Learning from mistakes: embedding qualitative methods in a pilot ARFID care pathway

Avoidant/Restrictive Food Intake Disorder, better known as ARFID, has been formally recognized in the Diagnostic and Statistical Manual of Mental Disorders since 2013, yet clinicians and patients alike have long struggled with a scarcity of evidence on how best to treat it. Now, a team of researchers and clinicians in the United Kingdom has published one of the most detailed evaluations to date of a pilot multidisciplinary care pathway for adults with ARFID, offering a candid, real-time portrait of what works, what does not, and how services can improve while they are still being built. The study, published in the Journal of Eating Disorders, demonstrates a novel methodological approach: embedding qualitative research directly into the “study” phase of the plan-do-study-act cycle, the iterative quality improvement framework widely used in healthcare, to capture the experiences of both staff and service users and translate them into concrete refinements.

The research was conducted by a team based at the SPIRED Research Clinic within Sussex Partnership NHS Foundation Trust, in collaboration with academics at the University of Exeter, the University of Bristol, the University of Surrey and the University of East Anglia. Led by corresponding author Cat Papastavrou Brooks, the team set out to address a gap that has persisted since ARFID entered the psychiatric lexicon. Although clinical guidelines recommend multidisciplinary care for the condition, formal evaluations of such pathways remain rare, leaving services to design treatment models largely in the dark. The Sussex pilot pathway, established within a community eating disorder service, aimed to provide adults with ARFID access to coordinated care from a specialist multidisciplinary team, or MDT, comprising professionals from different clinical backgrounds working together on assessment and treatment.

To understand how the pathway was functioning in practice, the researchers conducted semi-structured interviews with seventeen participants who had direct experience of it: seven service users and ten staff members. Semi-structured interviews are a qualitative method in which participants respond to a consistent set of open-ended questions while retaining the freedom to raise issues the interviewer may not have anticipated, making them well suited to exploring complex, poorly understood clinical territory. The interview data were then analyzed using reflexive, codebook thematic analysis, a hybrid approach that combines the researcher’s interpretive engagement with the data, characteristic of reflexive thematic analysis, with a structured codebook that helps organize themes systematically. This dual approach allowed the team to identify recurring patterns across interviews while remaining sensitive to the nuance and individuality of each account.

The findings, the researchers acknowledge, are neither flattering nor discouraging but somewhere in between, captured perhaps best by the words of a staff member quoted in the study’s title: “We might be making mistakes, but we’re learning, we’re trying, and we’re listening.” The analysis revealed significant gaps in both understanding and treating ARFID, gaps that produced feelings of uncertainty and frustration among clinicians and patients alike. ARFID differs from better-known eating disorders such as anorexia nervosa in that it is not driven by concerns about body shape or weight. Instead, individuals with the condition eat a severely limited range of foods or struggle to eat enough altogether, often due to sensory sensitivities to food texture, taste or smell, fear of aversive consequences such as choking or vomiting, or a simple lack of interest in eating. The consequences can include serious nutritional deficiencies, weight loss, dependence on nutritional supplements, and profound interference with daily life and social functioning.

The study found that the establishment of a specialist multidisciplinary team improved knowledge and efficiency in meaningful ways. Concentrating expertise within a dedicated team allowed staff to build ARFID-specific competence more rapidly than would have been possible if the condition remained a peripheral concern scattered across generalist services. However, the creation of the specialist MDT was not without costs. The analysis surfaced challenges related to team integration and resource allocation, raising questions familiar to health systems everywhere: how to embed a new specialist service within an existing organizational structure, how to share its expertise with the wider workforce, and how to allocate finite staff time and funding between a growing specialist pathway and the broader service portfolio.

For service users, the most salient themes concerned engagement and communication. The interviews identified barriers in both domains, and they demonstrated clearly the need for service user consultation and pathway co-creation, the practice of involving patients directly in designing the care systems intended to serve them. Patients emphasized that being involved in planning their own care was essential to a positive experience. This finding echoes a broader movement in modern healthcare toward shared decision-making and lived-experience involvement, but the study suggests it carries particular weight in ARFID care, where the condition itself is frequently misunderstood and where rigid, one-size-fits-all treatment models are especially likely to fail.

The study’s methodological contribution may prove as influential as its clinical findings. The plan-do-study-act cycle, known widely as the PDSA cycle, is a cornerstone of quality improvement in healthcare services. Teams plan a change, implement it, study its effects, and act on the results before beginning the cycle again. In many real-world applications, the “study” phase is the weakest link, often reduced to a handful of quantitative metrics that can miss the texture of patient and staff experience. By embedding rigorous qualitative methods within that phase, the Sussex team showed how rich interview data can inform the “act” stage with specificity and depth, converting the lived experiences of patients and clinicians into an actionable roadmap for change. The approach effectively turns the PDSA cycle into a vehicle not just for measuring whether a service works, but for understanding how and why, from the perspectives of the people inside it.

The lessons drawn from the analysis fed directly into that act stage. The researchers identified three priority areas for improving the pathway: service user consultation, treatment adaptations for neurodiversity, and cross-service working. The neurodiversity finding reflects a growing recognition of the substantial overlap between ARFID and neurodevelopmental conditions, particularly autism and attention deficit hyperactivity disorder, meaning that treatment approaches may need to be adapted to accommodate sensory sensitivities, communication preferences and different cognitive styles. Cross-service working, meanwhile, addresses the fragmentation that patients and staff reported when care needed to span primary care, specialist eating disorder services, and other parts of the health system.

The authors draw broader conclusions for the field. Future ARFID pathways, they argue, should prioritize service user agency and access to multidisciplinary care as foundational principles rather than optional enhancements. They also recommend that emerging services incorporate consultation from external eating disorder services as well as from experts by lived experience, drawing on knowledge accumulated elsewhere rather than repeating the same early missteps. The study is framed explicitly as a learning process, and its honesty about uncertainty is presented as a strength rather than a weakness: acknowledging gaps in understanding, the authors suggest, is the first step toward closing them.

The significance of the work extends well beyond a single NHS trust. ARFID affects people across the lifespan, and while much of the existing research has focused on children and adolescents treated in pediatric programs, this study addresses adults, a population for whom evidence and services remain especially thin. Adults with ARFID may have lived with restricted eating for decades, often misdiagnosed, dismissed or unrecognized, and the development of community-based pathways represents an important step toward closing a long-standing gap in eating disorder provision. By documenting both the promise and the friction of building such a service, the study offers a template that other services attempting to develop ARFID provision can adapt to their own contexts.

The research was registered as a service evaluation by the participating NHS trust’s Quality Improvement Support team and, while it received funding support through a Springboard Award with the NIHR ARC KSS research academy, the funding organization played no role in study design, data collection, analysis or publication. The article is published open access under a Creative Commons license, making the full findings freely available to clinicians, researchers, and people affected by ARFID worldwide. As services around the world grapple with how to respond to a condition that is increasingly recognized but still poorly served, the Sussex team’s message is simple and perhaps universally applicable: listen to patients, build diverse teams, expect mistakes, and treat every iteration as a chance to do better.