person to person [28]. Although simply registering emotions and overeating already appeared to heighten participants’ awareness, graphical data visualizations may offer additional support for some individuals, and the dashboards of m-health platforms make it possible to visually review emerging patterns during sessions. Existing m-health platforms are widely applicable across all major operating smartphone systems and are continuously updated, ensuring stable performance and alignment with evolving technological and clinical needs. As such, the use of existing m-health platforms will likely be less costly, testing will take less time, and technical issues are likely to remain limited, resulting in smoother and more effective implementation.

Finally, as not all participants were keen on using the smartphone app, we recommend offering the EMA application as an add-on tool for personalized treatment rather than incorporating it as an integral part of BED treatment. Potentially, personalized content in the EMA application, such as choosing relevant emotions and situations, may increase compliance. Given that this study is one of the first that examined the feasibility of an EMA application during BED treatment, more research is needed to replicate and extend our findings while incorporating these recommendations. It would be particularly helpful to evaluate and compare effectiveness of adding EMA or EMIs to CBT, as well as their potential for preventive efforts and for bridging the waiting-list period before treatment.