Escape Extinction and Negative Reinforcement in the Treatment of Pediatric Feeding Disorders: a Single Case Analysis

Abstract

Pediatric feeding disorders are common among children with developmental disabilities and can have detrimental effects on growth and development. An escape extinction and negative reinforcement-based approach to treating food refusal was examined in a child with cerebral palsy. A changing criterion treatment design was implemented, which allowed the child to exit the treatment area contingent upon the acceptance and ingestion of a pre-determined number of bites. Food acceptance ranged from one to three bites at baseline and exceeded the pre-set criteria for mastery, at 14 bites during the final intervention phase.

• The study will contribute to the current literature on negative reinforcement procedures used in the treatment of pediatric feeding problems.

• The study will contribute to increasing the availability of literature pertaining to pediatric feeding problems among children with complex disabilities such as cerebral palsy.

• The intervention is brief with components to the treatment package which increases utility and ease of implementation.

• The study demonstrates the applicability of changing criterion design within clinical settings.

Pediatric feeding disorders are common among children with developmental disabilities, which affects their ability to grow and thrive and presents numerous developmental challenges. These challenges may include medical complications, “oral-motor dysfunction,” and behavioral problems that interfere with the child’s ability to consume the nutrients necessary for appropriate development (Williams et al. 2010; Sharp et al. 2010).

Given the complexity of feeding problems, often times a behavioral approach to treatment is included in the treatment package. Much of the literature related to pediatric feeding problems that are not otherwise accounted for by medical complications suggest a lack of motivation as demonstrated by children engaging in escape-maintained behavior during meal time (Piazza et al. 2004; Vaz et al. 2011). However, contemporary research suggests the integration of negative reinforcement that involves the removal of the non-preferred foods and presentation of preferred foods as part of the treatment package. Vaz et al. (2011) increased self-feeding by the removal of non-preferred food after one self-fed bite. The literature includes research that evaluates a variety of reinforcement procedures that have shown to be effective when treating feeding problems (Dawson et al. 2003; Piazza et al. 2004).

There is currently a lack of research evaluating the efficacy of negative reinforcement by means of terminating mealtime and exiting the meal location. In this study, the contingency was developed so that after a required number of bites had been met, the participant would exit not only the meal but also the mealtime area. A changing criterion design was chosen due to the ease at which the number of acceptable bites can be modified and increased over the course of the intervention.

The participant in this study is an 8-year-old male diagnosed with cerebral palsy. He resides with both parents and is an only child. Due to his condition, the participant presents with several areas of deficit in the areas of gross motors skills and fine motor as well as limited expressive language. His receptive language, however, appears to be near age appropriate as compared to same-aged peers based on recent testing.

A functional analysis was conducted on food refusal and elopement from the meal time area. Results from the functional analysis indicated that the participant’s challenging mealtime behavior is maintained by escape. As such, a negative reinforcement protocol was developed that allowed the participant to escape the meal as well as the mealtime area contingent upon accepting a pre-determined number of bites. In other words, an escape extinction procedure was implemented during the intervention phases until the participant met the criteria for that phase.

The participant did not engage in preferred activities such as games or leisure activities after exiting the meal time area. The potential for interaction effects by way of positive reinforcement were controlled in that the participant seemingly did not engage in any preferred activities when exiting the meal-time area following the intervention, after the session criteria were met.

Foods that the participant historically refused according to parent reports and data collection were selected for the baseline session and intervention phases. Baseline data were collected for three feeding sessions for 30 min per session. A criterion was set at 5 bites in the first intervention phase, followed by 7, and then 10 in the phases that followed. Following 10 bites, the criterion was lowered to 7 bites again in order to strengthen the functional relationship and to further demonstrate experimental control. The criterion was set to 12 bites in the final phase of the intervention. In order for the participant to meet the criterion level, the specified number of bites needed to be ingested within a 30-min period. If the participant ingested all of the required bites before the 30 min period, then the participant was allowed to terminate meal time and leave the meal area.

Treatment of feeding problems includes an interdisciplinary team approach (Piazza et al. 2009). Therefore, a certified nutritionist was consulted during the treatment planning who helped design a diet that included a protein of poultry, fish, or red meat for each meal in addition to a starch and a vegetable. The participant’s daily caloric consumption increased as a result of the intervention. The nutritionist consultant indicated it to now be at a clinically appropriate level, as it had not been prior to intervention.

Figure 1 demonstrates the results of the intervention. In the baseline phase, the participant ingested an average of 2 bites across the three sessions. When the intervention began and the criterion was set to 5 bites, the participant ingested 5 bites during two sessions and exceeded the criterion set by ingesting 6 bites in one session. In the second intervention phase, criterion was set at 7 bites. The participant ingested fewer than the set criterion on one out of the three sessions. Criterion was met on two sessions which resulted in an increase in criterion after the ninth session. At this time, the participant met or exceeded criterion in all three sessions. In sessions 13 through 15, criterion was decreased to 7 bites and the participant adjusted his behavior as evidenced by ingesting 7 bites during two sessions and 8 bites during one session. The final criterion was set to 12 bites, which was met and exceeded the target criterion for the final phase. During all of the intervention sessions, the participant ingested all of the required bites in under 30 min.

Fig. 1.

Results of the intervention

The intervention proved to be effective for this child in increasing the number of bites accepted during meal time and therefore increasing his caloric intake to clinically appropriate levels for increased nourishment. The findings indicate that the escape-contingency served as an effective reinforcer to shape meal time behavior by way of food acceptance. Although foods were chosen based on parent reports, it may have strengthened the study to complete a preference assessment on those foods identified by parents, prior to the intervention and use those for the feeding sessions in reverse preference order. Additionally, in order to control for other variables, foods presented during feeding sessions were all of similar textures. Future research could be expanded to assess food acceptance with a variety of accepted flavors, textures, and consistencies. Researchers can also examine the impact of this intervention on challenging and replacement behaviors. This study contributes to the literature involving the treatment of feeding problems in that it demonstrates that escape extinction and negative reinforcement as a treatment for escape-maintained mealtime behaviors was found to be an efficient and effective intervention with relative ease of implementation. It also demonstrates that these interventions can be implemented with individuals who have complex needs such as those with cerebral palsy.

Methods

The participant in this study was an 8-year-old male with cerebral palsy referred for behavioral feeding treatment due to food refusal. It was also identified via a functional analysis that escape-maintained behavior presented not only during meal time, but reliably in the meal time area. Although the child suffered motoric complications due to his condition, he had been cleared for feeding therapy by his team of physicians. It was identified that he had the ability to chew, swallow, and complete all aspect of self-feeding. Sessions were conducted in the child’s home specifically in the dining area where most meals occurred, which contained a table, two chairs with arms, plates, and spoons or forks.

An observer trained in the protocol scored a bite acceptance as any instance of the child depositing a bite of food into his mouth. Any bites that occurred within a 30-min period, regardless of latency between bites, were scored as accepted. Each bite deposited was recorded with a tally mark, and the total was calculated at the end of each session. A second observer simultaneously but independently observed and took data during 27 % of the feeding sessions. Inter-observer agreement was calculated by dividing the smaller number of tallies from one observer by the larger number from the other and converting into a percentage. Mean level of agreement was 100 %.

Effects of an escape extinction and negative reinforcement-based procedure were examined using a changing criterion design. The first phase was baseline, where there were no behavioral interventions implemented. The child was simply presented with a plate of food and asked to eat. The behavior analyst conducted 5 sessions per week until the final session. The treatment was implemented over 15 days in total. After three sessions of baseline data collection, the intervention began. The first criterion level was set at 5 bites, which needed to be accepted in a 30-min period before the child was allowed to exit the meal as well as the meal area. Immediately following the presentation of the food, the behavior analyst implemented escape extinction by consistently presenting the food items each time the child attempted to push it away.

The child was provided frequent reminder and feedback such as, “you need two more bites before you can leave,” “nice work,” one more,” etc. It should be noted that the child was not force fed nor was the fork placed into the child’s mouth for him. He was required to pick up the food items with the fork and place it in his mouth independently in order for it to be marked as an accepted bite. Immediately following the acceptance and ingestion of the last bite required in each phase, the child was allowed access to other areas of the home and escape both the meal as well as the meal location—even if this occurred before the 30 min were completed. Potential effects of positive reinforcement were controlled for by noting that when leaving the meal area, the participant seemingly did not engage in any preferred activities.

Intervention was conducted for 15 sessions, all of which the parents were present for observation. After the completion of the intervention, the behavior analyst worked with the parents on role-play, modeling, and in vivo observation and feedback in order to facilitate generalization of the intervention to other feeders and so that they could implement in other settings. Maintenance was monitored by monthly routine visits for three consecutive months following the intervention; however, at the time of this writing maintenance data was not available.