National Clinical Programme for Eating Disorders: a pragmatic review of a new national eating disorder service in Ireland

Abstract

Objective:

The Health Service Executive National Clinical Programme for Eating Disorders (NCPED) launched a Model of Care for Eating Disorder Services in Ireland in 2018. Currently, one adult and two child and adolescent eating disorder services are operational out of a total of sixteen recommended. The three objectives of this paper are to describe the early (1) referral pattern, (2) level of service activity and (3) the level of service user satisfaction.

Method:

Monthly submitted service activity data from each service to the NCPED between March 2018 and October 2020 were retrospectively analysed. One hundred and fifty-nine carers and service users completed an experience of service questionnaire (ESQ). A descriptive analysis of referral pattern, level of service activity and ESQ was performed. A thematic analysis was performed on three qualitative questions on the ESQ.

Results:

There was substantial referral numbers to eating disorder services by 18 months (n = 258). The main referral source was community mental health teams. The majority (n = 222, 86%) of referrals were offered an assessment. The most common age profile was 10–17 years of age (n = 120, 54.1%), and anorexia nervosa was the most common disorder (n = 96, 43.2%). ESQ results demonstrate that most service users were satisfied with their service, and the main themes were carer involvement, staff expertise, therapeutic alliance and service access.

Conclusions:

This preliminary service activity and service user satisfaction data highlight several issues, including trends when setting up a regional eating disorder service, potential pitfalls of pragmatic data collection and the need for adequate information-technology infrastructure.

Introduction

Eating disorders (ED) are a group of recognised mental health conditions that include anorexia nervosa (AN), bulimia nervosa (BN), binge eating disorder (BED) and avoidant restrictive food intake disorder (ARFID) (American Psychiatric Association, 2013). Their aetiology is a complex mixture of biological, psychological and social factors, with heritability estimates in the range of 0.4–0.5 for AN and BN (Watson et al. 2019; Yao et al. 2021). Eating disorders have a lifetime prevalence of approximately 5% of the population, and the incidence in Ireland each year has been estimated at approximately 1800 cases (National-Working-Group-and-Clinical-Advisory-Group-HSE, 2018). This is expected to be higher following the coronavirus-19 (COVID-19) pandemic (Katzman, 2021).

Eating disorders impact on both mental and physical health, especially malnutrition in people with restrictive eating disorders such as AN and ARFID. Comorbidity, both mental and physical, is the highest seen in mental health services, and this includes both acute and long-term conditions such as osteoporosis, cardiac, gastrointestinal and fertility complications (Lucas et al. 1999; Pászthy, 2007; Halvorsen et al. 2012; Meczekalski et al. 2013; Hanachi et al. 2020; Auger et al. 2021). The mortality of AN is also the highest within mental health (Arcelus et al. 2011; Crow et al. 2012). In addition, there is the impact on psychosocial and brain development, given the relatively young age of onset. The outcomes from an eating disorder and recovery are improved by early access to assessment and evidence-based treatment (Bardone-Cone et al. 2019).

Traditionally, eating disorder care in Ireland has been provided through community mental health teams. In the context of the growing evidence base on early intervention and evidence-based treatment, the Health Service Executive (HSE) National Clinical Programme for Eating Disorders (NCPED) launched a national Model of Care for Eating Disorder Services in Ireland in 2018. In collaboration with the College of Psychiatrists of Ireland and Bodywhys (the national association for people affected by eating disorders in Ireland), the purpose is to improve the access, quality and value of eating disorder care in Ireland. The values of this model of care include developing a service that applies evidence-based practice, shared decision-making, effective communication and applies respect and dignity to all service users. Moreover, it is encapsulated to improving clinical outcomes with improved equity of access and an integrative care approach for service users suffering from an eating disorder (National-Working-Group-and-Clinical-Advisory-Group-HSE, 2018).

In developing services, providers must take into consideration four broad themes: the illness stage, illness severity, prior treatment history and motivation to receive further treatment (Currin and Schmidt, 2005; Lock et al. 2010; Peterson et al. 2016). As such, specialist outpatient treatment services for eating disorders have been demonstrated to be effective, timely and cost effective (Endacott et al. 2006; Painter et al. 2010).

The HSE Model of Care recommends a stepped model approach to eating disorder care which includes primary care and support services, out-patient, day-patient and in-patient programmes. The core group of clinicians recommended to provide a patient’s care as per the programme include a consultant psychiatrist/child psychiatrist, a therapist with eating disorder training, access to a dietitian and assistance with physical monitoring (e.g. general practitioner, physician or paediatrician).

The outpatient and day programmes will be provided by 16 regional specialist eating disorder teams – eight for adults and eight for children and adolescents nationally. The first regional eating disorder service for children and adolescents in Dublin (Linn Dara) commenced in March 2018 in CHO 7, the same year that, the St Vincent’s University Hospital (SVUH) adult team for CHO 6 was expanded in its outpatient service. A second community eating disorder service for children and adolescents, CAREDS Cork Kerry was established in mid-2019 for CHO 4. These three teams commenced incrementally in terms of posts and pathways. None is yet fully resourced in line with the Model of Care which in 2018, recommended a team of 14.4 whole time equivalents (WTE) for under 18’s and 13.4 WTEs for over 18’s in a sector population of 500 000–600 000. Moreover, this model of care will provide additional eating disorder services to those provided by existing community mental health teams in each area.

Within the NCPED, the interdisciplinary model of care supports the initiative that minimum psychological treatment competences of core multi-disciplinary team clinicians are required for each team. For an adult ED team, the relevant therapies include cognitive behavioural therapy for eating disorders (CBT-E), Maudsley model of anorexia nervosa treatment for adults (MANTRA), specialist supportive clinical management (SSCM), interpersonal therapy for bulimia nervosa (IPT-BN) and discipline appropriate skills in physical monitoring and observation (Kass et al. 2013; Ioannidis et al. 2020). For the under 18’s ED team, the relevant therapies are Family Based Treatment/Family Treatment for Anorexia Nervosa (FBT/FT-AN), Family Therapy for Bulimia Nervosa (FT-BN) and CBT-ED and Adolescent Focused Therapy (AFT). To support clinicians to achieve this level of expertise, there has been a significant programme of training, supervision and support within the programme.

Objectives

This paper describes the activity of the first three HSE Regional Specialist Eating Disorder Services over their initial 18 months. We hope that this will serve as a useful template for the future eating disorder teams and indeed for other new regional mental health services or programmes that may be established. Our three main objectives are:

1. 1. To describe the first 6, 12 and 18 months in terms of referral patterns and demand.

2. 2. To review the level of service activity of the first 18 months since each service commenced.

3. 3. To evaluate service user satisfaction with these services over the time period of the study.

Methods

Data collection

As part of the HSE NCPED evaluation process, each service submitted monthly service activity data on a proforma Excel file that was completed by each service’s clinical coordinator, or another clinician, and approved by the Consultant clinical lead on each team. For this level of evaluation, data collected were returned at a service rather than individual level. The first 18 months activity data of the three eating disorder services are included in this analysis, covering an 18-month period for each team between March 2018 and October 2020.

Service activity information

Referral information collected included: total number of first referrals, re-referrals, gender, referral source, referral outcome and consultation. Assessment information collected included age range, diagnostic group, accommodation, social and education, waiting time from referral to assessment, outcome of assessment, onward referrals, waiting time from assessment to first recommended therapy session and treatment modality. Discharge information collected included cases discharged and discharge destination.

Statistical analysis

Each monthly proforma Excel file was obtained with permission and collated into Excel. Data were validated and summarised to Excel. If a monthly variable cell was blank, it was assumed missing, or where appropriate, zero. The Excel spreadsheet was exported to STATA statistical package v17. Results are presented as total (n) number for each variable and as a percentage (%) of total referrals or total of referrals offered assessment. The difference between carer and young person responses and summed responses on ESQ were calculated using Wilcoxon Rank summed test, as the variables are ordinal, independent (carer vs child) and assumed non-parametric distribution. Statistical significance is reported as two-tailed, p < 0.05.

Experience of service questionnaire (ESQ)

The experience of service questionnaire (ESQ) is a short, free-to-use feedback tool developed for carers and young people by the Commission for Health improvement in the UK (Attride-Stirling, 2002; Brown et al. 2014). It is used by the NCPED as a measure of service user satisfaction at several time points during treatment: baseline or 6 weeks, 6 months and discharge. The ESQ has three versions with developmental adjustments: parent/carer, child version for children aged 9–11 years and the young adolescent aged 12–18 years of age. It consists of 12 statements of quality that are scored as a Likert scale between 0 (blank), 1 (“not true”), 2 (“partly true”) and 3 (“certainly true”). A total ESQ score represents the mean (standard deviation) of the summed scores of each participant. This provides information on respondents’ subjective satisfaction with the service. There are three qualitative questions: “What was really good about your care?”; “Was there anything you didn't like or anything that needs improving?” and “Is there anything else you want to tell us about the service you received?”

This analysis includes ESQ feedback from Linn Dara and CAREDS adolescent teams as it was not yet implemented fully in SVUH. The data are represented as a total (n) and percentage (%) of carers and their children who completed the ESQ for each question response. A thematic analysis was then performed on the responses of the three qualitative questions led by JF, with support from DOD and SMcD. This involved an inductive and semantic approach to identify responses by applying Braun and Clarke’s guidelines for qualitative analysis (Braun and Clarke, 2006).

Ethical approval

This is a retrospective service level activity evaluation. However, it also falls within the scope of local ethical approval under the Clinical Research Ethics Committee in Cork, Linn Dara, and the Clinical audit committee in St. Vincent’s University Hospital, Dublin. This was sought for each service as part of the wider ethical approval for clinical evaluation of the NCPED.

Results

Referral rate and service user demographics of the first 18 months service operating

The first 18 months of referral information were divided into three periods (first 6, 12, 18 months). Eighty-six percent (n = 222) of service users were female (Table 1). There were 242 new referrals and 16 re-referrals in an 18-month period (see Fig. A1 for total new referrals per consecutive month that the service was in operation). The child and adolescent mental health service was the main referral source with 42% of referrals (n = 109), followed by general practitioners (n = 90, 35%). A low percentage of referrals were received by liaison psychiatry (n = 6, 2.3%), emergency department (n = 6, 2.3%) and paediatrics (n = 6, 2.3%). Eighty-six percent (n = 222) of referrals were accepted for an assessment. Less than 11% (n = 27) of referrals were not accepted for an assessment.

Table 1. Total referral rate of the first 6, 12, 18 months to three HSE community eating disorder services in Ireland (N = 258)

a¹ – The number of service users at 18 months who were attending child and adolescent eating disorder service only. The remainder were attending the adult eating disorder service.

b² – The percentage of service users at 18 months who were attending child and adolescent eating disorder service only (a¹/18 month total for that variable * 100). The remainder percentage were attending the adult eating disorder service.

Abbreviations: GP, general practitioner; CAMHS, child and adolescent mental health service; CMHT, community mental health team; Tel, telephone; cons, consultation.

Profile of service user demographics following being offered assessment post referral of the first 18 months of three community eating disorder services in Ireland

The most common age profile was between 10 and 17 years of age (n = 120, 54.1%) (Table 2). Most service users lived with their family (n = 177, 79.7%) and most service users were actively in education (n = 139, 62.6%). Almost 5% (n = 10) of service users were unemployed. A small percentage of service users had current or past involvement with the child and family agency (n = 8, 3.6%).

Table 2. Characteristics of service user demographics following referral to three eating disorder services in Ireland (n = 222, total number of accepted referrals for assessment)

a¹ – The number of service users who were attending child and adolescent eating disorder service only. The remainder were attending the adult eating disorder service.

b² – The percentage of service users who were attending child and adolescent eating disorder service only (a¹/total for that variable * 100). The remainder percentage were attending the adult eating disorder service.

Abbreviations: ARFID, avoidant restrictive food intake disorder; OSFED, other specified feeding and eating disorder; UFED, unspecified feeding or eating disorder; ID, intellectual disability.

The most frequent eating disorder diagnosis was AN (n = 96, 43.2%), followed by BN (n = 23, 10.4%) and OSFED (n = 23, 10.4%) (Table 2). ARFID was the fourth most frequent eating disorder diagnosis (n = 13, 5.9%). Binge eating disorder was one of the least diagnosed eating disorders (n = 6, 2.7%). There was no pica or rumination disorder diagnosed. Fifty-four (24.3%) service users’ diagnosis data was missing. Sixty percent (n = 134) of service users had a comorbid psychiatric diagnosis. The three most frequent comorbid psychiatric diagnoses were anxiety disorder (n = 46, 20.7%), depressive disorder (n = 34, 15.3%), trauma related emotional difficulties (n = 13, 5.9%) and autism spectrum disorder (n = 17, 7.7%) (Table A1). Twenty-three (10.4%) service users had a co-morbid medical diagnosis. The most common comorbid medical diagnosis was a gastrointestinal disorder (n = 10, 4.5%). Almost 15% (n = 33) of service users had a history of past or current self-harm (Table A1). Only one (0.5%) service user had a borderline intellectual disability (Table A1). No service user had a mild, moderate or severe intellectual disability.

The service activity (waiting times, assessment outcomes, treatments offered) of the first 18 months of three community eating disorder services in Ireland

Thirty-seven (16.7%) service users were offered an assessment within seven days of the service accepting the referral (Table 3). The remaining service users were offered an assessment within four weeks (n = 106, 47.7%) or after four weeks (n = 59, 26.6%). Nine percent (n = 20) of information on service user waiting times from referral to assessment is missing. The most common assessment outcome was to open the case to the eating disorder team (n = 138, 62.2%). Only 20 (9%) service user cases were closed following assessment. Ten percent (n = 23) of service users care was in a shared care arrangement with other services. Almost 19% (n = 41) of assessment outcomes were missing. Seventy (31.5%) service users were provided treatment within seven days of assessment. Forty-seven service users (21.2%) were provided treatment within four weeks and the remaining service users (n = 132, 14.4%) received their first treatment after four weeks. Seventy-three (32.9%) service users’ data on waiting time from assessment to first treatment are missing. The most common first follow-up treatment offered was psychiatry review (n = 123, 55.4%) and dietetic review (n = 105, 47.3%). Other (psychological and non-psychological) (n = 68, 30.6%) treatments were the most common non-psychiatry and non-dietetic intervention. Fifty-three (23.9%) service users were offered FBT/FT and 22 (9.9%) were offered CBT-E as first follow-up. Fifty-one (23%) service users declined or did not take up the treatment option. Sixty-one percent (n = 136) of services users remained active clinical cases, while the remainder of service users were discharged to their General Practitioner (GP) (n = 54, 24.3%), community mental health team (n = 31, 14%) or other (n = 1, <1%). There was a total of 86 discharges in the first 18 months of service activity and most service users were discharged to their GP (n = 54, 62.7%), then their community mental health team (n = 31, 36%).

Table 3. The service activity (waiting times, assessment outcomes, treatments offered) of the first 18 months of three community eating disorder services in Ireland (n = 222, total number of accepted referrals for assessment)

a¹ – The number of service users who were attending child and adolescent eating disorder service only. The remainder were attending the adult eating disorder service.

b² – The percentage of service users who were attending child and adolescent eating disorder service only (a¹/total for that variable * 100). The remainder percentage were attending the adult eating disorder service.

Abbreviations: ED, eating disorder; FBT/FT, family-based therapy/family therapy; CBT-E, cognitive behavioural therapy for eating disorders; IPT, Interpersonal psychotherapy; SSCM, specialist supportive clinical management; AFT, adolescent-focused therapy; MANDRA, Maudsley model of anorexia nervosa treatment for adults; GP, general practitioner; CMHT/CAMHS, community mental health team/child and adolescent mental health services.

Service user satisfaction of two child and adolescent community eating disorder services in Ireland

There were 159 individuals who completed an ESQ, with a total of 184 ESQs completed across the various time points. More carers (n = 85) than young people (n = 74) completed questionnaires. Table A2 represents the demographics of carers and service users (young people) who completed the ESQ. Of the carers who completed an ESQ (n = 85), most of their children (service users) were female (n = 70, 83%), Irish Caucasian ethnicity (n = 82, 96.5%) and had a diagnosis of AN (n = 49, 57.6%). Of the young persons who completed an ESQ (n = 74), again the majority were female (n = 58, 78.4%), Irish Caucasian ethnicity (n = 70, 94.6%) and had a diagnosis of AN (n = 46, 62.2%). The most common age range was 15–18 years of age (n = 48, 56.4%).

Most carers within the first time point ESQ1 (n = 63, 95.5%) and second time point ESQ2 (n = 28, 90.3%) felt listened to by staff (Table 4). Carers for both time points (ESQ1:ESQ2) felt that the team was working well together to help their child (n = 65, 98.5%: n = 30, 96.8%), it was easy to talk (n = 63, 95.5%: n = 29, 93.6%) and that their views and worries were considered (n = 63, 95.5%: n = 29, 93.6%). One hundred percent (n = 66) of parents at the first time point would suggest the service to a friend. The lowest rated items for carers at both time points were convenience of timing of appointments (n = 47, 71.2%: n = 23, 74.2%) and convenience of location of appointment (n = 55, 83.3%: n = 24, 77.4%). The total mean scores (standard deviation) of both time points (ESQ1: ESQ2) for carers were 40.7 (1.8) and 40.5 (2.9), respectively.

Table 4. Results of Experience of Service Questionnaire of two child and adolescent eating disorder services (ESQ) (n = 159, no. participants) (no. questionnaires, 184)

^a ESQ1^, represents if a service user completed an experience service questionnaire between 6 weeks and 12 weeks of availing of the service.

^b ESQ2 represents if a service user completed a second experience service questionnaire at 6 months, 1 year or at the point of discharge.

^c Wilcoxon rank summed test. Two-tailed significance, significance is p < 0.05.

^d n represents the total number of service users (percentage of number of service users who completed questionnaire at that time point) and their carer who scored 3 which indicates they felt the statement is “certainly true”.

^e The total represents the mean (standard deviation) of the summed scores of each participant ESQ.

For young persons, 74% (n = 41) and 84% (n = 27) felt listened to by staff at the first time point (ESQ1) and second time point (ESQ2) respectively (Table 4). Less than 50% (n = 27) of young persons at ESQ1 and 62.5% (n = 20) of young persons at ESQ2 felt it was easy to talk to staff. Items rated most highly by young persons at both time points (ESQ1:ESQ2) were the team working well together (n = 47, 85.5%: n = 29, 90.6%), facilities (n = 47, 85.5%: n = 30, 93.8%), and feeling that they were treated well (n = 46, 83.6%: n = 30, 93.8%). The lowest rated item for young people at both time points was convenience of timing of appointments (n = 28, 50.9%: n = 18, 56.3%). The total mean scores (standard deviation) of both time points ESQ1 and ESQ2 for young persons were 30.8 (5.6) and 33.3 (3.3), respectively.

Carers were more likely to give a “certainly true” response at the first time point (ESQ1) than young persons for feeling listened to (z = 3.304, p < 0.001), easy to talk (z = 5.846, p < 0.001), treated well (z = 3.400, p < 0.001), working together (z = 2.730, p < 0.009) and to suggest the service to a friend (z = 4.672, p < 0.001), using Wilcoxon rank summed test. Carers were also more likely than young persons to give a “certainly true” response in describing the overall service experience (z = 4.343, p < 0.001) at the first time point (ESQ1). In contrast, young persons’ “certainly true” responses increased to a level at the second time point (ESQ2) that there was no statistically significant difference in “certainly true” responses between carers and young persons for feeling listened to by staff (z = 0.643, p = 0.744), treated well (z = −0.033, p = 1.000), staff working together (z = 0.993, p = 0.637) and being likely to suggest the service to a friend (z = 1.956, p = 0.099). There remained a statistically significant total mean score between carers and young person at ESQ1 (z = 9.346, p < 0.001) and ESQ2 (z = 6.343, p < 0.001) (i.e. overall satisfaction).

Thematic analysis of qualitative responses in experience of service questionnaire

The qualitative responses of the ESQ provided an opportunity to perform a qualitative assessment of the service factors valued by carers and children in recovery with an eating disorder (Fig. 1). Figure 1 represents a range of experiences, both negative and positive that were divided into themes from the qualitative questions “What was really good about your care?” and “Was there anything you didn't like or anything that needs improving?” Carer positive responses after analysis resulted in four overarching themes: therapeutic alliance, staff expertise, access to service and parental support. Each overarching theme had subthemes identified. Under the umbrella theme of therapeutic alliance, carers valued being listened to and understood, and that they perceived staff as caring, empathic and easy to talk to (e.g. “open, empathic and respectful of my daughters feelings”). Parents appreciated an open, patient centred approach, with the use of care planning and goal setting (e.g. “creation of goals, the way everyone was included”). Under the main theme of staff expertise, carers perceived staff as competent professionals and had an appreciation for the multidisciplinary team approach (e.g. “I felt that an expert was helping us”, “integrated team who all seemed to take an interest in my child and us as an individual family”). The theme of access to the service had the following subthemes: initial access to assessment and treatment, regularity of appointments and continuation of the service during COVID restrictions with blended telehealth and face to face approach (e.g. “seen so quickly and caught this problem early” and “excellent continuation of service during lock down”). Finally, carers felt they had carer support with the subthemes of carer empowerment through provision of psychoeducation and family therapy (e.g. “helped us understand her eating disorder and how to help us help her on a daily basis toward her recovery”). The young person’s positive responses resulted in three similar overarching themes: therapeutic alliance, access to service and competent support (e.g. “The people were very kind”, “They regularly met with us to keep track of progress”, and “It is really comfortable to talk to the staff as they know what I feel and are well capable of helping me”).

Fig. 1. Themes representing what was important to Service Users in two child and adolescent eating disorder services across both positive and negative experiences: thematic analysis of qualitative information from Experience of Service Questionnaire (ESQ). Young person’s thematic main themes. Abbreviations: MDT, multidisciplinary team; OT, occupational therapy; GP, general practitioner.

The carer negative responses resulted in the following themes: mode of therapy offered, the level of carer support and difficulties with access/facilities (e.g. “psychological input. Since X was not offered one to one intervention we had to pay for a private therapist”, “more focus on giving parents necessary skills to help”, “I would like this service to be available across the rest of this area” and “I think GPs need to be educated on the importance of early intervention and be aware of the service. My GP was unaware”). Moreover, young persons’ negative responses resulted in the following themes: communication with parents and service user involvement (e.g. “All the sharing from sessions with parents. It felt a bit excessive at times” and “I feel like my feelings weren't taken into account”).

Discussion

This article describes the formation of the HSE NCPED in the first three services (two children and adolescent and one adult) over their first 18 months of service activity. Furthermore, this article describes service user experience of their eating disorder service to date.

The referral rate for the first 12 months of three services was 126 referrals. It is envisaged that Ireland will have in the region of 1800 referrals per annum when all 16 eating disorder services are in operation (National-Working-Group-and-Clinical-Advisory-Group-HSE, 2018), with people accessing care from both eating disorder and local community mental health teams. The authors wish to highlight that this referral rate is most likely a significant underestimate of potential referrals. The reason for this is two services were not initially open to general practitioner referrals, just from other community mental health teams, and one was a pilot project for a number of months. This was due to the lack of staffing and training in the first year of service operation. Moreover, new services were not well known amongst healthcare professionals and general practitioners to refer readily in the early stages. We envisage that in time and following further service activity analysis, this referral rate will be significantly higher. Eighty-six percentage of referrals was accepted for assessment in the first 12 months; we again envisage as all eating disorder teams become busier this acceptance rate may be lower to meet demand.

As has been noted in other eating disorder research, AN was the most frequent eating disorder referred and diagnosed, followed by BN and OSFED (Smink et al. 2012). Interestingly, there was a high percentage of comorbid psychiatric diagnosis, in particular anxiety disorder (Kountza et al. 2018; Marucci et al. 2018). The percentage of self-harm behaviour was low in comparison to previously reported (Warne et al. 2021a , 2021b ). This could be explained by under reporting by service clinicians, such patients attending their local mental health team for a wider spectrum of care, or more likely due to this data being recorded at a single time point of a service users experience (i.e. assessment point). As such, a service user’s history may change during treatment and it will be helpful to record this at point of discharge also. This is a limitation of the study.

Eating disorders are best treated early and quickly (Currin and Schmidt, 2005). The current waiting time from referral to assessment reported in this study reflect the pragmatic difficulties in achieving key performance indicators. A significant portion of service users and their families were waiting greater than 4 weeks to be assessed. For adults, this may partly be explained by a legacy waiting list that pre-existed the NCPED and was being worked through. Small initial staffing numbers on all teams is also a factor. This reflects the needed service provision to meet current level of service standard.

Most service users received their initial treatment session within four weeks. Regrettably, some service users and their families had been waiting greater than four weeks to initiate treatment. The authors reflect on prior evidence that many service users have a duration of untreated eating disorder of up to two years and that prolonged wait-list time predicts a higher dropout rate from therapy and poorer outcomes (Carter et al. 2012; Brown et al. 2018; Austin et al. 2021; Flynn et al. 2021).

The core components of an eating disorder treatment are access to a psychiatrist and a therapist experienced in the treatment of eating disorders, with access when needed to dietitian and paediatric/medical consultation (Peterson et al. 2016). The most frequently offered initial treatment based on our service activity data is that of psychiatry review, dietetics and other (psychological and non-psychological) options. Only 53 (23.9%) service users were offered family-based therapy/family therapy straight away. The NCPED has a strong interprofessional approach to the delivery of first-line therapy to improve access times to evidence based therapies. Eating disorder clinicians from psychiatry, psychology, psychiatric nursing, social work, occupational therapy and family therapy have training and support in FBT, FT-AN, CBTE and MANTRA. We envisage as more clinicians are trained and certified, these evidence-based interventions will be offered more timely and readily in the future.

Over an 18-month period, 24 service users were referred for acute medical admission and only eight were referred for potential admission to an approved centre (inpatient mental health facility in Ireland). In Ireland, if a service user is referred to an approved centre, this reflects the severity of the service users’ condition as assessed by the eating disorder team, and this number only indicates that a referral was made, not that the service user was actually admitted to an inpatient approved centre which in the case of children and adolescents, is decided by inpatient teams.

Eighty-five (38.3%) service users were discharged over the 18-month period and positively more service users were discharged to their general practitioner than a generic mental health team.

The Child Outcomes Research Consortium (CORC) recommends using the ESQ, along with other measures, to understand families’ experiences alongside symptom reduction (Attride-Stirling, 2002). Brown and colleagues demonstrated that the ESQ differentiated well between less satisfied carers in terms of satisfaction with care, but does not show good precision in relation to satisfaction with the environment (Brown et al. 2014). The ESQ scoring tends to reflect overall positive or negative affect towards recovery (“halo effect”) (Brown et al. 2014). The carers and young persons’ experience of their service in this study is predominantly positive. It is interesting that there was a statistical difference in “certainly true” responses between carer and young person in the early phase of treatment and that this trend did not predominantly persist after months of treatment. This may be best explained by resistance from the young person in acknowledging they have an eating disorder and that the eating disorder itself (as an external object) was therapy interfering (Zeeck et al. 2005; Högdahl et al. 2016; Watson et al. 2017). This is exemplified by the insight of a service user’s quote, “How kind, funny and caring everyone was and how much they helped me. Even when I thought they were not helping, they were just keeping me safe”. This may also reflect a young person returning to psychological health during recovery and this area warrants future investigation. Similar to Sheridan and colleagues, service users appreciate access to service and treatment options as found in our thematic analysis (Sheridan and McArdle, 2016). Likewise, our findings that service users and their carers appreciate a good therapeutic alliance and staff expertise is similar to previously reported findings (Escobar-Kick et al. 2010). In the early stages of the NCPED roll out, there has only been service user satisfaction data available from the child and adolescent services. The NCPED plans to collect adult service user satisfaction data as the national programme is rolled out and more adult services become operationalised.

The NCPED model of care commented on a lack of knowledge of patient experience of recovery (National-Working-Group-and-Clinical-Advisory-Group-HSE, 2018). It has been suggested that patient satisfaction with treatment may improve clinical outcomes (and vice versa). Higher levels of carer and child satisfaction have been shown to enhance engagement and treatment completion (Stallard, 1995; Shapiro et al. 1997). As the NCPED evolves and more services become operational in Ireland, there will be ongoing wider service evaluation projects to evaluate service users’ and their families’ experience of a specialist community eating disorder service and to identify the factors valued in service satisfaction and recovery.

This study has several limitations. First, the data collected about service activity were collated monthly based on each service, not at an individual level. This study should be repeated prospectively to get a broadened understanding to make valid associations between variables. As such, we wish to emphasise the trend of variables (e.g. number of referrals, number diagnosed with AN) being returned and lack of individualised data on symptom severity or biological parameters (e.g. weight, height, median percentage body mass index). This may be being collated at regional service level. In terms of several service activity, there was a significant amount of missing data and this highlights the difficulty of collecting this data pragmatically and the lack of database and IT infrastructure in mental health services, making it complex for services to collate. The authors wish to highlight that collecting data is a vital need and requirement in evaluating such a clinical programme. Moreover, the researchers do not have access to adult service user satisfaction data at present. A limitation of the qualitative thematic analysis was that most service users provided short single sentences for review. This should be explored in the future by in depth interviews of service user satisfaction. As part of the NCPED model of care, it aims to deliver on the health service executive’s (HSE) three goals of quality and safety, good access and value for money (National-Working-Group-and-Clinical-Advisory-Group-HSE, 2018). As such, the working group within the NCPED highlight the ongoing need for a National Eating Disorder Register within Ireland, to ensure that a service user’s journey can be appropriately mapped and understood to improve service delivery and patient outcomes.

We did not have valid data on the number of service users and carers who declined to complete the ESQ. We acknowledge that families may have found it difficult to provide negative comments due to the lack of alternative available similar public services in Ireland. These data reflect a single time point of service activity, it does not reflect the many changes and nuances that may occur during the course of treatment in terms of service experience (e.g. the assessment diagnosis reported may not reflect the discharge diagnosis, change in treatment offered or level of comorbidity). As the NCPED is rolled out and each team becomes operational, the NCPED plan to address these limitations, including providing a more effective, individual level evaluation process of the eating disorder recovery in Ireland

In summary, the HSE national clinical programme for eating disorders is being implemented in Ireland. To date, the level of service activity and quality of service provided highlight that the NCPED regional eating disorder teams can provide early access to evidence-based care that can lead to meaningful change and recovery in the lives of individuals experiencing an eating disorder and their families. Albeit that is, if each service is appropriately staffed, trained, and resourced to meet the demands of its sector population.