Journal of Pediatric Psychology Advance Access published April 2, 2015
Journal of Pediatric Psychology, 2015, 1–10 doi: 10.1093/jpepsy/jsv029 Original Research Article
Distinct Influences of Anxiety and Pain Catastrophizing on Functional Outcomes in Children and Adolescents With Chronic Pain Susan T. Tran,1 PHD, Kristen E. Jastrowski Mano,2 PHD, Keri R. Hainsworth,3,4 PHD, Gustavo R. Medrano,5 PHD, Kimberly Anderson Khan,3,4 PSYD, Steven J. Weisman,3,4 MD, and W. Hobart Davies,6 PHD 1
Cincinnati Children’s Hospital Medical Center, 2University of Cincinnati, 3Medical College of Wisconsin, 4Children’s Hospital of Wisconsin, 5The Family Institute at Northwestern University, and 6University of Wisconsin-Milwaukee
All correspondence concerning this article should be addressed to Susan T. Tran, PHD, Division of Behavioral Medicine and Clinical Psychology, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Ave. MLC #3015, Cincinnati, OH 45229, USA. E-mail: [email protected] Received August 18, 2014; revisions received March 6, 2015; accepted March 11, 2015
Abstract Objectives Examine whether anxiety and pain catastrophizing are distinct constructs in relation to functional outcomes in pediatric chronic pain, and whether they differentially predict functional outcomes based on age. Methods In all, 725 youth (191 children, 534 adolescents) with chronic pain completed measures of pain characteristics, anxiety, pain catastrophizing, functional disability, and health-related quality of life (HRQOL). Structural equation modeling was used to examine interrelationships. Results Anxiety and pain catastrophizing were distinct. For both children and adolescents, pain catastrophizing predicted pain, functional disability, and HRQOL, and was a stronger predictor of pain intensity. For children, anxiety predicted HRQOL, and pain catastrophizing was a stronger predictor of functional disability. For adolescents, anxiety predicted functional disability and HRQOL, and anxiety was a stronger predictor of HRQOL. Conclusions There were age-related differences regarding whether anxiety or pain catastrophizing more strongly predicted specific functional outcomes. Assessment and intervention efforts should emphasize both anxiety and pain catastrophizing. Key words: anxiety; functional disability; pain catastrophizing; pediatric chronic pain.
Introduction Pediatric chronic pain is a relatively common condition among youth (King et al., 2011) that is often associated with decreased physical and psychological functioning, decreased satisfaction with health and life, and increased functional disability (Hunfeld et al., 2001; Merlijn et al., 2006; Palermo, 2000). Anxiety and pain catastrophizing are common psychological comorbidities in pediatric chronic pain, which influence functional outcomes (Cohen, Vowles, & Eccleston, 2010; Crombez et al., 2003; Vervoort, Goubert, Eccleston, Bijttebier, & Crombez, 2006; Wendland, Jackson, & Stokes, 2010). Researchers have noted the strong associations between these two constructs and called for research to distinguish the
role of each (Hermann, Hohmeister, Zohsel, Ebinger, & Flor, 2007; Holroyd, Drew, Cottrell, Romanek, & Heh, 2007; Jastrowski Mano et al., 2012; Vervoort, Eccleston, Goubert, Buysse, & Crombez, 2010) particularly in children and adolescents (Eccleston, Fisher, Vervoort, & Crombez, 2012); however, researchers have not yet adopted a statistical approach to clarify the relationships between these constructs. The clinical utility of both constructs may be clarified by the extent to which (1) each can be distinguished from the other, and (2) each is differentially related to important functional outcomes (Benore, D’Auria, Banez, Worley, & Tang, 2015; Drahovzal, Stewart, & Sullivan, 2006). The connection between pediatric chronic pain and anxiety has been well documented. Higher anxiety and internalizing symptoms
Published by Oxford University Press on behalf of the Society of Pediatric Psychology 2015. This work is written by US Government employees and is in the public domain in the US. 1
2 have been reported in youth with a variety of chronic pain complaints, such as recurrent abdominal pain (Campo et al., 2004; Galli et al., 2007), headaches (Galli et al., 2007; Mazzone, Vitiello, Incorpora, & Mazzone, 2006), and nonspecific musculoskeletal pain (O’Sullivan, Beales, Jensen, Murray, & Myers, 2011) compared with healthy control groups. Anxiety is often accompanied by physical arousal (Guite & Kazak, 2010), and the connection between anxiety and chronic pain is circular: physical pain may trigger anxious worrying, and the muscle tension associated with anxiety may heighten pain sensitivity (Beesdo et al., 2009; Simons & Kaczynski, 2012). Furthermore, in youth with chronic pain, high anxiety is related to impairments in functioning regardless of pain intensity (Cohen et al., 2010; Simons, Sieberg, & Claar, 2012). Even when an individual with chronic pain is not experiencing pain, he or she may anticipate having pain during an activity, worry about questions from peers about their pain, or worry about loss of functioning owing to pain (Powers, Gilman, & Hershey, 2006). Taken together, these studies suggest that anxiety plays a critical role in functional disability for youth with chronic pain. Pain catastrophizing, another construct related to negative expectancies, also plays a critical role in functional disability. Pain catastrophizing is a negative orientation toward painful stimuli including exaggerated or perseverative worry about the negative consequences of pain, perceived lack of control over symptoms, and an inability to cope with pain (Jones et al., 2003; Sullivan et al., 1995). Recently, it has been suggested that some aspects of pain catastrophizing are developmentally normal in childhood (i.e., not pathological, as is commonly believed; Eccleston et al., 2012); however, higher levels of pain catastrophizing have consistently been associated with poorer outcomes in pediatric chronic pain. The fear avoidance model (Asmundson, 1999), recently applied to pediatric chronic pain (Simons & Kaczynski, 2012), states that an individual with negative expectancies of pain, such as pain catastrophizing, is more likely to avoid activities that may flare pain resulting in increased functional disability over time. In populations of both school children and children with chronic pain, pain catastrophizing has been shown to be a significant predictor of functional disability over pain intensity (Crombez et al., 2003; Vervoort et al., 2006). Greater pain catastrophizing has also been associated with lower health-related quality of life (HRQOL) in youth with fibromyalgia (Libby & Glenwick, 2010) and in a community sample of children with chronic pain (Merlijn et al., 2006). While research has demonstrated that anxiety and pain catastrophizing are important constructs in chronic pain, it remains unclear what the defining characteristics of each are during childhood and adolescence. In fact, Eccleston and colleagues (2012) have recently suggested that pain catastrophizing in children is not the same as pain catastrophizing in adults. Specifically, after determining that the most highly endorsed items on the Pain Catastrophizing Scale for Children (PCS-C; a widely accepted measure of pain catastrophizing in youth aged 8–18 years) were those representing rumination, he suggested that “the PCS-C may be better understood as an index of worry about pain rather than catastrophic beliefs” (Eccleston et al., 2012: 1561). Worry is a hallmark symptom of generalized anxiety disorder, panic disorder, and other anxiety disorders (Craske, Rapee, Jackel, & Barlow, 1989). It is well-established that anxiety and pain catastrophizing are correlated with medium to large effect sizes in children both with (Benore et al., 2015; Jastrowski Mano et al., 2012) and without chronic pain (Hermann et al., 2007; Vervoort et al., 2010), and the relationship between anxiety and catastrophizing may be owing to shared underlying negative affectivity (MacDonald, Linton, & Jansson-Frojmark, 2008;
Tran et al. Noe¨l, Francis, Williams-Outerbridge, & Fung, 2012; Vervoort et al., 2006; Vlaeyen & Linton, 2000). Notwithstanding the above, it is generally accepted that anxiety and pain catastrophizing are theoretically distinct from one another in general (Benore et al., 2015; Eccleston et al., 2005), and there is some evidence to suggest that anxiety and pain catastrophizing have distinct roles in chronic pain, in particular. For example, pain catastrophizing predicted unique variance in functional disability over trait anxiety in a community sample of children (Vervoort et al., 2010) and variance in pain and disability over negative affectivity in a small clinical sample of youth with chronic pain (Vervoort et al., 2006). However, the picture is not always clear—Benore and colleagues (2015) found that changes in general anxiety and pain catastrophizing yielded similar relationships to functional outcomes in a pediatric chronic pain rehabilitation program. Notably, however, the researchers evaluated the roles of anxiety and pain catastrophizing in separate models, which does not allow for the accounting of shared variance between the related constructs. They call for clarification of the interconnected relationships between these “anxiety-related constructs” in a more parsimonious model. It has yet to be determined whether pain catastrophizing and anxiety are statistically distinct in a clinical pediatric chronic pain sample, and it is unclear how anxiety and pain catastrophizing are uniquely related to functional outcomes. Further, Simons and Kaczynski (2012) found that the relationship between fear of pain and avoidance of activities was significantly stronger for adolescents compared with younger children, suggesting important developmental differences in the way that similar constructs influence functioning. As suggested by Eccleston and colleagues (2012), developmental changes need to be accounted for, as cognitive development may influence how anxiety and pain catastrophizing manifest and influence functional outcomes. Accounting for potential changes in these relationships over child development would allow for more targeted and potentially more effective clinical interventions. Thus, two important questions emerge: (1) are anxiety and pain catastrophizing distinct constructs in pediatric chronic pain, both statistically and in relation to functional outcomes?, and (2) do anxiety and pain catastrophizing differentially predict functional outcomes as a function of age? The current study aimed to expand on recent literature by examining the potential commonalities and distinctions between anxiety and pain catastrophizing in a clinical sample of children and adolescents with chronic pain, and by examining the unique contributions of anxiety and pain catastrophizing with regard to important functional outcomes. We conducted exploratory factor analyses to determine which items showed the strongest discriminant validity for anxiety and pain catastrophizing. Based on past literature (Benore et al., 2015; Jastrowski Mano et al. 2012), we expected that some items would overlap in terms of conceptual similarity and item content; we removed those items to obtain unique anxiety and catastrophizing factors to predict functional outcomes. As improving functional disability and HRQOL are primary treatment goals for pediatric chronic pain (American Pain Society, 2001; Kashikar-Zuck et al., 2012; Zeltzer, Tsao, Bursch, & Myers, 2006), these constructs, along with pain intensity, were used as outcome variables. Models for younger children (ages 8–12 years) and adolescents (ages 13–18 years) were examined separately to account for potential agerelated differences. Based on previous literature, it was hypothesized that (1) anxiety and pain catastrophizing would be correlated, yet statistically distinct constructs, and (2) higher levels of anxiety and pain catastrophizing would uniquely predict higher functional disability, lower HRQOL, and higher pain intensity. Given their advanced cognitive capacity (Eccleston et al., 2012) and increased
Anxiety, Catastrophizing and Chronic Pain Outcomes agency (Simons & Kaczynski, 2012), it was hypothesized that (3) for adolescents, pain catastrophizing would demonstrate a stronger relationship with functional disability compared with anxiety.
3 chronic pain (Pielech et al., 2014). For this study, the total score was used for all analyses. Again, EFA was conducted separately for children and adolescents to determine which PCS-C items were most appropriate for SEM. See the Statistical Analysis and Results sections for details.
Methods Participants Participants included 725 youth (191 children age 8–12 years and 534 adolescents ages 13–18 years; M age ¼ 14.11 years, SD ¼ 2.45 years) from an outpatient interdisciplinary chronic pain center at a large midwestern U.S. children’s hospital. Youth were referred by a physician or specialized tertiary care clinic. Participants included 500 females (69%) and 225 males (31%); 75% were White, 9% were Black, 5% were Latino, and 11% identified as biracial or other. The most frequently reported primary pain locations were head (37%), abdomen (15%), lower extremity (15%), and back (14%). Seventeen percent of the sample reported a second pain location, and 4% reported a third pain location.
Functional Disability The Child Activity Limitations Questionnaire (CALQ; Hainsworth, Davies, Khan, & Weisman, 2007) is a 21-item youth self-report measure of functional disability. The respondent reports how difficult 21 activities are owing to pain on a 6-point Likert scale ranging from 0 (Not at all difficult) to 5 (Extremely difficult). It is a written measure based on of the Child Activity Limitations Interview (Palermo, Witherspoon, Valenzuela, & Drotar, 2004). The CALQ has demonstrated good internal consistency (a ¼ .91), and construct and discriminant validity as a self-report questionnaire (Hainsworth et al., 2007).
Data were gathered from a retrospective chart review of measures completed by youth before their intake appointment at the chronic pain center. Families were mailed a packet in advance of the appointment containing measures for the youth with chronic pain. Data were compiled into a de-identified database. The institutional review board approved this retrospective data review.
Health-Related Quality of Life The Pediatric Quality of Life Inventory (PedsQL 4.0; Varni, Seid, & Rode, 1999) is a youth self-report measure of HRQOL. The PedsQL 4.0 is a 23-item measure and has been validated for use with children and adolescents aged 8–18 years in both community and pediatric settings (Varni, Seid, & Kurtin, 2001). The total score ranges from 0 to 100, with higher scores indicating better HRQOL. For this investigation, the total score was used for all analyses.
Measures
Statistical Analysis
Patient Demographic Information At the pain clinic intake appointment, parents provided basic demographic information (e.g., patient’s age and gender).
Before testing the proposed models, data were screened for missing values and outliers. Descriptive statistics, Pearson correlations, and independentsamples t tests were conducted to describe the variables under review. We conducted t tests between younger children (8–12 years) and adolescents (13–18 years) across study variables to examine differences between the two age-groups; and between males and females to describe gender differences in the sample. To test the first hypothesis, which was that anxiety and pain catastrophizing would be correlated, yet statistically distinct constructs, the factor structure of the SCARED and PCS-C combined was examined using EFA, specifically principal axis factoring with varimax rotation. The EFAs were conducted on 54 items (41 SCARED and 13 PSC-C items), separately for children and adolescents. Eigen values, scree tests, and factor loadings were examined. Based on the literature (Benore et al., 2015; Jastrowski Mano et al., 2012), we anticipated some cross loading items (i.e., items demonstrating high factor loadings on both factors). Items were excluded if they loaded on both factors or had factor loadings 0.90, SRMR values
Journal of Pediatric Psychology, 2015, 1–10 doi: 10.1093/jpepsy/jsv029 Original Research Article
Distinct Influences of Anxiety and Pain Catastrophizing on Functional Outcomes in Children and Adolescents With Chronic Pain Susan T. Tran,1 PHD, Kristen E. Jastrowski Mano,2 PHD, Keri R. Hainsworth,3,4 PHD, Gustavo R. Medrano,5 PHD, Kimberly Anderson Khan,3,4 PSYD, Steven J. Weisman,3,4 MD, and W. Hobart Davies,6 PHD 1
Cincinnati Children’s Hospital Medical Center, 2University of Cincinnati, 3Medical College of Wisconsin, 4Children’s Hospital of Wisconsin, 5The Family Institute at Northwestern University, and 6University of Wisconsin-Milwaukee
All correspondence concerning this article should be addressed to Susan T. Tran, PHD, Division of Behavioral Medicine and Clinical Psychology, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Ave. MLC #3015, Cincinnati, OH 45229, USA. E-mail: [email protected] Received August 18, 2014; revisions received March 6, 2015; accepted March 11, 2015
Abstract Objectives Examine whether anxiety and pain catastrophizing are distinct constructs in relation to functional outcomes in pediatric chronic pain, and whether they differentially predict functional outcomes based on age. Methods In all, 725 youth (191 children, 534 adolescents) with chronic pain completed measures of pain characteristics, anxiety, pain catastrophizing, functional disability, and health-related quality of life (HRQOL). Structural equation modeling was used to examine interrelationships. Results Anxiety and pain catastrophizing were distinct. For both children and adolescents, pain catastrophizing predicted pain, functional disability, and HRQOL, and was a stronger predictor of pain intensity. For children, anxiety predicted HRQOL, and pain catastrophizing was a stronger predictor of functional disability. For adolescents, anxiety predicted functional disability and HRQOL, and anxiety was a stronger predictor of HRQOL. Conclusions There were age-related differences regarding whether anxiety or pain catastrophizing more strongly predicted specific functional outcomes. Assessment and intervention efforts should emphasize both anxiety and pain catastrophizing. Key words: anxiety; functional disability; pain catastrophizing; pediatric chronic pain.
Introduction Pediatric chronic pain is a relatively common condition among youth (King et al., 2011) that is often associated with decreased physical and psychological functioning, decreased satisfaction with health and life, and increased functional disability (Hunfeld et al., 2001; Merlijn et al., 2006; Palermo, 2000). Anxiety and pain catastrophizing are common psychological comorbidities in pediatric chronic pain, which influence functional outcomes (Cohen, Vowles, & Eccleston, 2010; Crombez et al., 2003; Vervoort, Goubert, Eccleston, Bijttebier, & Crombez, 2006; Wendland, Jackson, & Stokes, 2010). Researchers have noted the strong associations between these two constructs and called for research to distinguish the
role of each (Hermann, Hohmeister, Zohsel, Ebinger, & Flor, 2007; Holroyd, Drew, Cottrell, Romanek, & Heh, 2007; Jastrowski Mano et al., 2012; Vervoort, Eccleston, Goubert, Buysse, & Crombez, 2010) particularly in children and adolescents (Eccleston, Fisher, Vervoort, & Crombez, 2012); however, researchers have not yet adopted a statistical approach to clarify the relationships between these constructs. The clinical utility of both constructs may be clarified by the extent to which (1) each can be distinguished from the other, and (2) each is differentially related to important functional outcomes (Benore, D’Auria, Banez, Worley, & Tang, 2015; Drahovzal, Stewart, & Sullivan, 2006). The connection between pediatric chronic pain and anxiety has been well documented. Higher anxiety and internalizing symptoms
Published by Oxford University Press on behalf of the Society of Pediatric Psychology 2015. This work is written by US Government employees and is in the public domain in the US. 1
2 have been reported in youth with a variety of chronic pain complaints, such as recurrent abdominal pain (Campo et al., 2004; Galli et al., 2007), headaches (Galli et al., 2007; Mazzone, Vitiello, Incorpora, & Mazzone, 2006), and nonspecific musculoskeletal pain (O’Sullivan, Beales, Jensen, Murray, & Myers, 2011) compared with healthy control groups. Anxiety is often accompanied by physical arousal (Guite & Kazak, 2010), and the connection between anxiety and chronic pain is circular: physical pain may trigger anxious worrying, and the muscle tension associated with anxiety may heighten pain sensitivity (Beesdo et al., 2009; Simons & Kaczynski, 2012). Furthermore, in youth with chronic pain, high anxiety is related to impairments in functioning regardless of pain intensity (Cohen et al., 2010; Simons, Sieberg, & Claar, 2012). Even when an individual with chronic pain is not experiencing pain, he or she may anticipate having pain during an activity, worry about questions from peers about their pain, or worry about loss of functioning owing to pain (Powers, Gilman, & Hershey, 2006). Taken together, these studies suggest that anxiety plays a critical role in functional disability for youth with chronic pain. Pain catastrophizing, another construct related to negative expectancies, also plays a critical role in functional disability. Pain catastrophizing is a negative orientation toward painful stimuli including exaggerated or perseverative worry about the negative consequences of pain, perceived lack of control over symptoms, and an inability to cope with pain (Jones et al., 2003; Sullivan et al., 1995). Recently, it has been suggested that some aspects of pain catastrophizing are developmentally normal in childhood (i.e., not pathological, as is commonly believed; Eccleston et al., 2012); however, higher levels of pain catastrophizing have consistently been associated with poorer outcomes in pediatric chronic pain. The fear avoidance model (Asmundson, 1999), recently applied to pediatric chronic pain (Simons & Kaczynski, 2012), states that an individual with negative expectancies of pain, such as pain catastrophizing, is more likely to avoid activities that may flare pain resulting in increased functional disability over time. In populations of both school children and children with chronic pain, pain catastrophizing has been shown to be a significant predictor of functional disability over pain intensity (Crombez et al., 2003; Vervoort et al., 2006). Greater pain catastrophizing has also been associated with lower health-related quality of life (HRQOL) in youth with fibromyalgia (Libby & Glenwick, 2010) and in a community sample of children with chronic pain (Merlijn et al., 2006). While research has demonstrated that anxiety and pain catastrophizing are important constructs in chronic pain, it remains unclear what the defining characteristics of each are during childhood and adolescence. In fact, Eccleston and colleagues (2012) have recently suggested that pain catastrophizing in children is not the same as pain catastrophizing in adults. Specifically, after determining that the most highly endorsed items on the Pain Catastrophizing Scale for Children (PCS-C; a widely accepted measure of pain catastrophizing in youth aged 8–18 years) were those representing rumination, he suggested that “the PCS-C may be better understood as an index of worry about pain rather than catastrophic beliefs” (Eccleston et al., 2012: 1561). Worry is a hallmark symptom of generalized anxiety disorder, panic disorder, and other anxiety disorders (Craske, Rapee, Jackel, & Barlow, 1989). It is well-established that anxiety and pain catastrophizing are correlated with medium to large effect sizes in children both with (Benore et al., 2015; Jastrowski Mano et al., 2012) and without chronic pain (Hermann et al., 2007; Vervoort et al., 2010), and the relationship between anxiety and catastrophizing may be owing to shared underlying negative affectivity (MacDonald, Linton, & Jansson-Frojmark, 2008;
Tran et al. Noe¨l, Francis, Williams-Outerbridge, & Fung, 2012; Vervoort et al., 2006; Vlaeyen & Linton, 2000). Notwithstanding the above, it is generally accepted that anxiety and pain catastrophizing are theoretically distinct from one another in general (Benore et al., 2015; Eccleston et al., 2005), and there is some evidence to suggest that anxiety and pain catastrophizing have distinct roles in chronic pain, in particular. For example, pain catastrophizing predicted unique variance in functional disability over trait anxiety in a community sample of children (Vervoort et al., 2010) and variance in pain and disability over negative affectivity in a small clinical sample of youth with chronic pain (Vervoort et al., 2006). However, the picture is not always clear—Benore and colleagues (2015) found that changes in general anxiety and pain catastrophizing yielded similar relationships to functional outcomes in a pediatric chronic pain rehabilitation program. Notably, however, the researchers evaluated the roles of anxiety and pain catastrophizing in separate models, which does not allow for the accounting of shared variance between the related constructs. They call for clarification of the interconnected relationships between these “anxiety-related constructs” in a more parsimonious model. It has yet to be determined whether pain catastrophizing and anxiety are statistically distinct in a clinical pediatric chronic pain sample, and it is unclear how anxiety and pain catastrophizing are uniquely related to functional outcomes. Further, Simons and Kaczynski (2012) found that the relationship between fear of pain and avoidance of activities was significantly stronger for adolescents compared with younger children, suggesting important developmental differences in the way that similar constructs influence functioning. As suggested by Eccleston and colleagues (2012), developmental changes need to be accounted for, as cognitive development may influence how anxiety and pain catastrophizing manifest and influence functional outcomes. Accounting for potential changes in these relationships over child development would allow for more targeted and potentially more effective clinical interventions. Thus, two important questions emerge: (1) are anxiety and pain catastrophizing distinct constructs in pediatric chronic pain, both statistically and in relation to functional outcomes?, and (2) do anxiety and pain catastrophizing differentially predict functional outcomes as a function of age? The current study aimed to expand on recent literature by examining the potential commonalities and distinctions between anxiety and pain catastrophizing in a clinical sample of children and adolescents with chronic pain, and by examining the unique contributions of anxiety and pain catastrophizing with regard to important functional outcomes. We conducted exploratory factor analyses to determine which items showed the strongest discriminant validity for anxiety and pain catastrophizing. Based on past literature (Benore et al., 2015; Jastrowski Mano et al. 2012), we expected that some items would overlap in terms of conceptual similarity and item content; we removed those items to obtain unique anxiety and catastrophizing factors to predict functional outcomes. As improving functional disability and HRQOL are primary treatment goals for pediatric chronic pain (American Pain Society, 2001; Kashikar-Zuck et al., 2012; Zeltzer, Tsao, Bursch, & Myers, 2006), these constructs, along with pain intensity, were used as outcome variables. Models for younger children (ages 8–12 years) and adolescents (ages 13–18 years) were examined separately to account for potential agerelated differences. Based on previous literature, it was hypothesized that (1) anxiety and pain catastrophizing would be correlated, yet statistically distinct constructs, and (2) higher levels of anxiety and pain catastrophizing would uniquely predict higher functional disability, lower HRQOL, and higher pain intensity. Given their advanced cognitive capacity (Eccleston et al., 2012) and increased
Anxiety, Catastrophizing and Chronic Pain Outcomes agency (Simons & Kaczynski, 2012), it was hypothesized that (3) for adolescents, pain catastrophizing would demonstrate a stronger relationship with functional disability compared with anxiety.
3 chronic pain (Pielech et al., 2014). For this study, the total score was used for all analyses. Again, EFA was conducted separately for children and adolescents to determine which PCS-C items were most appropriate for SEM. See the Statistical Analysis and Results sections for details.
Methods Participants Participants included 725 youth (191 children age 8–12 years and 534 adolescents ages 13–18 years; M age ¼ 14.11 years, SD ¼ 2.45 years) from an outpatient interdisciplinary chronic pain center at a large midwestern U.S. children’s hospital. Youth were referred by a physician or specialized tertiary care clinic. Participants included 500 females (69%) and 225 males (31%); 75% were White, 9% were Black, 5% were Latino, and 11% identified as biracial or other. The most frequently reported primary pain locations were head (37%), abdomen (15%), lower extremity (15%), and back (14%). Seventeen percent of the sample reported a second pain location, and 4% reported a third pain location.
Functional Disability The Child Activity Limitations Questionnaire (CALQ; Hainsworth, Davies, Khan, & Weisman, 2007) is a 21-item youth self-report measure of functional disability. The respondent reports how difficult 21 activities are owing to pain on a 6-point Likert scale ranging from 0 (Not at all difficult) to 5 (Extremely difficult). It is a written measure based on of the Child Activity Limitations Interview (Palermo, Witherspoon, Valenzuela, & Drotar, 2004). The CALQ has demonstrated good internal consistency (a ¼ .91), and construct and discriminant validity as a self-report questionnaire (Hainsworth et al., 2007).
Data were gathered from a retrospective chart review of measures completed by youth before their intake appointment at the chronic pain center. Families were mailed a packet in advance of the appointment containing measures for the youth with chronic pain. Data were compiled into a de-identified database. The institutional review board approved this retrospective data review.
Health-Related Quality of Life The Pediatric Quality of Life Inventory (PedsQL 4.0; Varni, Seid, & Rode, 1999) is a youth self-report measure of HRQOL. The PedsQL 4.0 is a 23-item measure and has been validated for use with children and adolescents aged 8–18 years in both community and pediatric settings (Varni, Seid, & Kurtin, 2001). The total score ranges from 0 to 100, with higher scores indicating better HRQOL. For this investigation, the total score was used for all analyses.
Measures
Statistical Analysis
Patient Demographic Information At the pain clinic intake appointment, parents provided basic demographic information (e.g., patient’s age and gender).
Before testing the proposed models, data were screened for missing values and outliers. Descriptive statistics, Pearson correlations, and independentsamples t tests were conducted to describe the variables under review. We conducted t tests between younger children (8–12 years) and adolescents (13–18 years) across study variables to examine differences between the two age-groups; and between males and females to describe gender differences in the sample. To test the first hypothesis, which was that anxiety and pain catastrophizing would be correlated, yet statistically distinct constructs, the factor structure of the SCARED and PCS-C combined was examined using EFA, specifically principal axis factoring with varimax rotation. The EFAs were conducted on 54 items (41 SCARED and 13 PSC-C items), separately for children and adolescents. Eigen values, scree tests, and factor loadings were examined. Based on the literature (Benore et al., 2015; Jastrowski Mano et al., 2012), we anticipated some cross loading items (i.e., items demonstrating high factor loadings on both factors). Items were excluded if they loaded on both factors or had factor loadings 0.90, SRMR values
