Commentary: Conducting Randomized Controlled Trials of Psychological Treatment to Improve the Outcomes of Recurrent Abdominal Pain: Contributions and Challenges

Abstract

In this issue, Robins and colleagues presented the results of a randomized controlled trial (RCT) that documented the efficacy of cognitive behavioral treatment (CBT) for reducing the frequency of pain and school absences in a group of children with recurrent abdominal pain (RAP). This is an important clinical population for several reasons: (a) relatively large numbers of children and adolescents are affected with RAP (Campo, Jansen-McWilliams, Comer, & Kelleher, 1999); (b) RAP has significant functional impact on children’s school, physical activities, and health care utilization (Wasserman, Whitington, & Rivara, 1988); and (c) RAP persists into adulthood among a relatively large percentage of patients (Walker, Guite, Duke, Barnard, & Greene, 1998). Despite its clinical importance, RAP is a complex condition that is very difficult to manage within current models of medical treatment (Tarkan, 2004). Although RAP commonly presents to primary care physicians and pediatric gastroenterologists as a “medical” condition, this condition reflects a heterogeneous constellation of symptoms with a multifactorial etiology (e.g., somatic lifestyle, critical life events, learned response patterns, and family environment) and does not typically improve with medical treatment alone (Friedrich & Jaworski, 1995). This is also the case for other pediatric conditions that involve recurrent pain, for which psychological treatments are emerging as effective front line interventions. However, as Eccleston, Morley, Williams, Yorke, and Mastroyannopoulou (2002) have highlighted, few RCTs have evaluated psychological treatments for nonheadache chronic pain in children, and available trials have had significant limitations, including the failure to assess nonpain-related outcomes such as functioning, mood, and quality of life. Robins et al. have addressed this need by using a short-term five-session model of family centered CBT that replicated and extended previous work by Sanders, Shepherd, Cleghorn, and Woolford (1994). One clear strength of this study is the potential clinical relevance of the treatment, which is brief, focused, manualized, and reproducible. One guiding principle of the intervention approach was a social learning conceptual model of RAP that fits well with current scientific knowledge of the importance of family reinforcement in influencing children’s experience of pain and subsequent disability (Chambers, 2003). The intervention was designed to interfere with the vicious cycle of maladaptive family interactions (e.g., the child’s maladaptive response to pain and the response of family members to the child’s symptoms) (Friedrich & Jaworski, 1995; Kazak, Simms, & Rourke, 2002) by combining conjoint parentchild sessions and individual sessions with children. Based on data from empirically supported interventions (Sanders et al., 1989, 1994; Janicke & Finney, 1999) and studies of influences on the functional impact of abdominal pain (Walker et al., 1998), sample themes of individual sessions included education concerning recurrent pain and its antecedents and consequences, education and practice in pain management, cognitive strategies to reduce catastrophizing, and increasing adaptive communication and partnership between parents and children. The emphasis on reducing negative functional consequences of RAP was a particularly salient feature of this intervention. To evaluate their intervention model, Robins et al. used a comprehensive measurement approach based on multiple sources (e.g., parent and child) and methods (subjective self-report and objective data) that included