Jennifer J. Thomas, PhD
Associate Professor of Psychology, Department of Psychiatry, Harvard Medical School.
Co-director, Eating Disorders Clinical and Research Program, Massachusetts General Hospital, Boston, MA. Associate Editor for the International Journal of Eating Disorders.
Dr. Thomas has disclosed no relevant financial or other interests in any commercial companies pertaining to this educational activity.
CCPR: Avoidant/restrictive food intake disorder (ARFID) is still a fairly new disorder, added to the DSM-5 in 2013. What exactly is it?
Dr. Thomas: ARFID is a newly designated feeding or eating disorder where the person does not eat enough food for a reason unrelated to body image problems, environmental factors, cultural reasons, or medical conditions. ARFID occurs at any age—from very young children to adults—but its onset is typically earlier than other eating disorders and it often appears before puberty, where it can be confused with developmental pickiness. That’s a problem because ARFID is as common and serious as other eating disorders.
CCPR: What was the reason for including ARFID in the DSM-5?
Dr. Thomas: The ARFID diagnosis facilitates treatment and research. Prior to the DSM-5, at least half of patients with eating disorders had “eating disorder not otherwise specified.” Some had picky eating, weight loss, difficulty gaining weight, or fear of eating, but without body image disturbances. In the DSM-IV, “feeding disorder of infancy and early childhood” captured underweight children before age 6, which we now diagnose as ARFID. However, ARFID includes children with these symptoms who are not underweight and are of all ages.
CCPR: DSM-5 includes three kinds of ARFID. Can you describe them?
Dr. Thomas: The three prototypical presentations of ARFID are: 1) sensory sensitivity leading to highly selective eating, like you might see with autism spectrum disorder (ASD); 2) lack of hunger or lack of interest in eating or food, which can sometimes be confused with anorexia nervosa; and 3) a fear of aversive consequences like choking or vomiting, which is similar to other anxiety disorders. Patients can have more than one of these presentations.
CCPR: Can you give an example?
Dr. Thomas: I treated an adolescent male with all three presentations—he only consumed hot dogs, bread, and Ensure nutrition drinks. This wasn’t about his body image. He was terrified that any other foods might give him gastrointestinal distress, he didn’t like foods with unfamiliar tastes or textures, and he had lost touch with understanding when he was hungry.
CCPR: How common is ARFID?
Dr. Thomas: We don’t have large studies, but the research suggests that about 3% of children and adolescents may have ARFID (Kurz S et al, Eur Child Adolesc Psychiatry 2015;24(7):779–785). For adults, it’s less than 1%, similar to anorexia nervosa (Hilbert A et al, Int J Eat Disord 2021;54(3):399–408).
CCPR: How does ARFID affect everyday life for kids and their families?
Dr. Thomas: It limits a child’s ability to socialize normally, and it’s time consuming and frustrating for the family. A child might say, “I can’t go to summer camp because I don’t know about the food there.” Or, “I can’t eat anything at the cafeteria or dining hall.” Parents might report taking a suitcase of granola bars with them when the family goes on a trip. Or they’ll drive three hours to the nearest fast food chain where the child will eat.
CCPR: What about teens and young adults?
Dr. Thomas: Parties, dates, and work meetings involve eating. Teens and young adults can feel self-conscious about eating foods that other people see as “kid foods.” People with ARFID often eat bland, “white” foods like French fries, mac and cheese, or vanilla ice cream; they’ll also often stick to a particular brand. That is challenging when you’re expected to eat whatever is served.
CCPR: Are there screeners we can use for ARFID?
Dr. Thomas: The Eating Disorders in Youth-Questionnaire (EDY-Q) is validated for children ages 8–13 years (www.tinyurl.com/navfcjhb). Our structured interview, the Pica, ARFID, and Rumination Disorder Interview (PARDI), takes 45 minutes, which is too long for busy clinicians to use, and the Nine Item ARFID Screen (NIAS) isn’t validated in children (Bryant-Waugh R et al, Int J Eat Disord 2019;52(4):378–387). So the clinical interview is key for diagnosis.
CCPR: How do you ask about ARFID?
Dr. Thomas: Ask, “Do you think you have a problem with your eating? Do you think you’re not getting enough food overall, or enough different kinds of foods?” Sometimes young children don’t feel they have a problem at all. Ask the parents, who might tell you they go to 10 stores to find the only yogurt their child will eat. See if the child has fallen off their growth curve for height, weight, or body mass index (BMI). (Editor’s note: When it comes to body positivity, BMI is problematic, but in this context BMI matters from a medical perspective.)
CCPR: What do you look for on a growth curve?
Dr. Thomas: Growth curves can have errors, and growth spurts can happen at different ages. But I’d worry if a child has been a 50th percentile kid and now they’re at the 20th percentile. Think about the family’s stature—if both parents are tall and the child is much shorter than expected, that’s a red flag. ARFID is trickier with younger patients. The child’s weight might slowly go down, or maybe it’s always been low. Have they always been small, or have they always had ARFID? Some children are always particular, even about infant formula when they’re extremely young. Others are OK until 2 years old and then start declining food.
CCPR: Can ARFID obscure anorexia nervosa?
Dr. Thomas: Yes, sometimes ARFID obscures anorexia nervosa, and at other times anorexia nervosa develops from ARFID once a child hits puberty. You might have a 12-year-old girl with very low weight and a selective diet who isn’t worried about getting fat. Or that same 12-year-old may have anorexia all along and say, “I don’t like ice cream.” Another patient might gain weight then worry about their changing appearance, and we move from addressing ARFID to addressing body image. But in all of these cases, regardless of the diagnosis, a parent might not recognize the severity of the eating disorder, saying, “They don’t want to eat. We can’t force it.”
CCPR: Anorexia and ARFID can seem so similar. How can we tease out anorexia from ARFID?
Dr. Thomas: Look at the foods that the child prefers. Kids with ARFID eat more carbohydrates and fewer vegetables and proteins compared to healthy kids (Harshman SG et al, Nutrients 2019;11(9):2013). Kids with anorexia nervosa trend more toward fruits and vegetables and are afraid of eating mac and cheese or candy. If someone’s underweight and they come in and they’re eating mac and cheese and ice cream, I’m thinking ARFID. If they’re only eating salad, I’m thinking anorexia.
CCPR: What are the treatments for ARFID?
Dr. Thomas: Regardless of presentation (sensory sensitivity, lack of interest in eating or food, or fear of aversive consequences), the primary intervention is exposure to previously avoided foods. Clinicians use behavioral therapy, nutrition counseling, occupational therapy, and speech therapy, but few interventions are rigorously evaluated. At Mass General, we developed a cognitive behavioral therapy for ARFID (CBT-AR) for ages 10 and up (Thomas JJ et al, Curr Opin Psychiatry 2018;31(6):425–430). We borrow techniques used for pediatric feeding, anxiety, and eating disorders, and family-based treatment techniques. We track how much the patient is eating and whether they have the same food every day. We look at the details of diet restriction. If the person’s diet is dominated by grains and dairy without a lot of fruits and vegetables, we’ll give folks a long list of foods to consider adding to their diet.
CCPR: What other conditions co-occur with ARFID that we need to think about?
Dr. Thomas: More than 40% of children and teens with ARFID have a history of anxiety disorder (Kambanis PE et al, Int J Eat Disord 2020;53(2):256–265). There is also overlap with ASD (Yule S et al, J Acad Nutr Diet 2021;121(3):467–492). ARFID treatment often utilizes techniques from anxiety disorder treatment (eg, exposure and response prevention) to address fears around food, but some patients need treatment for other anxiety disorders after remitting from ARFID itself. In ARFID we also see low iron as well as nutritional deficiencies like scurvy that pediatricians rarely look for. Patients with comorbid ARFID and ASD can experience even more severe impact on their nutrition, growth, and weight.
CCPR: How can we explain ARFID to parents?
Dr. Thomas: Tell them, “ARFID is a disorder; it’s not just your child being picky or stubborn.” It is not anorexia nervosa, bulimia nervosa, or a body image issue. We walk families through our free online workbook (www.tinyurl.com/4nwtwuvm) and our treatment approach. We use pictures that represent the three presentations of ARFID: 1) for sensitivity and selectivity, a picture of a kid who doesn’t want broccoli; 2) for fear of aversive consequences, a picture of someone choking; and 3) for lack of interest in food, a photo of someone looking uninterested in their food. We ask the parents if their child is in one or more of these categories.
CCPR: How do you talk with the child about treatment?
Dr. Thomas: For kids with selective diets, rather than saying, “I think you need to add broccoli to your diet,” I might say, “Here are 50 vegetables. Which ones might you be willing to try or learn about?” We use nonjudgmental exposure: “What does this food look like? What does it feel like?” We have them do tastings (“What does it taste like?”) and then try to move from tasting to incorporating those foods into their diet. For children who have fears about choking or vomiting, we might create a hierarchy of foods they worry about and figure out how certain they are that the bad outcome will occur. We help them learn what would really happen when they eat those foods. For kids who aren’t interested in eating, we help them habituate to body sensations associated with feeling full (eg, chugging water to get used to feeling full). We borrow ideas from depression behavioral activation literature (eg, have them eat their favorite foods for pleasure). We also do relapse prevention.
CCPR: How long is treatment?
Dr. Thomas: For CBT-AR, we originally planned 20 sessions for non-underweight patients and up to 30 sessions for those who were underweight. Some patients need more sessions just to stay on their growth curve. Now, we are shortening CBT-AR, focusing on the most powerful interventions earlier in treatment with good results in as few as 15 sessions. We are developing a stepped model for families in which they can use an app to do exposures and add more support as needed.
CCPR: Are there other approaches to ARFID treatment?
Dr. Thomas: In our approach, kids drive variety of food and parents drive volume. Dr. Jim Lock developed a family-based treatment for ARFID, where parents lead the intervention. Dr. Will Sharp has a day hospital program using behavioral analysis at the bite level—take a spoon and follow the patient’s mouth until they accept it, giving immediate, tiny reinforcements. Dr. Rachel Bryant-Waugh has care pathways to determine when to refer to behavioral therapy versus other interventions, such as occupational therapy or speech therapy. The only randomized controlled trials (RCTs) are a small one from Dr. Sharp on intermediate outcomes of his day hospital program, and another small one from Dr. Lock on family-based treatment (Sharp WG et al, J Pediatr Gastroenterol Nutr 2016;62(4):658–663; Lock J et al, Int J Eat Disord 2019;52(6):746–751). Dr. Lock has a larger one ongoing as well. We hope our RCT of CBT-AR will be funded soon.
CCPR: Are there medications for ARFID or comorbid conditions?
Dr. Thomas: There are no FDA-approved medications for ARFID. Some clinicians try medications known for increasing appetite such as mirtazapine, cyproheptadine, and even antipsychotics. There are no RCTs, and in a couple of uncontrolled case series cyproheptadine and olanzapine appeared to help kids eat more and gain weight, but it is not clear that the kids had recovered from ARFID (Sant’Anna AM et al, J Pediatr Gastroenterol Nutr 2014;59(5):674–678; Brewerton TD et al, J Child Adolesc Psychopharmacol 2017;27(10):920–922). Our team might prescribe antianxiety or antidepressant medication for a co-occurring disorder or to help kids with mealtime. Our research shows that kids with ARFID have lower levels of ghrelin (an appetite-stimulating hormone) than equally weighted individuals with anorexia nervosa who are at similarly low weights, so it might be worth exploring an agonist (Becker KR et al, Psychoneuroendocrinology 2021;129:105243). (Editor’s note: We will examine a study on medication treatments for ARFID in an upcoming issue of CCPR.)
CCPR: What does relapse prevention look like?
Dr. Thomas: At the end of CBT-AR, we celebrate successes—restoring weight, correcting nutrition deficiencies, adding new foods—and we reinforce the strategies the kids used to change (eg, food logging, exposures). We identify ways to keep expanding their diet after treatment ends. We want them to continue the lifelong adventure of learning about new foods. If you stop trying new things in elementary school, think of everything you might miss out on. I didn’t start eating mushrooms until college and drinking coffee until graduate school, and now they are my favorites!
CCPR: What kind of outcomes are you seeing?
Dr. Thomas: About two-thirds of kids in our uncontrolled trial had remission from ARFID, as did about half of adults (Thomas JJ et al, Int J Eat Disord 2020;53(10):1636–1646; Thomas JJ et al, J Behav Cogn Ther 2021;31(1):47–55). I think we’re more successful with kids because ARFID starts early and may be chronic by adulthood. Also, with kids we can involve the parents to support change. Although ARFID can create a lot of difficulties for patients and families, I think there’s every reason to be hopeful that people of all ages can recover.
CCPR: Thank you for your time, Dr. Thomas.
PO Box 626, Newburyport MA 01950