A Retrospective Consecutive Controlled Case Series of Underspoon: A Modified-Bolus Placement to Address Behavior That Interfere With Swallowing

Abstract

Children diagnosed with pediatric feeding disorder (PFD) or avoidant/restrictive food intake disorder (ARFID) may present with comorbid oral-motor delays that often contribute to their failure to consume an adequate volume and/or variety of foods. Once the food enters the mouth, these children may exhibit additional problematic behavior such as expulsion and packing that further averts oral intake. Previous research has evaluated the impact of modified-bolus placement methods, including flipped spoon and NUK^® brush, in comparison to upright spoon on expulsion and packing. Whether responses are due to performance or skill deficits, researchers hypothesize that modified-bolus placement may assist with reducing the response effort associated with swallowing. This retrospective consecutive controlled case series assessed the clinical application of an additional modified-bolus placement method, underspoon, to increase food consumption and decrease problematic behavior that interfere with mouth clean for children with PFD or ARFID.

Keywords

underspoon modified-bolus placement utensil manipulation bite presentation pediatric feeding disorder ARFID expulsion packing

Feeding and swallowing problems resulting in a failure to orally consume an adequate volume and/or variety of foods to meet nutrient or energy needs warrant a diagnosis of pediatric feeding disorder (PFD) or avoidant/restrictive food intake disorder (ARFID; American Psychiatric Association, 2013; Goday et al., 2019). The diagnostic criteria for PFD require the presence of medical (e.g., vomiting), nutritional (e.g., reliance on formula orally or via feeding tube), feeding skill (e.g., coughing or choking during meals), or psychosocial dysfunction (e.g., inappropriate mealtime or refusal behavior, caregiver stress) for a minimum of 2 weeks (Goday et al., 2019). Pediatric feeding disorders occur in 4% of young children and the prevalence is much higher in children with developmental disabilities (Kovacic et al., 2021; Seiverling et al., 2018; Sharp et al., 2013).

Applied-behavior-analytic intervention is the most widely studied and empirically supported approach to address ARFID or PFD (Sharp et al., 2017; Sharp, Jaquess, et al., 2010). Common initial treatment components include escape prevention, differential reinforcement of alternative behavior (DRA), and stimulus fading (Ledford et al., 2018; Saini et al., 2019; Sharp et al., 2017; Volkert & Piazza, 2012) with the focus on reducing disruptive mealtime behavior, such as crying, tantrums, and/or pushing the food away, that prohibit food from entering the mouth (Piazza et al., 2003). Once the food enters the mouth, however, additional problematic behavior such as expulsion (e.g., spitting or wiping food out of the mouth) and packing (i.e., holding food in the mouth without swallowing; Sevin et al., 2002) may occur due to oral-motor dysfunction or motivational deficits that result in decreased oral intake.

An antecedent strategy to reduce expulsion and packing involves manipulation of food type and/or texture, (e.g., reducing food texture to smooth puree) as this may increase the likelihood of taste exposure and/or reduce the response effort associated with preparing the bolus for swallowing (Munk & Repp, 1994; Patel et al., 2005; Patel, Piazza, Santana, et al., 2002). Consequence-based strategies include re-presentation of expels, re-distribution of a packed bite, and differential reinforcement for mouth clean (a permanent product of swallowing) which may reduce the amount of escape accessed via expulsion and packing and/or increase the probability that the child swallows the bite (Patel, Piazza, Martinez, et al., 2002; Sevin et al., 2002). Unfortunately, the traditional method of presenting food on an upright spoon in combination with any of the aforementioned strategies may not reduce expulsion or packing to acceptable levels for many children with oral-motor deficits (e.g., Sharp et al., 2012; Volkert et al., 2019). Feeding skill dysfunction involves disruptions in the orchestrated movement of several muscles in the lips, palate, and tongue that promote swallowing (Prasse & Kikano, 2009), and can include lack of lip closure, an open mouth posture, and/or pooling saliva (e.g., Sharp, Harker, et al., 2010; Stubbs et al., 2018).

An alternative antecedent strategy to reduce expulsion and packing associated with oral-motor concerns is a modified-bolus placement (also known as utensil manipulation, alternate bite presentation). This most often involves the feeder depositing food directly onto the tongue. Researchers have systematically evaluated two modified-bolus placement methods, flipped spoon and NUK^® brush, in comparison to upright spoon. These methods have been shown to reduce expulsion (Dempsey et al., 2011; Girolami et al., 2007; Ibañez et al., 2021; Sharp et al., 2012; Sharp, Harker, et al., 2010; Volkert et al., 2019; Wilkins et al., 2014) and packing (Milnes et al., 2019; Stubbs et al., 2018; Volkert et al., 2011, 2019). Wilkins et al. (2014) reported that for 55% of children admitted to an intensive day treatment program for ARFID or PFD, escape extinction procedures (i.e., nonremoval of the spoon [NRS], re-presentation) were not sufficient to treat expulsion and packing. They posited that feeding skill dysfunction may be a predictor for the necessity of modified-bolus placement, but these authors did not characterize oral-motor-skill status for the participants in their study. Volkert et al. (2019) reported that 24% of patients received modified-bolus placement during an intensive multidisciplinary intervention (IMI) admission. These authors identified that this subset of children displayed significantly more moderate-to-severe oral-motor deficits than children who did not receive treatment involving modified-bolus placement.

A modified-bolus placement called underspoon has been deployed clinically at an IMI program; however, there is little to no published research evaluating its efficacy at reducing expels and packing compared to other bite-presentation methods. A study by Gibbons et al. (2007) described a similar procedure using the bottom of the spoon to address tongue thrust that resulted in expels for a 6-year-old girl with Down syndrome, but to the current investigators’ knowledge, the current study is the first systematic evaluation of underspoon. The primary objectives of the current study were to assess (a) the patient characteristics of those who received treatment with underspoon and (b) the clinical application of underspoon as a modified-bolus placement method for these patients. The investigators hypothesized that children who required modified-bolus placement would have deficits in oral-motor function, and that underspoon presentation may hold the same benefit as other modified bolus-placements (e.g., flipped spoon) to decrease expels and/or packing by compensating for oral-motor deficits while also promoting contact with the food (Volkert et al., 2019).

Method

Study Design

The investigators conducted a retrospective consecutive controlled case series to examine the generality (i.e., ability to replicate beyond the initial case demonstration) of experimental demonstrations of underspoon among consecutive cases in an IMI program over a 17 month period and explore and quantify characteristics of such cases (Hagopian, 2020). The investigators followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement and checklist to guide the reporting of research procedures (Von Elm et al., 2007). STROBE is a tool used to improve the quality and transparency of reporting case-control observational data and widely used in biomedical research (parallel to PRISMA for systematic reviews and meta-analyses).

Participants

For the current review, inclusion criteria required that patients admitted to the IMI program for treatment of ARFID or PFD between 01/01/2020 and 06/01/2021 (a) received an initial modified-bolus placement assessment that included a multielement or pairwise comparison of at least two bolus placement methods (i.e., underspoon vs. upright spoon) to target expulsion or packing at clinically problematic levels (60% or greater bite presentations) prior to the analysis, (b) involved bite presentations at a bolus size of at least 0.25 cc (amount of food on the spoon large enough to evaluate if the patient packed or expelled food), and (c) had behavioral data on relevant behavior (e.g., bite acceptance, inappropriate mealtime behavior, expels, packs) with a second observer independently scoring reliability on behavioral data for at least 20% of sessions during the comparison or a video recording available to meet this criterion. Investigators excluded the patient if an assessment of modified-bolus placement (a) was not conducted during the admission, (b) excluded upright or underspoon presentation methods, or (c) did not target expulsion or packing.

Eighty-six patients admitted to the IMI feeding program during the 17-month review period were considered potentially eligible for inclusion in the current study (Figure 1). Eight patients (9.3% of 86 potentially eligible individuals) received treatment involving a modified-bolus placement assessment of which the investigators further examined for eligibility. In three cases, the modified-bolus placement assessment did not include upright spoon and/or underspoon presentation methods, and in one case, the purpose of the modified-bolus placement was to address acceptance and unrelated to expels and packing. Two investigators (one of whom was blind to the purpose of the current study) independently reviewed and screened all participants. The first author calculated inter-coder agreement on inclusion by taking the total number of agreements between the two investigators and dividing by agreements plus disagreements, which yielded an agreement of 98.5%. After calculating inter-coder agreement, a third investigator served as a “tie-breaker” for instances in which the primary and secondary investigators disagreed. This resulted in four patients (4.6% of 86 potentially eligible individuals) who met final inclusion criteria.

Figure 1.

Flow diagram of participant eligibility.

Table 1 provides participant demographics for patients who met inclusion. At each child’s initial evaluation, the speech pathologist conducted a skill assessment (i.e., Beckman Oral Motor Evaluation) of oral-motor abilities in vivo. The speech pathologist characterized all participants as having moderate-to-severe oral-motor delays.

Table 1.

Participant Characteristics.

Participant	Age upon admission (years)	Sex assigned at birth	Race	Feeding concern/ ARFID diagnosis	Disability diagnosis	Medical history	Oral-motor status at initial evaluation
Kai	2.3	Male	White	A2, A3b	DD	In-utero drug exposure	Moderate
Ben	1.9	Male	Black or African American	A3b	ASD	None	Severe
Ezra	9.3	Male	Black or African American	A3a	ASD, I/DD	Cerebral palsy, prematurity, GERD, bronchopulmonary dysplasia, hypotonia, constipation, and pulmonary valve disease	Severe
Mila	4.3	Female	White	A3b	DD	Gastroparesis, constipation, and blindness	Moderate

Note. Includes data from N = 4 patients whose modified-bolus placement assessment compared upright spoon and underspoon presentation methods to address expels or packing. ARFID = avoidant/restrictive food intake disorder; A2 = significant nutritional deficiency; A3a = feeding tube dependence; A3b = oral formula dependence; ASD = autism spectrum disorder; I/DD = intellectual or developmental disability; GERD = gastroesophageal reflux disease.

Kai was a 2-year-old White male admitted to the program for significant nutritional deficiencies in Vitamin A and iron (ARFID A2) and oral formula dependence on a baby bottle containing a homemade blenderized formula comprised of water, whole milk, baby cereal, and baby food to meet 100% of his needs (ARFID A3b). Kai had a medical history significant for developmental disability and in-utero drug exposure. At the time of his initial evaluation, the speech pathologist identified Kai as having moderate oral-motor delays, characterized by moderately impaired lip strength, resulting in an open-mouth posture, excessive drooling, and poor lip closure around the spoon.

Ben was a 1-year-old Black male who was diagnosed with autism spectrum disorder (ASD) and admitted to the program for dependence on oral nutritional formula (whole milk and Pediasure) to meet 100% of his nutritional and energy needs (ARFID A3b). The speech pathologist identified Ben as having severe oral-motor delays at the time of his initial evaluation to the program characterized by severely impaired jaw strength and durational chew, and moderately impaired lingual variety of movement to control a bolus in the oral phase in preparation for the swallow.

Ezra was a 9-year-old Black male admitted to the program for 85% to 100% dependence on a homemade blenderized formula comprised of rice cereal, baby food, and Compleat® Pediatric formula through his gastrostomy tube (ARFID A3a). Ezra’s medical history was significant for ASD, intellectual and developmental disability, cerebral palsy, prematurity, gastroesophageal reflux disease, bronchopulmonary dysplasia, hypotonia, constipation, and pulmonary valve disease. The speech pathologist indicated Ezra presented with severe oral-motor delays, characterized by severely impaired jaw strength and durational chew, and severely impaired lingual variety of movement to control a bolus in the oral phase in preparation for the swallow.

Mila was a 4-year-old White female admitted to the feeding program for oral formula dependence (Kate Farms Pediatric Peptide 1.5) to meet 100% of her nutritional and energy needs (ARFID A3b). Her medical history was significant for developmental disability, gastroparesis, constipation, and blindness. At her initial evaluation, the speech pathologist reported moderate oral-motor delays characterized by moderately impaired jaw strength and durational chew, and moderately impaired lingual variety of movement to control a bolus in the oral phase in preparation for the swallow.

Setting and Treatment Approach

Participants were admitted to the IMI feeding program if they had a diagnosis of ARFID or PFD as evidenced by a failure to achieve expected weight gain, significant nutritional deficiencies, and/or dependence on eternal feeding or oral supplementation formula, and a history of persistent problematic mealtime behavior (e.g., crying, pushing away the spoon, vomiting) that contribute to reductions in oral consumption to meet caloric and nutritional needs. The multidisciplinary team consisted of a licensed psychologist, registered dietitian, speech pathologist or occupational therapist, nurse practitioner, pediatric gastroenterologist, and social worker. All meals occurred in a private treatment room with an adjacent observation room equipped with a one-way window. Treatment rooms contained a table, age- or developmentally appropriate seating (e.g., booster seat, highchair), serving tray, feeding utensils (e.g., small Maroon Spoon™, Beckman E-Z Spoon [Ben only], rubber-coated baby spoon), 1-cc syringes (Ben and Mila only), digital food scale, gram intake log, timer, and the child’s preferred items.

Intervention followed a manualized sequence of behavioral strategies developed through clinical practice and a review of the literature (see Sharp et al., 2016 for randomized clinical trial). For young children who presented as dependent on formula to meet energy and nutritional needs and who may not have the feeding skill to consume table food, the clinic standard began with pureed food to target volume. If children exhibited moderate levels of expels or packs at clinically problematic levels (i.e., between 40% and 100% for at least three sessions), standard of care typically included reducing the bolus of food presented, removing highly problematic foods, reinforcement for retention and swallowing (e.g., DRA for mouth clean) and/or preventing the child from wiping food out of the mouth or re-presentation (expels only). If expels or packs persisted, despite the previously described treatment components, the next step in treatment involved conducting a modified-bolus presentation assessment. The assessments involved single-case research design methodology to compare the efficacy of multiple presentation or bolus placement methods for reducing expels and packing and increasing mouth clean and is further described below.

Assessment of Modified-Bolus Placement

All presentation methods involved NRS during which the feeder presented a utensil with smooth puree food to the child’s mouth by touching the tip of the utensil to the child’s lips and waited for the child to open their mouth to deposit the entire bolus into the child’s mouth (food residue smaller than the size of a pea or grain of rice left on the spoon, depending on the amount of the initial bolus,). The feeder repeated the deposit if food a pea or grain of rice size, depending on the amount of the initial bolus, or larger remained on the spoon. The feeder removed the utensil immediately after the bite was deposited into the child’s mouth. The feeder conducted mouth clean checks 30 seconds after depositing the bite using a three-step prompting sequence (i.e., vocal, model, physical) to look inside the mouth. Specifically, 30 seconds after depositing the bite, the feeder said, “Show me.” If the child did not open his or her mouth 5 seconds after the verbal prompt, the feeder said, “Show me like this” and modeled an open mouth. If 5 seconds after the model prompt the child still did not open their mouth, the feeder used a rubber coated baby spoon to check for food in the child’s mouth. If food remained in the mouth during the check, they continued to check the child’s mouth every 30 seconds until the child swallowed and the mouth was clear of food. The feeder presented the next bite no sooner than the initial mouth check and once the mouth was clear of food. Sessions consisted of five bites.

During the modified-bolus placement assessment, the clinical team used visual analysis to select the most effective bolus placement method at reducing expels or packs. Specifically, teams attended to bolus placement methods that resulted in low levels (0%–20%) of expels (Kai, Ben, Ezra) or packs (Mila) and high levels of mouth clean (80%–100%). The treatment team monitored data on acceptance and IMBs, but they did not select the modified-bolus placement method based on these data due to the emphasis on taste exposure and consumption. After the team achieved improvements in volume and variety of foods using the modified-bolus placement, all four participants receive additional presentation assessments to determine if the participant could return to an upright spoon presentation (i.e., expels and/or packs remained low).

Upright spoon

For upright spoon, the default presentation method in clinic, the feeder either scooped the bolus from the individual bowl of food, leveled the bolus by dragging the bowl of the spoon against the edge of the food bowl, and scraped off any excess or overflow back into the cup using the edge of the cup or with a rubber coated baby spoon to ensure appropriate bolus size. If the feeder used a syringe, they collected food into a syringe from the bowl of food and used the syringe to squeeze the bolus on to the spoon. Next, the feeder presented the upright spoon to the lips midline and inserted the bowl of the small Maroon Spoon™ into the child’s open mouth, removing the spoon after the child closed their lips around the spoon. If the child exhibited an open mouth posture for 1 second after inserting the bowl of the spoon into the mouth, the therapist gently scraped the bolus onto the back of either the upper front teeth or upper lip prior to removing the spoon.

Underspoon

For underspoon (see Figure 2), the feeder placed the bolus toward the middle of the underside of the small Maroon Spoon™ by either scooping the bolus from the individual bowl of food, leveling the bolus by dragging the back of the spoon against the edge of the food bowl, and scraping off any excess with a rubber coated baby spoon to ensure appropriate bolus size or using a syringe for measurment and placement. The feeder used either a transverse digital grip (i.e., the feeder’s thumb on the top of the handle and index finger on the underside of the handle) or adult grip (i.e., the feeder’s index and middle finger on the top of the handle and thumb beneath the handle) to grasp the spoon mid-way between the neck and end of the spoon handle. Prior to inserting the spoon into the child’s mouth, the feeder held the spoon upside down (i.e., bowl of the spoon is down) and presented the upside-down spoon to the child’s lips at midline, rotated the spoon 180° at the lips (i.e., rotating either clockwise or counterclockwise using either hand outside of the child’s mouth) and inserted the bowl of the spoon into the mouth. The feeder deposited the bolus onto the tongue starting mid-way between the anterior and posterior section by applying gentle downward pressure and concurrently using a wiping motion to drag the spoon toward the bottom lips and out of the mouth. For underspoon using E-Z Spoon, the feeder used the same procedures outlined for underspoon but used a syringe to measure and place the bolus onto the underside of the E-Z Spoon.

Figure 2.

Written and visual procedures for underspoon and flipped spoon.

Flipped spoon

For flipped spoon (see Figure 2), the feeder prepared the bolus identical to upright spoon using either the scooping or syringe method. The feeder used a transverse digital grip to hold the small Maroon Spoon™ mid-way between the neck and the end of the spoon handle. They first presented the spoon upright to the child’s lip midline and inserted the bowl of the spoon into the child’s open mouth. The feeder then positioned the spoon halfway between the anterior and posterior section of the tongue and applied gentle downward pressure with the bowl of the spoon, flatting the tongue. The feeder then immediately rotated the spoon inward 180° in the child’s mouth (i.e., rotating clockwise if left-handed or counterclockwise if right-handed, flipping the spoon) and deposited the bolus onto the tongue using a wiping motion and applying gentle downward pressure by dragging the spoon toward the bottom lips/teeth and out of the mouth.

Assessment of Social Validity

In the last week of their child’s admission, we asked families to complete an eight-item questionnaire to assess caregiver experience, perceptions, and acceptability of treatment. Participation was voluntary and the questionnaire also presented opportunities for the family to deliver constructive feedback. Families rated six items on a 5-point Likert-type scale and two items on a 10-point Likert-type scale; better performance was depicted through higher scores. The final question specifically aimed to measure customer satisfaction, loyalty, and enthusiasm for the program by yielding a Net promoter score (Reichheld, 2003). Specifically, scores of 9 or 10 were classified as promoters of the program, defined as loyal and enthusiastic customers who will continue to refer others. Scores of 7 and 8 were classified as passive, defined as satisfied but unenthusiastic customers who may find services elsewhere. Scores of 6 or below were classified as detractors of the program, defined as unhappy customers who may speak negatively of the program. Thus, higher scores indicated greater overall satisfaction.

Coding and Response Measurement

Two investigators independently extracted and entered data, including demographic information, assessment procedures, and outcome measures, into separate Microsoft^® Excel^® spreadsheets to compare and calculate intercoder agreement (described below). Demographic information included age at admission, sex assigned at birth, race/ethnicity, disability status, ARFID subtype (for further subtype differentiation, see Sharp & Stubbs, 2019), medical history, oral motor status at the time of the initial evaluation (e.g., within functional limits, moderate-to-severe oral motor delays or deficits), and number of days of treatment completed in the program. Assessment procedure data included bolus presentation methods compared in the modified-bolus placement assessment, bolus size, foods presented, prior feeding behavioral protocols (e.g., NRS, DRA, re-presentation), the current protocol at time of assessment, syringe usage (i.e., whether the feeding therapist used a syringe to measure the amount of food presented for each bite), the presentation method selected by clinical team following the primary modified-bolus placement assessment (e.g., underspoon, flipped spoon), and number of sessions between assessments. Outcome measure data identified the bolus size, utensil, behavioral protocol, number of target foods, percentage of goals met at the end of admission, and performance at follow-up.

Mealtime behavior analyzed for this review included 5-second acceptance, inappropriate mealtime behavior ([IMB], all participants) mouth clean, expel (Ben, Kai, Ezra), and pack (Mila). The investigators defined 5-second acceptance as when the child opens their mouth, and the entire bite is deposited within 5 seconds. The definition of IMB included both head turns (i.e., child turns their head and/or body 45° past midline or greater than 2″ away during the bite presentation) and disruptions (i.e., anytime the child touches the spoon, plate, cup, food, or anywhere on the therapist’s hand/arm below the elbow and/or child blocks access to their mouth during the bite presentation). The investigators defined mouth clean as the child swallowing the entire bolus (except for a pea size or less if the bolus was 0.5 cc or larger and smaller than the size of a grain of rice if the bolus was less than 0.5 cc), within 30 seconds after the feeder deposited the bite. Expels were defined as food larger than the size of a pea if the bolus was 0.5 cc or larger and food smaller than the size of a grain of rice if the bolus was smaller than 0.5 cc outside the child’s mouth, which included instances when the food dripped out; the child actively pushed the food out with their tongue; removed the food with their hand; or wiped the food on their hand, arm, or clothing. The therapist scored a pack when the child held food (greater than pea size in their mouth if the bolus was at least 0.5 cc and smaller than a grain of rice if the bolus was smaller than 0.5 cc) 30 seconds after the feeder deposited the bite. If food was outside the child’s mouth at any time, including before the 30 seconds mouth check or after the feeding therapist scored a pack, the feeding therapist scored an expel if the amount was large enough. Mouth cleans, expels, and packs were converted into a percentage by dividing occurrence of the target behavior by the total number of bites entering the child’s mouth multiplied by 100.

Inter-Coder Agreement, Inter-Observer Agreement, and Procedural Integrity

The investigators applied the same inter-coder agreement methodology for participant demographics, assessment, and outcome as they did for inclusionary criteria. Inter-coder agreement was 90.9% for demographics, 98.8% for assessment coding, and 100% for outcomes.

Two independent observers scored standard behavioral data, including 5-second acceptance, mouth cleans, expels, packs, and IMB during the modified-bolus placement assessments or using a video recording of the assessment to allow for calculation of inter-observer agreement (IOA). Investigators calculated IOA on a trial-by-trial basis for each participant by dividing agreements (e.g., two independent observers scored expel) for each bite by agreements plus disagreements and converting that ratio to a percentage. For the first modified-bolus placement assessment, IOA was collected for 37.2% of trials, equaling 95.9% (range 88%–100%) across participants. For the secondary modified-bolus placement assessment IOA was collected for 22.3% of trials and equaled 98.8% (range 97%–100%). Ezra was the only participant with a tertiary modified-bolus placement assessment and IOA was collected for 100% of trials, equaling 100%.

Two investigators used a procedural integrity checklist (available from the first author upon request) to independently score whether the feeding therapist correctly or incorrectly implemented the protocol using video recordings on the primary modified-bolus placement assessment. Given the retrospective nature of this research, video recordings were only available for a subset of sessions for some participants, especially if the modified-bolus placement assessment spanned multiple meal blocks. Video records of the primary modified-bolus placement assessment were available for 26.3% of sessions for Ben, 24.1% of sessions for Kai, 35.6% of sessions for Mila, and 100% of sessions for Ezra. Each investigator collected procedural integrity for 100% of sessions in which video recordings were available. Investigators operationally defined correct implementation of upright spoon, underspoon, and flipped spoon using the descriptions in the general methods above, with the exception of details pertaining to presenting the spoon midline (all bite presentation methods), placement mid-way between the anterior and posterior section of the tongue (underspoon and flipped spoon) and applying pressure onto the tongue with the feeding utensil (underspoon and flipped spoon), due to challenges associated with retrospectively scoring procedural integrity using video recordings. The remainder of the checklist included items related to presenting the correct bolus size, delivering instruction (e.g., “Take a bite”), labeled praise (e.g., “Good job taking your bite”), ignoring inappropriate behavior during meals, re-presentation of expels within 3 seconds of an expulsion (if applicable), conducting a mouth clean check, appropriate use of reinforcement procedures (e.g., noncontingent access [NCA] vs. DRA), removing the utensil immediately following acceptance, and presenting the utensil at the next 30-second bite interval. Investigators scored a plus sign for items in which the therapist correctly implemented the protocol as intended and a minus sign for omission or commission errors in which the therapist did not implement protocol components as intended. We calculated a mean percentage for procedural integrity by dividing the number of items marked with a plus sign by the total number of possible items and multiplying by 100. Average procedural integrity for each participant were as follows: 93.3% (range 75%–100%) for Kai, 92.0% (range 75%–100%) for Ben, 100% for Ezra, and 89.6% (range 88%–100%) for Mila. Inter-coder agreement for procedural integrity equaled 93.7% (range 85%–98%) for Kai, 91.7% (range 86%–97%) for Ben, 92.3% (range 76%–100%) for Ezra and 100% for Mila.

Results

Modified-Bolus Placement Assessment Outcomes

Treatment teams implemented modified-bolus placement assessments with underspoon for 9.3%, but only 4.6% met inclusion, of patients admitted to the IMI feeding program. Three boys and one girl whose ages ranged from 23 to 112 months (M = 53.5, SD = 40.9) met inclusion and all participants were partially or fully dependent on formula orally or via tube to meet their nutritional and energy needs (ARFID A3b) and diagnosed with ASD and/or developmental disability. Modified-bolus placement methods used in the assessments included underspoon and/or flipped spoon compared to an upright spoon presentation. Treatment teams did not evaluate NUK^® brush in the modified-bolus placement assessments. Treatment teams implemented primary modified-bolus placement assessments using a multielement design for two participants (Kai and Ben) and a pairwise design for two participants (Ezra and Mila). Kai, Ben, and Mila underwent two modified-bolus placement assessments before returning to upright spoon and Ezra had three. Kai and Ezra’s (by meal block) secondary modified-bolus placement assessment also included a multielement design. Ben and Mila’s secondary and Ezra’s tertiary assessments used a pairwise design. However, the therapist did not employ the prescribed design for Mila’s secondary assessment.

Prior to the modified-bolus placement assessment, Kai’s treatment involved NRS, NCA to therapist attention and tangibles, and preventing wiping of food from the mouth (i.e., the feeder hovered their hands above the child’s hands to prevent contact between the child’s hands or arms and mouth/food in the mouth, presenting first a rice-size and then 0.25 cc of 16 total pureed foods on a large Maroon Spoon™. The treatment team implemented the modified-bolus placement assessment to treat expels that were high across foods when presented using upright spoon. Prior to the assessment, Kai also did not close his lips around the spoon to remove the bolus as the feeder retracted the spoon from his mouth.

Behavioral data from Kai’s primary and secondary modified-bolus placement assessments, three meals prior to discharge, and 3-month follow-up are displayed in Figures 3 and 4. During the modified-bolus placement assessment, the treatment team presented 0.25 cc of pureed foods on a small Maroon Spoon™ and compared upright spoon, flipped spoon, and underspoon across four meal blocks. The primary assessment included 13 of the 16 foods (avocado, apple cinnamon oatmeal, eggbeaters, green beans, peanut butter and jelly sandwich, mango, peas, brown rice, yams, pears, strawberry yogurt, broccoli and cheese, and spinach). Underspoon produced the highest levels of 5-second acceptance (M = 97.1%, range 80%–100%) and lowest levels inappropriate mealtime behavior (M = 5.7%, range 0%–20%) of the bite presentation methods. Acceptance was lower and more variable with upright spoon (M = 84.0%, range 60%–100%) and flipped spoon (M = 61.7%, range 20%–100%), and levels of inappropriate mealtime behavior were moderately low for upright spoon (M = 25.0%, range 0%–40%) and flipped spoon (M = 20.0%, range 0%–100%). Expels, which were the target response for this assessment, were lowest with underspoon (M = 14.3%, range 0%–40%) when compared with upright spoon (M = 68.0%, range 20%–100%) or flipped spoon (M = 28.3%, range 0%–50%), and underspoon resulted in higher mouth cleans (M = 82.9%, range 60%–100%) than upright spoon (M = 28%, range 0%–60%) or flipped spoon (M = 68.3%, range 40%–100%). Additionally, flipped spoon produced moderate levels of gagging (M = 50%, range 0%–100%) compared to upright spoon (M = 0%) and underspoon (M = 22.9%, range 0%–40%), a common side effect of flipped spoon that we observe clinically, although rarely reported in the literature (Ibañez et al., 2020). Data are available upon request.

Figure 3.

Percentage of 5-second acceptance and inappropriate mealtime behavior for Kai.

Figure 4.

Percentage of expels and mouth clean for Kai.

The treatment team moved forward with underspoon, and bolus faded first to 0.5 cc and then to 1 cc, rotating through all 16 foods. Following 140 sessions of underspoon, the treatment team conducted the secondary modified-bolus placement assessment using the same protocol presenting a 1-cc bolus on a small Maroon Spoon™ and compared upright spoon, underspoon, and flipped spoon. In the second modified-bolus placement assessment, one presentation method was implemented per meal block, across nine meal blocks. The therapist presented the 16 target foods equally across conditions (13 foods from the primary assessment and applesauce, chicken, and macaroni and cheese). The treatment team began with attempting to identify a bite presentation method that would allow for presenting a bolus larger than underspoon would allow (the amount cannot exceed 1 cc to ensure the proper formation of the bolus and placement), by comparing flipped spoon and upright spoon. Due to variable levels of acceptance with flipped spoon and variable expels with upright spoon, the team returned to the comparison with underspoon. For upright spoon, 5-second acceptance was higher (M = 92.2, range 40%–60%) and inappropriate mealtime behaviors were lower (M = 10.4, range 0%–60%) compared to other bite presentation methods. Although expels were highest with upright spoon (M = 27.8%, range 0%–100%) and flipped spoon (M = 9.1%, range 0%–60%) and lowest with underspoon (M = 0%), expels decreased across time to low levels for all presentation methods. Overall, underspoon produced higher and less variable levels of mouth clean (M = 100%) than flipped spoon (M = 90.9%, range 0%–100%) or upright spoon (M = 73.0%, range 0%–100%), but mouth clean also improved across time. The treatment team continued with upright spoon for the remainder of Kai’s 38-day admission as this allowed the team to continue bolus fading past a 1-cc bolus (the maximum bolus size allotted for the underspoon presentation method). By discharge, the treatment team had moved to presenting a slightly rounded bolus (puree food fills the entire bowl of the spoon with a small, rounded top, approximating 2 cc) of pureed food on a large Maroon Spoon™ and Kai’s expels (M 0%) and inappropriate mealtime behavior (M = 0%) were at zero levels with upright spoon presentations and 5-second acceptance (M = 100%) and mouth clean (M = 100%) also remained at high levels. Kai discharged consuming 17 new foods at a pureed texture and the treatment team met 100% of his admission goals. At a 3-month follow-up visit, Kai remained on upright spoon presentations and expels (M = 0%) and inappropriate mealtime behavior (M = 15%) remained low, and 5-second acceptance (M = 85%) and mouth cleans (M = 100%) were high.

Results of Ben’s primary and secondary modified-bolus placement assessments and three meals prior to discharge are displayed in Figures 5 and 6. Before conducting the modified-bolus placement assessment, Ben’s treatment package was comprised of NRS, NCA with tangibles, preventing wiping of food from the mouth, re-presentation of an expelled bite up to five times, and removal of three target foods with high expels. The therapist began treatment with presenting a rice-size bolus and bolus faded first to 0.13, 0.25, 0.5, and 0.75 cc bolus on a small Maroon Spoon™ across 16 pureed foods. Due to variability in expels at 0.5 cc, Ben’s treatment team collected data on expels per food to further assess if expels were higher with certain foods and subsequently removed grilled chicken, hamburger, and broccoli from his list of target foods, resulting in 13 total target foods (eggbeaters, sliced cheese, French fries, macaroni and cheese, white rice, waffle, mango, peach, pineapple, strawberry, carrot, and corn). The treatment team conducted a primary modified-bolus placement assessment for three meal blocks comparing three bolus presentation methods, including upright spoon, underspoon using a small Maroon Spoon™, and underspoon using an E-Z Spoon, to address expels that were high when foods were deposited using an upright spoon presentation. The therapist presented four foods, including sliced cheese, waffle, strawberry, and corn, in the primary assessment at 0.75 cc. Results of the assessment revealed that upright spoon produced notably higher levels of 5-second acceptance (M = 100%) and lower levels of inappropriate mealtime behavior (M = 66.7%, range 40%–100%) than both underspoon presentations. Underspoon with a small Maroon Spoon™ and underspoon with an E-Z spoon produced similarly low levels for 5-second acceptance (small Maroon Spoon™: M = 6.7%, range 0%–20%; E-Z Spoon: M = 13.3%, range 0%–40%) and high levels of inappropriate mealtime behavior (Small Maroon Spoon™: M = 93.3%, range 80%–100; E-Z Spoon: M = 86.7%, range 60%–100%). However, the treatment team selected underspoon using the small Maroon Spoon™ following the modified-bolus placement assessment as this method produced the lowest levels of expels (M = 26.7%, range 20%–40%) compared to upright spoon (M = 100%) and underspoon with an E-Z Spoon (M = 56.7%, range 40%–80%). Additionally, mouth cleans were also highest with underspoon presentations (small Maroon Spoon™: M = 76.7%, range 50%–100%; E-Z Spoon: M=73.3%, range 60%–80%) than upright spoon (M = 6.7%, range 0%–20%).

Figure 5.

Percentage of 5-second acceptance and inappropriate mealtime behavior for Ben.

Figure 6.

Percentage of expels and mouth clean for Ben.

The treatment team conducted a secondary modified-bolus placement assessment after 275 sessions of underspoon to determine if Ben could return to an upright spoon presentation. Prior to this, Ben’s treatment team had reduced the bolus to 0.05 cc and changed the reinforcement from NCA with tangibles to DRA for acceptance due to increases in the intensity of inappropriate mealtime behavior and reductions in acceptance. They continued to bolus fade in 0.1-cc increments until reaching 0.75 cc. In the second assessment, the therapist compared upright spoon and underspoon presentations across nine meal blocks and presented 0.75 cc measured using a syringe rotating through all 13 target foods. Results of the second assessment revealed that 5-s acceptance (upright spoon: M = 98.4%, range 80%–100%; underspoon: M = 89.7%, range 40%–100%) and inappropriate mealtime behavior (upright spoon: M = 4.8%, range 0%–40%; underspoon: M = 3.9%, range 0%–60%) were similar across presentation methods. Levels of expels remained low with underspoon (M = 1.7, range 0%–20%) and decreased over time with upright spoon (M = 33.5, range 0% to 100) but remained variable and in the moderate-to-low range. The treatment team continued treatment using upright spoon to allow for additional food to fit on the spoon. At discharge, Ben was consuming 1 cc of purees on a small Maroon Spoon™ and expels (M = 20.5, range 4%–39%) and inappropriate mealtime behavior (M = 2.9, range 0%–9%) remained moderately low, while 5-second acceptance (M = 85.0%, range 74%–96%) and mouth clean (M = 100%) remained at high levels. Ben discharged early, after 35 visits, due to the intensive program pausing treatment and canceling all non-urgent visits in response to the COVID-19 pandemic and did not participate in 3- or 6-month follow-up.

Prior to conducting a modified-bolus placement assessment, Ezra’s treatment included NRS, NCA with therapist attention and tangibles, and re-presentation of expels up to 10 times. The therapist first presented a rice-size bolus and bolus faded to 0.25 and 0.5 cc on a large Maroon Spoon™, rotating through 16 foods. Similar to Ben, Ezra’s treatment team also began collecting data on expels per food to detect foods that were frequently expelled. Data revealed high levels of expels across all food groups except fruit. Due to high expels with proteins, starches, and vegetables, and following consultation with the registered dietician, the treatment team removed eggbeaters, fish stick, lima beans, broccoli, white rice, and macaroni and cheese from the list of target foods, resulting in 10 total foods (chicken breast, turkey frank, strawberry, peach, pear, mango, green bean, corn, Spaghetti-O’s, grits). As Ezra retained fruit most frequently, the treatment team also began slowly increasing the variety of foods from each food group by rotating through all fruits and systematically replacing bites of fruit with other target foods from other food groups. Figures 7 and 8 depict behavioral data from Ezra’s three modified-bolus placement assessments, three meals prior to discharge, and 3- and 6-month follow-up data.

Figure 7.

Percentage of 5-second acceptance and inappropriate mealtime behavior for Ezra.

Figure 8.

Percentage of expels and mouth clean for Ezra.

Following the implementation of multiple antecedent strategies to reduce expels, the treatment team moved to conducting the primary modified-bolus placement assessment for one meal block comparing upright spoon and underspoon and presenting 0.25 cc of the four puree foods that included Spaghetti-O’s, turkey franks, peach, and green bean on a small Maroon Spoon™. The treatment team selected underspoon presentations as expels were lower (M = 20%, no range) than with upright spoon (M = 53.3%, range 40%–80%) during the primary modified-bolus placement assessment. Additionally, 5-second acceptance was higher with upright spoon (M = 100%) than underspoon (M = 86.7%, range 80%–100%) and inappropriate mealtime behavior was at zero levels for both presentation methods. In the 162 sessions following the primary modified-bolus placement assessment and prior to the second assessment, the treatment team changed the reinforcement contingencies from NCA with tangibles to DRA for acceptance and moved on to the next bite after 2 min without acceptance. The second assessment occurred across five meal blocks and compared upright spoon, underspoon, and flipped spoon using a 0.75 cc of eight puree foods (corn, chicken, peach, grits, pear, peas, Spaghetti-O’s, and strawberry). There was little differentiation in 5-second acceptance (upright spoon: M = 100%; underspoon: M = 98.12%, range 80%–100%; flipped spoon: M = 90.0%, range 60%–100%) and inappropriate mealtime behavior (upright spoon: M = 0%; underspoon: M = 1.8%, 0%–20%; flipped spoon: M = 3.3%, range 0%–20%) across bite presentation methods. Although mouth clean appeared high across presentations (upright spoon: M = 95%, range 80%–100%; underspoon: M = 100%; flipped spoon: M = 96.7%, range 80%–100%), results revealed expels were highest when foods were presented with upright spoon (M = 43.3%, range 0%–60%) compared to underspoon (M = 13.9, range 0%–40%) or flipped spoon (M = 21.7, range 20%–40%); thus, the treatment team selected flipped spoon to continue increasing the volume of food presented and retained. In the 552 sessions prior to the tertiary modified-bolus presentation assessment, the treatment team bolus faded to slightly rounded using flipped spoon and changed the reinforcement from DRA for acceptance to NCA with tangibles to improve the efficiency of the meal. In the third and final modified-bolus presentation assessment, a multielement comparison of a 0.5 cc bolus of four target foods (turkey frank, Spaghetti O’s, corn, and pear) on upright spoon and flipped spoon was conducted across one meal block in attempts to return to upright spoon. The treatment team continued forward with upright spoon as expels were at zero levels (M = 0%) for both bite presentation methods. Additionally, 5-second acceptance (M = 100%) and mouth clean (M = 100%) were at high levels and inappropriate mealtime behavior did not occur (M = 0%) for both upright and underspoon. At discharge (40 visits), the treatment team met 100% of Ezra’s discharge goals and he accepted 10 different foods at a rounded bolus on a large Maroon Spoon™ (transitioned from a small Maroon Spoon™). Expels and inappropriate mealtime behavior remained at zero levels (M = 0%), and 5-s acceptance and mouth clean remained high (M = 100%) prior to discharge and at both 3- and 6-month follow up visits.

Figures 9 and 10 display Mila’s behavioral data from the primary and secondary modified-bolus placement assessment and three meals prior to discharge. Mila’s family did not participate in any follow-up due to living out-of-state and their insurance covering only in-state services. For Mila, packing occurred at problematic levels and was the target behavior of interest. Mila’s treatment was comprised of NRS and DRA for clean mouth (i.e., swallowing at any time/without a limited hold) and the treatment team had bolus faded 16 foods from a rice-size bolus to 0.25 cc to 0.5 cc on a large Maroon Spoon™. For the primary modified-bolus placement assessment, which spanned eight meal blocks, the treatment team used Mila’s existing treatment with the additional component of removing the packed bite after 2 min of packing. During the primary assessment, the feeder conducted a multielement assessment comparing upright spoon with underspoon, rotating through 12 of the 16 target foods (green bean, yogurt, pancake, avocado, hot dog, peas, pasta, peach, French fry, string cheese, pineapple, and corn) at 0.5 cc on a small Maroon Spoon™. A 5-second acceptance was high for both upright (M = 100%) and underspoon (M = 96.4%, range 86%–100%) and inappropriate mealtime behavior was at zero levels (M = 0%). The treatment team discontinued the assessment and moved forward with underspoon due to lower levels of packing with underspoon (M = 15.7%, range 0%–43%) compared to upright spoon (M = 85.0%, range 60%–100%). Levels of mouth clean were higher with underspoon (M = 84.3%, range 57%–100%) than upright spoon (M = 15.0%, range 0%–40%).

Figure 9.

Percentage of 5-second acceptance and inappropriate mealtime behavior for Mila.

Figure 10.

Percentage of packing and mouth clean for Mila.

In the 154 sessions between primary and secondary assessments, the treatment team continued to bolus fade in 0.1-cc increments until reaching 0.9 cc. For the secondary modified-bolus placement assessment which spanned one meal block, the feeder compared upright spoon and underspoon and presented 0.9 cc of four foods (hot dog, wheat bread, strawberry, and peas). Results of the second assessment revealed that 5-second acceptance and mouth clean remained high with upright spoon and underspoon (M = 100%), and packing and inappropriate mealtime behavior remained at zero levels (M = 0%). The treatment team continued with upright spoon through discharge and packing remained low (M = 1.6%, range 0%–5%). Additionally, 5-second acceptance (M = 100%) and mouth clean (M = 98.4%, range 95%–100%) remained high and inappropriate mealtime behavior occurred at zero levels (M = 0%). By discharge (41 visits), the treatment team had bolus faded 19 foods to a rounded bolus and met 100% of Mila’s admission goals.

Social Validity Outcomes

Table 2 presents the eight items on the program satisfaction questionnaire, and the average and standard deviation for each item. Three out of four (75%) families completed the satisfaction questionnaire at the end of day treatment. One family’s day treatment admission was interrupted by the COVID-19 pandemic; therefore, the family did not receive a questionnaire. All three families reported they were satisfied or very satisfied with the program on all questions with scores ranging from 4.3 to 5 (out of 5). Caregivers collectively rated the highest possible score of 5 for responsiveness, improvement with feeding in clinic and at home, and improvement prior to starting the program. Additionally, all three families scored a 9 or 10 Net promoter score (Question #8) in which they were considered promoters of the program.

Table 2.

End of Admission Social Validity Survey Results.

Survey items	Score
Survey items	M	SD
1. Was your child’s care team responsive and understanding of your concerns during treatment?	5.0	0.0
2. If new feeding problems happen, how likely are you to come back to this program for treatment?	4.3	0.6
3. Did treatment help to improve your child’s feeding in clinic?	5.0	0.0
4. Did treatment help to improve your child’s feeding at home?	5.0	0.0
5. Do you feel that your child’s feeding problem has improved since you started the program?	5.0	0.0
6. Do you feel that your child's treatment was right for the feeding program?	4.7	0.6
7. Using a number between 0 and 10, where 0 is the worst visit possible and 10 is the best visit possible, how would you rate this visit?	9.0	1.0
8. How likely is it that you would recommend this clinic to a friend or colleague?	9.7	0.6

Note: Survey items 1 to 6 were on a 5-point Likert-type scale and items 7 and 8 were on a 10-point Likert-type scale.

Discussion

The current study is only the second to evaluate underspoon as a modified-bolus placement to address expels and packing for children with ARFID or PFD. All four children had moderate-to-severe oral-motor delays. Underspoon improved expels for three participants, and packs for one participant, when compared to upright spoon (all participants) and flipped spoon (Kai and Ezra only). Following treatment using underspoon, all participants successfully transitioned back to an upright spoon during their admission and this progress maintained at follow-up. Two participants (Kai and Ezra) transitioned to a large Maroon Spoon™ by the end of the program and two participants (Ben and Mila) remained on a small Maroon Spoon™. All participants were discharged with a bolus greater than the amount that could be presented with underspoon, and participants were eating an average of 15 new foods (range 10–19). Three participants (Kai, Mila, and Ezra) met 100% of their admission goals by the end of their admission. The program did not account for admission goals for Ben due to unexpectedly discharging early from the program due to the COVID-19 pandemic. Admission duration in days ranged from 35 to 41 days (M = 38.5, SD = 2.6). Caregivers also rated their admission as socially acceptable.

Some conclusions can be conservatively drawn from this retrospective CCCS. Kai, who had moderate oral-motor delays, displayed impaired lip strength, and the remaining three participants had severe oral-motor delays characterized by severely impaired jaw strength, durational chew, (i.e., chewing consecutively until food is masticated) and tongue lateralization. Parallels may be made between participants in the current study and the participant in the Gibbons et al. (2007) study, in that they were all dependent on formula to meet their nutritional needs and exhibited oral-motor delays that negatively impacted P.O. intake of pureed foods. Tongue thrust, as displayed by the participant in Gibbons et al. (2007), may have also severely impacted the child’s ability to lateralize, in addition to retaining food, although not explicitly stated. Thus, these conclusions align with the findings of Volkert et al. (2019) supporting the use of modified-bolus placement in addressing feeding concerns for children with moderate-to-severe oral-motor delays.

These data also suggest clinicians may try antecedent strategies first and then consequence-based strategies with the largest evidence base before implementing a modified-bolus placement assessment (Ledford et al., 2018; Volkert & Piazza, 2012). Prior to the primary modified-bolus placement assessment, treatment teams across participants evaluated a variety of antecedent strategies including bolus fading (all participants), removal of foods with higher levels of problematic behavior (Ben and Ezra), and systematically increasing the variety of foods presented (Ezra). They then evaluated re-presentation for two participants with whom expels were problematic (Ben and Ezra) and differential reinforcement of alternative behavior for clean mouth for Mila, with whom packing was problematic. Although Kai also displayed problematic levels of expels, the clinical team did not attempt re-presentation with Kai prior to beginning the modified-bolus placement assessment as expels were a result of open-mouth posture without lip closure (data available upon request). Furthermore, underspoon was evaluated more frequently than flipped spoon in the modified-bolus placement assessment and no evaluations involved NUK^® brush, although both modified-bolus placement methods possess existing peer-reviewed research (Volkert et al., 2019). A limitation of underspoon is that the therapist can only present up to 1 cc on the bottom of the spoon, thus constraining the amount of food that can be presented per bite and require transitioning to upright spoon or another modified-bolus placement method (e.g., NUK^® brush or flipped spoon) to present larger volumes of food. Clinicians may also consider additional behavioral side effects (e.g., gagging with Kai) and social validity when selecting bolus placement methods to compare, but these considerations were not reported when reviewing patient files. Given that these data represent a small subset of evaluations involving modified-bolus placement, best practice may be established through conclusions yielded from a systematic review or meta-analysis.

One limitation of the current study was that researchers collected procedural integrity on only the primary modified-bolus placement assessment for each participant. Given the retrospective nature of this study, video files were not available for all modified-bolus placement assessments thus limiting the number of sessions that investigators could score procedural integrity data on, if any. Lack of procedural integrity threatens the internal validity of the study and limits conclusions as failure to present these data do not allow an understanding of the degree to which the therapist implemented the assessment as planned. Additionally, investigators did not account for presenting the spoon midline, placement of the bolus (i.e., halfway between the anterior and posterior tongue) and how much pressure the feeder applied (i.e., gentle downward pressure) to the tongue across underspoon and flipped spoon when measuring procedural integrity due to challenges associated with accurately observing placement via video or scoring pressure. They also did not account for therapist error in using the correct experimental design (e.g., Mila’s second modified-bolus placement assessment was intended to be implemented using a pairwise design).

Another limitation of retrospective research is that it is currently unclear as to the mechanisms that lead to successful transition back to upright spoon. In our standard clinical practice, modified-bolus placement is commonly used with patients who exhibit moderate-to-severe oral motor deficits (Sharp et al., 2020; Volkert et al., 2019), as displayed by all participants in the current study. Kai and Mila, who had moderate oral-motor delays, experienced fewer sessions with underspoon prior to successfully returning to upright spoon than the remainder of the participants with severe oral-motor delays. Kai’s open-mouth posture and poor lip closure suggested that he would benefit from modified-bolus placement for removing the bolus from the spoon (e.g., Bloomfield et al., 2021; Stubbs et al., 2018). Of note, when exposed to underspoon, the participants did not require physical prompting (e.g., chin prompt). Past research reported using a chin prompt to address open-mouth posture and poor lip closure, such as with Kai (Wilkins et al., 2011), and with flipped spoon to reduce expels (Dempsey et al., 2011). However, although Kai exhibited no or poor lip closure (M = 0) during the primary modified bolus placement assessment and throughout the 140 sessions of underspoon (i.e., resulting in scraping the bolus on the upper teeth), lip closure improved without targeted intervention upon return to upright spoon during the secondary assessment (M = 93%, range 60%–100%). Future research may seek to prospectively explore additional oral-motor variables, including, but not limited to, mouth/lip closure, tongue coordination, and food retention, that contribute to decision-making associated with considering modified-bolus placement and returning to the upright spoon. Another avenue of prospective research may also be to better control for the presentation of foods. This may involve presenting a smaller subset of foods for the entire assessment to ensure uniformity across bite presentation methods (Ben, Ezra) or conducting the assessment until the therapist is able to present each food with each bite presentation method (Mila).

Future research may also evaluate the efficacy of underspoon at increasing acceptance of bites. Acceptance was moderately higher for Ezra and substantially higher for Ben when the therapist used upright spoon compared to underspoon. This is not entirely surprising as underspoon involves the additional step of flipping the spoon 180° at the lips prior to inserting the spoon into the child’s open mouth which can provide additional opportunities for inappropriate mealtime behavior to occur and requires that the child’s mouth remain open for both of these steps. For Kai, 5-second acceptance was higher with underspoon than upright spoon and flipped spoon likely due to an increased latency to deposit the bite. For upright spoon, the therapist delays scraping the bolus onto the teeth to provide a brief opportunity for lip closure and flipped spoon involves applying pressure on the tongue twice prior to depositing the bolus. Thus, the topography of food refusal (i.e., active or passive) that the child engages in may influence acceptance and the bolus placement method selection.

The current study supports the efficacy of underspoon as a modified-bolus placement method to treat expels and packing for children with ASD and/or related developmental disabilities and ARFID or PFD who exhibit oral-motor delays. Underspoon represents another modified-bolus placement method to consider, adding to the existing peer-reviewed research on flipped spoon and NUK^® brush to address behavior that interfere with mouth clean. However, more research is needed to establish decision-making rules surrounding the selection of bite presentation methods for assessment, the efficacy of underspoon compared to other modified-bolus placement methods, and potential procedural variations of underspoon.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

William G. Sharp

Valerie M. Volkert

Author Biographies

Meara X. H. McMahon, PhD, BCBA: Dr. McMahon is a Board Certified Behavior Analyst and postdoctoral fellow at Children’s Multidisciplinary Feeding Program where she also completed pre-doctoral internship.

Kristin L. Hathaway, PhD, BCBA: Dr. Hathaway is a licensed psychologist at Children’s Multidisciplinary Feeding Program and Board Certified Behavior Analyst.

Abby K. Hodges, PhD, BCBA: Dr. Hodges is a graduate of University of Denver, studying Child, Family, and School Psychology and currently completing her postdoctoral fellowship at the Marcus Autism Center with Emory School of Medicine.

William G. Sharp, PhD: Dr. Sharp is a clinical psychologist and director of Children’s Healthcare of Atlanta’s Multidisciplinary Feeding Program.

Valerie M. Volkert, PhD, BCBA-D: Dr. Volkert is a licensed psychologist, Board Certified Behavior Analyst, and Site Manager of Children’s Healthcare of Atlanta’s Multidisciplinary Feeding Program at the Marcus Autism Center.

References

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Author.

Bloomfield

B. S.

Wawrzonek

A. J.

Volkert

V. M.

(2021). Preliminary comparison of redistribution at different intervals to reduce packing in children with avoidant/restrictive food intake disorder. Behavioral Development, 26(1), 1–13. https://doi.org/10.1037/bdb0000098

Dempsey

Piazza

C. C.

Groff

R. A.

Kozisek

J. M.

(2011). A flipped spoon and chin prompt to increase mouth clean. Journal of Applied Behavior Analysis, 44(4), 961–965. https://doi.org/10.1901/jaba.2011.44-961

Gibbons

B. G.

Williams

K. E.

Riegel

K. E.

(2007). Reducing tube feeds and tongue thrust: Combining an oral-motor and behavioral approach to feeding. The American Journal of Occupational Therapy, 61(4), 384–391. https://doi.org/10.5014/ajot.61.4.384

Girolami

P. A.

Boscoe

J. H.

Roscoe

(2007). Decreasing expulsions by a child with a feeding disorder: Using a brush to present and re-present food. Journal of Applied Behavior Analysis, 40(4), 749–753. https://doi.org/10.1901/jaba.2007.749-753

Goday

P. S.

Huh

S. Y.

Silverman

Lukens

C. T.

Dodrill

Cohen

S. S.

Delaney

A. L.

Feuling

M. B.

Noel

R. J.

Gisel

Kenzer

Kessler

D. B.

Kraus

Camargo

Browne

Phalen

J. A.

(2019). Pediatric feeding disorder: Consensus definition and conceptual framework. Journal of Pediatric Gastroenterology and Nutrition, 68(1), 124–129. https://doi.org/10.1097/MPG.0000000000002188

Hagopian

L. P.

(2020). The consecutive controlled case series: Design, data-analytics, and reporting methods supporting the study of generality. Journal of Applied Behavior Analysis, 53(2), 596–619. https://doi.org/10.1002/jaba.691

Ibañez

V. F.

St Peters

K. P.

Paul

J. A.

Vollmer

T. R.

(2021). Further evaluation of modified-bolus-placement methods during initial treatment of pediatric feeding disorders. Journal of Applied Behavior Analysis, 54(1), 287–308. https://doi.org/10.1002/jaba.748

Ibañez

V. F.

Peterson

K. M.

Crowley

J. G.

Haney

S. D.

Andersen

A. S.

Piazza

C. C.

(2020). Pediatric prevention: Feeding disorders. Pediatric Clinics of North America, 67(3), 451–467. https://doi.org/10.1016/j.pcl.2020.02.003

10.

Kovacic

Rein

L. E.

Szabo

Kommareddy

Bhagavatula

Goday

P. S.

(2021). Pediatric feeding disorder: A nationwide prevalence study. The Journal of Pediatrics, 228, 126–131.e123. https://doi.org/10.1016/j.jpeds.2020.07.047

11.

Ledford

J. R.

Whiteside

Severini

K. E.

(2018). A systematic review of interventions for feeding-related behaviors for individuals with autism spectrum disorders. Research in Autism Spectrum Disorders, 52, 69–80. https://doi.org/10.1016/j.rasd.2018.04.008

12.

Milnes

S. M.

Piazza

C. C.

Ibañez

V. F.

Kozisek

J. M.

(2019). A comparison of Nuk presentation and Nuk redistribution to treat packing. Journal of Applied Behavior Analysis, 52(2), 476–490. https://doi.org/10.1002/jaba.532

13.

Munk

D. D.

Repp

A. C.

(1994). Behavioral assessment of feeding problems of individuals with severe disabilities. Journal of Applied Behavior Analysis, 27(2), 241–250. https://doi.org/10.1901/jaba.1994.27-241

14.

Patel

M. R.

Piazza

C. C.

Layer

S. A.

Coleman

Swartzwelder

D. M.

(2005). A systematic evaluation of food textures to decrease packing and increase oral intake in children with pediatric feeding disorders. Journal of Applied Behavior Analysis, 38(1), 89–100. https://doi.org/10.1901/jaba.2005.161-02

15.

Patel

M. R.

Piazza

C. C.

Martinez

C. J.

Volkert

V. M.

Santana

C. M.

(2002). An evaluation of two differential reinforcement procedures with escape extinction to treat food refusal. Journal of Applied Behavior Analysis, 35(4), 363. https://doi.org/10.1901/jaba.2002.35-363

16.

Patel

M. R.

Piazza

C. C.

Santana

C. M.

Volkert

V. M.

(2002). An evaluation of food type and texture in the treatment of a feeding problem. Journal of Applied Behavior Analysis, 35(2), 183–186. https://doi.org/10.1901/jaba.2002.35-183

17.

Piazza

C. C.

Patel

M. R.

Gulotta

C. S.

Sevin

B. M.

Layer

S. A.

(2003). On the relative contributions of positive reinforcement and escape extinction in the treatment of food refusal. Journal of Applied Behavior Analysis, 36(3), 309–324. https://doi.org/10.1901/jaba.2003.36-309

18.

Prasse

J. E.

Kikano

G. E.

(2009). An overview of pediatric dysphagia. Clinical Pediatrics, 48(3), 247–251. https://doi.org/10.1177/0009922808327323

19.

Reichheld

F. F.

(2003). The one number you need to grow. Harvard Business Review, 81, 46–54.

20.

Saini

Jessel

Iannaccone

J. A.

Agnew

(2019). Efficacy of functional analysis for informing behavioral treatment of inappropriate mealtime behavior: A systematic review and meta-analysis. Behavioral Interventions, 34(2), 231–247. https://doi.org/10.1002/bin.1664

21.

Seiverling

Towle

Hendy

H. M.

Pantelides

(2018). Prevalence of feeding problems in young children with and without autism spectrum disorder: A chart review study. Journal of Early Intervention, 40(4), 335–346. https://doi.org/10.1177/1053815118789396

22.

Sevin

B. M.

Gulotta

C. S.

Sierp

B. J.

Rosica

L. A.

Miller

L. J.

(2002). Analysis of response covariation among multiple topographies of food refusal. Journal of Applied Behavior Analysis, 35(1), 65–68. https://doi.org/10.1901/jaba.2002.35-65

23.

Sharp

W. G.

Berry

R. C.

McCracken

Nuhu

N. N.

Marvel

Saulnier

C. A.

Klin

Jones

Jaquess

D. L.

(2013). Feeding problems and nutrient intake in children with autism spectrum disorders: A meta-analysis and comprehensive review of the literature. Journal of Autism and Developmental Disorders, 43(9), 2159–2173. https://doi.org/10.1007/s10803-013-1771-5

24.

Sharp

W. G.

Harker

Jaquess

D. L.

(2010). Comparison of bite-presentation methods in the treatment of food refusal. Journal of Applied Behavior Analysis, 43(4), 739–743. https://doi.org/10.1901/jaba.2010.43-739

25.

Sharp

W. G.

Jaquess

D. L.

Morton

J. F.

Herzinger

C. V.

(2010). Pediatric feeding disorders: A quantitative synthesis of treatment outcomes. Clinical Child and Family Psychology Review, 13(4), 348–365. https://doi.org/10.1007/s10567-010-0079-7

26.

Sharp

W. G.

Odom

Jaquess

D. L.

(2012). Comparison of upright and flipped spoon presentations to guide treatment of food refusal. Journal of Applied Behavior Analysis, 45(1), 83–96. https://doi.org/10.1901/jaba.2012.45-83

27.

Sharp

W. G.

Stubbs

K. H.

(2019). Avoidant/restrictive food intake disorder: A diagnosis at the intersection of feeding and eating disorders necessitating subtype differentiation. International Journal of Eating Disorders, 52(4), 398–401. https://doi.org/10.1002/eat.22987

28.

Sharp

W. G.

Stubbs

K. H.

Adams

Wells

B. M.

Lesack

R. S.

Criado

K. K.

Simon

E. L.

McCracken

C. E.

West

L. L.

Scahill

L. D.

(2016). Intensive, manual-based intervention for pediatric feeding disorders: Results from a randomized pilot trial. Journal of Pediatric Gastroenterology and Nutrition, 62(4), 658–663. https://doi.org/10.1097/MPG.0000000000001043

29.

Sharp

W. G.

Volkert

V. M.

Scahill

McCracken

C. E.

McElhanon

(2017). A systematic review and meta-analysis of intensive multidisciplinary intervention for pediatric feeding disorders: How standard is the standard of care? The Journal of Pediatrics, 181, 116–124.e114. https://doi.org/10.1016/j.jpeds.2016.10.002

30.

Sharp

W. G.

Volkert

V. M.

Stubbs

K. H.

Berry

R. C.

Clark

M. C.

Bettermann

E. L.

McCracken

C. E.

Luevano

McElhanon

Scahill

(2020). Intensive multidisciplinary intervention for young children with feeding tube dependence and chronic food refusal: An electronic health record review. The Journal of Pediatrics, 223, 73–80.e2. https://doi.org/10.1016/j.jpeds.2020.04.034

31.

Stubbs

K. H.

Volkert

V. M.

Rubio

E. K.

Ottinger

(2018). A comparison of flipped-spoon presentation and redistribution to decrease packing in children with feeding disorders. Learning and Motivation, 62, 103–111. https://doi.org/10.1016/j.lmot.2017.03.007

32.

Volkert

V. M.

Piazza

C. C.

(2012). Pediatric feeding disorders. In Sturmey

Hersen

(Eds.), Handbook of evidence-based practice in clinical psychology (pp. 456–481). Wiley. https://doi.org/10.1002/9781118156391.ebcp001013

33.

Volkert

V. M.

Sharp

W. G.

Clark

M. C.

Ormand

Rubio

E. K.

McCracken

Bryan

(2019). Modified-bolus placement as a therapeutic tool in the treatment of pediatric feeding disorders: Analysis from a retrospective chart review. Journal of Speech, Language, and Hearing Research, 62(9), 3123–3134. https://doi.org/10.1044/2019_jslhr-s-18-0268

34.

Volkert

V. M.

Vaz

P. C. M.

Piazza

C. C.

Frese

Barnett

(2011). Using a flipped spoon to decrease packing in children with feeding disorders. Journal of Applied Behavior Analysis, 44(3), 617–621. https://doi.org/10.1901/jaba.2011.44-617

35.

Von Elm

Altman

D. G.

Egger

Pocock

S. J.

Gøtzsche

P. C.

Vandenbroucke

J. P

. (2007). The strengthening the reporting of observational studies in epidemiology (STROBE) statement: Guidelines for reporting observational studies. Bulletin of the World Health Organization, 85, 867–872.

36.

Wilkins

J. W.

Piazza

C. C.

Groff

R. A.

Volkert

V. M.

Kozisek

J. M.

Milnes

S. M.

(2014). Utensil manipulation during initial treatment of pediatric feeding problems. Journal of Applied Behavior Analysis, 47(4), 694–709. https://doi.org/10.1002/jaba.169

37.

Wilkins

J. W.

Piazza

C. C.

Groff

R. A.

Vaz

P. C. M.

(2011). Chin prompt plus re-presentation as treatment for expulsion in children with feeding disorders. Journal of Applied Behavior Analysis, 44(3), 513–522. https://doi.org/10.1901/jaba.2011.44-513