Developing a web-based add-on intervention for parents of adolescents with anorexia nervosa: a mixed methods preliminary study

Abstract

Background

The aim of this study was to develop a web-based psycho-educational intervention for parents of adolescents with anorexia nervosa and carry out a mixed method evaluation of its acceptability and feasibility.

Methods

After developing the intervention, we recruited parents from 19 families of adolescents with AN engaged in family-based therapy at an outpatient specialist center for eating disorders. Participants were given 6-month access to the intervention. Acceptability, feasibility, parental well-being and mental health, and family functioning were collected at baseline and again three months after baseline.

Results

Among the 11 parents from families who participated and accessed the intervention, the majority rated the intervention as satisfying (median = 6, IQR = 5–7), and easy to use (median = 7, IQR = 6.5–8.3). Frequency of use was about once a month. No significant differences between baseline and follow-up were found for parental well-being and mental health. We observed an increase in parental self-efficacy with a large effect size (p = 0.07, rs = -0.52), suggesting a potentially meaningful association. Parents identified positive aspects of the intervention as ease of access, comprehensive content and a positive impact on their approach to their child’s illness.

Conclusions

This study suggests that an add-on digital intervention to family-based treatment for parents of adolescents with AN is acceptable and feasible. Satisfaction, ease of navigation, and finding answers to questions were positively rated. Web-based or computerized psychoeducation for parents can be a valuable addition to treatment for adolescent AN.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s40337-026-01570-5.

Plain language summary

Anorexia nervosa (AN) is a serious eating disorder that affects not only the young person but also impacts on the family. Parents play a key role in supporting their child's recovery, yet they often lack accessible resources to help them understand and manage the challenges they face. This study aimed to develop an online psychoeducational intervention for parents of adolescents with AN and to assess its acceptability and feasibility.

We recruited parents from 19 families whose children were receiving specialist outpatient family-based therapy for AN. Parents were given six-month access to the intervention. Acceptability, feasibility, parental wellbeing, mental health, and family functioning were measured at baseline and three months later.

Most parents found the intervention satisfying and easy to use, valuing its ease of access, breadth of content, and positive impact on how they understood their child's illness. We also observed a potentially meaningful improvement in parental confidence, although this did not reach statistical significance, likely due to the small sample size.

This study suggests that an online educational intervention for parents of adolescents with AN is acceptable and feasible as an addition to family-based treatment. Digital psychoeducation for parents could be a valuable addition to treatment for adolescent AN.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s40337-026-01570-5.

Background

Anorexia Nervosa (AN) is a mental health disorder characterized by eating restriction and food avoidance, leading to abnormally and dangerously low weight, that commonly starts around puberty and affects 2% or more of adolescents [1–3]. AN is a deadly disorder, it is among the psychiatric disorders with the highest mortality, and represents the leading cause of chronic morbidity in young women [4, 5]. The course of AN can be protracted unless effective treatment is given in adolescence [6]. The prognosis of AN is better during adolescence than in adulthood, with around 50–60% of adolescents fully recovering from AN, and around 30% partially recovering at one-year post-treatment [7, 8]. Good outcomes have been associated with shorter duration of illness, early intervention, early weight gain, and family warmth and support [9–11].

Family-based treatment specifically developed for AN in adolescents, known as Family-based treatment (FBT) or Maudsley-based family therapy (MB-FT), has been shown to be effective across several countries [7, 12]. These approaches support the idea that the family is the best resource to help restore healthy weight and attain recovery in adolescents with AN. Several studies have shown that including families in treatment within the first three years of the illness produces better outcomes for the patient at one and five years [13, 14]. However, managing a child with AN can be difficult, leading to higher levels of depression and anxiety in caregivers, as well as higher levels of expressed emotion (e.g., criticism, hostility, overprotection), which contributes to the persistence of AN [15, 16]. Standard MB-FT reduces parental expressed emotion, which has a positive impact on the child [12, 17, 18]. However, parents value the support from professionals beyond the once-weekly treatment sessions [19, 20]. Therefore, adjunct interventions for caregivers of individuals with AN can be beneficial [21, 22].

Such interventions have shown several benefits, including potential improvements in child outcomes, mainly by increasing caregivers’ self-efficacy and reducing their distress and burden in taking care of a child with AN [23–28]. Consistent with this idea, a recent study has shown that parental self-efficacy can be a predictor of weight gain in adolescents with AN [29]. These findings suggest that one way of improving the effectiveness of standard FBT could be add-on interventions aimed primarily at reducing carer anxiety/depression and increasing carer self-efficacy. Given the sharp rise in the use of digital technologies and telehealth in mental health care, particularly post-COVID, digital or eHealth interventions adjunct to standard therapy might be a promising alternative [30–34]. One such example is ‘SUCCEAT’, a structured 8-week program delivered via online modules in which carers access one new module weekly, complete interactive exercises, and report on their experiences [35].

The primary aim of this study was to develop an online intervention for parents of adolescents suffering from AN, and evaluate its acceptability and feasibility using quantitative and qualitative methodologies. Secondary outcomes included parental well-being and mental health, family functioning, and parental efficacy (pre-post intervention). Below, we briefly describe the intervention development and the pilot evaluation of this intervention.

Methods

Participants

We recruited consecutive families whose adolescents with AN were in treatment at AliNEA, the outpatient unit for eating disorders of the Geneva University Hospitals (HUG), Switzerland. Recruitment was carried out between January and October 2022. We included families of adolescents with an active AN diagnosis (according to DSM-5) [36], who had access to the Internet (computer or smartphone), French-speaking, and were receiving standard outpatient treatment in the service, which consisted of Maudsley-Based Family Therapy (MBTF). The treatment consists of weekly sessions with a therapist and a medical check in (weekly during the initial phase of treatment, and biweekly/monthly in Phase 2 and 3), the average duration of treatment in the service is between 6 and 9 months. Both parents are invited to weekly sessions, however attendance varies. Eighteen mothers and 15 fathers (19 families) were included in the study after providing informed consent.

Intervention development

A web-based psycho-education tool called PARENTS (Parents of Adolescents suffering with AN focused on: Resources, Empathy, Nutrition, Techniques, and Support) was developed by a team of clinicians and researchers (ET, HS, FE and NM) and a carer (EM). The development process was carried out in 4 steps (see Fig. 1).

Fig. 1.

Four key phases of intervention development: (1) focus groups with 13 parents to identify core themes; (2) development of content (text, images, videos, audio testimonies); (3) creation of the digital platform; and (4) refinement and review prior to online deployment

Step 1: focus groups

We conducted two initial focus groups, involving 5 parents in the first session and 8 in the second, totaling 13 parents of adolescents with AN. The sessions were facilitated by two psychologists, following a topic guide; the focus of the topic guide was to identify parental understanding and needs in relation to: (a) how best to help their child, (b) looking after themselves, and (c) general questions about any needs they might have in relation to caring for their child and any resources available to them. Four main topics and sub-themes were identified and are summarized in Table 1.

Table 1.

Summary of topics and sub-themes included in the intervention

Main topics	Sub-themes
A) Adolescence and AN	• Basics of adolescence • Psychoeducation regarding AN   o What is it?   o What’s going on in my child’s head and body   o Effects and treatment of undernutrition   o Psychological and physical impacts of AN
B) Managing AN at home	• Meal management   o Child’s emotions and behaviors at the table • Strategies at mealtime • Roles and support of parents during mealtime • Management of other eating disorder symptoms • Management of other difficult times (outside mealtime)
C) Nutrition	• Description of a balanced diet   o Why it is good for you • Refeeding and changes
D) Parental well-being/taking care of yourself	• Ways and strategies of dealing with adversity • How to cultivate what sustains you • Self-compassion • Distraction • Soak up the good • Zen enough or 5% better • Other tools for resilience and well-being • Mistakes, blame and acceptance

Step 2: content development

The content was developed by the study team and an external collaborator- a carer (EM). Media such as images, videos, and audio testimonies were also produced by the study team.

Step 3: digital development

The digital platform was developed using a hierarchical tree structure to ease navigation and adapted to be accessed on many supports (computers, tablet PCs, smartphones). Internet security requirements of the University Hospitals of Geneva were adhered to. Access was restricted, and each user had a personal identifier and a password.

Step 4: Refinement and final content review

The content and digital platform went through a process of refinement and review by the study team and external collaborator before we finalized the format and content of the intervention before online deployment. The intervention was delivered through a secure online platform that provided participants with unrestricted access to psychoeducational content. Access to the platform was restricted to authorized participants via a protected network. Participants were able to navigate freely and select topics of interest. The platform followed a hierarchical structure, organized into four top-level thematic areas, each further divided into multiple sub-sections (see Table 1). Educational materials were presented in various formats, including written descriptions, informational videos, images, and audio testimonies. Participants were encouraged to navigate the platform freely and explore topics of personal interest. Therapists responsible for the follow-up of participating families were advised to promote the use of the intervention.

Measures

Socio-demographic data were collected from case-notes. Parents completed a range of measures.

Acceptability and feasibility

Acceptability and feasibility were assessed using an ad-hoc Likert-scale questionnaire including evaluation of: (a) satisfaction with the intervention (from 0 = ”not at all satisfied” to 10 = “Extremely satisfied”), (b) ease of use of the online platform (from 0=“Very difficult” to 10=“Very easy”), (c) the ability to find answers to questions (from 0 = “not at all” to 10 = “All of my questions”), and (d) frequency of use of the intervention (ranging from 0 = “Never”, 5 = “About once a month, to 10 = “Every day”). Number of connections per user (derived directly from the intervention platform) was also used as a measure of feasibility.

Parental well-being and mental health

Parental well-being and mental health were assessed with the General Health Questionnaire (GHQ-28) [37], a 28-item questionnaire designed to assess different aspects of mental health. It consists of four sub-scales that measure health (somatic symptoms, anxiety and insomnia, social dysfunction, and depression), each scale consists of 7 questions. GHQ-28 scoring methods are based on the 4-point Likert scale, and a lower score indicates a better mental state. Psychometric studies have shown its strong validity, fidelity, and sensitivity [37]. Furthermore, we assessed parents’ mental health using the Depression, Stress, and Anxiety Scale – short version (DASS-21). The DASS-21 is a self-report questionnaire with 21 items with a 4-point Likert scale to assess symptoms of depression, anxiety, and stress. Cut-off scores are used to categorize symptom severity into conventional levels such as normal, moderate, and severe [38].

Family functioning

Family functioning was assessed using the Family Assessment Device (FAD) [39]. The FAD is a self-report questionnaire with 60 items to assess aspects of family structure, organization, and interaction. The FAD is scored on a 4-point Likert scale, where higher scores equal a greater impairment in family functioning. The measure yields six specific subscales: Problem Solving (quality and directness of problem-solving strategies), Communication (clarity and directness of verbal information exchanged), Roles (clarity and appropriateness of role distinctions), Affective Responsiveness (openness and appropriateness of emotional information), Affective Involvement (clarity and appropriateness of emotional involvement with other family members), and Behavior Control (clarity and appropriateness of rules). The FAD also includes a distinct General Functioning subscale that includes items related to the other subscales and assesses the overall health/impairment within the family. Cut-off scores are available to distinguish between clinically significant and non-clinical presentations [39].

Family Attitude Scale

In addition, we assessed parents’ attitudes towards their child with the Family Attitude Scale (FAS), a questionnaire with 30 items rated on a 5-point Likert scale measuring critical or hostile attitudes and behaviors towards another family member, with higher scores reflecting greater levels of criticism, hostility, and emotional over-involvement [40].

Parent versus Anorexia

Finally, we used the 7-item Parents versus Anorexia (PvA) questionnaire, rated on a 5-point Likert scale, to evaluate parental efficacy in parents of children with ED, higher scores indicate greater parental efficacy. This questionnaire measures the impact of new therapeutic strategies used in the Maudsley model of family-based treatment [41].

Among the questionnaires used in our study, the GHQ-28, DASS-21, FAD, and FAS were validated in French [42–45]. The PvA was translated into French by a native speaker, followed by back translation and verification by two independent researchers (AAJ, ET) (See Supplementary Material 1).

The GHQ-28, DASS-21, FAD, FAS, and PvA questionnaires were administered to parents at baseline (T0) and after 3 months (T1). Acceptability and feasibility were only assessed at T1.

Procedures

After the initial assessment in the outpatient unit, during which patients received an AN diagnosis, parents were informed about the project. Subsequently, following the first therapy session, parents were asked to provide informed consent.

Parents of adolescents completed T0 questionnaires, and unique logins were assigned. After 3 months, parents were required to fill out the T1 questionnaires. Additionally, they were invited to participate in a qualitative interview to share their experiences of the intervention. Interviews were conducted via an online provider (Zoom) by a researcher involved in the study (AAJ).

All questionnaires were completed electronically using REDCap (REDCap version 13.6.1). Figure 2 includes a study flowchart.

Fig. 2.

CONSORT diagram showing the number of families at each study phase: (1) assessed for eligibility; (2) allocated to the intervention; (3) completed follow-up; and (4) included in data analysis

Data analysis

Quantitative

Descriptive statistics were carried out for baseline variables and our primary outcomes, acceptability and feasibility (Median, IQR, and Range). Secondary outcomes were analyzed using Wilcoxon signed-rank tests comparing each variable of interest between T0 and T1. We also calculated effect sizes.

Spearman’s Rank-Order Correlations were performed to determine the correlation between the number of connections to the web-based intervention and parental mental well-being measured at T1.

Adolescents who were included in the study and were briefly transferred to in-patient refeeding during standard outpatient treatment (N = 3) are included in the analyses. All analyses were performed in IBM SPSS Statistics (Version 26).

Qualitative

Semi-structured interviews were performed with parents at the end of the study via video-conference. Each session lasted approximately 30 min and included open-ended questions [see Supplementary Material 2]. The qualitative data obtained were analyzed using thematic analysis, following the six-phase approach outlined by Braun and Clarke: (1) familiarisation through repeated reading and note-taking; (2) systematic coding of meaningful data features; (3) grouping codes into initial themes; (4) reviewing and refining themes for coherence; (5) defining and naming themes; and (6) producing a final report with illustrative extracts and analytic interpretation [46]. In addition, two validation checks comparing coding decisions to assess consistency were performed after phases 3 and 4 by two independent coders (AAJ and FW), with FW serving as the second independent coder.

Results

Out of twenty-two families who were approached, 19 families (N_mother = 18, N_father = 15) consented and completed the baseline questionnaires at T0. After 3 months (T1), 15 families (N_mother = 14, N_father = 3) filled out the follow-up questionnaires. The majority of mothers completed the study, while fathers had a larger dropout rate (N null= 12 (80%)). Participants who did not complete T1 measures (N = 15) were excluded from comparative analyses.

Main outcomes

After reviewing usage data, 4 families (N_mother = 4, N_father = 1) were removed from the analyses as they never logged onto the web-based intervention. The final sample for analysis consisted of 11 families (N_mother = 10, N_father = 2). Socio-demographic details are described in Table 3.

Table 3.

Socio-demographic and other characteristics of the participants included in the full analyses (N = 11 families)

Individual-level variables	N	Percent	Median (IQR)
Parent participating
Mother	10	83.3	–
With ED history	3	30.0	–
Father	2	16.7	–
With ED history	0	0
Age in years			50 (48–51)
Marital status
Married	6	50.0	–
Divorced	4	33.3	–
Unmarried	1	8.3	–
Widowed	1	8.3	–
Employment
Tertiary sector	10	83.33	–
Unemployed	1	8.33	–
Unknown	1	8.33	–
Secondary or primary sector	0	0	–
Adolescents
Sex
Female	10	90.1	–
Male	1	9.1	–
Age in years	–	–	15 (14–16)
BMI	–	–	15.9 (14.9–16.6)
Duration of illness (in months)	–	–	18 (16–22)
Hospitalized during treatment	3	27.3	–

Feasibility and acceptability

Of the 22 families approached, 19 agreed to participate (86%). At T1, 4 of the 19 families were lost to follow-up, leaving 15 families. Among these 15 families, 4 were excluded for not accessing the intervention, resulting in 11 families (73%) who accessed the intervention. In relation to acceptability at follow-up (T1), parents rated their satisfaction with the PARENTS intervention as satisfying (median = 6, IQR = 5–7). Concerning the ease of use of the intervention, this was rated as rather easy (median = 7, IQR = 6.5–8.3). Parents’ ability to find answers to questions through the intervention was rated as high (median = 6, IQR = 3-7.3) and the average frequency of use of the intervention, as estimated by the parents, was rated as “once or twice” (median = 3, IQR = 3–5) during the study period. In addition, measures of the number of connections per user showed that parents accessed the PARENTS intervention 3–4 times within 3 months (median = 3.5, IQR = 1–7, range = 1–11). Detailed results are reported in Table 4.

Table 4.

Acceptability, parental mental health, and well-being prior and after use of PARENTS (N_mother = 10, N_father = 2) (Wilcoxon signed-rank tests and effect sizes)

Variables	Baseline	Follow-up
	Median (IQR)	Median (IQR)	z	p	r
Acceptability
Satisfaction		6.00 (5.00–7.00)
Ease of navigation		7.00 (6.50–8.25)
Find answer to questions		6.00 (3.00-7.25)
Frequency of use		3.00 (3.00–5.00)
Number of accesses to the online platform		3.5 (1.00–7.00)
DASS21
Depression	13.00 (8.50–20.00)	15.00 (6.50–29.50)	− 0.410	0.682	-0.12
Anxiety	5.00 (0.05-9.50)	4.00 (0.00-11.50)	–1.081	0.280	-0.31
Stress	18.00 (16.00–20.00)	13.00(10.00-27.50)	− 0.276	0.782	-0.08
GHQ-28
Global score	25.50 (19.00-33.50)	32.00 (13.25–43.25)	− 0.864	0.387	-0.25
FAD
General functioning	1.83 (1.60–2.23)	1.88 (1.60–2.29)	− 0.119	0.906	-0.03
FAS
Global score	87.00 (71.75–96.75)	87.50 (75.50–103.00)	− 0.628	0.530	-0.18
PvA
Total score	19.50 (17.50-21.75)	20.50 (19.25–24.50)	-1.785	0.074	-0.52

For ease of interpretation, correlation coefficients are categorized as follows: r = 0.1 represents a small effect, r = 0.3 signifies a moderate effect, and r > = 0.5 denotes a large effect

Secondary outcomes

No significant changes between baseline and follow-up were found for DASS-21, GHQ, FAD, and FAS scores. The PvA showed a non-significant increase with a large effect size (p = 0.074, rs = -0.52). Details are reported in Table 4.

Correlations between actual use of the intervention and outcomes

The number of accesses to the online platform per user demonstrated a positive correlation with parental efficacy at follow-up (r = 0.47, p = 0.059), indicating a non-significant but potentially meaningful association. Detailed results are reported in Table 5.

Table 5.

Correlations between the number of connections per user and mental health and well-being among parents (N = 11 families) (Spearman’s Rank-Order Correlation)

Variables	Number of connection/per users	DASS21 score depression	DASS21 score anxiety	DASS21 score stress	GHQ global score	FAD general functioning	FAS score	PvA score
Connections per users	1	0.424*	0.352	0.336	0.023	0.210	−0 0.098	0.466*

** p < 0.05, * p < 0.1

Qualitative interviews

All families included in the study were invited to participate in the interviews. Of the 11 families in the final sample, four consented to take part, resulting in four mothers participating in separate individual sessions. Thematic analysis of qualitative interviews revealed three main themes: Content, Usability, and Therapeutic benefits.

Content

The first theme extracted from the interviews relates to the content of the intervention. All the interviews highlighted positive aspects regarding the content. The latter was described as ‘thorough and useful’, particularly chapters on Developing your own resilience. Parents identified useful coping strategies in the intervention and successfully applied them with their child: “I don’t know, the strategies of simply, for example, hugging her to reassure her, rather than trying to lecture her or telling her for the fifteenth time why it’s good for her to eat that. Finally… We really found the good stuff. And then the resilience side, it’s true that it was especially good. As parents, it was really good to be there […]” (Interview 3).

Parents felt confident while reading the content since they knew that it was created and adapted by health professionals specializing in the ED field. They described how the addition of different content formats (text, images, audio clips, and videos) felt refreshing. Furthermore, they mentioned that having access to information in writing outside of the therapeutic sessions was very helpful since they felt therapy can be at times emotionally overwhelming: “In any case, I thought it was important to have very clear information. Because perhaps when you receive all this information orally during the treatment, which is well done, but you’re so emotionally involved that maybe you can’t register everything. And it’s written out quite completely. It seemed very clear to me. It’s true that it allows you to have just a little distance.” (Interview 3).

Negative aspects regarding the content were also mentioned. Some sections, such as Adolescent development or Nutrition, were less used by participants: “I used a little less what was on: “Why do we have to eat?“. Really, all that was basic explanations of nutrition… because I already knew that. And the development [chapter on adolescence] all that, that I flew over it. To find out if there was anything else I could learn now. And then everything that was really psychological and psychosomatic, all of that.” (Interview 1).

Parents described finding some information in the intervention similar to what was discussed in therapy sessions. For example, parents thought that the content lacked specific strategies. Finally, participants highlighted the lack of explicit information about what could be shared with their child as problematic: “Apart from the videos I think, because Marine [therapist’s alias] said we could watch them together, I didn’t know what we could watch together. So I think it would be good to indicate “Advise to watch with your child” or to read with your child if there are things like that, why not.” (Interview 1).

In terms of suggestions, parents would appreciate more content in formats other than text, such as audio testimonials from other parents. In addition, preambles summarizing the audio or video content were suggested. Lastly, a chapter regarding relapse risk prevention could be useful: “What did I look for… When there were relapses of crisis in fact. At one point in the beginning, when Marie [patient’s alias] started to eat, it was fine. And then, there were once or twice when all of a sudden it just didn’t work. It [the situation] was blocked and it’s true that there I got a little… I went to look twice on the site to see if I could find some solutions. And it was pretty general. It wasn’t sharp enough for what I was looking for […] She was in crisis. Suddenly, she no longer wanted to eat. She was crying over her plate. It was bad. That’s what, it was especially these crises.” (Interview 1).

Usability

The second theme extracted was regarding usability. The ease of use of the intervention was unanimously reported by participants. Navigation was considered easy, even for individuals who are less familiar with technology. The design with the use of pastel colors and dinner scenery as the home page worked well for parents. Accessibility at any time/place and on many supports (computers, tablet PCs, smartphones) was appreciated: “It’s very comfortable on the computer, because the screen is so big. I used it a lot on smartphones, especially on public transport or when I was in the waiting room. But it’s more difficult to scroll, because the text didn’t display too well on the screen, even if I turned the screen and it wasn’t practical to find the line you’d stopped at before. It was more complicated on the smartphone, yeah. But I used it a lot [on the smartphone].” (Interview 1).

In fact, parents mentioned using the intervention like a “refresher course” or “toolbox” whenever they felt like it. Finally, having the choice between different types of content (text, images, audio clips, and videos) was very welcomed. Availability of content in a written, video, and audio format was viewed very positively by parents: “[…] it could be like a kind of toolbox you have if you suddenly have a question, if you want to remember something. Knowing that you’re never safe from the possibility of something happening again. We hope not, but…” (Interview 3).

Limited-time access to the intervention was mentioned as a barrier. Parents described a wish for non-limited time access: for example, before (while they are on the waiting list), during the time between assessment and the start of treatment, for a longer duration (we restricted use to 6 months due to the nature of the study), and even after treatment: “Just before, actually [before being in treatment]. Because you know, when there were problems with Marie [patient’s alias], you know, there was a waiting list. In other words, we didn’t have a place at first. And I’m thinking that maybe if someone had given me the site at that time… It would have saved me having to go and look at fifty thousand disaster scenarios on the web. And then maybe I could have calmed down a bit better: “Okay, now you calm down and read this.” Maybe there could be some advice before the first sessions.” (Interview 2).

Additionally, parents mentioned some difficulties in navigation, i.e., not having a bookmark or placeholder to remind themselves where they had left off, or finding specific sections, which could be tedious and time-consuming: “On this platform, I would have liked to have a box where I could ask a question or put in a word. Marie [patient’s alias] didn’t experience any vomiting, but she could type “vomiting” and then have a whole series of things appear on the screen. Marie [patient’s alias] would rip her cheeks out with her fingernails. She was physically attacking herself because of the anxiety. I wish I’d known, like, if I typed in “anguish” something like that, and it opened up a couple of subjects like that […] Yeah, so if we’ve got time we’ll have a look at it first. But it’s true that the two times I found myself on the site I was in a moment of crisis and I thought well… But I wasn’t sure. I didn’t want to look for everything. In fact, if I’d had that search box…” (Interview 2).

Finally, suggestions were made to improve usability, such as adding a search bar function, or having more reminders by e-mail or from their therapist to use the intervention: “And it’s true that at certain moments, the therapist can remind you of certain elements that may be on the platform. Perhaps consider consulting the platform to review that specific video. It could be interesting; maybe you’d have a different perspective on it.” (Interview 3).

Therapeutic benefits of the intervention

The last theme extracted from the interviews was regarding the therapeutic benefits of the intervention. The intervention seemed to be an important tool for the parents on an emotional level. Parents often described feeling scared and anxious because their child is suffering from AN. They commented on how the intervention attenuated some of the negative feelings they had and gave them a sense of calm: “[…] I found it [a sense of appeasement], but it was really useful to have it written down and then to be able to read it, I honestly found that it soothed me a lot […] To read things that were quite difficult, but at the same time it was soothing to be able to step back a bit. […] And I must say that it was very calming to have this resource…” (Interview 3). Parents highlighted having doubts and uncertainties and were looking for answers to their questions: “Really for me, I found everything on the site useful. Because like I said, I’m looking for answers. I’m looking for questions that go round in my head. The answer is there [in the intervention]. Especially as it’s a situation we can’t talk about with everyone.” (Interview 3).

Furthermore, the intervention helped overcome other negative feelings, such as isolation: “But when I got to the site and started reading, I started looking at the symptoms and everything. I discovered names [meant: the illness name] and I was telling myself that there are others too. That means I’m not alone. […] It’s true, really, thank you for the parents who’ve shared, it’s very important. At first, I refused. I was ashamed, I don’t know how to explain it. […] But now I can talk to everyone.” (Interview 4).

Many parents described that they were looking for answers during treatment, which they found stressful at times. The intervention supported them by providing them with information that they could rely on in times of need. They were able to find answers and strategies to face numerous situations: “I’ve been trying to reflect, as it’s been a while. What stood out to me the most was using strategies, especially during meals or when there were tearful moments. Figuring out which words to use to reassure the child— that was very helpful.” (Interview 3). Finally, every parent highlighted benefits from the digital intervention, saying they would recommend its use to other parents facing a similar situation: “I would recommend it to everyone who has doubts. And I really mean it, of course, yes” (Interview 1).

Discussion

The aim of this study was to develop an online add-on psychoeducation intervention for parents of adolescents with AN and to evaluate its acceptability and feasibility. A high proportion of invited families (86%) expressed willingness to participate, suggesting initial interest and perceived relevance of the intervention. Overall, the findings suggest that the intervention was feasible in terms of recruitment, and the intervention was acceptable to participants, as reflected by good satisfaction and positive participant feedback. Although a difference was apparent between mothers and fathers, the intervention was generally well-received. This distinction raises important questions such as a sex-based distinction in parental involvement in children’s care. Systematic reviews support such sex differences in which mothers are generally more implicated in caregiving [47].

Among completers, more than half of the participants explored the content of the intervention and 67% accessed the online platform at least once a month. However, overall use was limited, suggesting that further consideration of strategies to enhance engagement may be warranted, such as incorporating periodic reminders or prompts to encourage more regular use and integrating the platform more explicitly into the clinical workflow to reinforce its relevance. Five parents did not access the intervention at all, possibly due to a perceived lack of need for additional support, practical constraints such as insufficient time, emotional overload related to caregiving, or a preference for direct clinician support over digital materials. Identifying these barriers can help inform future adaptations aimed at improving accessibility and uptake. Parents rated navigating the content of the intervention as rather easy. Ease of navigation has been shown as an important aspect of usability and a key factor of success for digital interventions [48]. Parents’ reported ratings of frequency of use of the intervention were lower than we expected. During qualitative interviews, parents described different strategies in accessing the online content: some browsed the content in its entirety to gather as much information as possible about AN, while others accessed the material only when they felt they needed it. Furthermore, parents mentioned that the use of reminders (by email or by the therapist) could help them improve use of the intervention. Such help may be necessary since motivation can be a significant barrier to the utilization of a digital intervention [49]. Moreover, literature supports that prompts can improve engagement in digital interventions [50, 51].

We showed no significant changes in secondary outcomes, i.e. parental mental health and well-being between baseline and follow-up. Interestingly, no significant changes were observed in this measure, despite families concurrently participating in Maudsley-Based Family Therapy (MBTF) during the same period. However, results showed a small increase in parents’ self-efficacy in taking care of their child suffering with AN after 3 months, although this change did not reach statistical significance. This difference is substantiated by a positive correlation between the number of connections to the digital platform and self-efficacy. Studies show that higher self-parental efficacy might be a key factor in positive outcomes of children suffering with AN [52, 53].

In terms of parental experiences, three themes were extracted from qualitative interviews: content, usability, and therapeutic benefits of the intervention. The content is a crucial part of any psychoeducational intervention. One of the barriers to treatment can be the lack of information among caregivers, for which psychoeducation can be useful [22]. Parents of children suffering from AN may have many doubts and questions when faced with such a complex illness. Parents mentioned in the interviews and the acceptability and feasibility questionnaires that the intervention helped them answer many of those questions. In addition, the diversity in content format (text, images, audio clips, and videos) could have enhanced satisfaction and engagement with the intervention [54]. An add-on intervention like ours, which includes psychoeducation, can help families get a better understanding of the illness. Usability is another key component of a digital intervention. Users highlighted the ease of use of the intervention and how access from multiple platforms was helpful. Adding a search bar to find topics would enhance the usability of the digital intervention. Parents observed many positive changes following the use of the intervention and mentioned that the limited-time access was restrictive and highlight a wish to be able to access the online content whilst waiting for treatment to start or after treatment. Finally, parents’ experience highlighted the therapeutic benefits of the intervention; psychological impacts such as reducing anxiety and providing a sense of relief were mentioned. Furthermore, the intervention seemed to address parental feelings of isolation, loneliness experienced by caregivers of individuals with mental illness is well-recognized in the literature [55]. Other benefits witnessed by parents were improved communication and interaction with their child thanks to guidelines and strategies found in the intervention. They noted they were able to use the intervention in times of need and for specific situations (e.g., how to handle a mealtime).

Strengths and limitations

Our study needs to be understood in the context of relevant limitations, among these the fact that this was a naturalistic study, therefore our sample size was small. However, given that we aimed to test the acceptability and feasibility of the use of the platform, we were able to achieve good access to the intervention and a relatively high completion rate. However, unbalanced increments of these acceptability scales limit the interpretation of results, particularly when data gathers around the middle of the scale. This reduces the scale potential to discriminate the responses and could lead to an underestimation of the results. Mothers were more likely to complete participation, as is often the case in treatment for adolescent mental health problems [47]. Else, it might reflect a culture-specific issue in southern European countries. Important strengths to highlight are the high rate of families who agreed to participate in the study (86%), and the high levels of completers (73%). Outcomes of the study reflect a positive attitude towards a digital intervention, which does not add a burden to caregivers and can be accessed at any time.

Conclusion

In conclusion, this study suggests an add-on digital intervention to FBT for parents of adolescents with AN is acceptable and feasible. Acceptability and feasibility of the digital intervention were generally positive; satisfaction with the intervention was supported in both quantitative and qualitative data. Ease of navigation and finding answers to questions were positively rated. However, the frequency of use was low. Despite the small sample size, we observed a non-significant increase in parents’ self-efficacy following the intervention, with a large effect size. This effect may have been due to the concurrent family therapy received during the same period, however we did not see any increase in other measures of parental well-being or family functioning, as one would expect secondary to family therapy for AN. Literature regarding the benefits of digital interventions in adolescent eating disorders remains limited [56]. Lock et al. (2021) demonstrated the potential of an online guided self-help version of family-based treatment, reporting medium to large effect sizes across several outcomes, including parental self-efficacy, with improvements sustained at a 3-month follow-up. In line with these findings, our study provides preliminary evidence that such interventions may offer meaningful support to families of children with eating disorders. Further studies using a randomized design are needed to better ascertain the independent impact of this type of intervention.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AN

Anorexia nervosa

ED

Eating disorder

GHQ-28

General health questionnaire

DASS-21

Depression, stress, and anxiety scale – short version

FAD

Family assessment device

FAS

Family attitude scale

MBTF

Maudsley-based family therapy

PvA

Parents versus anorexia

Author contributions

NM, FE, HS and ET were involved in the conceptualization and study design. NM, HS, ET and CS prepared the educational material. AAJ, ET, FW and ABC participated in the data collection and analysis, and manuscript preparation. NM reviewed the content. All authors reviewed and approved the final manuscript.

Funding

Open access funding provided by Copenhagen University. This research was supported by Fondation Privée des Hôpitaux Universitaires de Genève (reference RP02-10).

Data availability

The data that support the findings of this study are available on request from the corresponding author.

Declarations

Ethics approval and consent to participate

All procedures complied with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. Ethics approval for this study was granted by the Cantonal Research Ethics Commission (CCER) of Geneva, Switzerland (Project-ID 2021 − 01663). Informed consent was obtained from all individuals participants included in the study.

Consent for publication

Written informed consent for publication was obtained from all individual participants included in the study.

Competing interests

The authors declare no competing interests.