Piazza et al. (2021) Chapter 25 - Pediatric Feeding Disorders PDF

CHAP TER 25 A Behavior-Analytic Approach to Pediatric Feeding Disorders Cathleen C. Piazza and Caitlin A. Kirkwood Feeding is unique relative to other behaviors dis- Not surprisingly, nature has even provided some cussed in this book, because food1 functions as organisms wi...

CHAP TER 25 A Behavior-Analytic Approach to Pediatric Feeding Disorders Cathleen C. Piazza and Caitlin A. Kirkwood Feeding is unique relative to other behaviors dis- Not surprisingly, nature has even provided some cussed in this book, because food1 functions as organisms with the ability to adapt to unpredict- an appetitive stimulus, and feeding is ubiquitous able changes in food availability. For example, across species. Organisms require relatively con- some bird species that typically exhibit a narrow stant caloric intake, and metabolic output cannot range of foraging behavior develop strategies to exceed caloric intake for long. Many organisms identify alternative food sources when the food have adapted unique mechanisms for the habitat in their typical habitat becomes scarce (Diquelou, in which they feed that promote consistent acqui- Griffin, & Sol, 2016). sition of calories and nutrients. For example, the Even when changes occur that disrupt food hyacinth macaw, a native parrot of South Ameri- availability, motivation to feed is typically not ca, has a beak that can exert hundreds of pounds disrupted (Diquelou et al., 2016). In fact, profes- of pressure per square inch; this beak is ideal for sionals in the United States are often so confident cracking the hard shells of the palm nuts that are about the reinforcing properties of food that their a major component of the bird’s diet (Borsari & response to a caregiver of a poorly growing child is Ottoni, 2005). Some organisms have adaptations “The child will eat when he [or she] gets hungry.” that allow them to reduce metabolic needs, store But are there exceptions to this time-worn adage? nutrients, or both to respond to predictable chang- And if so, why does this happen and what are the es in their habitat that affect food availability. For stimulus conditions under which it occurs? example, black bears suppress their basal meta- bolic rate by 25% during the 5–7 months per year that they hibernate (Toien et al., 2011). Despite an DEFINITION organism’s adaptations to feed in a specific habitat or during predictable environmental changes, un- We use the term feeding disorder for children who predictable changes in environmental conditions, do not consume sufficient calories, hydration, or such as drought, may threaten the organism’s abil- nutrition to gain weight and grow, to maintain ity to access calories and nutrients consistently. hydration, or to meet their nutritional needs for macro- and micronutrients. Feeding disorders are 1 Weuse the term food here to mean an organism’s source of heterogeneous and may include refusal to eat; energy (kilocalories) and nutrients. refusal to eat certain types or textures of food; 427 428 S U B S P E C I A LT IE S IN A P P L IE D B E H AV I O R A N A LY S I S dependence on a limited or developmentally in- thrust, pocketing food, poor lip closure, refusal, appropriate source of nutrition, such as bottle de- spitting out food, and vomiting (Arvedson & pendence in a 3-year-old child; and skill deficits, Brodsky, 2002). such as inability to self-feed or transition to age- or developmental-stage-appropriate textures. The Di- agnostic and Statistical Manual of Mental Disorders, DEPENDENT VARIABLES fifth edition (DSM-5; American Psychiatric As- sociation, 2013) uses the term avoidant/restrictive Feeding consists of a complex chain of behav- food intake disorder to refer to the following: (1) iors that includes placing bites or drinks into the A child exhibits a feeding or eating disturbance mouth; lateralizing the food to the molars and characterized by persistent failure to meet appro- chewing, if necessary; forming a bolus (a mass of priate nutritional needs, energy needs, or both, food or drink) on the tongue; elevating the tongue; with significant weight loss, significant nutritional and propelling the bolus to the pharynx (Derkay deficiency, need for enteral feeding or oral nutri- & Schechter, 1998). A child with a feeding disor- tional supplements, or obvious interference with der may have difficulties with one or more of the psychosocial functioning; (2) the disturbance is behaviors in the chain. Thus intervention may not better explained by a lack of available food or target one or more problematic feeding behaviors, by an associated culturally permissible practice; (3) and intervention for one behavior may affect the the eating disturbance does not occur only during occurrence of other behaviors. We provide the the course of anorexia nervosa or bulimia nervosa, reader with operational definitions and a discus- and there is no evidence of a disturbance in the sion of dependent variables that researchers have child’s experience of body weight or shape; and (4) used; we realize that the list is not exhaustive and the eating disturbance is not attributed to a con- that the field does not have standard operational current medical condition or better explained by definitions for these behaviors. another mental disorder, and if the eating distur- Acceptance, the occurrence of the bite or drink bance occurs in the context of another condition entering the mouth, is often the first in the chain or disorder, its severity exceeds what is routinely of feeding behaviors that researchers target for in- associated with that condition or disorder and tervention. In LaRue et al. (2011), observers scored warrants additional clinical attention. the occurrence of acceptance if a child leaned to- A diagnostic nosology such as DSM-5 is help- ward the spoon or cup and opened his or her mouth ful in describing the characteristics of a feeding in the absence of negative vocalizations and inap- disorder, but it tells us little about why feeding propriate behavior, so that the feeder could deposit disorders occur or how to develop effective inter- the entire bolus of food (except for an amount ventions. Although multiple children may have smaller than the size of a pea) or any amount of liq- the same topographical expression of a feeding uid in the mouth within 5 seconds of presentation. disorder, such as poor weight gain, the etiology of Notice that LaRue et al.’s definition included qual- the feeding disorder may be different across chil- ifiers such as “leaned toward the spoon or cup in dren. For example, one poorly growing child may the absence of negative vocalizations” and “within have delayed gastric emptying, complain of being 5 seconds of presentation.” LaRue et al. included full, and refuse to eat after consuming a small these qualifiers to ensure that observers were amount of food (Stein, Everhart, & Lacy, 2015). measuring children’s rather than feeders’ behav- Another child, by contrast, may have oral–motor ior. Observers scored acceptance if a child leaned skill deficits. Although the child eats frequently forward soon after the bite or drink presentation and appears motivated to eat (e.g., asks for food), and opened his or her mouth so that the feeder meals are lengthy, and the child cannot consume could put the bite or drink in the child’s mouth. sufficient calories to gain weight and grow. A third By contrast, observers did not score acceptance child may vomit frequently, and poor weight gain if the child’s mouth was open because he or she is an indirect result of gastroesophageal reflux dis- was crying, which gave the feeder the opportunity ease. Within those broad topographical presenta- to put the bite or drink in the mouth. Ideally, in- tions of feeding disorders, individual children may tervention increases feeding compliance, and dis- exhibit different behaviors that contribute to their tinguishing between child compliance and feeder feeding disorders, such as excessive saliva produc- behavior is important for determining progress. tion, inability to chew, inability to lateralize the Observers in LaRue et al. scored the occurrence of tongue, an open-mouth posture, persistent tongue acceptance for each bite or drink presentation and  A Behavior-Analytic Approach to Feeding Disorders 429 converted occurrences to a percentage after divid- tion the utensil was in arm’s reach of the child. ing the occurrences of acceptance by the number Controlling for the presence of the utensil may be of presented bites or drinks. important in cases where the duration of utensil Children in Peterson, Piazza, and Volkert presence may differ markedly from phase to phase (2016) self-fed, and “observers scored the occur- or from condition to condition. rence of acceptance when the child used the uten- Expulsion is food or liquid exiting the mouth. sil or his fingers to put the entire bite of food in his For example, Wilkins, Piazza, Groff, and Vaz mouth within 8 s of presentation, not including (2011) defined expulsion for liquids as each time placement of the bite in the mouth during re-pre- any amount of liquid (pea size or larger) that the sentation” (p. 5). Peterson et al. converted occur- child had not swallowed was visible outside the lips rences of acceptance to a percentage after dividing after any amount of liquid had passed the plane of the number of accepted bites by the number of the lips; and expulsion for solids as each time any presented bites. Peterson et al. described the bite amount of food (pea size or larger) that the child presentation as one piece of food measuring 0.6 had not swallowed was visible outside the lips after × 0.6 × 0.6 centimeters, and the feeder restricted the entire bolus of food had passed the plane of the operant by presenting only one bite at a time. the lips. Observers in Wilkins et al. scored the fre- Observers scored acceptance only if the child put quency of expulsion and presented the data as ex- the entire bite in his mouth within 8 seconds of pulsions per bite. As with many feeding behaviors, the feeder placing the bite in front of the child. the opportunity to engage in expulsion will affect Acceptance, in this case, conveyed information rates of expulsion, and the clinician should con- about how much food entered the child’s mouth, sider which method of data presentation is most and it accounted for the number of bites available appropriate to address the clinical problem or re- to the child to accept from session to session. search question. For example, the clinician might Inappropriate mealtime behavior may refer to calculate expulsions per bite if opportunity to numerous behaviors that disrupt feeding and that expel is fixed (e.g., based on number of presented differ along many dimensions. Some researchers bites), but expulsions per opportunity if opportu- have addressed this dilemma by conceptualizing nity to expel varies, such as during re-presentation. inappropriate mealtime behavior as behavior a Mouth clean is a product measure of swallowing child emits with his or her external body parts, that we often use, because swallowing is difficult to such as the head, hands, arms, and legs, that measure reliably in our clinical experience. For ex- prevents solids, liquids, or both from entering or ample, observers in Kadey, Piazza, Rivas, and Zeleny remaining in the mouth. These researchers have (2013) scored “mouth clean if no food larger than a separated inappropriate mealtime behavior from pea was in [the child’s] mouth, unless the absence behaviors the child emits with his or her mouth, of food was the result of expulsion” (p. 540). The such as expulsion and packing, and behaviors feeder in Kadey et al. checked the child’s mouth 30 that have a putative physiological basis, such as seconds after the bite entered the mouth by saying, coughing, gagging, and vomiting. For example, “Show me.” The mouth check gave observers the observers in Ibañez, Piazza, and Peterson (2019) opportunity to score mouth clean or pack, which is scored inappropriate mealtime behavior “when the converse of mouth clean. For example, observ- the utensil was in arm’s reach of the child and the ers in Wilkins et al. (2014) scored “pack if the entire child turned his head 45° or greater away from the bite (with the exception of food smaller than a pea) utensil during a bite or drink presentation; used entered the child’s mouth, and food larger than a his hand to contact the utensil, food or drink, or pea was in the child’s mouth at the 30-s check” the feeder’s hand or arm anywhere from the elbow (p. 4). Kadey et al. and Wilkins et al. converted down while the feeder was presenting the bite or mouth clean or pack, respectively, to a percentage drink; threw food, liquids, or utensils; or blocked after dividing the number of occurrences of mouth his mouth with his hand, bib, or toys” (p. 1009). clean or pack, respectively, by the number of bites Notice that Ibañez et al.’s operational definition that entered the child’s mouth. Note that the de- qualifies that the utensil must be in arm’s reach of nominator in Kadey et al. and Wilkins et al. was the child, because the defined behavior can occur number of bites that entered the mouth, meaning only when a utensil is present. Observers scored that no opportunities for mouth clean or pack oc- the frequency of inappropriate mealtime behavior curred if no bites entered the child’s mouth; this is and converted it to a rate by dividing the number important for readers to remember when interpret- of inappropriate mealtime behaviors by the dura- ing data from studies that use this measure. Note 430 S U B S P E C I A LT IE S I N A P P L IE D B E H AV I O R A N A LY S I S also that Kadey et al. and Wilkins et al. included example, Rommel et al. (2003) characterized the a caveat that the child could have a mouth clean feeding disorders of 700 children referred for as- if a small amount of food remained in the mouth, sessment and treatment of severe feeding difficul- because even typical feeders may have residue in ties as medical (86%), oral–motor (61%), and/or the mouth after swallowing. The researchers based behavioral (18%). Combined causes, such as medi- the size of the residue (pea size in these studies) cal, behavioral, and/or oral–motor, for the feeding on the amount of presented food. In both studies, disorder occurred in over 60% of children. Other bolus sizes were a level small maroon spoon for six researchers have found that feeding disorders have children, a half-level small maroon spoon for four neurological (62%), structural (53%), behavioral children, and a level baby spoon for two children in (43%), cardiorespiratory (34%), and metabolic Wilkins et al., and a level small maroon spoon for (12%) causes, with most children having causes the child in Kadey et al. The size of the acceptable in two or more categories simultaneously (Burk- residue might change if the presented bolus was low, McGrath, Valerius, & Rudolph, 2002; Davis, larger or smaller. For example, the acceptable resi- Bruce, Cojin, Mousa, & Hyman, 2010). Feeding due in Volkert, Peterson, Zeleny, and Piazza (2014) disorders are also prevalent among specific diag- was the size of a grain of rice, because the presented nostic groups, such as children with autism spec- bolus was a piece of food measuring 0.6 × 0.6 × 0.6 trum disorder, cerebral palsy, and Down syndrome centimeters. (Bandini et al., 2010). Although chewing is an important feeding skill, The high prevalence of medical conditions and there is a relative dearth of applied-behavior-ana- oral–motor dysfunction in children with feeding lytic research on teaching children with feeding disorders suggests that biological factors play an disorders to chew. In a notable exception, Volkert important role in the etiology and maintenance of et al. (2014) scored a chew “each time the child’s feeding disorders. Children with chronic medical teeth and/or jaw completed one up-and-down mo- problems that affect the digestive system directly, tion with the teeth parted at least 1.3 cm following such as congenital defects of the gastrointestinal a verbal or model prompt while food was visible tract, delayed gastric emptying, food allergies, anywhere in the mouth except the center of the gastroesophageal reflux disease, malabsorption, tongue or between the front teeth” (p. 708). One or metabolic disorders, may associate eating with limitation of this definition is that it did not dif- fatigue, nausea, pain, or a combination. For exam- ferentiate rotary chews from an immature chewing ple, children with gastroesophageal reflux disease pattern such as munching (Volkert, Piazza, Vaz, & may associate eating with the pain that occurs Frese, 2013). One unique and important feature when excess acid erupts into the esophagus. Nau- of Volkert et al. (2014), however, was that the re- sea plays an important role in the development searchers included a measure of mastication. They of food aversions (Schafe & Bernstein, 1996), defined mastication as food with pieces no larger and when nausea is paired with eating, aversions than 0.2 × 0.2 centimeters in a liquid medium at to tastes may develop after only one or a few tri- the mastication check. The mastication check als, may generalize to different foods, and may be was like the mouth check described by Kadey et highly treatment-resistant. al. (2013), but observers determined whether the Researchers estimate that feeding disorders child had masticated rather than swallowed the occur in 40–70% of children with chronic medical bite. One important future direction for research conditions (Davis et al., 2010; Douglas & Byron, on pediatric feeding disorders is to develop more 1996; Lukens & Silverman, 2014; Thommessen, sophisticated measures of chewing, perhaps using Heiberg, Kase, Larsen, & Riis, 1991), suggesting automated methods (Hadley, Krival, Ridgel, that the presence of other chronic medical prob- Hahn, & Tyler, 2015), which will allow investiga- lems, such as bronchopulmonary dysplasia, may tors to determine whether a child’s chewing skills contribute to the etiology of feeding disorders. In- are appropriate for the child’s age, developmental fants with complex medical histories are subject to level, or a combination. numerous invasive diagnostic tests and procedures that involve manipulation of the face and mouth, such as laryngoscopy. Such a child may come to ETIOLOGY associate the presentation of items to the mouth with discomfort, pain, or both. From the child’s Research has suggested that the etiology of feeding perspective, a spoon may be indistinguishable from disorders is complex and multifactorial (Rommel, a laryngoscope or other devices that professionals Meyer, Feenstra, & Veereman-Wauters, 2003). For use during invasive medical procedures and tests.  A Behavior-Analytic Approach to Feeding Disorders 431 Caregivers of chronically hospitalized and medi- that the cause of feeding disorders is multifacto- cally fragile children often report oral aversions rial (Rommel et al., 2003). The goal of evaluation that affect feeding and other activities associated should be to determine whether anatomical, medi- with the face and mouth, such as face washing and cal, or oral–motor skill deficits contribute to the tooth brushing. child’s feeding disorder and whether the child is Oral–motor dysfunction may include difficul- a safe oral feeder. Members of such a team might ties sucking, difficulties with bolus propulsion, include a behavior analyst, a dietitian, a gastro- inability to lateralize food from one side to an- enterologist or the child’s primary physician, and other, difficulties swallowing, and tongue thrust, an occupational or speech therapist. A behavior and these problems may affect a child’s ability and analyst should not underestimate the negative motivation to eat (Darrow & Harley, 1998). The consequences of feeding a child before appropri- child’s refusal to eat may cause or exacerbate pre- ate evaluation. At a minimum, the behavior ana- existing oral–motor dysfunction and further con- lyst should consult the child’s primary physician, tribute to the child’s failure to develop appropriate describe the proposed course of assessment and oral–motor skills. That is, a child who refuses to treatment, and obtain medical clearance to start eat does not have the opportunity to practice the therapy. Failure to identify a medical condition or skill of eating and does not develop the oral–motor oral–motor skill deficits before beginning therapy skills to become a competent eater. could result in worsening of the feeding disorder or When eating is paired with an aversive experi- even death. For example, introducing food variety ence, the child may develop refusal behavior such to promote proper nutrient intake is a reasonable as batting at the spoon, head turning, or covering goal for a child with a feeding disorder, but a re- the mouth to avoid eating. These behaviors may action to an unidentified food allergy could cause increase in frequency as a function of caregiver anaphylaxis, which is a severe, potentially life- responses to child behavior during meals. Bor- threatening event (Sicherer & Sampson, 2010). rero, Woods, Borrero, Masler, and Lesser (2010) In addition, some children with oral–motor dys- conducted observations of 25 children with feed- function do not manage specific consistencies of ing disorders and their caregivers to describe and solids, liquids, or both, and other children are not quantify the caregivers’ responses to the children’s safe oral feeders with solids or liquids of any consis- inappropriate mealtime behavior. Researchers tency. Oral–motor dysfunction may be associated compared conditional probabilities of a caregiver with aspiration due to solids or liquids entering the providing attention, escape, or a tangible item fol- airway, which can cause medical problems such as lowing refusal or acceptance to the unconditioned pneumonia. An evaluation by a swallow special- probabilities of each event. Observations indicated ist, usually an occupational or speech therapist, that caregivers coaxed, removed the spoon, threat- can determine whether oral–motor dysfunction ened to take away preferred items, presented pre- may be causing or contributing to a child’s feed- ferred foods, or engaged in a combination of these ing disorder and determine whether the child is events following refusal behavior. Escape in the a safe oral feeder (Schwarz, Corredor, Fisher-Me- form of spoon removal or meal termination and dina, Cohen, & Rabinowitz, 2001). A dietitian attention in the form of coaxing, reprimands, and calculates the child’s caloric, hydrational, and nu- statements of concern most frequently followed tritional needs and determines whether the child refusal behavior. Similarly, Piazza, Fisher, et al. requires diet modifications. For example, a child (2003) observed caregivers and children with feed- with a glycogen storage disorder requires careful ing disorders during meals. Caregivers responded monitoring of blood sugar levels and a diet that to inappropriate mealtime behavior with one or restricts simple sugars. Drops in blood sugar levels more of the following consequences: (1) providing may cause seizures, coma, and even death (Goto, escape from bites of food or the meal, (2) coaxing Arah, Goto, Terauchi, & Noda, 2013). or reprimanding (e.g., “Eat your peas, they are good Results of the interdisciplinary evaluation may for you”), or (3) providing a toy or preferred food. indicate that the child needs medical treatment, consistency or texture manipulations, a special diet, or a combination, and these interventions EVALUATION may resolve the child’s feeding disorder. Some chil- dren, however, may not start feeding or may not A behavior analyst should consider referring a feed well even after interdisciplinary intervention, child with a feeding disorder for evaluation by particularly if they engage in refusal behavior. In an interdisciplinary team, given the evidence these cases, a qualified behavior analyst can ma- 432 S U B S P E C I A LT IE S I N A P P L IE D B E H AV I O R A N A LY S I S nipulate mealtime antecedents and consequences pared variations of extinction matched to the to determine whether they affect the child’s refusal results of each functional analysis. Both analyses behavior. identified escape and attention as reinforcement Piazza, Fisher, et al. (2003) used analogue func- for inappropriate mealtime behavior, and es- tional analyses to assess the effects of caregiver cape and attention extinction were necessary to consequences on child behavior. Inappropriate achieve a clinically acceptable outcome for one mealtime behavior, such as batting at the spoon participant. For the other two participants, the and head turning, produced attention, such as Piazza, Fisher, et al. procedure identified multiple coaxing and brief verbal reprimands, during the reinforcers for inappropriate mealtime behavior, attention condition; a break from the bite or drink but the Najdowski et al. procedure identified only presentation during the escape condition; access to escape as the reinforcer for inappropriate mealtime a tangible item, such as a preferred food, during behavior. For those two participants, the interven- the tangible condition; and no differential conse- tion matched to the reinforcers identified by the quence during the control condition. Escape from Piazza, Fisher, et al. procedure produced a clinically bite or drink presentations functioned as negative acceptable outcome, but the intervention matched reinforcement for the inappropriate mealtime be- to the reinforcer identified by the Najdowski et al. havior of 9 of the 10 children who showed differ- procedure did not. ential responding during the functional analyses. The differential responding demonstrated by Access to adult attention or tangible items func- participants across test and control conditions tioned as positive reinforcement for the inappro- in Bachmeyer et al. (2019), Girolami and Scotti priate mealtime behavior of 8 of the 10 children (2001), Najdowski et al. (2008), and Piazza, Fisher, who showed differential responding during the et al. (2003) suggest that even if the etiology of a functional analyses. Girolami and Scotti (2001) pediatric feeding disorder is multiple and complex, found that escape from food presentation and environmental events may reinforce inappropriate mealtime demands for two children, and contin- mealtime behavior. This finding is important, be- gent access to toys and attention for one child, cause (1) we may not be able to identify the cause functioned as reinforcement for mealtime behav- of the child’s feeding disorder; (2) even if we iden- ior problems such as aggression and spitting out tify the cause, that cause may be immutable, such food. Najdowski et al. (2008) trained caregivers to as a history of prematurity; and (3) the underlying conduct functional analyses in which a caregiver cause may not be related to the condition(s) that placed a plate of nonpreferred food on the table maintains the behavior (Iwata et al., 1982/1994). in front of a child in the attention, escape, and We can, however, change how we respond to the tangible conditions and preferred food in the con- child’s inappropriate mealtime behavior, and such trol condition. The caregiver washed dishes in the changes may be effective as treatment. attention and control conditions and provided at- tention following inappropriate mealtime behav- ior in the attention condition. The caregiver sat TREATMENT next to the child; provided continuous prompts to “Take a bite”; followed the child with the plate if Evaluations of treatments based on theories of op- the child left the chair; used three-step prompting; erant conditioning have formed the bulk of the in- and removed the bite if the child engaged in inap- tervention research on pediatric feeding disorders. propriate mealtime behavior in the demand con- Kerwin (1999) surveyed peer-reviewed medical dition. Unlike in the Piazza, Fisher, et al. (2003) and psychological journals to identify studies that and Girolami and Scotti (2001) studies, escape reported on psychosocial or behavioral interven- was the only reinforcer identified for inappropriate tions for pediatric feeding disorders. She used the mealtime behavior. Najdowski et al. noted that the modified criteria of the Task Force on Promotion experimental preparation they used might have and Dissemination of Psychological Procedures accounted for the difference in findings. (1995) to identify methodologically rigorous stud- Bachmeyer, Kirkwood, Criscito, Mauzy, and ies that met the criteria and to classify interven- Berth (2019) conducted two functional analyses tions for pediatric feeding disorders that were well with three children with feeding disorders: one established, probably efficacious, or promising. with the procedure Piazza, Fisher, et al. (2003) de- Analysis of the 29 studies that met the criteria scribed and one with the procedure Najdowski et indicated that the only well-established interven- al. (2008) described. Bachmeyer et al. then com- tions were behavioral interventions that included  A Behavior-Analytic Approach to Feeding Disorders 433 (1) positive reinforcement of appropriate feeding pediatric feeding disorders. Researchers have used behavior and ignoring inappropriate mealtime be- differential reinforcement in several ways that havior, and (2) positive reinforcement of appropri- include, but are not limited to, immediate (Patel, ate feeding behavior and physical guidance of the Piazza, Martinez, Volkert, & Santana, 2002) or appropriate feeding behavior (e.g., Ahearn, Ker- delayed (Kern & Marder, 1996; Riordan et al., win, Eicher, Shantz, & Swearingin, 1996; Kerwin, 1984) reinforcement with stimuli that research- Ahearn, Eicher, & Burd, 1995; Linscheid, Oliver, ers selected arbitrarily (Kern & Marder, 1996; Blyler, & Palmer, 1978; Piazza, Anderson, & Fish- Casey, Cooper-Brown, Wacker, & Rankin, 2006) er, 1993; Riordan, Iwata, Finney, Wohl, & Stan- or with systematic preference assessments (Buck- ley, 1984; Riordan, Iwata, Wohl, & Finney, 1980; ley, Strunck, & Newchok, 2005), or with tokens Stark, Powers, Jelalian, Rape, & Miller, 1994). the child could exchange for meal discontinuation Volkert and Piazza (2012) extended Kerwin (Kahng, Boscoe, & Byrne, 2003). For example, Pe- (1999), using the same criteria to identify stud- terson, Volkert, and Zeleny (2015) used differen- ies of interventions for pediatric feeding disorders tial reinforcement for two children with a feeding and to categorize the level of empirical support for disorder to increase self-drinking from a cup. Re- those interventions. Volkert and Piazza identified searchers conducted a multiple-stimulus-without- 74 studies that met the inclusion criteria. Analysis replacement assessment before each session and of those studies showed that differential reinforce- used the three most preferred stimuli as reinforce- ment of alternative behavior, escape extinction ment for self-drinking. The researchers increased and putative escape extinction, and physical guid- the amount of liquid in the cup after the child’s ance for self-feeding were well-established inter- self-drinking with the smaller amount increased ventions; non-nutritive sucking (Field et al., 1982; with differential reinforcement. Sehgal, Prakash, Gupta, Mohan, & Anand, 1990) Stark et al. (1996) randomly assigned nine chil- and oral stimulation (Fucile, Gisel, & Lau, 2002; dren with cystic fibrosis to a behavioral interven- Rocha, Moreira, Pimenta, Ramos, & Lucena, tion or a wait-list control group and used calories 2007) were probably efficacious interventions; and consumed and weight gain as the dependent vari- oral support (Boiron, Nobrega, Roux, Henrot, & ables. The multicomponent behavioral interven- Saliba, 2007; Einarsson-Backes, Deitz, Price, Glass, tion included caregiver praise, a star chart, and & Hays, 1994), stimulus fading (e.g., Shore, Bab- access to privileges for appropriate feeding be- bit, Williams, Coe, & Snyder, 1998), and simul- havior. Caloric intake increased to 1,032 calories taneous presentation without escape extinction per day and mean weight gain was 1.7 kilograms (e.g., Ahearn, 2003; Buckley & Newchok, 2005; for the group receiving the behavioral interven- Piazza et al., 2002) were promising interventions. tion, compared to 244 calories and 0 kilograms, Like Kerwin, Volkert and Piazza found that most respectively, for the control group. Participants empirically supported interventions were behav- maintained higher levels of caloric intake relative ior-analytic. Unlike Kerwin, however, they found to baseline at 3- and 6-month follow-ups. Other that three interventions (non-nutritive sucking, researchers have used differential reinforcement oral stimulation, and oral support) did not incor- alone and in combination with other procedures, porate behavior-analytic principles or procedures. such as response cost (Kahng, Tarbox, & Wilke, These studies appeared in journals from the fields 2001), to increase acceptance of solids (Werle, of developmental medicine, occupational therapy, Murphy, & Budd, 1993), liquids (Kelley, Piazza, otorhinolaryngology, and pediatrics, and focused & Fisher, 2003), or both (Roth, Williams, & Paul, on increasing oral intake in premature infants. For 2010). One study demonstrated increased accep- the purposes of this chapter, we review interven- tance and decreased self-injurious behavior during tions aimed at increasing acceptance, decreasing noncontingent reinforcement (Wilder, Normand, inappropriate mealtime behavior, decreasing ex- & Atwell, 2005). pulsion, increasing mouth clean and decreasing Manipulating antecedents is another method packing, and teaching chewing skills. that researchers have used to increase the food or liquid acceptance of children with feeding dis- orders. For example, Meier, Fryling, and Wallace Acceptance (2012) and Patel et al. (2007) used high-probability Results of the analysis by Volkert and Piazza (2012) instructions, such as “Put an empty spoon in your showed that differential reinforcement of alternative mouth,” that were like those of the target behavior behavior was a well-established intervention for (e.g., “Take a bite”) to increase acceptance. Simul- 434 S U B S P E C I A LT IE S I N A P P L IE D B E H AV I O R A N A LY S I S taneous presentation is an antecedent procedure in package. Piazza, Patel, Gulotta, Sevin, and Layer which the feeder presents preferred food and non- (2003) attempted to clarify the relative contribu- preferred food together, like a nonpreferred pea tions of positive reinforcement and putative escape on a preferred chip. For example, Ahearn (2003) extinction. Differential positive reinforcement of added condiments to increase vegetable consump- mouth clean, in which the feeder provided atten- tion, and Tiger and Hanley (2006) added choco- tion and tangible items for mouth clean and inap- late to milk to increase milk consumption. Piazza propriate mealtime behavior produced escape, was et al. (2002) compared simultaneous and sequential not effective for increasing acceptance or decreas- presentation in an extension of Kern and Marder ing inappropriate mealtime behavior. Acceptance (1996). The feeder presented a bite of nonpreferred increased and inappropriate mealtime behavior food on a bite of preferred food in the simultaneous decreased only when the feeder implemented pu- condition or presented the bite of preferred food if tative escape extinction. Inappropriate mealtime the child ate the bite of nonpreferred food in the behavior and negative vocalizations were lower for sequential condition. Acceptance increased for some children during differential reinforcement two of three participants in the simultaneous but and putative escape extinction relative to putative not the sequential condition. Acceptance for the escape extinction alone, but the differences were third participant increased when the feeder imple- often small or not replicated in subsequent phases. mented simultaneous presentation and physical Piazza, Patel, et al. concluded that putative escape guidance, but not when the feeder implemented extinction was necessary to increase acceptance sequential presentation and physical guidance. and decrease inappropriate mealtime behavior, but that differential positive reinforcement for mouth clean may have contributed to lower levels of in- Acceptance and Inappropriate Mealtime Behavior appropriate mealtime behavior, negative vocaliza- Before researchers began conducting functional tions, or both for some children when combined analyses of inappropriate mealtime behavior, they with putative escape extinction. A study by Reed developed procedures for putative escape extinc- et al. (2004) produced similar results for noncon- tion, based on the assumption that escape from tingent reinforcement and putative escape extinc- bites or drinks functioned as negative reinforce- tion, except that the effects of noncontingent re- ment for inappropriate mealtime behavior. The inforcement on inappropriate mealtime behavior procedures researchers have studied most often and negative vocalizations were less robust than are nonremoval of the spoon (Hoch, Babbitt, Coe, those for differential positive reinforcement in Pi- Krell, & Hackbert, 1994) and physical guidance azza, Patel, et al. (2003). (Ahearn et al., 1996), and both involve discon- Peterson, Piazza, Ibañez, and Fisher (2019) con- tinuing the hypothesized response–reinforcer rela- ducted a randomized controlled trial to evaluate tion. Ahearn et al. (1996) showed that nonremov- the effects of a behavior-analytic intervention al of the spoon and physical guidance increased (noncontingent reinforcement and nonremoval of acceptance. During nonremoval of the spoon, the the spoon) relative to a wait-list control to deter- feeder held the bite near the child’s lips until the mine whether the food selectivity of young chil- feeder could deposit the bite into the mouth. Dur- dren with autism spectrum disorder would resolve ing physical guidance, the feeder applied gentle over time without intervention. The researchers pressure to the mandibular junction of the jaw to randomly assigned three children to the behavior- open the mouth and deposited the bite if the child analytic intervention and three children to the did not accept it. Ahearn et al. assessed caregiver wait-list control group. Consumption increased for acceptability for the interventions by asking each the three children in the behavior-analytic group, caregiver which treatment he or she preferred. All but not for the children in the wait-list control caregivers chose physical guidance, which was as- group. sociated with fewer corollary behaviors (such as Researchers based the behavior-analytic inter- disruptions) for all children and with shorter meal ventions described above on the assumption that durations for two of the three children. escape from bites or drinks would function as re- Studies by Hoch et al. (1994) and Ahearn et inforcement for inappropriate mealtime behavior al. (1996) are representative of other studies on in the absence of a formal functional analysis putative escape extinction, in that researchers (Ahearn et al., 1996; Cooper et al., 1995; Hoch included differential or noncontingent reinforce- et al., 1994). By contrast, Bachmeyer et al. (2009) ment (Cooper et al., 1995) in an intervention used a functional analysis to determine that es-  A Behavior-Analytic Approach to Feeding Disorders 435 cape from bites or drinks and adult attention re- an increase in acceptance and mouth clean and inforced the inappropriate mealtime behavior of decreased inappropriate mealtime behavior. The four children with feeding disorders. Bachmeyer researchers gradually increased the ratio of non- et al. then evaluated the effects of variations of preferred to preferred food until the children were extinction that matched one or both functional consuming the nonpreferred food alone. reinforcers: (1) escape extinction and attention Mueller et al. (2004) used a fading component following inappropriate mealtime behavior, (2) to achieve the targeted outcome; other researchers attention extinction and escape following inap- have used various fading procedures with nonre- propriate mealtime behavior, and (3) escape ex- moval of the spoon or physical guidance. Groff, tinction and attention extinction. Results showed Piazza, Volkert, and Jostad (2014) used a syringe that variations of extinction that discontinued to deposit solids and liquids into the mouth of a delivery of the reinforcers for inappropriate meal- child who clenched his teeth during presentation, time behavior identified by the functional analysis gradually increased the volume of solids and liq- (escape and attention), were necessary to reduce uids in the syringe, and then faded from syringe to inappropriate mealtime behavior to clinically ac- spoon for solids and syringe to cup for liquids. Dur- ceptable rates and to increase acceptance to high, ing fading, the researchers taped the syringe to a stable levels. LaRue et al. (2011) tested the effects spoon or a cup, so the tip of the syringe protruded of a negative-reinforcement-based intervention from the tip of the spoon or lip of the cup by 5 cen- (differential negative reinforcement and nonre- timeters, and moved the tip of the syringe and the moval of the spoon with re-presentation) with 11 tip of the spoon or lip of the cup closer together. children whose inappropriate mealtime behavior The child began eating from a spoon and drinking was maintained by escape. Mouth clean produced from the cup without the syringe during probe ses- a 30-second break from bite or drink presentations sions the feeder conducted between fading steps. during differential negative reinforcement. Ac- Other dimensions on which researchers have con- ceptance increased and inappropriate mealtime ducted fading include bite number (Najdowski, behavior decreased when the feeder implemented Wallace, Doney, & Ghezzi, 2003), bottle to spoon nonremoval of the spoon and re-presentation; dif- (Johnson & Babbitt, 1993), high- to low-probabil- ferential negative reinforcement for mouth clean ity demands (Penrod, Gardella, & Fernand, 2012), had no effect on behavior, either alone or in com- liquid to baby food (Bachmeyer, Gulotta, & Piazza, bination with nonremoval of the spoon and re- 2013), preferred to nonpreferred liquid type (Lu- presentation. iselli, Ricciardi, & Gilligan, 2005), liquid volume The studies on escape extinction reviewed (Hagopian, Farrell, & Amari, 1996), portion size above included consequence manipulations such (Freeman & Piazza, 1998), spoon distance (Rivas, as differential reinforcement for acceptance and Piazza, Patel, & Bachmeyer, 2010), food variety nonremoval of the spoon (e.g., Piazza, Patel, et al., (Valdimarsdottir, Halldorsdottir, & Sigurådóttir, 2003; Reed et al., 2004). Researchers also have 2010), spoon to cup (Babbitt, Shore, Smith, Wil- tested antecedent manipulations with escape ex- liams, & Coe, 2001), and texture (Luiselli & Glea- tinction or putative escape extinction to increase son, 1987; Shore et al., 1998). acceptance. Patel et al. (2006) combined high- probability instructions that were like those for Alternative Interventions the low-probability instruction with nonremoval of the spoon to increase acceptance. Dawson et al. Recent research has compared the effects of be- (2003), by contrast, showed that high-probability havior-analytic interventions for pediatric feed- instructions did not differentially affect levels of ing disorders to ones that are popular among acceptance when combined with nonremoval of non-behavior-analytic professionals. For example, the spoon. Addison et al. (2012) compared the effects of a Mueller, Piazza, Patel, Kelley, and Pruett (2004) behavior-analytic intervention to a sensory-inte- used blending, a variation of simultaneous presen- gration intervention. A speech therapist and two tation, with differential or noncontingent posi- occupational therapists developed individual- tive reinforcement and nonremoval of the spoon ized sensory-integration interventions for the two to increase consumption for two children with a participants. Acceptance and amount consumed feeding disorder. Blending or mixing preferred and increased and inappropriate mealtime behavior nonpreferred foods, such as preferred yogurt mixed decreased during the behavior-analytic but not with nonpreferred green-bean puree, produced the sensory-integration intervention. Peterson et 436 S U B S P E C I A LT IE S IN A P P L IE D B E H AV I O R A N A LY S I S al. (2016) conducted a randomized controlled trial seating, in which the feeder reclined the highchair of a behavior-analytic versus a modified sequential- from its upright position. Expulsion decreased and oral-sensory approach to treat the food selectivity of was equivalent for the two interventions. Patel, Pi- six young children with autism spectrum disorder. azza, Santana, and Volkert (2002) evaluated the Peterson et al. randomly assigned three children effects of type and texture of food (Munk & Repp, to the behavior-analytic intervention and three 1994) on expulsion. Rates of expulsion were high- children to the modified sequential-oral-sensory er when the feeder presented meat relative to other intervention, compared the effects across novel, foods, and expulsion decreased when the research- healthy target foods, and tested for generalization ers lowered the texture of meat. of intervention effects. Consumption of novel, Utensil manipulation is another strategy research- healthy target foods increased for the children in ers have used to decrease expulsion (Dempsey, Pi- the behavior-analytic intervention group, but not azza, Groff, & Kozisek, 2011; Gulotta, Piazza, Patel, for the children in the modified sequential-oral- & Layer, 2005; Hoch, Babbitt, Coe, Duncan, & sensory group. Peterson et al. then implemented Trusty, 1995; Volkert, Vaz, Piazza, Frese, & Bar- the behavior-analytic intervention with the chil- nett, 2011). For example, Girolami, Boscoe, and dren previously assigned to the modified sequen- Roscoe (2007) showed that presenting and re-pre- tial-oral-sensory intervention group. Consumption senting bites on a Nuk, which is a bristled utensil of novel, healthy target foods increased during the caregivers use to initiate toothbrushing with in- behavior-analytic intervention, and Peterson et al. fants, reduced expels relative to presentation and observed a potential generalization effect for foods re-presentation of bites on a spoon or on a spoon that had been exposed to the modified sequential- and a Nuk, respectively. Wilkins et al. (2014) com- oral-sensory intervention. pared presentation of bites on a spoon or a Nuk with 12 children during initial intervention. Ac- ceptance increased and inappropriate mealtime Expulsion behavior decreased for 8 of 12 children. Five of the Results of functional analysis studies show that es- 8 had lower levels of expulsion, and 4 of the 8 had cape from bites or drinks functions as negative re- higher levels of mouth clean, when the feeder pre- inforcement for inappropriate mealtime behavior, sented bites on the Nuk. but inappropriate mealtime behavior is probably not the only behavior that produces escape. For Mouth Clean and Pack example, Coe et al. (1997) used nonremoval of the spoon to increase the acceptance of two children Hoch et al. (1994) proposed that the feeder should with a feeding disorder and observed simultane- provide reinforcement for a behavior that occurs ous increases in expulsion. Coe et al. hypothesized early in the chain of feeding behaviors, such as that negative reinforcement in the form of escape acceptance, and then shift reinforcement to a be- from swallowing food reinforced expulsion. Re-pre- havior that occurs later in the chain, such as swal- sentation, or scooping up expelled food and placing lowing. To that end, Patel, Piazza, Martinez, et al. it back into the mouth or getting a new bite of the (2002) compared the effects of differential positive same food, resulted in near-zero levels of expulsion. reinforcement for acceptance versus mouth clean. Although Coe et al. (1997) and Sevin, Gulotta, When differential reinforcement did not increase Sierp, Rosica, and Miller (2002) demonstrated acceptance or mouth clean, the feeder added pu- the effectiveness of re-presentation, others have tative escape extinction. Acceptance and mouth found that re-presentation was not effective con- clean increased and inappropriate mealtime be- sistently. For example, Wilkins et al. (2011) added havior decreased. Patel et al. concluded that the a chin prompt when nonremoval of the spoon plus point in the chain in which the feeder provided re-presentation did not decrease expulsion. During differential reinforcement was not as important as the chin prompt, the feeder placed gentle upward putative escape extinction for increasing accep- pressure on the child’s chin as the feeder deposited tance and mouth clean and decreasing inappropri- the bite or drink during re-presentation, which ate mealtime behavior. reduced expulsion. Shalev, Milnes, Piazza, and Two consequence-based interventions re- Kozisek (2018) compared a modified chin prompt, searchers have used to increase mouth clean and in which the feeder waited for the child’s jaw to decrease packing are redistribution (Girolami et relax and then placed gentle upward pressure on al., 2007; Gulotta et al., 2005; Levin, Volkert, the chin while depositing the drink, with reclined & Piazza, 2014; Sevin et al., 2002; Stubbs, Volk-  A Behavior-Analytic Approach to Feeding Disorders 437 ert, Rubio, & Ottinger, 2017) and a chaser (Vaz, had mouth clean with some pureed foods but not Piazza, Stewart, Volkert, & Groff, 2012). For ex- others. The researchers pureed food in a smoothie ample, Sevin et al. (2002) used a Nuk brush dur- blender during a second texture assessment, and ing redistribution to remove packed food from the levels of mouth clean were higher than with the participant’s mouth and place the food on the other textures. tongue. Redistribution increased mouth clean and Finally, researchers have used fading to increase decreased packing. Volkert et al. (2011) obtained mouth clean and decrease pack. For example, the similar results by using a flipped spoon during redis- goal for the child in Patel, Piazza, Kelly, Ochsner, tribution. The feeder removed packed food with a and Santana (2001) was to increase his intake of spoon, inserted the spoon with the bite into the a calorically dense beverage, Carnation Instant participant’s mouth, turned the spoon 180°, and Breakfast with whole milk. The child refused the dragged the bowl of the spoon along the tongue breakfast drink, but he did drink water. Therefore, toward the lips to deposit the previously packed the researchers added and gradually increased the bite. The feeder in Vaz et al. (2012) gave the child amount of the drink powder in water and subse- a chaser (a liquid or solid the child consistently quently replaced the water with milk. Other di- accepted and swallowed) to reduce packing. The mensions along which researchers have faded to feeder presented the chaser either immediately increase mouth clean and decrease pack are liquid after he or she deposited the target bite into the to baby food (Bachmeyer et al., 2013), spoon to cup child’s mouth for two children, or 15 seconds after (Groff, Piazza, Zeleny, & Dempsey, 2011), syringe he or she deposited the target bite for a third child. to spoon, and syringe to cup (Groff et al., 2014). Researchers in the studies described above used a Nuk or a flipped spoon to redistribute packed Chewing food. Researchers have also evaluated the effects of utensil manipulation as an antecedent inter- Chewing is a skill that emerges in typically eating vention (Dempsey et al., 2011; Sharp, Odom, & children as the caregiver increases the texture of Jaquess, 2012; Stubbs et al., 2017). For example, presented food, which is usually around 12 months Sharp, Harker, and Jaquess (2010) compared the of age. In our experience, many children with feed- effects of presentation on an upright spoon, a ing disorders do not begin chewing at or after 12 flipped spoon, or a Nuk. Levels of mouth clean in- months of age when the caregiver increases food creased for the flipped spoon and Nuk but were not texture. Nevertheless, caregivers often base the clinically acceptable. Other studies have shown texture of presented food on a child’s age rather clinically acceptable increases in mouth clean and than the child’s chewing skills. A mismatch be- decreases in packing with flipped-spoon presen- tween the texture of presented food and the child’s tation (Rivas, Piazza, Kadey, Volkert, & Stewart, chewing skills increases the risk of aspiration, par- 2011; Sharp et al., 2012; Stubbs et al., 2017), Nuk ticularly if the child swallows the food without presentation (Gulotta et al., 2005; Sevin et al., masticating it sufficiently (Patel et al., 2005). Chil- 2002), or a combination of flipped spoon and chin dren who lack appropriate chewing skills may de- prompt (Dempsey et al., 2011). velop inappropriate compensatory behavior, such Texture or food consistency is another anteced- as using the tongue to push food against the roof ent variable that affects levels of mouth clean and of the mouth. We often see this behavior emerge pack (Bachmeyer et al., 2013; Kadey et al., 2013; when the caregiver presents meltable solids, such Patel, Piazza, Layer, Coleman, & Swartzwelder, as cookies, crackers, and chips. The child learns 2005; Sharp & Jaquess, 2009). For example, Kadey that he or she can use the tongue to moisten and et al. (2013) assessed food texture and food type break apart the meltable solid, and this behavior to identify potential causes of one young girl’s does not change when the caregiver presents foods packing. First, the researchers compared levels of that do not melt or break apart with saliva, such as mouth clean with chopped food (table food cut meats. These children reach an impasse in which into small pieces), wet ground food (small chunks they consume meltable solids and small amounts of food in a wet medium), and pureed food (table of more difficult foods, such as pizza, but they can- food blended until smooth). The results showed not advance any further. They often have exces- that levels of mouth clean were highest with pu- sive meal lengths and do not consume sufficient reed food, but even those levels were not accept- calories for weight gain and growth, because their able clinically. When the researchers presented chewing skills are not efficient and effective. We foods individually, they determined that the child have found that teaching a child to chew is the 438 S U B S P E C I A LT IE S I N A P P L IE D B E H AV I O R A N A LY S I S strategy that is most effective for advancing tex- implement a feeding intervention with above 90% ture. integrity; the researchers used verbal and written For example, our group evaluated a multicom- instructions, modeling, video review, and perfor- ponent intervention to increase chews per bite, as- mance feedback during and after in-home feeding sess mastication, and eliminate early swallowing, services. Mueller et al. (2003) evaluated four dif- which observers scored at the mastication check ferent multicomponent training packages to in- if no food was visible in the mouth and the food crease intervention integrity for caregivers imple- was not absent because of expulsion (Volkert et al., menting pediatric feeding interventions. In Study 2014). Caregivers served as feeders and used grad- 1, written protocols, verbal instructions, therapist uated verbal, model, and physical prompting to modeling, and rehearsal training increased care- teach the children in sequential steps to chew (1) givers’ intervention integrity to high levels. Muel- on an empty 7.6-centimeter piece of airline tubing ler et al. then examined the effects of the train- that was 0.6 centimeters in diameter, (2) on a bite ing package’s components in Study 2. Mueller et of food measuring 0.6 × 0.6 × 0.6 centimeters in al. assigned six caregivers to one of three train- the tube, (3) on a strip of food measuring 0.6 × 0.6 ing conditions with two caregivers per condition: × 5.1 centimeters on half of a tube, and (4) on a written protocols and verbal instructions; written strip of food measuring 0.6 × 0.6 × 5.1 centimeters. protocols, verbal instructions, and modeling; and Final steps included presenting a bite of food mea- written instructions, verbal instructions, and re- suring 0.6 × 0.6 × 0.6 centimeters and increasing hearsal. Each training package produced high lev- bite size for one child. els of intervention integrity and maintenance over Volkert et al. (2013) noted that one limitation a 3-month period. Other researchers have taught of the chewing literature is that studies have not caregivers to use general and specific prompts measured mastication. Ensuring that a child has (Pangborn, Borrero, & Borrero, 2013; Werle et al., masticated accepted food is important for mini- 1993), functional-analysis procedures (Najdowski mizing aspiration risk. Volkert et al. used a vocal et al., 2003, 2008), intervention (e.g., demand fad- prompt, “Chew X times,” for one child and provid- ing, differential reinforcement, escape extinction; ed praise if the child met the chew criterion. The Anderson & McMillan, 2001; Najdowski et al., feeder checked 30 seconds after the bite entered 2003, 2010; Pangborn et al., 2013; Seiverling, Wil- the child’s mouth to determine if the child masti- liams, Sturmey, & Hart, 2012), and data collection cated the bite, which Volkert et al. defined as food (Najdowski et al., 2003). with pieces no larger than 0.2 × 0.2 centimeters in a liquid medium after chewing. Chews per bite and mastication increased during the intervention. CONCLUSION The negative health consequences of a feeding Caregiver Training disorder can be serious and substantial and may Caregiver training is one of the most, if not the include dehydration, growth limitation, severe most, important aspect of intervention for pediat- malnourishment, and substantial weight loss ric feeding disorders, as caregivers typically serve (Babbitt et al., 2001; Palmer & Horn, 1978; Piazza as feeders or are present at mealtime. Werle et & Carroll-Hernandez, 2004). Deficits in calories, al. (1993) used several training techniques, such nutrition, or both can cause long-term behavior, as discussion, handouts, role plays, behavioral re- health, and learning problems (Freedman, Dietz, hearsal, verbal feedback, and occasional videotape Srinivasan, & Berenson, 1999). Young children review, to train three caregivers to use specific and may be at greatest risk for the negative impact of general prompts and positive reinforcement. Re- a feeding disorder, as the most damaging effects of sults indicated increased offerings of target foods inadequate caloric intake, poor nutrition, or both and specific prompts for two caregivers, with an occur before age 5, which is a period of critical additional increase in positive attention for a brain development (Winick, 1969). Feeding disor- third caregiver. Werle et al. observed a simultane- ders may also affect a child’s social development, as ous increase in acceptance of target foods across children with feeding disorders often miss impor- children and decreases in food refusal as training tant social opportunities, such as birthday parties, continued. because of their inability or unwillingness to eat. Anderson and McMillan (2001) trained two Feeding disorders often have a negative impact on caregivers of a child with a feeding disorder to families as well, as they may cause caregiver stress  A Behavior-Analytic Approach to Feeding Disorders 439 and depression (Franklin & Rodger, 2003; Singer, we know little about what causes these behaviors. Sing, Hill, & Jaffe, 1990) and are financially costly Are they part of a chain of escape and avoidance to the families and to society (Nebraska Legisla- behaviors, and as we extinguish one behavior, ture, 2009; Williams, Riegel, Gibbons, & Field, does another behavior emerge to take its place? 2007). Are they the result of an oral–motor skill deficit, The etiology of feeding disorders is multiply in which the child lacks the skills to manage sol- controlled and complex (Rommel et al., 2003), and ids or liquids effectively? Are they the result of a the behaviors that constitute a feeding disorder combined etiology? The systematic, data-based ap- are heterogeneous. Current diagnostic nosologies proach that behavior analysts use is ideal for an- describe the characteristics of a feeding disorder swering these questions, but we have yet to apply but are not prescriptive. Historically, researchers them to pediatric feeding disorders. have hypothesized that escape from feeding func- Finally, intervention for a feeding disorder re- tions as negative reinforcement for inappropriate quires knowledge that extends far beyond applied mealtime behavior, based on the results of studies behavior analysis. A behavior analyst should rec- in which putative escape extinction was effective ognize when to consult with another professional, for increasing acceptance and decreasing inap- such as an allergist, pediatric gastroenterologist, propriate mealtime behavior (Cooper et al., 1995; or speech and language pathologist. Inadequate Hoch et al., 1994; Kerwin et al., 1995; Patel, Piazza, training can lead to mistakes in therapy that can Martinez, et al., 2002; Piazza, Patel, et al., 2003; have serious consequences, such as anaphylaxis Reed et al., 2004). Functional analysis studies have due to cross-contamination, aspiration when a confirmed that escape functioned as negative rein- child is not a safe oral feeder for the presented food forcement for the inappropriate mealtime behav- or liquid, or choking because the presented tex- ior of most children in those studies (Allison et al., ture is inappropriate for the child’s chewing skills. 2012; Bachmeyer et al., 2009; Girolami & Scot- “Knowing what you don’t know” is an essential ti, 2001; Kirkwood, Piazza, & Peterson, in press; skill for behavior analysts working with children LaRue et al., 2011; Najdowski et al., 2008; Piazza, with feeding disorders. Fisher, et al., 2003), but some studies have found that adult attention and tangible items functioned as reinforcement too (Bachmeyer et al., 2009; Gi- REFERENCES rolami & Scotti, 2001; Piazza, Fisher, et al., 2003). There are only a few studies of function-based in- Addison, L. R., Piazza, C. C., Patel, M. E., Bachmeyer, terventions (Allison et al., 2012; Bachmeyer et al., M. H., Rivas, K. M., Milnes, S. M., et al. (2012). A 2009; Bachmeyer et al., 2019; Kirkwood et al., in comparison of sensory integrative and behavioral press; LaRue et al., 2011; Najdowski et al., 2003). therapies as treatment for pediatric feeding disorders. Results of these studies suggest that we can use the Journal of Applied Behavior Analysis, 45, 455–471. Ahearn, W. H. (2003). Using simultaneous presentation results of a functional analysis of inappropriate to increase vegetable consumption in a mildly selec- mealtime behavior to prescribe effective interven- tive child with autism. Journal of Applied Behavior tions to increase acceptance and decrease inap- Analysis, 36, 361–365. propriate mealtime behavior. Ahearn, W. H., Kerwin, M. E., Eicher, P. S., Shantz, The complexity of feeding disorders, however, J., & Swearingin, W. (1996). An alternating treat- necessitates that we refine our functional-analysis ments comparison of two intensive interventions for procedures to identify the characteristics of the food refusal. Journal of Applied Behavior Analysis, 29, feeding environment that establish escape or other 321–332. stimuli as reinforcement. Results of studies on fad- Allison, J., Wilder, D. A., Chong, I., Lugo, A., Pike, J., & ing suggest that we can alter the properties of Rudy, N. (2012). A comparison of differential rein- stimuli associated with high levels of acceptance forcement and noncontingent reinforcement to treat and low levels of inappropriate mealtime behavior food selectivity in a child with autism. Journal of Ap- plied Behavior Analysis, 45, 613–617. to produce appropriate feeding. Can we identify American Psychiatric Association. (2013). Diagnostic those antecedent variables a priori? Most children and statistical manual of mental disorders (5th ed.). emit some form of appropriate behavior. How can Arlington, VA: Author. we identify that behavior and use it to prescribe an Anderson, C. M., & McMillan, K. (2001). Parental use effective feeding intervention? of escape extinction and differential reinforcement Similarly, although behavior analysts have eval- to treat food selectivity. Journal of Applied Behavior uated procedures to reduce expulsion and packing, Analysis, 34, 511–515. 440 S U B S P E C I A LT IE S I N A P P L IE D B E H AV I O R A N A LY S I S Arvedson, J. C., & Brodsky, L. (2002). Pediatric swal- Peck, S. M., Richman, D., et al. (1995). Use of com- lowing and feeding: Assessment and management (2nd ponent analyses to identify active variables in treat- ed.). Albany, NY: Singular. ment packages for children with feeding disorders. Babbitt, R. L., Shore, B. A., Smith, M., Williams, K. E., Journal of Applied Behavior Analysis, 28, 139–153. & Coe, D. A. (2001). Stimulus fading in the treat- Darrow, D. H., & Harley, C. M. (1998). Evaluation of ment of adipsia. Behavioral Interventions, 16, 197–207. swallowing disorders in children. Otolaryngologic Bachmeyer, M. H., Gulotta, C. S., & Piazza, C. C. Clinics of North America, 31, 405–418. (2013). Liquid to baby food fading in the treatment Davis, A. M., Bruce, A., Cocjin, J., Mousa, H., & of food refusal. Behavioral Interventions, 28, 281–298. Hyman, P. (2010). Empirically supported treatments Bachmeyer, M. H., Kirkwood, C. A., Criscito, A. B., for feeding difficulties in young children. Current Mauzy, C. R., & Berth, D. P. (2019). A comparison of Gastroenterology Reports, 12, 189–194. functional analysis methods of inappropriate meal- Dawson, J. E., Piazza, C. C., Sevin, B. M., Gulotta, C. time behavior. Journal of Applied Behavior Analysis, S., Lerman, D., & Kelley, M. L. (2003). Use of the 52, 603–621. high-probability instructional sequence and escape Bachmeyer, M. H., Piazza, C. C., Frederick, L. D., Reed, extinction in a child with food refusal. Journal of Ap- G. K., Rivas, K. D., & Kadey, H. J. (2009). Functional plied Behavior Analysis, 36, 105–108. analysis and treatment of multiply controlled inap- Dempsey, J., Piazza, C. C., Groff, R. A., & Kozisek, J. M. propriate mealtime behavior. Journal of Applied Be- (2011). A flipped spoon and chin prompt to increase havior Analysis, 42, 641–658. mouth clean. Journal of Applied Behavior Analysis, 44, Bandini, L. G., Anderson, S. E., Curtin, C., Cermak, S., 961–965. Evans, E. W., Scampini, R., et al. (2010). Food selec- Derkay, C. S., & Schechter, G. L. (1998). Anatomy and tivity in children with autism spectrum disorders and physiology of pediatric swallowing disorders. Otolar- typically developing children. Journal of Pediatrics, yngologic Clinics of North America, 31, 397–404. 157, 259–264. Diquelou, M. C., Griffin, A. S., & Sol, G. D. (2016). Boiron, M., Nobrega, L. D., Roux, S., Henrot, A., & The role of motor diversity in foraging innovations: Saliba, E. (2007). Effects of oral stimulation and oral A cross-species comparison in urban birds. Behavioral support on non-nutritive sucking and feeding perfor- Ecology, 27, 584–591. mance in preterm infants. Developmental Medicine Douglas, J. E., & Byron, M. (1996). Interview data on and Child Neurology, 49, 439–444. severe behavioural eating difficulties in young chil- Borrero, C. S. W., Woods, J. N., Borrero, J. C., Masler, dren. Archives of Disease in Childhood, 75, 304–308. E. A., & Lesser, A. D. (2010). Descriptive analysis of Einarsson-Backes, L. M., Dietz, J., Price, R., Glass, R., & pediatric food refusal and acceptance. Journal of Ap- Hays, R. (1994). The effect of oral support on suck- plied Behavior Analysis, 43, 71–88. ing efficiency in preterm infants. American Journal of Borsari, A., & Ottoni, E. B. (2005). Preliminary obser- Occupational Therapy, 48, 490–498. vations of tool use in captive hyacinth macaws (Ano- Field, T., Ignatoff, E., Stinger, S., Brennan, J., Greenberg, dorhynchus hyacinthinus). Animal Cognition, 8, 48–53. R., Widmayer, S., et al. (1982). Nonnutritive sucking Buckley, S. D., & Newchok, D. K. (2005). An evalua- during tube feedings: Effects on preterm neonates in tion of simultaneous presentation and differential an intensive care unit. Pediatrics, 70, 381–384. reinforcement with response cost to reduce packing. Franklin, L., & Rodger, S. (2003). Parents’ perspectives Journal of Applied Behavior Analysis, 38, 405–409. on feeding medically compromised children: Impli- Buckley, S. D., Strunck, P. G., & Newchok, D. K. (2005). cations for occupational therapy. Australian Occupa- A comparison of two multicomponent procedures to tional Therapy Journal, 50, 137–147. increase food consumption. Behavioral Interventions, Freedman, D. S., Dietz, W. H., Srinivasan, S. R., & Be- 20, 139–146. renson, G. S. (1999). The relation of overweight to Burklow, K. A., McGrath, A. M., Valerius, K. S., & cardiovascular risk factors among children and ado- Rudolph, C. (2002). Relationships between feeding lescents: The Bogalusa Heart Study. Pediatrics, 103, difficulties, medical complexity, and gestational age. 1175–1182. Nutrition in Clinical Practice, 17, 373–378. Freeman, K. A., & Piazza, C. C. (1998). Combining Casey, S. D., Cooper-Brown, L. J., Wacker, D. P., & stimulus fading, reinforcement, and extinction to Rankin, B. E. (2006). The use of descriptive analysis treat food refusal. Journal of Applied Behavior Analy- to identify and manipulate schedules of reinforce- sis, 31, 691–694. ment in the treatment of food refusal. Journal of Be- Fucile, S., Gisel, E., & Lau, C. (2002). Oral stimula- havioral Education, 15, 41–52. tion accelerates the transition from tube to oral Coe, D. A., Babbitt, R. A., Williams, K. E., Hajimihalis, feeding in preterm infants. Journal of Pediatrics, 141, C., Snyder, A. M., Ballard, C., et al. (1997). Use of 230–236. extinction and reinforcement to increase food con- Girolami, P. A., Boscoe, J. H., & Roscoe, N. (2007). sumption and reduce expulsion. Journal of Applied Decreasing expulsions by a child with a feeding dis- Behavior Analysis, 30, 581–583. order: Using a brush to present and re-present food. Cooper, L. J., Wacker, D. P., McComas, J. J., Brown, K., Journal of Applied Behavior Analysis, 40, 749–753.  A Behavior-Analytic Approach to Feeding Disorders 441 Girolami, P. A., & Scotti, J. R. (2001). Use of ana- multicomponent treatment for food refusal. Journal of log functional analysis in assessing the function of Applied Behavior Analysis, 34, 93–96. mealtime problem behavior. Education and Training Kelley, M. E., Piazza, C. C., & Fisher, W. W. (2003). in Mental Retardation and Developmental Disabilities, Acquisition of cup drinking using previously refused 36, 207–223. foods as positive and negative reinforcement. Journal Goto, A., Arah, O. A., Goto, M., Terauchi, Y., & Noda, of Applied Behavior Analysis, 36, 89–93. M. (2013). Severe hypoglycemia and cardiovascular Kern, L., & Marder, T. J. (1996). A comparison of simul- disease: Systematic review and meta-analysis with taneous and delayed reinforcement as treatments for bias analysis. British Journal of Medicine, 347, 1–11. food selectivity. Journal of Applied Behavior Analysis, Groff, R. A., Piazza, C. C., Volkert, V. M., & Jostad, 2, 243–246. C. M. (2014). Syringe fading as treatment for feed- Kerwin, M. E. (1999). Empirically supported treatments ing refusal. Journal of Applied Behavior Analysis, 47, in pediatric psychology: Severe feeding problems. 834–839. Journal of Pediatric Psychology, 24, 193–214. Groff, R. A., Piazza, C. C., Zeleny, J. R., & Dempsey, J. Kerwin, M. E., Ahearn, W. H., Eicher, P. S., & Burd, R. (2011). Spoon-to-cup fading as treatment for cup D. M. (1995). The costs of eating: A behavioral eco- drinking in a child with intestinal failure. Journal of nomic analysis of food refusal. Journal of Applied Be- Applied Behavior Analysis, 44, 949–954. havior Analysis, 28, 245–260. Gulotta, C. S., Piazza, C. C., Patel, M. R., & Layer, S. A. Kirkwood, C. A., Piazza, C. C., & Peterson, K. M. (in (2005). Using food redistribution to reduce packing press). A comparison of function- and nonfunction- in children with severe food refusal. Journal of Ap- based extinction for inappropriate mealtime behav- plied Behavior Analysis, 38, 39–50. ior. Journal of Applied Behavior Analysis. Hadley, A. J., Krival, K. R., Ridgel, A. L., Hahn, E. C., & LaRue, R. H., Stewart, V., Piazza, C. C., Volkert, V. M., Tyler, D. J. (2015). Neural network pattern recogni- Patel, M. R., & Zeleny, J. (2011). Escape as reinforce- tion of lingual–palatal pressure for automated detec- ment and escape extinction in the treatment of feed- tion of swallow. Dysphagia, 30, 176–187. ing problems. Journal of Applied Behavior Analysis, Hagopian, L. P., Farrell, D. A., & Amari, A. (1996). 44, 719–735. Treating total liquid refusal with backward chaining Levin, D. S., Volkert, V. M., & Piazza, C. C. (2014). A and fading. Journal of Applied Behavioral Analysis, 29, multi-component treatment to reduce packing in 573–575. children with feeding and autism spectrum disorders. Hoch, T. A., Babbitt, R. L., Coe, D. A., Duncan, A., & Behavior Modification, 38, 940–963. Trusty, E. M. (1995). A swallow induction avoidance Linscheid, T. R., Oliver, J., Blyler, E., & Palmer, S. procedure to establish eating. Journal of Behavior (1978). Brief hospitalization for the behavioral treat- Therapy and Experimental Psychiatry, 26, 41–50. ment of feeding problems in the developmentally Hoch, T. A., Babbitt, R. L., Coe, D. A., Krell, D. M., & disabled. Journal of Pediatric Psychology, 12, 451–459. Hackbert, L. (1994). Contingency contacting: Com- Luiselli, J. K., & Gleason, D. L. (1987). Combining sen- bining positive reinforcement and escape extinction sory reinforcement and texture fading procedures to procedures to treat persistent food refusal. Behavior overcome chronic food refusal. Journal of Behavior Modification, 18, 106–128. Therapy and Experimental Psychiatry, 18, 149–155. Ibañez, V. F., Piazza, C. C., & Peterson, K. M. (2019). Luiselli, J. K., Ricciardi, J. N., & Gilligan, K. (2005). Liq- A translational evaluation of renewal of inappropri- uid fading to establish milk consumption by a child ate mealtime behavior. Journal of Applied Behavior with autism. Behavioral Interventions, 20, 155–163. Analysis, 52, 1005–1020. Lukens, C. T., & Silverman, A. H. (2014). Systematic Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., & review of psychological interventions for pediatric Richman, G. S. (1994). Toward a functional analysis feeding problems. Journal of Pediatric Psychology, 39, of self-injury. Journal of Applied Behavior Analysis, 27, 903–917. 197–209. (Reprinted from Analysis and Intervention in Meier, A. E., Fryling, M. J., & Wallace, M. D. (2012). Developmental Disabilities, 2, 3–20, 1982) Using high-probability foods to increase the accep- Johnson, C. R., & Babbitt, R. L. (1993). Antecedent tance of low-probability foods. Journal of Applied Be- manipulation in the treatment of primary solid food havior Analysis, 45, 149–153. refusal. Behavior Modification, 17, 510–521. Mueller, M. M., Piazza, C. C., Moore, J. W., Kelley, M. Kadey, H., Piazza, C. C., Rivas, K. M., & Zeleny, J. E., Bethke, S. A., Pruett, A., et al. (2003). Training (2013). An evaluation of texture manipulations parents to implement pediatric feeding protocols. to increase swallowing. Journal of Applied Behavior Journal of Applied Behavior Analysis, 36, 545–562. Analysis, 46, 539–543. Mueller, M. M., Piazza, C. C., Patel, M. R., Kelley, M. Kahng, S., Boscoe, J. H., & Byrne, S. (2003). The use E., & Pruett, A. (2004). Increasing variety of foods of an escape contingency and a token economy to consumed by blending nonpreferred foods into pre- increase food acceptance. Journal of Applied Behavior ferred foods. Journal of Applied Behavior Analysis, 37, Analysis, 36, 349–353. 159–170. Kahng, S., Tarbox, J., & Wilke, A. E. (2001). Use of a Munk, D. D., & Repp, A. C. (1994). Behavioral assess- 442 S U B S P E C I A LT IE S I N A P P L IE D B E H AV I O R A N A LY S I S ment of feeding problems of individuals with severe Penrod, B., Gardella, L., & Fernand, J. (2012). An evalu- disabilities. Journal of Applied Behavior Analysis, 27, ation of a progressive high-probability instructional 241–250. sequence combined with low-probability demand Najdowski, A. C., Wallace, M. D., Doney, J. K., & Ghe- fading in the treatment of food selectivity. Journal of zzi, P. M. (2003). Parental assessment and treatment Applied Behavior Analysis, 45, 527–537. of food selectivity in natural settings. Journal of Ap- Peterson, K. M., Piazza, C. C., Ibañez, V. F., & Fisher, W. plied Behavior Analysis, 36, 383–386. W. (2019). Randomized controlled trial of an applied Najdowski, A. C., Wallace, M. D., Penrod, B., Tarbox, J., behavior analysis intervention for food selectivity in Reagon, K., & Higbee, T. S. (2008). Caregiver-con- children with autism spectrum disorder. Journal of ducted experimental functional analyses of inappro- Applied Behavior Analysis, 52, 895–917. priate mealtime behavior. Journal of Applied Behavior Peterson, K. M., Piazza, C. C., & Volkert, V. M. (2016). Analysis, 41, 459–465. Comparison of a modified sequential oral sensory ap- Najdowski, A. C., Wallace, M. D., Reagon, K., Penrod, proach to an applied behavior analytic approach in B., Higbee, T. S., & Tarbox, J. (2010). Utilizing a the treatment of food selectivity in children with au- home-based parent training approach in the treat- tism spectrum disorders. Journal of Applied Behavior ment of food selectivity. Behavioral Interventions, 25, Analysis, 49, 485–511. 89–107. Peterson, K. M., Volkert, V. M., & Zeleny, J. R. (2015). Nebraska Legislature. (2009). Fiscal note: Legislative fis- Increasing self-drinking for children with feeding cal analyst estimate. LB 243, Revision 01. disorders. Journal of Applied Behavior Analysis, 48, Palmer, S., & Horn, S. (1978). Feeding problems in chil- 436–441. dren. In S. Palmer & S. Ekvall (Eds.), Pediatric nutri- Piazza, C. C., Anderson, C., & Fisher, W. W. (1993). tion in developmental disorders (pp. 107–129). Spring- Teaching self-feeding skills to patients with Rett syn- field, IL: Charles C Thomas. drome. Developmental Medicine and Child Neurology, Pangborn, M. M., Borrero, C. S. W., & Borrero, J. C. 35, 991–996. (2013). Sequential application of caregiver training Piazza, C. C., & Carroll-Hernandez, T. A. (2004) As- to implement pediatric feeding protocols. Behavioral sessment and treatment of pediatric feeding disor- Interventions, 28, 107–130. ders. In R. E. Tremblay, R. G. Barr, & R. Peters (Eds.), Patel, M. R., Piazza, C. C., Kelley, M. L., Ochsner, C. A., Encyclopedia on early childhood development [Online]. & Santana, C. M. (2001). Using a fading procedure Montreal: Centre of Excellence for Early Childhood to increase fluid consumption in a child with feeding Development. Retrieved July 29, 2013, from www.ex- problems. Journal of Applied Behavior Analysis, 34, cellence-earlychildhood.ca/documents/Piazza-Carroll- 357–360. HernandezANGxp.pdf. Patel, M. R., Piazza, C. C., Layer, S. A., Coleman, R., & Piazza, C. C., Fisher, W. W., Brown, K. A., Shore, B. A., Swartzwelder, D. M. (2005). A systematic evaluation Patel, M. R., Katz, R. M., et al. (2003). Functional of food textures to decrease packing and increase oral analysis of inappropriate mealtime behaviors. Journal intake in children with pediatric feeding disorders. of Applied Behavior Analysis, 36, 187–204. Journal of Applied Behavior Analysis, 38, 89–100. Piazza, C. C., Patel, M. R., Gulotta, C. S., Sevin, B. M., Patel, M. R., Piazza, C. C., Martinez, C. J., Volkert, V. & Layer, S. A. (2003). On the relative contribution M., & Santana, C. M. (2002). An evaluation of two of positive reinforcement and escape extinction in differential reinforcement procedures with escape the treatment of food refusal. Journal of Applied Be- extinction to treat food refusal. Journal of Applied Be- havior Analysis, 36, 309–324. havior Analysis, 35, 363–374. Piazza, C. C., Patel, M. R., Santana, C. M., Goh, H., Patel, M. R., Piazza, C. C., Santana, C. M., & Volkert, V. Delia, M. D., & Lancaster, B. M. (2002). An