ارزیابی چندگانه خط معیار درمان هراس از غذا در یک پسر جوان

A multiple-baseline design was used to evaluate the effectiveness of a behavioral treatment program using modeling, graduated exposure, and contingency management to treat food phobia in a 4-year-old boy. In addition, a treatment component involving time-out and re-introduction of the initial request to consume the target food was added to reduce vomiting behavior that developed during the course of treatment. The volume and range of foods consumed by the participant increased, and observer-rated anxiety and vomiting decreased over the course of treatment. The results of this controlled evaluation suggest that this treatment program was responsible for the observed changes, which were maintained at 6-month follow-up.

Although many specific phobias have been described in the psychopathology and treatment literatures, perhaps none has as serious health implications as the excessive and persistent fear and avoidance of chewing, swallowing, or choking on food or fluids. This phenomenon has been described in the child and medical literatures, where it is typically called “food phobia,” as well as in the adult psychological literature, where it is commonly referred to as “choking phobia” (McNally, 1994) or “food aversion” (DeSilva & Rachman, 1987). It is likely that the term food phobia has been used most often with children because they are often unable to verbalize precisely what it is they fear (i.e., choking on food rather than food per se). Indeed, recent reports of both limited literatures suggest that this phenomenon (which I will refer to as “food phobia” in order to remain consistent with the child literature) is characterized by the fear and avoidance of chewing or swallowing food or fluids, which most often occurs directly following a conditioning experience involving choking on food or vomiting (Chatoor, Conley, & Dickson, 1988; McNally, 1994; Singer, Ambuel, Wade, & Jaffe, 1992). The prevalence and long-term course of food phobia is unknown, as this problem has received little empirical attention. Pediatric feeding disorders in general occur in 25% of all children, and are associated with significant weight loss, medical problems, and the development of behavioral disorders such as bulimia and anorexia nervosa (Manikam & Perman, 2000; Woolston, 1991; Nock, in press). Given the negative physical and psychological outcomes associated with pediatric feeding problems in general, and the dearth of research available on food phobia in particular, there is a great need for empirical work focused on this problem. Most of the literature on food phobia consists of individual case studies, several of which have provided initial support for behavioral treatment approaches (see McNally, 1994 for a review). However, there are several key limitations of this literature. First, the majority of studies have included adult participants. Thus, there is limited information about the efficacy of these approaches with children and adolescents—and a particular lack of information about food phobia in pre-pubertal children (Chorpita, Vitali, & Barlow, 1997; Singer, Ambuel, Wade, & Jaffe, 1992). Second, virtually all of the treatment research on food phobia consists of uncontrolled evaluations, limiting the conclusions that can be drawn about the effectiveness of the treatments employed. A notable exception is a recent study by Chorpita and colleagues (1997) that used a multiple-baseline experimental design to demonstrate the effectiveness of a behavioral treatment program with a 13-year-old girl with food phobia. This study provides an impressive demonstration of the effectiveness of graduated exposure and contingency management in the treatment of food phobia; however, the authors concluded that there is a need to implement and evaluate such treatment strategies with younger children. Given developmental differences in cognitive ability and in typical methods of treatment delivery (e.g., individual treatment versus inclusion of parents in treatment), it is unlikely that the effectiveness of adolescent and adult treatment programs will generalize to young children without major modifications. Indeed, there is a great need to develop and evaluate more developmentally sensitive treatment programs for child behavior problems (Kazdin, 2000; Ollendick & Vasey, 1999). To address these limitations, the current study evaluated the effectiveness of a behavioral treatment program using therapist and parent modeling, graduated exposure, and contingency management to treat food phobia in a 4-year-old boy. This study employed a multiple-baseline design across different food categories to provide a controlled evaluation of the effectiveness of this treatment program. Moreover, the participant began vomiting immediately after swallowing certain foods, so a treatment component involving time-out from reinforcement and a re-introduction of the initial request to consume the target food was implemented to address this problem.

Data representing the volume (number of servings) and range (four categories) of food consumed by the participant each week over the 27-week treatment period and at 6-month follow-up are presented in Fig. 1. The figure depicts the number of servings of each category of food consumed by the participant during each week of the study. In this design, data were collected and graphed separately for each of the four food categories, and the intervention was applied to the consumption of each of these four food categories sequentially. Within the context of the multiple-baseline design used in this study, it could be concluded that the treatment program caused an increase in food consumption if the number of servings of food in each category increased when, and only when, the treatment program focused on that food category, thus ruling out the possibility that other factors (e.g., history, maturation, statistical regression, relationship factors) were responsible for any observed changes. In contrast to research employing multiple subjects, in which statistical criteria are used to evaluate whether there is an average treatment effect, single-case experimental designs typically employ criteria that apply to the visual inspection of treatment data to evaluate whether a reliable, consistent treatment effect exists. These criteria have been described in detail elsewhere and include the presence of a change in the magnitude (i.e., mean and level) and rate (i.e., trend and latency) of the target behavior(s) across treatment phases (see Kazdin (1982) and Kazdin (1984)). The satisfaction of the criteria for visual inspection require marked treatment effects that are generally considered more stringent than those for achieving statistical significance (Kazdin, 1982).