Adapting family-based treatment for primary care (FBT-PC) for treating child and adolescent restrictive eating disorders in rural settings: a case series Authors: Jocelyn Lebow, Ivana Croghan, Jennifer St Sauver, Robert M Jacobson, Maegen Storm, Seth W Gregory, Sarah Kruse, Aya Abdelrahman Siddig, Jared Masrud, Leslie Sim Published: 2025 ABSTRACT AbstractBackgroundAccess to eating disorder care for child and adolescent patients is challenging especially for patients from rural and/or lower-resourced health systems. In order to improve the availability of eating disorder treatment, we developed Family-Based Treatment for Primary Care (FBT-PC), for primary care delivery. Although research has shown that FBT-PC has a significant effect on young patients’ eating disorder symptoms, the model has not been evaluated in rural settings. The purpose of this study was to establish proof-of concept of an adaptation of FBT-PC for rural settings, in preparation for a pilot feasibility/acceptability trial.MethodsThis study describes adaptations made to the FBT-PC protocol for implementation in small town and rural areas, as well as a case series of six children and adolescents with restrictive eating disorders who received FBT-PC in these settings.ResultsFindings from this case series establish proof-of-concept for FBT-PC implementation in rural primary care clinics. Specifically, the majority of patients either completed treatment (n = 3) or were referred to standard FBT (n = 1) which is consistent with the intention of FBT-PC as offering a lower-step eating disorder intervention.ConclusionsThese promising preliminary cases support further evaluation of FBT-PC in rural settings. Future research is needed to assess clinical outcomes and acceptability of the intervention. Trial Registration Clinicaltrials.gov #22-011880Keywords: Adolescent, Anorexia nervosa, Feeding and eating disorder, Primary health care, Rural health Background The substantial morbidity and mortality of child and adolescent eating disorders is well-established [1]. Though evidence-based eating disorder interventions exist [2], access to such treatment is limited [3, 4], particularly in rural and/or under-resourced communities [5, 6]. Widespread healthcare disparities and delays in receiving care may contribute to worsening of illness burden and patient outcomes [7–9]. As such, expanding access to evidence-based eating disorder treatment is an urgent priority of the field. Though efforts to train more mental health providers are underway [10], longstanding difficulties disseminating evidence-based treatments for uptake among mental health professionals, including lack of resources, time, professional supports, and/or provider motivation, knowledge, and self-efficacy, suggest the need to reconsider how, and by whom, eating disorder care can be delivered [11–13]. There is a notable dearth of studies evaluating the comparative outcomes of evidence-based behavioral interventions for any psychiatric disorders delivered to rural versus urban adolescent populations. However, an initial study establishing the feasibility of telehealth delivery of the first-line outpatient treatment for adolescent eating disorders, Family-Based Treatment (FBT), to rural communities in Australia, suggest the modality has promise in rural communities, when they are able to access the care [14]. It is with this in mind that we developed Family-Based Treatment for Primary Care (FBT-PC), an outpatient intervention for child and adolescent restrictive eating disorders intended for delivery in the primary care setting [15]. FBT-PC follows principles and uses strategies from FBT [16]. FBT is a behavioral intervention in which trained eating disorder providers empower caregivers to take charge of their child’s eating using strategies such as prioritizing weight restoration, separating the illness from the patient, reducing caregiver and patient guilt and blame, and using a medical metaphor to increase families’ understanding of the illness. These same tenets form the basis of FBT-PC [16]. However, in FBT-PC trained medical primary care providers (PCPs) rather than eating disorder specialists, deliver care aimed at increasing parent/guardian self-efficacy to reverse their child’s dietary restriction and weight loss. Though these providers have access to consultation with an FBT-certified mental health provider, all interventions are delivered by PCPs. Additional adaptations in FBT-PC include shortening the intervention from three phases to two. In standard FBT, phase 1 includes empowering caregivers to take charge of all meals and encourage weight restoration, phase 2 focuses on returning developmentally appropriate control over eating to the patient, and phase 3 includes topics such as education about typical adolescent development and relapse prevention [16]. In FBT-PC, phase 1 is very similar to that of standard FBT, with a focus on caregiver-led refeeding to drive patient weight restoration, but phase 2 combines both transitioning control back to the patient, as well as relapse prevention. FBT-PC also differs in that the treatment is delivered in briefer sessions designed to fit in the average primary care visit framework (e.g. 45/30/20 minutes vs. 60/50 minutes in specialty care), and that measurement of weight and other vitals is completed by nurses or medical assistants, in line with standard primary care rooming procedures. Research has established that FBT-PC, when evaluated in a Midwest healthcare system located in a small city, has a significant effect on weight restoration, eating disorder remission, patient quality of life, and parent/guardian self-efficacy [17]. Patients were typically seen for an average of 7.82 sessions (though this ranged from 1 to 32 sessions) [17]. Findings also suggest that embedding the intervention in primary care expands access by reaching a more racially and socioeconomically diverse group of young patients earlier in their illness course [18]. Despite these promising outcomes, the utility of FBT-PC in addressing healthcare disparities hinges on the ability to disseminate the intervention to lower-resourced systems. At present, all evaluations of the model have been conducted in primary care practices associated with well-funded medical systems in urban settings [17, 19]. As such, it is unclear whether the intervention can be implemented in other primary care settings, particularly those that serve rural populations. The aim of this study is to describe modifications made to the FBT-PC protocol for implementation in three community-based primary care clinics in small towns and rural areas. In this way, we hope to establish proof-of concept for an adaptation of FBT-PC for rural settings, in preparation for a larger pilot feasibility/acceptability trial. Outcomes of interest include rates of patient retention and treatment completion. We will present a case series of six children and adolescents with restrictive eating disorders who received FBT-PC from three newly-trained PCPs in these settings. Methods Providers and settings FBT-PC providers included two pediatricians and one pediatric nurse practitioner from separate rural and small-town primary care clinics. Two were female and one was male. They had been in medical practice for 1–13 years but none had any previous eating disorder training. All volunteered to be trained in FBT-PC, expressing interest due to a perceived need for these services in their areas, and lack of local resources to which to refer patients. Patients were recruited from within the primary care clinics in which the FBT-PC providers were based. These clinics served rural patients, as defined by Office of Management and Budget criteria, as well as small town/city patients. The first clinic was located in Faribault, MN, serving a majority rural patient population, including 1339 patients ages 7–18. The second clinic was located in Onalaska, WI. This clinic serves the small city of La Crosse, WI and surrounding rural areas, and includes 6874 patients 7–18 years old. The final clinic was located in Red Wing, MN which serves a small-town and rural catchment area, including approximately 4171 patients 7–18 years old. Study procedures Cases in this series included the first six consecutively enrolled patients from these three settings. All patients ages 7–18 receiving primary care in the Mayo Clinic health care system were eligible for the study. Participants had a diagnosis of a DSM-5 restrictive eating disorder (anorexia nervosa or other specified feeding or eating disorder (OSFED) characterized by dietary restriction and/or weight loss or failure to make expected weight gains), with the exclusion of avoidant/restrictive food intake disorder. All enrolled patients had at least one caregiver who was also willing to participate in the study. Patients were excluded who had received FBT in the past, and/or who reported active suicidality. In addition, we further excluded patients if their enrolled caregiver had active substance dependence, mental health concerns that made it difficult to engage in treatment, or were perpetrators of abuse. All patients were medically and psychiatrically stable at the start of treatment. Diagnoses were made using DSM-5 criteria following clinical interviews with PCPs. Treatment completion was determined by PCPs in consultation with a clinical psychologist. This determination was based on patients no longer meeting criteria for a DSM-5 eating disorder, and having successfully completed phase 1 and phase 2 interventions. Goal weights were not calculated for FBT-PC, however providers were trained in using developmental growth curves to evaluate weight suppression, and recognizing weight restoration based on resolution of physical and psychological symptoms, ability to maintain a stable weight independently, and marked decreases or elimination of eating disordered cognitions. This study was approved by the Mayo Clinic Institutional Review Board (IRB). The IRB approved a waiver of informed assent, but all parents/guardians completed informed oral consent on behalf of the patient, as well as signed a Health Insurance Portability and Accountability Act (HIPAA) authorization form. Adaptation process A full description of the FBT-PC intervention and training has been previously published [15]. In brief, FBT-PC borrows key intervention strategies from standard FBT to increase parent/guardian self-efficacy to reverse their child’s dietary restriction and weight loss [16]. After consulting with stakeholders (PCPs, primary care leadership, administrative partners, nurse managers, and a rural advisory board comprised of patients receiving primary care in rural areas), the FBT-PC protocol was adapted for smaller, lower-resourced clinics located in rural areas. Adaptations to the FBT-PC model are described using the Framework for Reporting Adaptations and Modifications to Evidence-based Implementation Strategies (FRAME-IS) model [20]. This is an implementation science framework that separates behavioral intervention adaptations into changes in (1) context, (2) content, (3) training, and (4) evaluation [20]. Context Changes Context Changes The main context change was adapting the FBT-PC intervention for implementation in rural clinics with no access to on-site hospital or specialty services. Stakeholders from these settings expressed that, as their clinics had considerable constraints in physical space, it was not feasible to have an on-site study coordinator. Thus, the research protocol incorporated remote study procedures, including a telephone consent process. Another context change concerned the length of the sessions and timing of interventions. For two of the three FBT-PC providers, all their primary care visits were templated for 20 min. Administrators in these settings allowed FBT-PC providers to block two consecutive appointments to see study patients. This change necessitated an abbreviated intake (40 min instead of 60 min in the original FBT-PC iteration). As such, some session content was shifted from intake to subsequent follow-up visits. To make the intervention financially feasible, visits were billed based on time, which allowed the systems to support this change to templates without any productivity losses. 2.Content Changes Content Changes Study staff met with a rural clinical advisory board, which included current and former adult primary care patients who all lived in rural settings. They provided insight on specific considerations for engaging rural populations, including identification of additional barriers. In particular, they identified concerns about parents being willing to limit patients’ physical activity or prioritize meals and/or session attendance when youth were needed to help on the farm. These barriers were incorporated into case examples discussed during FBT-PC training and case consultation. Additionally, potential recruitment sources were identified by the board, including Future Farmers of America and 4-H clubs in the area. 3.Training Changes Training Changes Like the original FBT-PC iteration, training and ongoing case consultation were conducted over 3 lunch hours so as to minimize the time burden and loss of clinical productivity for the practice. However, these meetings were provided entirely virtually, instead of in-person. This change allowed providers from a geographically diverse group of practices to attend all trainings and meetings with a minimal impact on their busy schedule. In total, providers participated in 3 h of training and subsequent biweekly hourlong case consultation for the duration of the study period. Given that these PCPs were the only people with any eating disorder expertise in their respective practices, having regular virtual case consultation was intended to help them feel supported in working with clinically unique patients with challenging-to-treat disorders, and allowed for informal fidelity checks (e.g. conversations about which FBT-PC tenets and strategies were being used in session) while minimizing additional burden on providers. 4.Evaluation Changes Evaluation Changes Due to space constraints in the rural clinics, all study assessments were monitored remotely. While previous evaluations of the FBT-PC had on-site nursing staff members administer all measures as part of their clinical workflow, these rural clinics’ workflows did not allow for staff to administer or collect measures. As such, a study coordinator, who was located off site monitored patient visits and sent out measures remotely via RedCap, directly to patients and caregivers prior to their appointment. Results Patient characteristics This case series was comprised of 6 patients (mean age = 13.5 years, range 10–17 years; 100% white), including 3 males and 3 females. Patients were enrolled consecutively. Three patients were from Faribault, MN, two were from Red Wing, MN and one was from the Onalaska, WI service areas. Each newly-trained FBT-PC provider treated at least one patient, but they were not evenly distributed across the providers. Of these patients, 4 were privately insured, 1 had Medicaid insurance, and 1 had no insurance on record (their insurance had lapsed approximately 6 months before initiating care). This was the first course of eating disorder treatment for all patients. Half the sample (n = 3) were diagnosed with anorexia nervosa, while the rest were diagnosed with other specified feeding and eating disorder (OSFED). Of the patients with anorexia nervosa, two were diagnosed with restricting subtype, while 1 met criteria for binge/purge subtype. The 3 patients who were diagnosed with OSFED met criteria for a clinically significant restrictive eating disorder but did not meet the DSM-5 anorexia nervosa diagnostic criterion C. Mean BMI percentile (BMI%) at baseline was 46.59 (range = 0.13–81.61). Baseline weight suppression was calculated by subtracting current BMI% from historical highest BMI%, (M = 25.1%; range = 14.35–41.3). Of these cases, three completed the full course of treatment, as defined by patients no longer meeting criteria for a DSM-5 eating disorder and successfully completing phase 1 and phase 2 FBT-PC interventions. Additionally one patient was referred to outpatient standard FBT, one patient was lost to follow-up, and one patient was administratively withdrawn from the study due to the onset of an acute medical condition, unrelated to her eating disorder. The patients who completed FBT-PC treatment received an average of 9 sessions. All three gained weight from baseline to the end of treatment (M = 7.5 kg, range = 3.9–13.3 kg) with an average increase in BMI% of 14.49 percentile points (range = 6.01–23.47), and average reduction in weight suppression of −14.87 BMI% (range = − 5.96 to −23.7). See Table 1 for a summary of patient characteristics and treatment outcomes. Table 1Patient and treatment characteristicsPatient 1Patient 2Patient 3Patient 4Patient 5Patient 6Age (years)171016111314Race/EthnicityWhiteWhiteWhiteWhiteWhiteWhiteGenderFemaleFemaleFemaleMaleMaleMaleED DiagnosisOSFEDOSFEDAN-RAN-ROSFEDAN-BPBaseline BMI%45.54.2881.04670.1381.61Weight suppression (BMI %)41.314.9914.8829.735.414.35Sessions attended3471624Clinical outcomeAdministrative Withdrawal Standard FBTGraduationGraduationLost to Follow-upGraduationChange in weight (baseline to end to treatment)n/an/a3.9 kg5.3 kgn/a13.3 kgChange in weight suppression (BMI%; baseline to end of treatment)n/an/a− 5.96− 23.7n/a− 14.95AN-R = anorexia nervosa, restricting subtype; AN-BP = anorexia nervosa, binge/purge subtype; BMI% = Body Mass Index percentile; ED = eating disorder; OSFED = other specified feeding and eating disorder Patient and treatment characteristics AN-R = anorexia nervosa, restricting subtype; AN-BP = anorexia nervosa, binge/purge subtype; BMI% = Body Mass Index percentile; ED = eating disorder; OSFED = other specified feeding and eating disorder Case series The following descriptions of the first 6 patients consecutively enrolled in the study include patient and family characteristics, session content, and adaptations to facilitate FBT-PC for these rural youth. Case 1 Case 1 was a 17-year-old girl with prior diagnosis of attention deficit/hyperactivity disorder (ADHD) who was seen for FBT-PC by a pediatric nurse practitioner, who was also her PCP. The provider identified the eating disorder in the course of a primary care visit for ADHD medication management. At this time, the patient had lost over 10 pounds in 6 months. She reported eating 1–2 meals on some days, but often went 24 h without eating. She described feeling faint with positional changes, as well as presyncope with notable tunnel vision. The patient reported symptoms of depression and anxiety, as well as concentration impairment that had worsened, despite her remaining adherent to her ADHD medication regimen. Her PCP communicated that these symptoms were likely linked to her low weight and malnutrition. As the patient reported no body image concerns, she was initially diagnosed with OSFED. However, thoughtout treatment the PCP continued to evaluate for whether anorexia nervosa-restricting subtype was a better fit. At baseline, she weighed 57.1 kg, which corresponded to a BMI% of 45.5. Given that she had a history of tracking in the 83–86th BMI%, this represented significant weight suppression. This patient attended sessions with her mother and maternal grandmother, with whom she resided. Though the family expressed few concerns about the patient’s eating, she significantly restricted her intake and struggled to eat regularly despite feedback that this was needed. In spite of the family’s uncertainty about the patient’s eating disorder symptoms, their positive and longstanding relationship with their PCP allowed them to continue in care and quickly implement FBT-PC interventions, including monitored dinners and breakfasts at home, and school lunch monitoring. The PCP delivered education about the high calorie needs of an underweight and hypermetabolic patient and encouraged the family to supplement meals with a regimen of daily milkshakes. Because the family had financial stressors, the PCP helped brainstorm ways to integrate higher calorie foods at lower cost. The PCP emphasized the importance of presenting a united front between the grandmother and mother, as they were not always on the same page with regard to caloric and monitoring needs for the patient. Their joint attendance at all sessions allowed the mother and grandmother to work towards consensus about mealtime expectations. The family was seen weekly, but after three sessions, the patient tested positive for an acute medical condition unrelated to her eating disorder, which necessitated her withdrawal from the study. This was reported and documented as a study adverse event, but was determined by the data safety monitoring board to be unrelated to study interventions. The patient continued to follow with her PCP for medical care, and the PCP collaborated with the patient’s specialty medical team to keep track of her eating and weight in the context of her eating disorder history. Case 2 Case 2 was a 10-year-old girl with historical ADHD and anxiety diagnoses who was seen by an FBT-PC provider who was not her usual PCP. Study staff identified her as a potentially eligible patient through chart review processes that were built into the study protocol. Once this patient was identified as having an unexpected decline from her historical growth curve, staff reached out to the PCP prior to the patient’s appointment for ADHD management. When asked, the family expressed significant concern about the patient’s eating, and noted that, though the patient had historically been a picky eater, her dietary repertoire was narrowing rapidly. The patient reported restricting due to abdominal pain during eating, “feeling like her stomach was being squeezed”. Parents reported significant conflict with the patient at mealtimes, and that she struggled to stay at the table, particularly when her father was busy with planting season and had to miss meals. The patient’s baseline weight was 23.2 kg, and her BMI% was 4.28, significantly lower than her historical BMI% which ranged from 19 to 21. She was diagnosed with OSFED. At this session the patient also received a diagnosis of trichotillomania, due to reported symptoms of compulsive hair pulling. The PCP talked about short-term habit reversal strategies for these symptom, but emphasized the importance of addressing the eating symptoms prior to beginning a course of CBT or any psychotropic medications for this disorder, as evidence supports the fact that malnutrition negatively impacts the effectiveness of these interventions. The patient and her parents attended four sessions. The PCP worked with parents to separate the illness from the patient, and remain consistent, warm, and neutral during mealtime behaviors, including yelling, crying, and the patient hiding in the house. The patient’s anxiety and trichotillomania symptoms worsened during the course of treatment. This distressed the parents who were concerned that the meal expectations and monitoring were responsible for increasing symptoms. In response, the provider provided education on the impact of malnutrition on anxiety and coached them to remain neutral. In spite of this, the patient struggled to gain weight and by session 3, the family and provider collaboratively discussed that she might need to transition to standard FBT delivered by a trained eating disorder specialist. This would allow the family to have longer sessions each week along with the reassurance that a mental health provider was monitoring the patient’s anxiety symptoms. The patient was referred to standard FBT via virtual visits and transitioned to this care without any weight loss in the interim. Case 3 Case 3 was a 16-year-old girl who presented to her PCP, who was also an FBT-PC provider, for concerns regarding weight loss coupled with symptoms of early satiety and bloating, as well as dizziness with positional changes. During the visit, it was discovered that she rarely finished more than half of a typical portion at meals or snacks, citing feelings of being too full. She also refused to eat with her family and usually ate meals alone in her room. Her dietary restriction continued even after her PCP counseled her on the the importance of increasing her intake in spite of these physical symptoms. Though she initially denied any body image concerns or dysmorphia, she became visibly upset and tearful at the prospect of having to gain weight. She presented at a weight of 69.3 kg, and a BMI% of 81.04, which was 14.88 percentile points lower than her historical growth curve. She was diagnosed with anorexia nervosa, restricting subtype and attended FBT-PC sessions with her mother, who relayed recommendations to father whose work schedule made it impossible for him to attend sessions. The patient was extremely unhappy about the prospect of monitored meals and initially slept through breakfast and avoided coming home for dinners. The PCP encouraged the mother to place more expectations around mealtimes and engage in more rigorous monitoring. The PCP also empowered the mother to limit the patient’s exercise and activity until her weight was steadily increasing. The mother and PCP made a collaborative decision that the patient had to restore all the weight she lost before she was allowed to play school softball. The patient was highly motivated to return to this activity before the season ended. After a week of resistance, the patient became more compliant with expectations, stating that she was “giving in, because [she] knows eating is unavoidable”. After 2 sessions of weight gain, the patient expressed increased distress over her body image. The provider normalized the j-shaped recovery curve (e.g. initial worsening of behaviors with the initiation of treatment before symptom improvement is seen) that is typical in eating disorder treatment. She coached the family on separating the illness and responding to these concerns with neutrality. After 7 visits across 3 months, the patient had gained a total of 3.9 kg, which corresponded to a BMI% of 87. She and her parents reported she was eating completely independently and intuitively, without body image concerns or restriction. She resumed activities with no issues around overexercise or increased symptoms. The provider discussed relapse prevention and dismissed the patient from eating disorder care, with a plan to follow up with the provider in her capacity as a PCP, in one month. See Fig. 1 for Patient 3’s BMI curve. BMI curves for cases who completed FBT-PC (gray box indicates the treatment period) Case 4 Case 4 was an 11-year-old boy with comorbid anxiety and eating concerns. He was referred by his psychologist, who had been treating him for anxiety, after she became concerned about the patient’s restrictive eating in the context of wanting to “be healthy”, eliminating all foods he used to enjoy, marked fear of weight gain, negative body image, and preoccupation with weight and shape. The patient presented to his PCP at a weight of 47.8 kg, and a BMI% of 67, with a history of tracking near the 86th BMI%. He was diagnosed with anorexia nervosa, restricting subtype. This patient’s treatment course was complicated by his anxiety symptoms, including social anxiety around interacting with same-age peers and consequent avoidance of activities outside the home. Though these symptoms predated the onset of the eating disorder, his rapid weight loss and refusal to eat meals made it difficult for him to engage in anxiety treatment. The provider discussed how malnutrition can undermine progress in CBT. His FBT-PC provider encouraged the family to prioritize eating disorder treatment and weight restoration as an important precursor to successful anxiety care. The provider reminded parents that food is medicine and though this treatment would be initially distressing, it is necessary to avoid negative health outcomes or even death. While the parents struggled to tolerate the patient’s distress, the PCP kept them focused on separating the illness from their child, remaining consistent in refeeding efforts, and agnostic to the cause of the eating disorder. As the patient approached the end of weight restoration, he began to exhibit more avoidance behaviors around school attendance. As such, his PCP referred him to a specialty intensive pediatric anxiety program. This program indicated that at his former low weight he would not have been eligible for their program, but now that he was weight restored, they could accept him. After 16 sessions of FBT-PC across 8 months, corresponding with full resolution of dietary restriction and intrusive thoughts about eating, as well as a weight gain of 8.3 kg and return to the 80th BMI%, the family graduated from eating disorder care. See Fig. 1 for his BMI curve. Case 5 Case 5 was a 13-year-old boy who was identified during a primary care appointment for abdominal discomfort. At this time he endorsed a highly restrictive list of preferred foods, as well as early satiety and low appetite. He was experiencing significant weight suppression with a baseline BMI% of 0.13 representing a decline from the 35th BMI%. The patient was diagnosed with OSFED. The patient attended 2 sessions in the company of his mother. The family struggled with conceptualizing his symptoms as an eating disorder, reporting that the patient’s father did not believe that boys could have eating disorders. As the father did not attend sessions, the PCP educated the mother about how eating disorders present in community samples and how they often go undetected in boys because of misunderstandings and stereotypes. The PCP emphasized the medical nature of the illness, suggesting that their son was not choosing the eating disorder symptoms and behaviors, and that he had a serious illness. She attempted to reduce parents’ blame towards the patient and each other and encouraged the father to come to the sessions in order to help the parents develop a united front in addressing the eating disorder. After 2 sessions the family did not show for their scheduled appointment and did not respond to outreach attempts by clinic staff. Case 6 Case 6 was a 14-year-old boy who presented to his PCP who was also an FBT-PC provider, for concerns of anxiety, irritability, and mood swings. During this appointment, he reported eating no snacks and only one meal per day. He also described frequent “binge” episodes in which he felt like he lost control over his eating, and would eat things such as chips, or ice cream that he otherwise avoided. Though he denied purging, parents stated he would often take long showers after meals and they heard sounds of vomiting during this time. The patient denied purposeful efforts to lose weight, however he had lost 33 pounds over 3.5 months. He presented at a weight of 77.8 kg, and a BMI% of 81.61, which was below where he historically tracked, at the 95th BMI%. Based on these findings, he was diagnosed with anorexia nervosa, binge/purge subtype. The provider saw the patient and a parent for 4 visits (parents alternated attendance), focusing on empowering parents to prioritize weight gain, including monitoring three high calorie meals and snacks per day, as well as postprandial monitoring. The implementation of this regular meal plan immediately eliminated the patient’s binge episodes and all potential purging. Due to parents’ work schedules, the family was unable to come for weekly visits, and consequently were seen every 3 weeks over a total of 3 months. The provider supplemented these visits with regular electronic portal messaging to help parents problem solve issues with meal monitoring and keep them on track between visits. The patient responded rapidly to care, and gained 13.3 kg over the course of 12 weeks, which put him back on his curve, at a BMI% of 95.6. After 3 months, he was eating consistently and flexibly with only developmentally appropriate meal support from parents. At the time he discharged from FBT-PC treatment, as his symptoms were in full remission. His provider planned to follow him in his primary care clinic on an as-needed basis to ensure he maintained this progress. See Fig. 1 for his BMI curve. Discussion Of the six cases described in this series, three (50%) completed FBT-PC, one was referred to standard FBT, and two withdrew from the program. Though no conclusions about effectiveness can be drawn from these cases, benchmark studies of FBT-PC in larger health systems suggest that approximately 56% of patients successfully complete treatment [15], so this is largely in line with those data. FBT-PC was developed to address the treatment-access gap that characterizes eating disorder care and disproportionately impacts underresourced health care systems, such as those in rural areas [4–6]. As FBT-PC was originally tested in well-resourced primary care clinics in urban settings [17], this open case series describes the adaptation process of FBT-PC within three primary care clinics that serve rural patients. Stakeholders identified the need for important changes to the context, content, training, and evaluation to increase the disseminability of the FBT-PC protocol. These included reducing resources needed to implement the study by developing remote consent and measurement protocols, shifting session content so it could be delivered in 40-minute sessions, adapting training to be relevant for the unique needs of rural patients, and delivering training and case consultation virutally. After incorporating these changes, we conducted this case series to inform future adaptations and establish proof-of-concept for FBT-PC implementation in rural settings. Though conclusions that can be drawn from this small case series are limited, data support proof-of-concept for implementation of FBT-PC in these settings. The majority of the patients recruited from these rural primary care clinics did not present to their initial primary care appointment with expressed eating disorder or weight loss concerns. Instead, most cases were identified at appointments for the management of psychiatric symptoms (e.g., anxiety, ADHD). Only one of the patients in the case series expressed presenting concerns about her eating, weight, and early satiety. That patients and families did not present with specific concerns or insight into their eating disorder represents a notable barrier to identifying eating disorders in rural primary care practice. Most patients in this case series had at least one, if not multiple, comorbid mental health diagnosis, further complicating the process of providers identifying, and families accepting the eating disorder diagnosis and engaging in treatment (see cases 2 and 4). Psychiatric symptoms can mask eating disorder symptoms and contribute to delayed detection, which is associated with poorer outcomes [21]. In addition, untreated or undertreated eating disorders can undermine psychosocial or pharmacological treatments for comorbid psychiatric disorders. As such, it is crucial that PCPs in rural settings focus on reversing malnutrition, regardless of comorbidities, and despite the fact that families can understandably be distracted by and/or question whether addressing comorbid symptoms should be the priority. It is notable that these PCPs were able to surmount these barriers and retain three of the patients for a successful course of FBT-PC, until they no longer met criteria for a DSM-5 eating disorder, despite patients’ psychiatric comorbidities. Another barrier to engaging rural families in eating disorder care is the profound stigma surrounding these conditions, as well as pervasive misunderstandings about who they affect. For example, in case 5, the parents’ belief that boys cannot have eating disorders and the stigma associated with this notion likely contributed to their prematurely dropping out of treatment. Compared to patients treated in primary care, families presenting to specialty eating disorder clinics are at an advantage, as they are inherently further along in their acceptance of the diagnosis, given their pursuit of eating disorder treatment. It is possible that stigma and misunderstandings about eating disorders may represent a particularly prominent barrier for families seen in rural primary care clinics, given poorer health literacy among these populations compared to their urban counterparts [22]. Future iterations of FBT-PC for rural settings need to consider novel ways for PCPs to address these barriers to better engage families in care. Next steps might include developing low impact “pre-treatment” interventions or resources, to be implemented in general primary care, designed to improve family awareness and understanding of how eating disorders present in their population, as well as reduce stigma about the disorders. There are several limitations to this study, all of which are inherent to the rural settings in which it was conducted. Most notable is the racially homogeneous sample. While this is not unexpected given the racial makeup in patients served by these clinics, the fact this case series included all white patients underscores the established need for additional outreach attempts to patients of color who may benefit from eating disorder treatment. This is particularly important given the historical overrepresentation of white patients in eating disorder treatment and research. The gender distribution of this study, which had equal numbers of cisgender male and female patients, increases the generalizability of this case series and highlights that with training, PCPs can identify eating disorders in patients who fall outside of common eating disorder stereotypes. However, the fact that this sample was entirely comprised of cisgender patients, suggests that additional work is needed to maximize sample diversity with regards to gender identity as well as race. Finally, because these PCPs volunteered to be part of this study, selection bias is possible. As such, findings may not generalize to a different clinic setting that utilized FBT-PC providers who are less intrinsically motivated to deliver eating disorder care. Conclusion Despite these limitations, data from this case series establish proof-of-concept for FBT-PC implementation in rural primary care clinics. Specifically, the majority of patients either completed treatment or were referred to standard FBT, which is consistent with the intention of FBT-PC as offering a lower step of eating disorder care and/or a bridging intervention while families await treatment. Providers were able to implement the intervention within the parameters of their clinic workflows and resources. Future research is needed to evaluate clinical outcomes and acceptability metrics from the practice as a whole. Acknowledgements The authors would like to thank the patients, families, and primary care providers whose experiences made this study possible. Abbreviations Attention deficit/hyperactivity disorder Body mass index percentile Family-based treatment Family-based treatment for primary care The framework for reporting adaptations and modifications to evidence-based implementation strategies Other specified feeding and eating disorder Primary care provider Author contributions JL conceptualized the study, designed the FBT-PC intervention, helped analyze the data and was a major contributor in writing the manuscript. IC, JSS, LS and RJ helped conceptualize the study, provided guidance around study design and evaluated and contributed to the manuscript. MS, SG and SK helped conceptualize the study, received FBT-PC training and provided FBT-PC services for patients, as well as contributed to the manuscript. AAS and JM helped with study coordination and data collection. All authors read and approved the final manuscript. Funding Research reported in this publication was supported by the National Institute of Mental Health under award number R34MH128296. Data availability The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate All study activities were reviewed and approved by the Mayo Clinic Institutional Review Board. Consent for publication Informed consent and assent were obtained from all human subjects participants. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.