The Sensitivity to Change and Responsiveness of the Adult Responses to Children’s Symptoms in Children and Adolescents With Chronic Pain
Find articles by Melanie Noel
Find articles by Nicole Alberts
Shelby L. Langer
Find articles by Shelby L. Langer
Rona L. Levy
Find articles by Rona L. Levy
Lynn S. Walker
Find articles by Lynn S. Walker
Tonya M. Palermo
2 Center for Child Health, Behavior & Development, Seattle Children’s Research Institute,
5 Departments of Anesthesiology, Pediatrics, and Psychiatry, University of Washington School of Medicine
1 Department of Psychology, University of Calgary and Alberta Children’s Hospital Research Institute,
2 Center for Child Health, Behavior & Development, Seattle Children’s Research Institute,
3 School of Social Work, University of Washington,
4 Department of Pediatrics, Vanderbilt University School of Medicine, and
5 Departments of Anesthesiology, Pediatrics, and Psychiatry, University of Washington School of Medicine
All correspondence concerning this article should be addressed to Melanie Noel, P h D, Department of Psychology, University of Calgary and the Alberta Children’s Hospital Research Institute, 2500 University Dr. N. W., Calgary, AB, T2N 1N4. E-mail: email@example.com
Received 2015 Apr 5; Revised 2015 Sep 15; Accepted 2015 Sep 17.
Copyright © The Author 2015. Published by Oxford University Press on behalf of the Society of Pediatric Psychology. All rights reserved. For permissions, please e-mail: firstname.lastname@example.org
This article has been cited by other articles in PMC.
Objective To examine the sensitivity to change and responsiveness of the Adult Responses to Children’s Symptoms (ARCS) among parents of youth with chronic pain. Methods Participants included 330 youth (89 children aged 7–11 years, 241 children aged 12–17 years) and their parents who participated in randomized controlled trials of family-based cognitive-behavioral therapy for chronic pain. Child pain and disability, parental emotional functioning, and parental responses to child pain were assessed at baseline and posttreatment. Results The Protect and Monitor scales of the ARCS were sensitive to change following intervention for both developmental groups, with clinically meaningful reductions in these behaviors, thereby demonstrating responsiveness. Among the adolescent sample, greater change on some ARCS scales was associated with better parental emotional functioning and lower child pain at posttreatment. Conclusions Findings support the sensitivity to change and responsiveness of the Protect and Monitor scales among parents of youth with chronic pain.
Keywords: adolescents, ARCS, children, chronic pain, monitor, parental responses, pediatric pain, protect.
The familial context of pediatric chronic pain is integral to our understanding of how pain problems develop, are experienced, and become maintained during childhood and adolescence ( Asmundson, Noel, Petter, & Parkerson, 2012 ; Goubert & Simons, 2014 ; Palermo & Chambers, 2005 ). Parental responses to children’s chronic pain complaints play a pivotal role in both child and parent functioning and have increasingly become an important target in evidence-based interventions to reduce pain and improve functioning in children and adolescents with chronic pain ( Levy et al., 2010 ; Palermo, Wilson, Peters, Lewandowski, & Somhegyi, 2009 ). The most widely used measure of parental responses to child pain is the Adult Responses to Children’s Symptoms (ARCS; Van Slyke & Walker, 2006 ), which was originally developed to assess three types of parental behaviors: protective (i.e., positive or negative reinforcement for pain complaints), minimizing (i.e., displaying anger or frustration or a lack of concern in response to pain complaints), and encouraging and monitoring (i.e., questioning child about symptoms or encouraging distraction from the child’s pain) responses.
Observational and Treatment Studies Using the Original Three-Factor ARCS
The majority of research using the ARCS has been cross-sectional and administered only the Protect scale, owing to a cited lack of evidence for the validity and poorer internal consistency of the other scales ( Claar, Guite, Kaczynski, & Logan 2010 ; Langer, Romano, Levy, Walker, & Whitehead, 2009 ; Simons, Claar, & Logan, 2008 ). Observational studies with youth spanning the developmental periods of childhood and adolescence suggest that parental protective responses are linked to a variety of maladaptive child outcomes including poorer school functioning ( Logan, Simons, & Carpino, 2012 ) and higher functional disability ( Claar et al., 2010 ; Langer et al., 2009 ; Sieberg, Williams, & Simons, 2011 ). Consistent with pediatric models of pain, such as the Pediatric and Interpersonal Fear Avoidance Models ( Asmundson et al., 2012 ; Goubert & Simons, 2014 ) and the Pain Empathy Model ( Goubert et al., 2005 ), parents may engage in protective behaviors in an effort to reduce their own distress. Indeed, consistent with these theories, parents’ catastrophic thinking about child pain and emotional distress have been shown to be related to the extent to which they engage in protective responses to child pain (e.g., Langer, Romano, Mancl, & Levy, 2014 ; Simons, Smith, Kaczynski, & Basch, 2015 ). Research examining the other ARCS scales suggests that in addition to protective behaviors, minimizing behaviors are linked to maladaptive child outcomes such as increased somatic symptoms ( Claar, Simons, & Logan, 2008 ; Simons et al., 2008 ). Similarly, recent research suggests that encouraging and monitoring responses may also be maladaptive. Among youth who catastrophize frequently about their pain, these responses may heighten their attention to pain and signal threat, thereby inadvertently exacerbating or maintaining functional disability ( Cunningham et al., 2014 ). Taken together, this research underscores the importance of examining each of the scales of the ARCS, which provide assessment of different aspects of parental behavior.
Modifying parental responses to child pain may have a powerful influence on parent and child outcomes. In a recent longitudinal examination, Welkom, Hwang and Guite (2013) found that decreases in protective behaviors between an initial evaluation and 2-month follow-up were linked to lower child and parent catastrophizing about child musculoskeletal pain, which in turn was associated with improvements in child functional disability over time. Indeed, parent operant training to reduce protective and other pain-attending parent responses to child pain has been increasingly incorporated into evidence-based cognitive-behavioral therapy (CBT) interventions to reduce pain and improve functioning among youth with chronic pain ( Eccleston et al., 2014 ). Protective behaviors as measured by the ARCS have been shown to decrease among parents of youth receiving family-based CBT interventions for pediatric chronic pain ( Levy et al., 2010 , 2013 ; Palermo et al., 2009 , 2016 ) and pain-related chronic illness ( McCormick, Reed-Knight, Lewis, Gold, & Blount, 2010 ). While this research provides some support for the sensitivity of the original ARCS Protect scale to change following intervention, the sensitivity of all of the ARCS scales is currently unknown.
Measure Refinement Work With the ARCS
The ARCS was developed as an extension to the Illness Behavior Encouragement Scale ( Walker & Zeman, 1992 ) to capture a wider range of parental behaviors than only solicitous responses. Originally, three scales (Protect, Minimize, Encourage & Monitor) were factor-analytically derived and found to be related yet distinct ( Van Slyke & Walker, 2006 ). The ARCS Protect scale was the only scale to be formally validated, showing strong associations with self-reported parent behaviors (a diary version of the measure), child gastrointestinal symptoms, and higher health care utilization and costs among youth (8–15 years) with functional abdominal pain (FAP; Walker, Levy, & Whitehead, 2006 ). Initial examination of the measure’s factorial validity using confirmatory factor analysis of the originally proposed three-factor structure resulted in slight modifications to the measure (removal of items) among youth (8–17 years) with abdominal and other chronic pain conditions ( Claar et al., 2010 ). Recently, Noel and colleagues (2015) extended this previous validation work by examining the factorial validity of other factor structures of the ARCS beyond the original three-factor model. Moreover, the authors took a developmental approach to analyses and examined the factorial validity of the ARCS separately in parents of children versus adolescents with a variety of chronic pain conditions or pain-related chronic illness (inflammatory bowel disease). Developmental differences in the factor structure of the ARCS were found, with all factor structures diverging from the originally proposed three-factor model. A four-factor model (Protect, Monitor, Minimize, Distract) provided the best fit to the data for children and the entire combined (adolescents and children) sample, whereas a five-factor model, with an additional Solicitousness factor (e.g., doing the child’s chores, giving child special privileges) fit the data best for adolescents. While this work provided a robust means of measure refinement and revealed developmental differences in patterns of parent responses across childhood and adolescence, many questions remain regarding the validity of the factor structure proposed by Noel et al. Specifically, it is unknown whether this factor structure of the ARCS scales are able to demonstrate meaningful changes following intervention directed at modifying parent behavior, and whether changes can be detected in child and adolescent samples. Moreover, the relationship between parental behavior change assessed with the Noel et al. child and adolescent factors and parent and child pain and functional outcomes at posttreatment is also unknown. Given the critical role of parent behaviors in the experience of pediatric chronic pain ( Palermo & Chambers, 2005 ), their potential role in the development and maintenance of chronicity ( Asmundson et al., 2012 ; Goubert & Simons, 2014 ), and their role in treatment ( Eccleston et al., 2014 ), the treatment utility (i.e., sensitivity to change, responsiveness) and validity of the Noel et al. factor structure of the ARCS scales warrant empirical investigation.
Although there is great confusion and controversy in the medical literature regarding measurement of change ( Streiner, Norman, & Cairney, 2014 ), we define sensitivity to change in a way that is consistent with the definitions put forth by Streiner et al. (2014) and Liang (2000) . Specifically, sensitivity to change “taps an instrument’s ability to measure any degree of change” ( Streiner et al., 2014 , p. 258). We also define responsiveness as a measure’s ability to assess change that is clinically significant. It has been argued that these aspects of measurement, although underexamined, are an essential measurement property along with validity and reliability ( Kirshner & Guyatt, 1985 ). Moreover, we, along with others, conceive of sensitivity and responsiveness in this regard as a form of construct validity ( Streiner et al., 2014 ).
Study Aims and Hypotheses
The primary aim of the present study was to examine the sensitivity of the Noel et al. ARCS factors to change following intervention among parents of children (Study 1) and adolescents (Study 2) undergoing family-based CBT for pediatric chronic pain. A secondary aim was to examine the responsiveness of the ARCS, or its ability to capture clinically significant change in parent responses to child pain. In addition, we examined whether changes on the Noel et al. ARCS scales were related to parent and child functional outcomes at posttreatment. Given differences in the methodology (e.g., timing of assessment), study samples (e.g., type of pain, inclusion/exclusion criteria, gender composition), and treatment modality (e.g., face-to-face cognitive-behavioral treatment vs. remotely delivered, Internet-based cognitive-behavioral treatment), the child and adolescent samples and treatments were examined separately. Furthermore, as a result of these differences, comparisons were not made between the two samples.
Given that the parent operant component of both interventions specifically aimed to reduce solicitous/protective and pain-attending parental behaviors, we hypothesized that the ARCS Protect and Monitor scales would show sensitivity to change following intervention across both developmental groups (children, adolescents), with greater reductions in these behaviors among families receiving intervention versus education control. We hypothesized that changes in parental behaviors would be clinically meaningful as indicated by a large magnitude of change from pre- to postintervention. Based on existing research, we hypothesized that reductions in protective, solicitous, monitoring, and minimizing responses would be related to lower parental emotional distress as well as child pain and functional disability at posttreatment, thereby providing additional evidence for the validity of these scales. We expected that increases in distracting responses would be linked to lower parent emotional distress and child pain and disability based on operant theories of chronic pain, which posit that such responses function to decrease pain behaviors ( Fordyce, 1976 ; Joliffe & Nicholas, 2004 ).
The current study examined the validity and treatment utility (i.e., sensitivity to change, responsiveness) of the ARCS separately in a group of children with FAP (Study 1) and adolescents with mixed chronic pain conditions (Study 2) and their parents who participated in randomized controlled trials (RCTs) of family-based CBT for chronic pain. This allowed us to examine psychometric properties of the ARCS using developmentally sensitive scoring procedures based on the factor structure recently proposed by Noel et al. (2015) . The methodologies used in each study are described in turn below. For the purposes of this article, we examined assessments completed at baseline and following intervention. We did not use long-term follow-up data for several reasons. The studies from which the data were drawn are ongoing and did not have complete data available for all participants and thus inclusion of follow-up data would have reduced our available sample size. In addition, the assessment time points at follow-up were variable across samples and included some differences in measurements.
Study 1 (Child Sample)
Participants in Study 1 included 89 parent–child dyads (61 girls, 28 boys; 94.4% mothers) enrolled in an RCT of CBT for FAP. All children had FAP and ranged in age from 7–11 years ( M age = 9.28, SD = 1.26). Baseline participant demographic details can be found in Table I . Similar to previous research ( Noel et al., 2015 ), the age cutoffs for the child sample (i.e., 7–11 years) were based on age categorizations recommended by the Standards for Research in Child Health ( Williams et al., 2012 ) and are also supported by developmental theories of cognitive development ( Piaget, 1972 ). Participants in the larger trial included youth between the ages of 7 and 17 years; however, for the purposes of examining the child-specific ARCS scoring system in the present study, only children (i.e., 7–11 years) were included to avoid confounding age with trial type. Moreover, we did not include the older children from Sample 1 in the analyses because this RCT primarily consisted of children under the age of 12 years and we believed that adding a small number of adolescents from this RCT to the larger RCT consisting of primarily adolescents (>12 years) would have introduced additional heterogeneity and a potential confound.
Child age (year), mean (SD)
Open in a separate window
The following inclusion criteria were used: (1) child was between the ages of 7 and 11 years; (2) child had experienced three or more episodes of abdominal pain during a 3-month period that was severe enough to interfere with activities; (3) child and parent had cohabited for the past 5 years or, in cases divided by custody, for at least half of the child's lifetime. Exclusion criteria included: (1) positive physical or laboratory findings that would explain the abdominal pain; (2) child chronic disease (e.g., Crohn's, ulcerative colitis, pancreatitis, diabetes, epilepsy, celiac sprue); (3) child lactose intolerance as diagnosed by the attending physician; (4) child major surgery within the past year; (5) child developmental disability requiring full-time special education or impairing ability to communicate; and (6) parent or child non-English-speaking ability. The data included from this study were collected as part of the site-specific institutional review board (IRB)-approved protocol.
Procedure and Interventions
Participants were enrolled in a RCT of in-person family-based CBT for children with FAP and their parents. Some of these children were included in the paper by Noel et al. (2015) that examined the factorial validity of the revised ARCS. The majority of participants in the child sample were recruited from the Gastrointestinal (GI) clinics at Seattle Children’s Hospital, Seattle Children’s Hospital satellite clinics, and the Atlantic Health System. They were approached in the clinic and screened by a research coordinator or nurse. A minority of participants responded to IRB-approved flyers posted around the community (e.g., local schools). Participants recruited in this way contacted the research team directly, after which they were screened. They also underwent a work-up by a GI doctor at Seattle Children’s Hospital before being enrolled in the study. Recruitment took place over a 4-year period from 2004 to 2009.
Children and their parents were randomly assigned to either a three-session intervention of social learning and cognitive-behavioral treatment targeting parents' responses to their children's pain complaints and children's coping responses, or a three-session education control intervention to control for time and attention. The content of this intervention is fully described elsewhere ( Levy et al., 2010 , 2013 ). All participants met with a trained therapist for three in-person sessions that were spaced approximately 1 week apart and lasted about 60 min each. Participants had the option of meeting with a therapist in the hospital clinic or in their homes. Therapists conducted both individual parent and conjoint parent–child interventions within sessions. The intervention aimed to reduce maladaptive parent responses (e.g., protective and pain-attending responses) to their child's pain complaints. More specifically, the intervention consisted of three major components which included: (1) relaxation training, (2) strategies aimed at assisting parents and children to modify responses to illness and to increase wellness behaviors, and (3) cognitive restructuring of dysfunctional thoughts regarding symptoms and their implications for functioning. Participants also engaged in skill practice within and between sessions via homework assignments.
Participants assigned to the education control group completed three sessions with the same amount of therapist time as included in the intervention. Content of the sessions focused on education about GI system anatomy and function, information about nutrition guidelines, and additional food-related information. Participants also completed homework assignments that were similar in time and effort as those assigned in the intervention group.
Child assessments, conducted by phone, were completed by research nurses working in the pediatric clinical research center. All nurse assessors were blind to randomized treatment condition. Child assessments were scheduled in advance. Children were mailed an answer key with the possible response options, and the nurse then called them at the scheduled time to administer the questions over the phone. All parents completed assessments on paper in their homes and mailed them back to the researchers.
Measures (described in turn below) were administered at baseline and posttreatment. The posttreatment assessment occurred approximately 4 weeks following the start of intervention. Although follow-up assessments also occurred at 3, 6, and 12 months following intervention ( Levy et al., 2010 , 2013 ); for the purposes of the present study, only the baseline and posttreatment assessment time points were used. Families were provided monetary incentives for completion of assessments.
Parental Responses to Child Pain
The parent-report version of the Adult Responses to Children's Symptoms (ARCS; Van Slyke & Walker, 2006 ) with a pain-specific stem was administered to assess parental responses to their children’s pain complaints. Using a 5-point Likert-type scale with the anchors “never” and “always,” parents indicate how often they have engaged in various behaviors when their child had pain. Based on recent research ( Noel et al., 2015 ) revealing that a four-factor model (Protect, Minimize, Monitor, and Distract) for children was superior to the three-factor model proposed in previous literature ( Claar et al., 2010 ; Van Slyke & Walker, 2006 ), the current paper used these four newly derived ARCS scales (for details regarding the scoring systems, refer to Noel et al., 2015 ). Scale scores were computed as averages. Internal consistency for the scales at baseline and posttreatment were as follows: Protect (baseline α = .84, posttreatment α = .88), Monitor (baseline α = .80, posttreatment α = .87), Minimize (baseline α = .59, posttreatment α = .64), and Distract (baseline α = .64, posttreatment α = .78).
Parent Anxiety and Depressive Symptoms
The Brief Symptom Inventory (BSI; Derogatis, 1993 ) was administered to assess symptoms of depression and anxiety among parents. The BSI is a 53-item self-report measure derived from the Symptom Checklist-90-Revised and reflects levels of psychological distress. Using a 5-point scale with the anchors “not at all” and “extremely,” parents indicate the degree to which they were bothered by particular symptoms. Subscale scores can be derived on nine symptom scales: somatization, obsessive-compulsive, interpersonal sensitivity, depression, anxiety, hostility, phobia, paranoia, psychoticism. The depression and anxiety scales were used in the current study. The BSI has demonstrated good validity and reliability ( Derogatis & Melisaratos, 1983 ) and has been used to assess distress among parents of youth with chronic pain (e.g., Connelly, Anthony, & Schanberg, 2012 ). In the current study, the BSI depression and anxiety subscales showed high internal consistency at baseline (α = .89, α = .85, respectively) and posttreatment (α = .85, α = 79, respectively).
Child Functional Disability
The Functional Disability Inventory (FDI) ( Claar & Walker, 2006 ; Walker & Green, 1991 ) was administered to assess child functional disability. Children rate the extent to which a variety of daily activities in multiple settings (e.g., home, school, recreational, and social contexts) have been difficult or posed physical challenges to them during the past week using a 0 ( no trouble ) to 4 ( impossible ) scale. Higher scores are indicative of greater pain-related disability. The FDI has been shown to be a valid, reliable, and clinically useful measure of functional disability among youth with chronic pain ( Claar & Walker, 2006 ; Walker & Greene, 1991 ). In the current study, the FDI showed high internal consistency at baseline (α = .90) and posttreatment (α = .87).
Child Pain Intensity
Children self-reported on their average level of pain intensity experienced over the past week using the Faces Pain Scale-Revised (FPS-R; Hicks, von Baeyer, Spafford, van Korlaar, & Goodenough, 2001 ). This FPS-R consists of six drawings of gender-neutral faces showing “no pain” (neutral face) to “most pain possible” pain expressions (from left to right). Faces are scored 0, 2, 4, 6, 8, and 10, with higher values indicative of greater pain intensity. The FPS-R is a psychometrically sound self-report measure of pain intensity among children between the ages of 4 and 12 years, and is recommended for assessment of pain intensity in children with chronic pain ( Stinson, Kavanagh, Yamada, Gill, & Stevens, 2006 ).
Study 2 (Adolescent Sample)
Participants in Study 2 included 241 parent–adolescent dyads (183 girls, 58 boys; 92.5% mothers). Adolescents had a variety of chronic pain conditions and ranged in age from 12 to 17 years ( M age = 14.97, SD = 1.34). Baseline participant demographic details can be found in Table I . Similar to previous research ( Noel et al., 2015 ), age cutoffs for the adolescent group (i.e., 12–17 years) were justified on empirical ( Williams et al., 2012 ) and theoretical ( Piaget, 1972 ) grounds. Participants in the larger trial included some youth between the ages of 10 and 17 years; however, for the purposes of the present study, only adolescents (i.e., 12–17 years) were included to avoid confounding age with trial type. Some of these adolescents were included in the paper by Noel et al. (2015) that examined the factorial validity of the revised ARCS.
The following inclusion criteria were used: (1) adolescent was between the ages of 12 and 17 years; (2) chronic idiopathic pain was present for the previous 3 months; (3) pain occurred at least once per week; (4) parent reported that pain interfered with at least one area of daily functioning; (5) the adolescent received a new patient evaluation in an interdisciplinary pain clinic. Exclusion criteria included: (1) adolescent had a serious comorbid chronic medical condition such as diabetes, cancer, or sickle cell disease; (2) adolescent had a developmental disability per parent report; (3) parent or adolescent was non-English speaking; (4) family did not have regular access to the Internet on a desktop or laptop computer; (5) adolescent had received more than four sessions of CBT for pain management within the past 6 months. The data included from this study were collected as part of a site-specific IRB-approved protocol.
Procedure and Interventions
Participants were enrolled in a RCT of Internet-delivered cognitive behavioral therapy for pain management. Participants were recruited over a 3-year period from one of 14 participating interdisciplinary pediatric pain clinics at academic medical centers across the United States and Canada. Adolescents and their parents were randomized to either the active treatment arm or an education control group (to control for time and attention). The content of this intervention was adapted from a pilot version of the treatment program that is described elsewhere ( Palermo et al., 2009 ). The intervention is composed of eight modules. Similar to the child intervention described in the methods for Study 1, intervention content was based on cognitive-behavioral and social-learning theories that aimed to improve behavioral self-regulation and cognitive skills of adolescents and reduce maladaptive (e.g., protective, pain-attending) parental responses to the adolescent’s pain. Parents and adolescents each accessed separate versions of the program Web site. Modules delivered to adolescents included the following: education about chronic pain; recognizing stress and negative emotions; relaxation, deep breathing, and imagery; cognitive skills (including coping); sleep and lifestyle; school attendance and pacing; staying active; and maintenance and prevention. The modules delivered to parents included the following: education about chronic pain; recognizing stress and negative emotions; operant strategies I; operant strategies II; modeling; sleep and lifestyle; communication; and maintenance and prevention. Participants engaged in skill practice between sessions through homework assignments and were assigned online coaches who provided tailored feedback on their assignments as well as overall progress in treatment. Each module took approximately 30 min to complete, and participants were asked to complete one module per week. On average, the total treatment duration was approximately 9 hr per family over 10–12 weeks.
Participants within the education control group continued with the standard medical care that had been prescribed for their pain problem. Adolescents and their parents were also provided access to an education version of the study Web site, which included static educational material about pediatric chronic pain management, drawn from publicly available educational Web sites. As such, participants in the education control group did not have access to behavioral and cognitive skills training, which was available in the treatment condition. Measures were administered online at baseline and posttreatment. The posttreatment assessment occurred approximately 12 weeks following the start of intervention. This study is ongoing with follow-up assessments at 6 and 12 months following intervention; for the purposes of the present study, only the baseline and posttreatment assessment time points were used. Families received incentives in the form of gift cards for completion of study assessments.
Similar to Study 1, several measures were used to assess the same constructs: parental responses to child pain complaints, parent anxiety and depressive symptoms, and child functional disability and pain intensity.
Parental Responses to Child Pain
As described in the methods for Study 1, the parent-report version of the Adult Responses to Children's Symptoms (ARCS; Van Slyke & Walker, 2006 ) with a pain-specific stem was used in this study. The current study used the adolescent-specific ARCS scales based on the factor structure proposed by Noel and colleagues (2015) . Specifically, five scales were examined: Protect, Solicitousness, Minimize, Monitor, and Distract (for details regarding the scoring systems, refer to Noel et al., 2015 ). Scale scores were computed as averages. Internal consistency for the scales at baseline and posttreatment were as follows: Protect (baseline α = .78, posttreatment α = .82), Solicitousness (baseline α = .68, posttreatment α = .68), Monitor (baseline α = .77, posttreatment α = .84), Minimize (baseline α = .66, posttreatment α = .73), and Distract (baseline α = .64, posttreatment α = .68).
Parent Anxiety and Depressive Symptoms
The Bath Adolescent Pain–Parental Impact Questionnaire (BAP-PIQ; Jordan, Eccleston, McCracken, Connell, & Clinch, 2008 ) was administered to assess symptoms of anxiety and depression among a subsample of parents, given that the measure was added to the protocol after the trial started. The BAP-PIQ assesses eight domains of parental functioning and behavior in the context of parenting an adolescent with chronic pain: depression, anxiety, child-related catastrophizing, self-blame and helplessness, partner relationship, leisure functioning, parental behavior, and parental strain. For comparison across the two trials, the anxiety and depression scales were used in the present study. Items are rated on a 5-point frequency scale ranging from “never” to “always,” using a reference period of the past 2 weeks. Previous research suggests that the measure has good internal consistency (i.e., Cronbach’s alpha values ranged from .76 to .89) and temporal reliability (the majority of BAP-PIQ scales had r s ≥ .70) over a 14-day period as well as acceptable validity ( Jordan et al., 2008 ). Specifically, acceptable levels of convergent validity were evidenced through correlations with similar measures of parental functioning (e.g., well-being, depression, parental stress, parental response to child pain, and quality of relationships). In the current study, the depression and anxiety subscales showed high internal consistency at baseline (α = .87, .90, respectively) and posttreatment (α = .88, .91, respectively).
Adolescent Functional Disability
The Child Activity Limitations Interview (CALI; Palermo, Lewandowski, Long, & Burant, 2008 ) was administered to adolescents via a 7-day online diary to assess functional disability. Items are rated on a 5-point scale with anchors 0 ( no difficulty ) and 4 ( extreme difficulty ). Higher scores are indicative of greater functional limitations. An average disability score was calculated for use in the present study. The CALI has been found to be a valid measure of pain-related disability in adolescents ( Palermo et al., 2008 ; Palermo, Witherspoon, Valenzuela, Drotar, 2004 ) and has been used as an outcome measure in previous studies of psychological interventions for pediatric chronic pain ( Palermo et al., 2009 ).
Adolescent Pain Intensity
A 7-day online diary was administered to assess adolescents’ self-reported pain intensity over that period. An average pain intensity score at each assessment time point was calculated across the 7-day period. Pain intensity was rated on an 11-point numerical rating scale (NRS), with the anchors 0 = no pain and 10 = worst pain . Higher ratings indicated more intense pain. NRSs have been shown to demonstrate good convergent and discriminant validity among youth with chronic pain ( Ruskin et al., 2014 ).
Data Analysis Across Studies
First, based on the recent scoring systems for the factor structure identified by Noel and colleagues (2015) , we calculated ARCS scale scores for the child and adolescent samples at baseline (T1) and postintervention (T2).
To address our primary research question: Are the newly derived ARCS scales sensitive to change following intervention? , we conducted a total of four one-way repeated measures analyses of variance (ANOVAS) for the child sample (one for each of the four child sample ARCS scales: Protect, Monitor, Minimize, Distract) and five one-way repeated measures ANOVAS for the adolescent sample (one for each of the five adolescent ARCS scales: Protect, Solicitousness, Monitor, Minimize, Distract). Each analysis included treatment group as a between-subjects factor and time (baseline, posttreatment) as a repeated measures factor. Examination of the interaction term produced by these analyses allowed us to determine whether treatment and control groups differed on the magnitude of change in ARCS scores. We adjusted all p -values for these analyses to control for family-wise error rate at 5% ( Wright, 1992 ), taking into account the overall number of analyses across developmental groups (nine in total). As indicators of the responsiveness of the Noel et al. ARCS scales, we calculated Effect Sizes (ES) separately for treatment and control groups within each developmental group. These parameters have been used as indicators of a measure’s responsiveness to intervention in previous research ( Lowe, Kroenke, Herzog, & Grafe, 2004 ; Pardasaney et al., 2012 ). ES was calculated as M 2 -M 1 /SD 1 (SD 1 = standard deviation of time 1; Kazis, Anderson, & Meenan, 1989 ). ES is a standardized indicator of a measure’s power to detect true change, with larger values reflecting higher sensitivity to change ( Liang, 2000 ). Moreover an effect size of 0.5, indicating a “moderate” effect size, has been cited as a reasonable approximation to a threshold indicative of important or meaningful change ( Streiner et al., 2014 ).
Finally, to examine whether change in parental behavior on the Noel et al. ARCS scales corresponded to child and parent functional outcomes at posttreatment, partial correlations were conducted between the ARCS scale change scores and posttreatment outcome measures of child functional disability, child pain intensity, and parent depressive and anxiety symptoms, while controlling for time 1 scores. Once again, these analyses were conducted separately for the child and adolescent samples using the recently proposed developmentally sensitive scoring systems based on separate factor structures for children and adolescents ( Noel et al., 2015 ). For these analyses, we calculated raw change scores for each ARCS scale score (T1 – T2), such that positively signed scores reflected greater improvement or reductions in maladaptive parental responses (Protective, Solicitous, Monitoring, and Minimizing) over time. As Distracting responses were hypothesized to be adaptive, positively signed scores on this scale indicated reductions in adaptive parental responses.
Are the ARCS Scales Based on the Noel et al. Factor Structure Sensitive to Change Following Intervention?
Mean scores from the repeated measures ANOVA analyses based on the Noel et al. factor scoring of the ARCS are shown in Table II . For the child sample, there were significant interactions between time (baseline, posttreatment) and treatment group for the ARCS Protect and Monitor scale scores ( F (1, 87) = 9.33, p < .05,
= .097, observed power = .86; F (1, 87) = 23.69, p < .001,
= .21, observed power = .98, respectively). Specifically, parents reported significantly greater reductions in protective and monitoring responses from pre- to postintervention among families who received active intervention compared with families who received education control. However, there were not significant interactions between time and treatment group for the ARCS Minimize and Distract ARCS scale scores ( F (1, 87) = .246, p > .05,
= .003, observed power = .08; F (1, 87) = .044, p > .05,
= .001, observed power = .06, respectively). This suggests that parents receiving intervention did not report significantly greater reductions in minimizing and distracting responses from baseline to posttreatment as compared with parents receiving education control. Support for the responsiveness of the Noel et al. ARCS scales was found using ES calculated for treatment and control groups. As expected, ES for the Protect and Monitor scales exceeded 0.5, indicating that they demonstrated clinically meaningful change from pre- to postintervention. ES calculated for the child group are shown in Table III .
Mean (SD) ARCS Scale Scores Across Developmental Group, Treatment Condition, and Time
Revised ARCS scale
Effect Sizes of ARCS Scale Scores Across Treatment Condition and Developmental Group
Revised ARCS scale
Child sample only (n = 89)
Adolescent sample only (N = 241)
Note . a Effect sizes > .50 indicate clinically meaningful change ( Streiner et al., 2014 ).
Mean scores from the repeated-measures ANOVA analyses based on the Noel et al. factor scoring of the ARCS are shown in Table II . For the adolescent sample, there were significant interactions between time (baseline, posttreatment) and treatment groups on the Protect and Monitor ARCS scale scores ( F (1, 239) = 16.31, p < .001,
= .064, observed power = .98; F (1, 239) = 9.87, p < .05,
= .040, observed power = .88, respectively). Specifically, parents reported significantly greater reductions in protective and monitoring responses from pre- to postintervention among families who received intervention compared with those who received education control. Conversely, the magnitude of change in Solicitousness, Minimize, and Distract ARCS scale scores over time did not differ between treatment and education control groups ( F (1, 239) = 1.39, p > .05,
= .006, observed power = .22 for Solicitousness; F (1, 239) = 4.57, p > .05,
= .019, observed power = .57 for Minimize; and F (1, 239) = 2.93, p > .05,
= .012, observed power = .40 for Distract). ES for the Protect and Monitor scales exceeded 0.5, indicating that these particular scales showed clinically meaningful change from pre- to postintervention. ES calculated for the adolescent group are shown in Table III .
Is Change on the ARCS Related to Relevant Parent and Child Outcomes After Treatment?
Changes on Protect, Monitor, Minimize, and Distract scales were not related to any of the parent or child posttreatment outcomes within the child sample. In other words, reductions in these behaviors over the course of treatment were not related to parent or child outcomes at posttreatment.
Changes/reductions on the Protect, Monitor, and Solicitousness scales were related to lower parent depressive ( r = −.33, p < .001; r = − .20, p < .05; r = − .38, p < .001, respectively) and anxiety ( r = − .22, p < .01; r = − .31, p < .001; r = −.36, p < .001, respectively) symptoms at posttreatment. Changes/reductions on the Protect and Solicitousness scales were related to lower child pain intensity ( r = −.16, p < .05; r = −.18, p < .01) at posttreatment. Changes/reductions on the Minimize and Distract scales were not related to any parent and child functional outcomes at posttreatment.
This is the first study to examine the sensitivity of the ARCS scales to change within the context of treatment. The strengths of our approach are the inclusion of two separate samples, including children and adolescents, who received family-based CBT interventions for chronic pain. Inclusion of both samples allowed for the use of a developmentally sensitive scoring system for the ARCS recently put forth by Noel and colleagues (2015) . Findings support the utility of this system for scoring the ARCS Protect and Monitor scales in clinical trials of psychological treatments for pediatric chronic pain among children and adolescents.
The results of this research suggest that the Protect and Monitor scales using the child and adolescent scoring system are sensitive to change following family-based CBT and that these changes are clinically meaningful (thereby supporting the responsiveness of these scales) and detectable immediately following treatment. However, the ARCS is multifactorial, and we were interested in whether the other scales were sensitive to change. Our findings suggest that the Distract and Minimize (as well as Solicitousness for adolescents) scales may not be sensitive to change or responsive. This may reflect limitations of these ARCS scales or it might simply reflect on the effects of the CBT interventions studied. Therefore, in studies of family-based CBT interventions that focus on teaching parents to alter their responses to children, our findings suggest that we may need better ways of measuring distracting and minimizing responses, or we may need to enhance these aspects of the intervention.
The majority of work on the ARCS has examined the Protect scale to the exclusion of the other scales. Findings are largely consistent across treatment and observational studies: Greater parental protective responses are generally associated with higher levels of child functional disability ( Claar et al., 2010 ; Langer et al., 2009 ; Sieberg et al., 2011 ) and with higher parental emotional distress and catastrophizing ( Langer et al., 2009 , 2014 ; Sieberg et al., 2011 ; Simons et al., 2015 ). The ARCS Child and Adolescent scale scoring put forth by Noel and colleagues (2015) used robust measure refinement including the addition of factors, removal of several problematic (i.e., cross or poorly loading) items, and as a result, different proposed developmental scoring systems for children and adolescents. Therefore, this research offers an important next step in the psychometric testing of this frequently used measure and provides evidence for the construct validity of the Protect and Monitor scales. Consistent with previous research on the former Protect scale ( Levy et al., 2010 , 2013 ), the Noel et al. Protect scale was sensitive to change following intervention supporting its validity and utility in treatment studies. However, inspection of the parameters of responsiveness (i.e., ES) suggests that the Monitor scale may be most sensitive to change from pre- to postintervention. We recommend that future research use this child and adolescent scoring systems tailored to the developmental group under study and include the Monitor scale in addition to the Protect scale given its demonstrated sensitivity, responsiveness, and importance in the present study. Additional research is needed to examine these child and adolescent scales and their relationships with other parent and child outcomes not included in the present study (e.g., parent and child catastrophizing about child pain, fear and acceptance of pain, family functioning).
Findings suggest differences between the child and adolescent samples with respect to the relationships found between changes on the ARCS scales and posttreatment measures of other parent and child outcomes targeted in CBT. While this could reflect developmental differences in the mechanisms and timing of change in interventions for children versus adolescents, it could also reflect differences in methodology, treatment modality and content, and other participant/sample characteristics between studies. Indeed, for these reasons, caution should be made when attempting to compare findings across developmental groups. Both family-based CBT interventions included a parent training component based on operant theories of chronic pain and conceptualized behavior change as involving changes in parent behaviors (reductions in protective and pain-attending behaviors) that would then subsequently lead to reductions in child disability and pain. Thus, associations between changes in ARCS scale scores and changes in parent and child outcomes in the adolescent, but not the child, sample might reflect differences in the timing of the postassessment. Indeed, the posttreatment assessment for the child sample occurred approximately 4 weeks following the start of intervention as compared with 12 weeks later in the adolescent sample. Assuming changes in protective and monitoring responses were maintained in the weeks and months following treatment, changes in child pain and functional disability would be expected to lag behind change in parent behaviors. These findings emphasize the need to examine mechanisms and directionality of changes in our treatment studies, which will be aided by inclusion of multiple assessments at various points in time during and following intervention. Future research should examine sensitivity to change and responsiveness of the Noel et al. ARCS scales at longer term follow-up (e.g., 6 months, 1 year). In addition, the two study samples examined youth with different types of chronic pain. The child sample was entirely composed of youth with abdominal pain, whereas the adolescent sample consisted of youth with a variety of different chronic pain conditions. While the factorial validity of the ARCS was examined among youth with a variety of chronic pain conditions ( Noel et al., 2015 ), there may be differences in the experience and expression of different types of pain that may evoke different parental responses. Further study of the ARCS in various pediatric chronic pain samples is warranted.
This study had limitations that can be addressed in future research. Although both CBT interventions in Study 1 and Study 2 were based on cognitive behavioral and social learning theories of pediatric chronic pain and involved youth and parent training, there were differences between the two interventions that could have introduced heterogeneity and imprecision into the results. Specifically, the child treatment was delivered in a face-to-face format and included only three sessions completed over 3–4 weeks, whereas the adolescent treatment was self-guided, remotely delivered via the Internet, and included eight sessions/modules completed over 10–12 weeks. Differences in the length of the treatment and timing of outcome assessment may have contributed to the different pattern of findings across samples. In addition, parents and children attended sessions together for most of the child intervention, whereas adolescents and their parents completed online modules separately. While the latter structure seems appropriately developmentally tailored to the degree of autonomy and reliance of parents that would be expected during adolescence, this difference in intervention delivery could have influenced the results. Relatedly, although both child and adolescent treatments included core components targeting parent responses to child pain, as well as cognitive restructuring and relaxation skills for youth, the adolescent treatment also included additional treatment material addressing school attendance, physical activity, and sleep, domains that have been found to be important areas of impact and disability in pediatric chronic pain ( Fales, Palermo, Law, & Wilson, 2014 ; Law, Dufton, & Palermo, 2012 ; Logan et al., 2012 ). Nevertheless, it is striking that even despite these differences in treatment components and delivery, the Protect and Monitor scales showed sensitivity to change following intervention as well as responsiveness across both developmental groups. This speaks to the robustness of this finding and the utility of this measure in general, and these scales in particular, in trials of psychological interventions for pediatric chronic pain at various developmental stages.
In addition, the ARCS, among chronic pain and illness groups, primarily assesses parental behaviors that likely serve to increase attention to and reinforcement for pain, thereby increasing child pain. Therefore, the measure captures maladaptive behaviors in this context. The field of chronic pain has made great strides in developing measures to assess adaptive psychological and behavioral responses to child pain (acceptance, psychological flexibility; McCracken & Gauntlett-Gilbert, 2011 ; Smith, Sieberg, Odell, Randall, & Simons, 2015 ) and we argue that these constructs should be assessed in addition to maladaptive parent behaviors to provide a more comprehensive assessment of parent factors. Moreover, assessment of pain in Study 1 was limited by reliance on participant recall, which is subject to bias and not as robust as prospective reporting in daily diaries. In addition, we acknowledge that there are limitations inherent in solely relying on parent-report of behaviors in the context of treatment aiming to modify those behaviors. Future research should strive to use multiple informants (children, parents) of these parental responses. We also note that there are limitations inherent in the use and interpretability of difference scores (owing to potential for unreliability and regression to the mean), and this is a topic of considerable debate in the literature ( Edwards, 2001 ). Finally, some of the ARCS scales (e.g., Solicitousness, Minimize, Distract) were found to have low reliability, which could in part account for failure to demonstrate significant sensitivity to change and responsiveness in this study. Future psychometric research on the ARCS could consider the addition of items to these particular scales and further examination of the measure at the item level.
In conclusion, the findings support the treatment utility (i.e., sensitivity to change, responsiveness) of the child and adolescent ARCS Protect and Monitor scales for use with parents of children and adolescents in trials of family-based CBT for pediatric chronic pain. Across interventions targeting children and adolescents with a variety of types of chronic pain and their parents, families receiving intervention showed clinically meaningful reductions in maladaptive parental responses to pain from pre- to posttreatment. Taken together, these findings support the child and adolescent ARCS Protect and Monitor scales as developmentally sensitive, valid, and responsive measures with strong treatment utility.
At the time that this research was conducted, Dr. Noel was funded by a Post-PhD Fellowship Award from the Canadian Institutes of Health Research and was a trainee member of Pain in Child Health, a Strategic Training Initiative in Health Research of the Canadian Institutes of Health Research. Research reported in this publication was supported by the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of Health under Award Numbers R01HD062538 and K24 HD060068 (PI: Tonya M. Palermo) and R01HD036069 (PI: Rona L. Levy).
Conflicts of interest : None declared.
Asmundson G. J. G., Noel M., Petter M., Parkerson H. ( 2012. ). Pediatric fear-avoidance model of chronic pain: Foundation, application, and future directions . Pain Research and Management , 17 , 397 – 405 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Claar R. L., Guite J. W., Kaczynski K. J., Logan D. E. ( 2010. ). Factor structure of the adult responses to children’s symptoms: Validation in children and adolescents with diverse chronic pain conditions . Clinical Journal of Pain , 26 , 410 – 417 . [ PubMed ] [ Google Scholar ]
Claar R. L., Simons L. E., Logan D. E. ( 2008. ). Parental response to children’s pain: The moderating impact of children’s emotional distress on symptoms and disability . Pain , 138 , 172 – 179 . [ PubMed ] [ Google Scholar ]
Claar R. L., Walker L. S. ( 2006. ). Functional assessment of pediatric pain patients: psychometric properties of the functional disability inventory . Pain , 121 , 77 – 84 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Connelly M., Anthony K. K., Schanberg L. E. ( 2012. ). Parent perceptions of child vulnerability are associated with functioning and health care use in children with chronic pain . Journal of Pain and Symptom Management , 43 , 953 – 960 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Cunningham N., Lynch-Jordan A., Barnett K., Peugh J., Sil S., Goldschneider K., Kashikar-Zuck S. ( 2014. ). Child pain catastrophizing mediates the relation between parent responses to pain and disability in youth with functional abdominal pain . Journal of Pediatric Gastroenterology and Nutrition , 59 , 732 – 738 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Derogatis L. R. ( 1993. ). The Brief Symptom Inventory (BSI): Administration, scoring, and procedures . Manual (3rd ed.). Minneapolis, MN: National Computer Systems . [ Google Scholar ]
Derogatis L. R., Melisaratos N. ( 1983. ). The brief symptom inventory: An introductory report . Psychological Medicine , 13 , 595 – 605 . [ PubMed ] [ Google Scholar ]
Eccleston C., Palermo T. M., Williams A. C., Lewandowski Holley A. C. dC., Morley S., Fisher E., Law E. ( 2014. ). Psychological therapies for the management of chronic and recurrent pain inchildren and adolescents . Cochrane Database of Systematic Reviews , 5 , CD003968 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Edwards J. R. ( 2001. ). Ten difference score myths . Organizational Research Methods , 4 , 265 – 287 . [ Google Scholar ]
Fales J., Palermo T. M., Law E. F., Wilson A. C. ( 2014. ). Sleep outcomes in youth with chronic pain participating in a randomized controlled trial of online cognitive-behavioral therapy for pain management . Behavioral Sleep Medicine , 12 , 1 – 17 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Fordyce W. E. ( 1976. ). Behavioural methods for chronic pain and illness . St Louis: : Mosby; . [ Google Scholar ]
Goubert L., Craig K. D., Vervoort T., Morley S., Sullivan M. J., Williams A. C. dC., Cano A., Crombez G. ( 2005. ). Facing others in pain: the effects of empathy . Pain , 118 , 285 – 288 . [ PubMed ] [ Google Scholar ]
Goubert L., Simons L. E. ( 2014. ). Cognitive styles and processes in paediatric pain . In McGrath P. J., Stevens B. J., Walker S. M., Zempsky W. T., editors. (Eds). Oxford Textbook of Paediatric Pain . London: : Oxford University Press; . [ Google Scholar ]
Hicks C. L., von Baeyer C. L., Spafford P. A., van Korlaar I., Goodenough B. ( 2001. ). The Faces Pain Scale–Revised: Toward a common metric in pediatric pain measurement . Pain , 93 , 173 – 183 . [ PubMed ] [ Google Scholar ]
Joliffe C. D., Nicholas M. K. ( 2004. ). Verbally reinforcing pain reports: An experimental test of the operant model of chronic pain . Pain , 107 , 167 – 175 . [ PubMed ] [ Google Scholar ]
Jordan A., Eccleston C., McCracken L. M., Connell H., Clinch J. ( 2008. ). The Bath Adolescent Pain- Parental Impact Questionnaire (BAP-PIQ): Development and preliminary psychometric evaluation of an instrument to assess the impact of parenting an adolescent with chronic pain . Pain , 137 , 478 – 487 . [ PubMed ] [ Google Scholar ]
Kazis L. E., Anderson J. J., Meenan R. F. ( 1989. ). Effect sizes for interpreting changes in health status . Medical Care , 27 , S178 – S189 . [ PubMed ] [ Google Scholar ]
Kirshner B., Guyatt G. ( 1985. ). A methodological framework for assessing health indices . Journal of Chronic Disease , 38 , 27 – 36 . [ PubMed ] [ Google Scholar ]
Langer S. L., Romano J. M., Levy R. L., Walker L. S., Whitehead W. E. ( 2009. ). Catastrophizing and parental response to child symptom complaints . Children’s Health Care , 38 , 169 – 184 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Langer S. L., Romano J. M., Mancl L., Levy R. L. ( 2014. ). Parental catastrophizing partially mediates the association between parent-reported child pain behavior and parental protective responses . Pain Research and Treatment , 2014 , 751097 . doi: 10.1155/2014/751097 [ PMC free article ] [ PubMed ] [ Google Scholar ]
Law E., Dufton L., Palermo T. M. ( 2012. ). Daytime and nighttime sleep patterns in adolescents with chronic pain and healthy adolescents . Health Psychology , 31 , 830 – 833 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Levy R. L., Langer S. L., Walker L. S., Romano J. M., Christie D. L., Youssef N., DuPen M., Ballard S. A., Labus J., Welsh E., Feld L. D., Whitehead W. E. ( 2013. ). Twelve month follow-up of cognitive behavioral therapy for children with functional abdominal pain . JAMA Pediatrics , 167 , 178 – 184 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Levy R. L., Langer S. L., Walker L. S., Romano J. M., Christie D. L., Youssef N., DuPen M. M., Ballard S. A., Welsh E. M., Jeffery R. W., Young M., Coffey M. J., Whitehead W. E. ( 2010. ). Cognitive-behavioral therapy for children with functional abdominal pain and their parents decreases pain and other symptoms . American Journal of Gastroenterology , 105 , 946 – 956 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Liang M. H. ( 2000. ). Longitudinal construct validity: Establishment of clinical meaning in patient evaluation instruments . Medical Care , 38 ( Suppl II ), S84 – S90 . [ PubMed ] [ Google Scholar ]
Logan D. E., Simons L. E., Carpino E. A. ( 2012. ). Too sick for school? Parent influences on school functioning among children with chronic pain . Pain , 153 , 437 – 443 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Lowe B., Kroenke K., Herzog W., Grafe K. ( 2004. ). Measuring depression outcome with a brief self-report instrument: Sensitivity to change of the Patient Health Questionnaire (PHQ-9) . Journal of Affective Disorders , 81 , 61 – 66 . [ PubMed ] [ Google Scholar ]
McCormick M., Reed-Knight B., Lewis J. D., Gold B. D., Blount R. L. ( 2010. ). Coping skills for reducing pain and somatic symptoms in adolescents with IBD . Inflammatory Bowel Disease , 16 , 2148 – 2157 . [ PubMed ] [ Google Scholar ]
McCracken L. M., Gauntlett-Gilbert J. ( 2011. ). Role of psychological flexibility in parents of adolescents with chronic pain: Dvelopment of a measure and preliminary correlation analyses . Pain , 152 , 780 – 785 . [ PubMed ] [ Google Scholar ]
Noel M., Palermo T. M., Essner B., Zhou C., Levy R. L., Langer S., Sherman A., Walker L. S. ( 2015. ). A developmental analysis of the factorial validity of the Adult Responses to Children’s Symptoms (ARCS) in children versus adolescents with chronic pain and pain-related chronic illness . Journal of Pain , 16 , 31 – 41 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Palermo T. M., Chambers C. T. ( 2005. ). Parent and family factors in pediatric chronic pain and disability: An integrative approach . Pain , 119 , 1 – 4 . [ PubMed ] [ Google Scholar ]
Palermo T. M., Law E. F., Fales J., Bromberg M. H., Jessen-Fiddick T., Tai G. ( 2016. ). Internet-delivered cognitive-behavioral treatment for adolescents with chronic pain and their parents: A randomized controlled multicenter trial . Pain , 157 , 174 – 185 . doi:10.1097/j.pain.0000000000000348 [ PMC free article ] [ PubMed ] [ Google Scholar ]
Palermo T. M., Lewandowski A. S., Long A. C., Burant C. J. ( 2008. ). Validation of a self-report questionnaire version of the Child Activity Limitations Interview (CALI): The CALI-21 . Pain , 139 , 644 – 652 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Palermo T. M., Wilson A. C., Peters M., Lewandowski A., Somhegyi H. ( 2009. ). Randomized controlled trial of an Internet-delivered family cognitive-behavioral therapy intervention for children and adolescents with chronic pain . Pain , 146 , 205 – 213 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Palermo T. M, Witherspoon D., Valenzuela D., Drotar D. ( 2004. ). Development and validation of the Child Activity Limitations Interview: a measure of pain-related functional impairment in school-age children and adolescents . Pain , 109 , 461 – 470 . [ PubMed ] [ Google Scholar ]
Pardasaney P. K., Latham N. K., Jette A. M., Wagenaar R. C., Ni P., Slavin M. D., Bean J. F. ( 2012. ). Sensitivity to change and responsiveness of four balance measures for community-dwelling older adults . Physical Therapy , 92 , 388 – 397 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Piaget J. ( 1972. ). Intellectual evolution from adolescence to adulthood . Human Development 15 , 1 – 12 . [ Google Scholar ]
Ruskin D., Lalloo C., Amaria K., Stinson J. N., Kewley E., Campbell F., Brown S. C., Jeavons M., McGrath P. A. ( 2014. ). Assessing pain intensity in children with chronic pain: convergent and discriminant validity of the 0 to 10 numerical rating scale in clinical practice . Pain Research and Management , 19 , 141 – 148 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Sieberg C. B., Williams S., Simons L. E. ( 2011. ). Do parent protective responses mediate the relation between parent distress and child functional disability among children with chronic pain? Journal of Pediatric Psychology , 36 , 1043 – 1051 . [ PubMed ] [ Google Scholar ]
Simons L. E., Claar R. L., Logan D. L. ( 2008. ). Chronic pain in adolescence: parental responses, adolescent coping, and their impact on adolescent’s pain behaviors . Journal of Pediatric Psychology , 33 , 894 – 904 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Simons L. E., Smith A., Kaczynski K., Basch M. ( 2015. ). Fear of your child's pain: The parent fear of pain questionnaire . Pain , 156 , 694 – 702 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Smith A. M., Sieberg C. B., Odell S., Randall E., Simons L. E. ( 2015. ). Living life with my child’s pain: The parent pain acceptance questionnaire (PPAQ) . Clinical Journal of Pain , 31 , 633 – 641 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Stinson J. N., Kavanagh T., Yamada J., Gill N., Stevens B. ( 2006. ). Systematic review of the psychometric properties, interpretability and feasibility of self-report pain intensity measures for use in clinical trials in children and adolescents . Pain , 125 , 143 – 157 . [ PubMed ] [ Google Scholar ]
Streiner D. L., Norman G. R., Cairney J. ( 2014. ). Health measurement scales: A practical guide to their development and use ( 5th ed .). Oxford: : GBR; . [ Google Scholar ]
Van Slyke D. A., Walker L. S. ( 2006. ). Mothers’ responses to children’s pain . Clinical Journal of Pain , 22 , 387 – 391 . [ PubMed ] [ Google Scholar ]
Walker L. S., Greene J. W. ( 1991. ). The functional disability inventory: Measuring a neglected dimension of child health status . Journal of Pediatric Psychology , 16 , 39 – 58 . [ PubMed ] [ Google Scholar ]
Walker L. S., Levy R. L., Whitehead W. E. ( 2006. ). Validation of a measure of protective parent responses to children’s pain . Clinical Journal of Pain , 22 , 712 – 716 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Walker L. S., Zeman J. L. ( 1992. ). Parental response to child illness behavior . Journal of Pediatric Psychology , 17 , 49 – 71 . [ PubMed ] [ Google Scholar ]
Welkom J. S., Hwang W. T., Guite J. W. ( 2013. ). Adolescent pain catastrophizing mediates the relationship between protective parental responses to pain and disability over time . Journal of Pediatric Psychology , 38 , 541 – 550 . [ PMC free article ] [ PubMed ] [ Google Scholar ]
Williams K., Thomson D., Seto I., Contopoulos-Ioannidis D. G., Ioannidis J. P., Curtis S., Constantin E., Batmanabane G., Hartling L., Klassen T. , & StaR Child Health Group . ( 2012. ). Standard 6: Age groups for pediatric trials . Pediatrics , 129 , S153 – 160 . [ PubMed ] [ Google Scholar ]
Wright S. P. ( 1992. ). Adjusted P-valu