- Email: [email protected]
Anxiety Measures Predict Health-Related Quality of Life in Children and Adolescents with Cyclic Vomiting Syndrome
Anxiety Measures Predict Health-Related Quality of Life in Children and Adolescents with Cyclic Vomiting Syndrome Sally E. Tarbell, PhD1,2, and B U. K. Li, MD3 Objective To evaluate the relationship between anxiety and health-related quality of life (HRQoL) in children and adolescents with cyclic vomiting syndrome (CVS). Study design Forty children aged 8-18 years diagnosed with CVS and 40 parents completed the Screen for Child Anxiety Related Emotional Disorders (SCARED) and the child and parent forms of the Pediatric Quality of Life Generic Core Scale, a measure of HRQoL. Results Eleven of the 40 children (27%) by self-report and 6 of 40 (15%) by parent-proxy report met the clinical cutoff for an anxiety disorder on the SCARED. Parent and child SCARED ratings were moderately correlated (intraclass correlation coefficient 0.68; P < .001). Child-rated HRQoL (mean SD, 74.3 15.2) and parent-rated HRQoL (mean, 72.1 14.6) were lower than healthy norms (P < .001). Disease severity (mean duration of CVS episodes, 3 2.4 days), annual frequency of CVS episodes (mean, 8.2 15.3), chronicity of CVS (mean, 5.8 3.4 years), and delay in diagnosis (mean, 2.4 1.9 years) were not associated with child-reported HRQoL; however, child SCARED scores accounted for approximately 50% of the variance in child-reported HRQoL (adjusted R2 = 0.49; df = 1, 38; P < .001). Conclusion Children and adolescents with CVS appear to be at increased risk for anxiety. Anxiety symptoms are a stronger predictor of HRQoL than disease characteristics in children and adolescents with CVS. Assessment and treatment of anxiety in children and adolescents with CVS may have a positive impact on HRQoL. (J Pediatr 2015;167:633-8).
C
yclic vomiting syndrome (CVS) is a functional gastrointestinal disorder characterized by recurring, stereotypic episodes of high intensity vomiting lasting for hours to days, often accompanied by symptoms of unrelenting nausea, retching, and severe abdominal pain.1 Between episodes, patients are typically healthy and resume normal activities.1 CVS is diagnosed based on the consensus criteria developed by an international multidisciplinary committee.2 Typically an extensive diagnostic evaluation is conducted to exclude other serious medical causes that can mimic its presentation, including intestinal malrotation, hydronephrosis, metabolic disorders, and increased intracranial pressure. Health-related quality of life (HRQoL) in children with CVS is low compared with that in children with irritable bowel syndrome and healthy peers, and similar to that in children with organic gastrointestinal disorders.3 This situation is similar to that of children with functional abdominal pain, who have lower HRQoL compared with healthy children and similar HRQoL as children with inflammatory bowel disease and gastroesophageal reflux.4 Anxiety is associated with CVS5,6 and other pediatric functional gastrointestinal disorders, including functional abdominal pain7-9 and nausea-predominant dyspepsia.10 Preliminary research indicates that anxiety can have a significant impact on HRQoL in children with such medical conditions as chronic pain11 and epilepsy.12 To date, no published study has evaluated the relationship between anxiety and HRQoL in pediatric CVS. The aim of the present study was to evaluate the relationship between child self-reports and parent-proxy reports of the child’s anxiety symptoms and disease characteristics and HRQoL in a cross-sectional study.
Methods Children were recruited from a specialty clinic for CVS at a children’s hospital. Those aged 8-18 years who met the international consensus criteria for CVS2 were invited to participate. Diagnostic criteria included: (1) recurrent severe, discrete episodes of vomiting; (2) normal health between episodes; (3) duration of vomiting of hours
CBT CVS HRQoL ICC PedsQL SCARED
Cognitive behavioral therapy Cyclic vomiting syndrome Health-related quality of life Intraclass correlation coefficient Pediatric Quality of Life Generic Core Scale Screen for Child Anxiety Related Emotional Disorders
From the 1Department of Psychiatry and Behavioral Sciences, Children’s Hospital Colorado; 2Department of Psychiatry, University of Colorado School of Medicine, Aurora, CO; and 3Division of Pediatric Gastroenterology, Hepatology and Nutrition, Department of Pediatrics, Medical College of Wisconsin, Milwaukee, WI The authors declare no conflicts of interest. 0022-3476/$ - see front matter. Copyright ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2015.05.032
633
THE JOURNAL OF PEDIATRICS
www.jpeds.com
to days; and (4) no apparent cause of vomiting, as well as supportive criteria that the episodes be stereotypical and selflimited. Children who were not English speakers or who had other major medical or developmental disorders were excluded. Demographic and medical information was collected by parent interview and review of the medical record. Anxiety Anxiety symptoms were assessed with the Screen for Child Anxiety Related Emotional Disorders (SCARED; http://psychiatry.pitt.edu/research/tools-research/assessmentinstruments).13 This 41-item questionnaire, based on the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition criteria for anxiety disorders in children, is validated for use in children aged 8 years and older. The SCARED screens for symptoms associated with specific anxiety disorders, as well as for behaviors, such as school avoidance, that are common across anxiety disorders in children. Total SCARED scores range from 0 to 82, with a score of $25 indicative of clinically significant anxiety symptoms. Both the child self-report and parent-proxy versions of the SCARED were used. The total SCARED scores were used to evaluate the relationship of anxiety symptoms to childand parent-rated HRQoL in the child participants. HRQoL Children aged 8-18 years completed the age-appropriate Pediatric Quality of Life Generic Core Scale (PedsQL),14 a 23item measure that yields a total HRQoL score and subscale scores of physical-, emotional-, social-, and school-related quality of life, as well as a composite psychosocial measure comprising the emotional, social, and school subscales. The PedsQL 4.0 versions for children (aged 8-12 years) and adolescents (aged 13-18 years) were used. Both the total HRQoL and the subscale scores range from 0 to 100, with higher scores indicating better HRQoL. Parents completed the age-appropriate PedsQL for their child or adolescent. The PedsQL inventories are validated instruments with satisfactory internal reliability,15,16 including for those with CVS.3 Comparison group data were derived from a study by Varni et al,17 which included a healthy control group and caregivers. The hospital’s Institutional Review Board approved this study. Parents and children provided written consent/assent before participating. Statistical Analyses Data analyses were performed with SPSS version 22.0 (IBM, Armonk, New York). The Cronbach a was used to evaluate the internal reliability of the PedsQL and the SCARED in the study sample. Intraclass correlation coefficients (ICCs) were used to evaluate agreement between parent and child reports of anxiety symptoms and HRQoL. ICC values were interpreted as follows: 0.40, poor agreement; 0.41-0.60, moderate agreement; 0.61-0.80 good agreement; and 0.81 and higher, excellent agreement. The 634
Vol. 167, No. 3 t test was used to examine differences in HRQoL between children with CVS and their parents’ proxy reports of HRQoL and published norms for healthy controls. Effect sizes were calculated using the Cohen d, with 0.20 indicating a small effect, 0.50 indicating a moderate effect, and 0.80 indicating a large effect. Stepwise multiple linear regression was used to examine associations among child self-reported and parent-proxy–reported anxiety symptoms, CVS characteristics, and HRQoL. Missing data for any dependent variable were excluded from the analyses. The significance level was set at P = .01 to control for type 1 error.
Results Of the 68 eligible families with CVS, 58 were enrolled in the study (85%). Data analyses were performed for the 40 families with complete data (Figure; available at www.jpeds.com). There were no differences between the families who did not enroll or did not complete the questionnaires in terms of child age, sex, or ethnicity. Parents completed the parent-proxy version of the SCARED, and 40 children completed the self-report version of the SCARED. Forty children completed the age-appropriate self-report version of the PedsQL (8-12 years, n = 20; 13-18 years, n = 20), and 40 parents completed the age-appropriate PedsQL for their child. Characteristics of the study sample are summarized in Table I (available at www.jpeds.com). Comorbid Conditions Nine children had a comorbid functional gastrointestinal disorder, including 3 with constipation (7%), 2 with abdominal migraine (5%), and 1 with irritable bowel syndrome (2%). Other medical conditions in the study group included allergies (n = 6; 15%), asthma (n = 4; 10%), postural orthostatic tachycardia syndrome (n = 3; 7%), and others (n = 2; 5%). Information on functional gastrointestinal symptoms, such as abdominal pain and nausea, was not gathered, owing to the overlap of such symptoms with CVS. Eight children had a history of 1 or more psychiatric diagnoses, including attention deficit hyperactivity disorder (n = 5; 12%), anxiety disorder (n = 3; 7%), and depression (n = 2; 5%). One child was receiving a selective serotonin reuptake inhibitor, but did not have a specific psychiatric diagnosis. Information on concurrent psychotherapy was not gathered. Medications A majority of the children were receiving 1 or more medications for CVS (n = 26; 65%), including amitriptyline in 17 (42%), propranolol in 11 (27%), L-carnitine and/or coenzyme Q10 in 10 (25%), cyproheptadine in 4 (10%), and others in 2 (5%). One subject was using marijuana for symptom control. Twenty-six children (65%) were prescribed an abortive medication (eg, antiemetic, sedative, analgesic) to Tarbell and Li
ORIGINAL ARTICLES
September 2015 take at the onset of a CVS attack. Six children (15%) were receiving no medications for CVS. Anxiety Symptoms The SCARED demonstrated excellent internal reliability for children with CVS (Cronbach a = 0.92) and for parentproxy reports (Cronbach a = 0.92). There was good agreement between parent and child ratings of the child’s anxiety symptoms on the SCARED (ICC = 0.68; 95% CI, 0.40-0.83; P < .001). The children’s mean score on the SCARED was 18.2 12.7, with 11 children (27%) meeting the clinical cutoff for anxiety symptoms (ie, total score of $25). The parents’ mean SCARED score was 14.2 11.3, and 6 children by parent report (15%) met the clinical cutoff for anxiety symptoms. Child-Reported HRQoL: Comparison with Published Data on Healthy Children The PedsQL demonstrated excellent internal consistency for child and adolescent self-reports (Cronbach a = 0.91-0.92). Overall, the lowest score for children with CVS was in the school domain, and the highest was in the social domain (Table II). Children with CVS reported significantly lower HRQoL than published normative data from healthy controls for the PedsQL total score and the physical, emotional, and psychosocial composite scores.17 Effect sizes for these differences ranged from small to large (Cohen d = 0.39-1.03). The social domain was not significantly different between children with CVS and healthy children (Table II). Parent-Proxy–Reported HRQoL: Comparison with Child-Reported HRQoL and Published Data Internal consistency for the parent-proxy–reported PedsQL for children and adolescents was excellent (Cronbach a = 0.90 for children, 0.92 for adolescents). As with the child’s self-report, the lowest scores were in the school domain, and the highest scores were in the social domain (Table II). Intraclass correlations were performed to evaluate agreement between parents and children on the PedsQL. Agreement was moderate to good across the subscale scores as well as for the PedsQL total score (ICC = 0.65-0.85; all significant at P < .01). Parent-proxy reported HRQoL for the children with CVS was significantly lower than that of healthy controls for the PedsQL total score and the physical, emotional, and psychosocial composite scales. Effect sizes for these differences ranged from small to large (Cohen d = 0.320.92). The social domain was not significantly different between parents of children with CVS and parents of healthy children (Table II). Relationship of CVS Characteristics and Anxiety Symptoms to Child- and Parent-Reported HRQoL Stepwise linear regression analyses were performed to determine the extent to which CVS characteristics (episode frequency, episode duration, chronicity, delay in diag-
Table II. PedsQL scale scores for child self-report and parent-proxy report compared with healthy group norms CVS Scale Child self-report Total Physical Psychosocial Emotional Social School Parent-proxy report Total Physical Psychosocial Emotional Social School
Healthy
n
Mean
SD
40 40 40 40 40 40
74.26 78.50 72.00 70.87 82.62 62.50
15.18 16.10 16.46 18.07 20.79 19.58
39 39 40 40 40 40
72.08 76.94 69.26 64.59 78.87 64.31
14.62 16.71 15.46 17.92 18.72 19.68
Mean
SD
Effect size; Cohen d
5972 5962 5963 5961 5948 5908
82.87* 86.86* 80.73* 78.21† 84.04 79.92*
13.16 13.88 14.70 18.64 17.43 16.93
0.65 0.60 0.59 0.39 0.08 1.03
10 070 10 050 10 071 10 044 10 036 8466
81.34* 83.26† 80.22* 80.28* 82.15 76.91*
15.92 19.98 15.84 16.99 20.08 20.16
0.58 0.32 0.69 0.92 0.16 0.63
n
Effect sizes are for comparisons between the CVS and healthy samples. The Cohen d was used to calculate effect sizes, with 0.20-0.30 designated as a small effect, 0.50 as a moderate effect, and 0.80 as a large effect. The t test was used to examine differences between children with CVS and the healthy comparison group. Healthy comparison group data are reprinted with permission from Varni et al.17 Significant differences between the CVS and healthy comparison groups: *P < .001. †P < .01.
nosis) and anxiety symptoms (total SCARED score) can predict the PedsQL total score. Separate analyses were conducted for the child and the parent data. A strong positive correlation was found between the child-rated total SCARED score and the self-reported PedsQL total score (r = 0.71), and the regression model predicted 49% of the variance (adjusted R2 = 0.49) (Table III). Only the SCARED total score entered the regression model as a significant predictor of the PedsQL total score, and the model was significant (F = 38.65; df = 1, 38; P < .001). Similarly, a positive correlation was found between the parent-proxy–rated total SCARED score and the parentproxy–rated PedsQL total score (r = 0.44) (Table IV). The regression model predicted 17% of the variance in the PedsQL total score (adjusted R2 = 0.17). Again, only the SCARED total score entered the regression model as a significant predictor of the total parent-proxy–rated PedsQL total score, and the model was significant (F = 8.89; df = 1, 37; P = .005). The regression analysis was repeated for the child data with the emotion domain scale removed from the PedsQL total score to control for possible confounding of the predictor variable, SCARED total score, with the dependent variable, PedsQL total score. The CVS characteristics and SCARED total score were used as predictors of this modified PedsQL total score. Again, the SCARED total score was the sole variable that entered the regression model as a significant predictor of this modified PedsQL total score (r = 0.68). The regression model predicted 45% of the variance in the modified PedsQL total score (adjusted R2 = 0.45). The model was significant (F = 32.76; df = 1, 38; P < .001).
Anxiety Measures Predict Health-Related Quality of Life in Children and Adolescents with Cyclic Vomiting Syndrome
635
THE JOURNAL OF PEDIATRICS
www.jpeds.com
Vol. 167, No. 3
Table III. Predictors of HRQoL in children with CVS by self-report Stepwise multiple regression model summaries for child predictors of HRQoL, child self-report Model 1
r
R2
Adjusted R2
SE of the estimate
R2 change
F change
df1
df2
Significance of F change
0.71*
0.50
0.49
10.83
0.50
38.65
1
38
<.001
Excluded variables Model 1, variable included: child SCARED total score
Beta in
Delay in diagnosis (mo) Duration of CVS episode (d) Episode frequency (annual) Chronicity (mo since diagnosis)
t
0.02 0.11 0.012 0.131
Significance
0.14 0.99 0.10 1.15
.89 .33 .92 .25
*Dependent variable: child-reported PedsQL total score.
Discussion Anxiety symptoms were significantly associated with HRQoL in the children with CVS. Quantitative disease characteristics, including illness chronicity, delay in CVS diagnosis, and frequency and duration of CVS episodes, did not contribute significantly to the amount of variance in HRQoL beyond that accounted for by anxiety symptoms. This was true even when the emotional subscale was removed from the PedsQL total score to control for possible confounding of the predictor and outcome variables; anxiety symptoms still accounted for a large portion of the variance in HRQoL. The strong association between anxiety and HRQoL also has been found in a study of children with epilepsy, in which anxiety, depression, and school achievement accounted for 51% of the variance in HRQoL,12 similar to the findings of the present study. This strong relationship between psychological symptoms and HRQoL underscores the importance of screening for psychiatric symptoms as part of the medical evaluation of children with CVS. The high rate of anxiety identified in these children (with 27% evidencing clinically significant anxiety symptoms by self-report) is consistent with findings from studies of children with other functional gastrointestinal disorders, especially functional abdominal pain.7,18 The relationship between anxiety and functional gastrointestinal disorders
is not well understood. Several explanations are offered. It is known that individuals with anxiety are hypervigilant for and differentially attend to interoceptive sensory information, allowing an augmented awareness of gastrointestinal discomfort.19 The heightened sympathetic arousal associated with anxiety states20 could lead to unpleasant gastrointestinal symptoms, including nausea and abdominal pain. CVS also is hypothesized to be a migraine variant for as many as 82% of children.21,22 The strong associations of migraine, anxiety, and mood disorders23,24 suggests a possible common underlying pathophysiology.24 Symptoms of CVS are very distressing to the affected children and their families, possibly leading to anticipatory anxiety related to the often-unpredictable nature of these severe vomiting attacks. The association of CVS with anxiety symptoms likely is multifactorial, given the numerous possible mechanisms through which anxiety symptoms and CVS could be associated. Further research is needed to better elaborate shared mechanisms from a genetic as well as metabolic, autonomic, neurocognitive, and behavioral standpoints. There is preliminary evidence of mutations in the mitochondrial DNA control region in children with CVS whose families share an increased incidence of other functional disorders, such as chronic fatigue, fibromyalgia, and irritable bowel syndrome.25,26 In addition, there is evidence of dysautonomia in children with CVS,27,28 functional abdominal pain,29,30 and anxiety disorders.20,31
Table IV. Predictors of HRQoL in children with CVS by parent-proxy report Stepwise multiple regression model summary for parent predictors of HRQoL, parent-proxy report Model 1
r
R2
Adjusted R2
SE of the estimate
R2 change
F change
df1
df2
Significance of F change
0.44*
0.19
0.17
13.30
0.17
8.89
1
37
.005
Excluded variables Model 1, variable included: parent-rated SCARED total score Delay in diagnosis (mo) Duration of CVS episode (d) Episode frequency (annual) Chronicity (mo since diagnosis)
Beta in 0.04 0.02 0.21 0.16
t 0.27 0.12 1.40 1.06
Significance .79 .90 .17 .29
*Dependent variable: parent-reported PedsQL total score.
636
Tarbell and Li
September 2015 Selective attention to threat-related information is a cognitive bias seen in children with anxiety disorders32 that also has been found in children with functional abdominal pain.33 Studies are currently underway to evaluate this tendency in youths with CVS. If such a cognitive bias is found, then attention bias modification, an emerging treatment for children with anxiety,34,35 also may be evaluated in children with CVS. Finally, there is a substantial literature supporting the use of cognitive behavioral therapy (CBT) for the treatment of anxiety in children,36 and emerging evidence suggesting that CBT can be effective in treating children with functional abdominal pain.37,38 There is a case report of successful symptom reduction in a child with CVS using CBT and heart rate variability biofeedback.39 These behavioral approaches warrant further study in CVS and its associated comorbidities, especially anxiety. Overall, the total HRQoL reported by the children and their parents was significantly lower than healthy norms as we found in a larger sample of children aged 2-18 years in a previous study.3 However, unlike in previous studies where the duration of the CVS episode was significantly associated with lower child-reported HRQoL, in the present study disease characteristics were not significantly associated with total HRQoL once entered into a predictor equation alongside anxiety symptoms. This may be attributable to the robust association between anxiety symptoms and HRQoL that was not investigated in the previous study. The lowest HRQoL subscore for children with CVS was in the school domain, whereas the highest score was in the social domain and did not differ from healthy norms by both child and parent report. The low school HRQoL may be attributable to the high average number of school days missed due to CVS (12; range, 0-60). Parents and children converged in their reports of HRQoL, supporting the reliability of these findings. This moderate-to-good agreement between parents’ and children’s ratings of HRQoL may reflect the objective nature of the characteristic vomiting attacks, where the parents are actively involved in symptom management. This study has several limitations. Even though we were able to recruit 85% of eligible participants for the study, the final sample was truncated owing to incomplete parent and/or child questionnaires. Although the use of published normative group data did not allow for precise demographic matching, the published data represented a large, heterogeneous population and provided reliable reference data for our study sample. Our study sample had limited ethnic diversity, being primarily Caucasian, in agreement with previous reports indicating that CVS may be more common in Caucasians.3,40,41 Furthermore, the participating families seen in a clinic specific for CVS may be more severely affected than the typical pediatric patients with CVS, leading them to seek out subspecialty care. Given the lack of studies of the severity of CVS episodes in children from a general pediatric practice compared with those from a subspecialty practice, we cannot judge the representativeness of our sample in terms of the larger affected population. Finally, owing to the study’s cross-sectional design, we are not able to
ORIGINAL ARTICLES comment on the impact of anxiety symptoms on HRQoL over time. Research on functional abdominal pain associated with anxiety has shown differences in the trajectory of the course of the illness,7 and thus a longitudinal evaluation of our CVS population could provide insight into how the treatment of anxiety may affect HRQoL over time. In summary, our study identifies anxiety symptoms as the strongest predictor of HRQoL in children and adolescents with CVS by both self-report and parent-proxy report. Because lower ratings of HRQoL have been linked to increased pediatric health care utilization and costs,42 assessment of HRQoL can be a tool for both monitoring and improving patient care. Most importantly, these data point toward potentially important clinical impacts—first that a biopsychosocial approach, including assessment of anxiety, needs to be part of the standard of care of children with CVS, and second, that identification and successful treatment of anxiety comorbidity could play a key role in improving clinical outcomes and reducing the substantial impact of this functional gastrointestinal disorder on the child, the family, and the health care system. There are several brief anxiety screeners that can be completed by parents and children and scored in a few minutes (http://psychiatry.pitt. edu/research/tools-research/assessment-instruments; http:// www.childfirst.ucla.edu/Resources.html; www.scaswebsite. com).13,43-45 The fact that these scales are available online at no cost and can be administered by clinic personnel with oversight by a mental health professional supports the feasibility of such screening. n Submitted for publication Feb 27, 2015; last revision received Apr 15, 2015; accepted May 19, 2015. Reprint requests: Sally E. Tarbell, PhD, Department of Psychiatry and Behavioral Sciences, Children’s Hospital Colorado, 13123 East 16th Ave, B130, Aurora, CO 80045. E-mail: [email protected]
References 1. Li BUK, Balint JP. Cyclic vomiting syndrome: evolution of our understanding of a brain-gut disorder. Adv Pediatr 2000;47:117-60. 2. Li BUK, Lefevre F, Chelimsky G, Boles RG, Nelson SP, Lewis DW, et al. North American Society for Pediatric Gastroenterology, Hepatology and Nutrition consensus statement on the diagnosis and management of cyclic vomiting syndrome. J Pediatr Gastroenterol Nutr 2008;47: 379-93. 3. Tarbell S, Li BUK. Health-related quality of life in children and adolescents with cyclic vomiting syndrome: a comparison with published data on youth with irritable bowel syndrome and organic gastrointestinal disorders. J Pediatr 2013;163:493-7. 4. Youssef N, Murphy T, Langseder A, Rosh J. Quality of life for children with functional abdominal pain: a comparison study of patients’ and parents’ perspectives. Pediatrics 2006;117:54-9. 5. Tarbell S, Li BUK. Psychiatric symptoms in children and adolescents with cyclic vomiting syndrome and their parents. Headache 2008;48: 259-66. 6. Forbes D, Withers G, Silburn S, McKelvey R. Psychological and social characteristics and precipitants of vomiting in children with cyclic vomiting syndrome. Dig Dis Sci 1999;44:19S-22S. 7. Shelby G, Shirkey K, Sherman A, Beck J, Haman K, Shears A, et al. Functional abdominal pain in childhood and long-term vulnerability to anxiety disorders. Pediatrics 2013;132:475-82.
Anxiety Measures Predict Health-Related Quality of Life in Children and Adolescents with Cyclic Vomiting Syndrome
637
THE JOURNAL OF PEDIATRICS
www.jpeds.com
8. Yacob D, DiLorenzo C, Bridge J, Rosenstein P, Onorato M, Bravender T, et al. Prevalence of pain-predominant functional gastrointestinal disorders and somatic symptoms in patients with anxiety or depression disorders. J Pediatr 2013;163:767-70. 9. Dufton L, Dunn M, Compas B. Anxiety and somatic complaints in children with recurrent abdominal pain and anxiety disorders. J Pediatr Psychol 2009;34:176-86. 10. Tarbell S, Shaltout H, Wagoner A, Diz D, Fortunato J. Relationship among nausea, anxiety and orthostatic symptoms in pediatric patients with chronic unexplained nausea. Exp Brain Res 2014;232: 2645-50. 11. Mahrer N, Monta~ no Z, Gold J. Relations between anxiety sensitivity, somatization and health-related quality of life in children with chronic pain. J Pediatr Psychol 2012;37:806-16. 12. Stevanovic D, Jancic J, Lakic A. The impact of depression and anxiety symptoms on the health-related quality of life of children and adolescents with epilepsy. Epilepsia 2011;52:e75-8. 13. Birmaher B, Brent D, Chiappetta L, Bridge J, Monga S, Baugher M. Psychometric properties of the Screen for Child Anxiety Related Emotional Disorders (SCARED): a replication study. J Am Acad Child Adolesc Psychiatry 1999;38:1230-6. 14. Varni JW, Seid M, Knight T, Uzark K, Szer I. The PedsQL 4.0 Generic Core Scales: sensitivity, responsiveness, and impact on clinical decision-making. J Behav Med 2002;25:175-93. 15. Varni J, Limbers C, Burwinkle T. How young can children reliably and validly self-report their health-related quality of life?: an analysis of 8,591 children across age subgroups with the PedsQL 4.0 Generic Core Scales. Health Qual Life Outcomes 2007;5:1-13. 16. Varni J, Limbers C, Burwinkle T. Parent proxy-report of their children’s health-related quality of life: an analysis of 13,878 parents’ reliability and validity across ages subgroups using the PedsQL 4.0 Generic Core Scales. Health Qual Life Outcomes 2007;5:1-10. 17. Varni J, Burwinkle T, Rapoli M, Seid M, Skarr D. The PedsQL 4.0 as a pediatric population health measure: feasibility, reliability, and validity. Ambul Pediatr 2003;3:329-41. 18. Campo J, Bridge J, Ehmann M. Recurrent abdominal pain, anxiety and depression in primary care. JAMA Pediatr 2004;113:817-24. 19. Barlow D. Anxiety and its disorders: The nature and treatment of anxiety and panic. New York: Guilford Press; 2004. 20. Cohen H, Benjamin J. Power spectrum analysis and cardiovascular morbidity in anxiety disorders. Auton Neurosci 2006;128:1-8. 21. Li BUK, Murray RD, Heitlinger LA, Robbins JL, Hayes JR. Is cyclic vomiting syndrome related to migraine? J Pediatr 1999;134:567-72. 22. Stickler GB. Relationship between cyclic vomiting syndrome and migraine. Clin Pediatr 2005;44:505-8. 23. Hamelsky S, Lipton R. Psychiatric comorbidity of migraine. Headache 2006;46:1327-33. 24. Merikangas KR, Stevens DE. Comorbidity of migraine and psychiatric disorders. Neurol Clin 1997;15:115-23. 25. Boles RG, Adams K, Li BUK. Maternal inheritance in cyclic vomiting syndrome. Am J Med Genet A 2005;133:71-7. 26. Boles RG, Powers ALR, Adams K. Cyclic vomiting syndrome plus. J Child Neurol 2006;21:182-8.
638
Vol. 167, No. 3 27. To J, Issenman RM, Kamath MV. Evaluation of neurocardiac signals in pediatric patients with cyclic vomiting syndrome through power spectral analysis of heart rate variability. J Pediatr 1999;135:363-6. 28. Chelimsky TC, Chelimsky GC. Autonomic abnormalities in cyclic vomiting syndrome. J Pediatr Gastroenterol Nutr 2007;44:326-30. 29. Sowder E, Gevirtz R, Shapiro W, Ebert C. Restoration of vagal tone: a possible mechanism for functional abdominal pain. Appl Psychophysiol Biofeedback 2010;35:199-206. 30. Chelimsky G, Boyle J, Tusing L, Chelimsky T. Autonomic abnormalities in children with functional abdominal pain: coincidence or etiology? J Pediatr Gastroenterol Nutr 2001;33:47-53. 31. Boyce W, Quas J, Alkon A, Smider N, Essex M, Kupfer D. Autonomic reactivity and psychopathology in middle childhood. Br J Psychiatry 2001;179:144-50. 32. Muris P, Rapee R, Meesters C, Schouten E, Geers M. Threat perception abnormalities in children: the role of anxiety disorder symptoms, chronic anxiety, and state anxiety. J Anxiety Disord 2003;17:271-87. 33. Boyer MC, Compas BE, Stanger C, Colletti RB, Konik BS, Morrow SB, et al. Attentional biases to pain and social threat in children with recurrent abdominal pain. J Pediatr Psychol 2006;31:209-20. 34. Eldar S, Apter A, Lotan D, Edgar K, Naim R, Fox N, et al. Attention bias modification treatment for pediatric anxiety disorders: a randomized controlled trial. Am J Psychiatry 2012;169:213-30. 35. Hakamata Y, Lissek S, Bar-Haim Y, Britton J, Fox N, Leibenluft E, et al. Attention bias modification treatment: a meta-analysis towards the establishment of a novel treatment for anxiety. Biol Psychiatry 2010; 68:982-90. 36. Silverman W, Pina A, Viswesvaran C. Evidence-based psychosocial treatments for phobic and anxiety disorders in children and adolescents. J Clin Child Adolesc Psychol 2008;37:105-30. 37. Brent M, Lobato D, LeLeiko N. Psychological treatments for pediatric functional gastrointestional disorders. J Pediatr Gastroenterol Nutr 2008;48:13-21. 38. Fisher E, Heathcote L, Palermo T, de C Williams AC, Lau J, Eccleston C. Systematic review and meta-analysis: psychological therapies for children with chronic pain. J Pediatr Psychol 2014;39:866-86. 39. Slutsker B, Konichezky A, Gothelf D. Breaking the cycle: cognitive behavioral therapy and biofeedback training in a case of cyclic vomiting syndrome. Psychol Health Med 2010;15:625-31. 40. Li BUK, Misiewicz L. Cyclic vomiting syndrome: a brain-gut disorder. Gastroenterol Clin North Am 2003;32:997-1019. 41. Pareek N, Fleisher DR, Abell TA. Cyclic vomiting syndrome: what a gastroenterologist needs to know. Am J Gastroenterol 2007;102:2832-40. 42. Seid M, Varni J, Segall D, Kurtin P. Health-related quality of life as a predictor of pediatric healthcare costs: a two-year prospective cohort analysis. Health Qual Life Outcomes 2004;2:1-10. 43. Chorpita B, Moffitt C, Gray J. Psychometric properties of the Revised Child Anxiety and Depression Scale in a clinical sample. Behav Res Ther 2005;43:309-22. 44. Spence S. Structure of anxiety symptoms among children: a confirmatory factor-analysis. J Abnorm Psychol 1997;1997:280-97. 45. Spence S, Barrett P, Turner C. Psychometric properties of the Spence Children’s Anxiety Scale. J Anxiety Disord 2003;17:605-25.
Tarbell and Li
ORIGINAL ARTICLES
September 2015
Figure. Subject enrollment.
Table I. Patient characteristics Characteristics
Value
Age, y, mean (SD) Sex, n (%) Male Female Ethnic minority, n (%) Annual number of CVS attacks, mean (SD) Duration of CVS episodes, d, median; mean (SD) Annual emergency department visits for CVS, range; mean (SD) Chronicity (onset of symptoms to visit), y, mean (SD) Delay in diagnosis (symptom onset to diagnosis of CVS), y, mean (SD) School days missed per year, range; mean (SD)
12.5 (2.7) 20 (50) 20 (50) 2 (5) 8.2 (15.3) 2; 3 (2.4) 0-12; 1.3 (2.5) 5.8 (3.4) 2.4 (1.9) 0-60; 12.3 (14.9)
Anxiety Measures Predict Health-Related Quality of Life in Children and Adolescents with Cyclic Vomiting Syndrome
638.e1
C
yclic vomiting syndrome (CVS) is a functional gastrointestinal disorder characterized by recurring, stereotypic episodes of high intensity vomiting lasting for hours to days, often accompanied by symptoms of unrelenting nausea, retching, and severe abdominal pain.1 Between episodes, patients are typically healthy and resume normal activities.1 CVS is diagnosed based on the consensus criteria developed by an international multidisciplinary committee.2 Typically an extensive diagnostic evaluation is conducted to exclude other serious medical causes that can mimic its presentation, including intestinal malrotation, hydronephrosis, metabolic disorders, and increased intracranial pressure. Health-related quality of life (HRQoL) in children with CVS is low compared with that in children with irritable bowel syndrome and healthy peers, and similar to that in children with organic gastrointestinal disorders.3 This situation is similar to that of children with functional abdominal pain, who have lower HRQoL compared with healthy children and similar HRQoL as children with inflammatory bowel disease and gastroesophageal reflux.4 Anxiety is associated with CVS5,6 and other pediatric functional gastrointestinal disorders, including functional abdominal pain7-9 and nausea-predominant dyspepsia.10 Preliminary research indicates that anxiety can have a significant impact on HRQoL in children with such medical conditions as chronic pain11 and epilepsy.12 To date, no published study has evaluated the relationship between anxiety and HRQoL in pediatric CVS. The aim of the present study was to evaluate the relationship between child self-reports and parent-proxy reports of the child’s anxiety symptoms and disease characteristics and HRQoL in a cross-sectional study.
Methods Children were recruited from a specialty clinic for CVS at a children’s hospital. Those aged 8-18 years who met the international consensus criteria for CVS2 were invited to participate. Diagnostic criteria included: (1) recurrent severe, discrete episodes of vomiting; (2) normal health between episodes; (3) duration of vomiting of hours
CBT CVS HRQoL ICC PedsQL SCARED
Cognitive behavioral therapy Cyclic vomiting syndrome Health-related quality of life Intraclass correlation coefficient Pediatric Quality of Life Generic Core Scale Screen for Child Anxiety Related Emotional Disorders
From the 1Department of Psychiatry and Behavioral Sciences, Children’s Hospital Colorado; 2Department of Psychiatry, University of Colorado School of Medicine, Aurora, CO; and 3Division of Pediatric Gastroenterology, Hepatology and Nutrition, Department of Pediatrics, Medical College of Wisconsin, Milwaukee, WI The authors declare no conflicts of interest. 0022-3476/$ - see front matter. Copyright ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2015.05.032
633
THE JOURNAL OF PEDIATRICS
www.jpeds.com
to days; and (4) no apparent cause of vomiting, as well as supportive criteria that the episodes be stereotypical and selflimited. Children who were not English speakers or who had other major medical or developmental disorders were excluded. Demographic and medical information was collected by parent interview and review of the medical record. Anxiety Anxiety symptoms were assessed with the Screen for Child Anxiety Related Emotional Disorders (SCARED; http://psychiatry.pitt.edu/research/tools-research/assessmentinstruments).13 This 41-item questionnaire, based on the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition criteria for anxiety disorders in children, is validated for use in children aged 8 years and older. The SCARED screens for symptoms associated with specific anxiety disorders, as well as for behaviors, such as school avoidance, that are common across anxiety disorders in children. Total SCARED scores range from 0 to 82, with a score of $25 indicative of clinically significant anxiety symptoms. Both the child self-report and parent-proxy versions of the SCARED were used. The total SCARED scores were used to evaluate the relationship of anxiety symptoms to childand parent-rated HRQoL in the child participants. HRQoL Children aged 8-18 years completed the age-appropriate Pediatric Quality of Life Generic Core Scale (PedsQL),14 a 23item measure that yields a total HRQoL score and subscale scores of physical-, emotional-, social-, and school-related quality of life, as well as a composite psychosocial measure comprising the emotional, social, and school subscales. The PedsQL 4.0 versions for children (aged 8-12 years) and adolescents (aged 13-18 years) were used. Both the total HRQoL and the subscale scores range from 0 to 100, with higher scores indicating better HRQoL. Parents completed the age-appropriate PedsQL for their child or adolescent. The PedsQL inventories are validated instruments with satisfactory internal reliability,15,16 including for those with CVS.3 Comparison group data were derived from a study by Varni et al,17 which included a healthy control group and caregivers. The hospital’s Institutional Review Board approved this study. Parents and children provided written consent/assent before participating. Statistical Analyses Data analyses were performed with SPSS version 22.0 (IBM, Armonk, New York). The Cronbach a was used to evaluate the internal reliability of the PedsQL and the SCARED in the study sample. Intraclass correlation coefficients (ICCs) were used to evaluate agreement between parent and child reports of anxiety symptoms and HRQoL. ICC values were interpreted as follows: 0.40, poor agreement; 0.41-0.60, moderate agreement; 0.61-0.80 good agreement; and 0.81 and higher, excellent agreement. The 634
Vol. 167, No. 3 t test was used to examine differences in HRQoL between children with CVS and their parents’ proxy reports of HRQoL and published norms for healthy controls. Effect sizes were calculated using the Cohen d, with 0.20 indicating a small effect, 0.50 indicating a moderate effect, and 0.80 indicating a large effect. Stepwise multiple linear regression was used to examine associations among child self-reported and parent-proxy–reported anxiety symptoms, CVS characteristics, and HRQoL. Missing data for any dependent variable were excluded from the analyses. The significance level was set at P = .01 to control for type 1 error.
Results Of the 68 eligible families with CVS, 58 were enrolled in the study (85%). Data analyses were performed for the 40 families with complete data (Figure; available at www.jpeds.com). There were no differences between the families who did not enroll or did not complete the questionnaires in terms of child age, sex, or ethnicity. Parents completed the parent-proxy version of the SCARED, and 40 children completed the self-report version of the SCARED. Forty children completed the age-appropriate self-report version of the PedsQL (8-12 years, n = 20; 13-18 years, n = 20), and 40 parents completed the age-appropriate PedsQL for their child. Characteristics of the study sample are summarized in Table I (available at www.jpeds.com). Comorbid Conditions Nine children had a comorbid functional gastrointestinal disorder, including 3 with constipation (7%), 2 with abdominal migraine (5%), and 1 with irritable bowel syndrome (2%). Other medical conditions in the study group included allergies (n = 6; 15%), asthma (n = 4; 10%), postural orthostatic tachycardia syndrome (n = 3; 7%), and others (n = 2; 5%). Information on functional gastrointestinal symptoms, such as abdominal pain and nausea, was not gathered, owing to the overlap of such symptoms with CVS. Eight children had a history of 1 or more psychiatric diagnoses, including attention deficit hyperactivity disorder (n = 5; 12%), anxiety disorder (n = 3; 7%), and depression (n = 2; 5%). One child was receiving a selective serotonin reuptake inhibitor, but did not have a specific psychiatric diagnosis. Information on concurrent psychotherapy was not gathered. Medications A majority of the children were receiving 1 or more medications for CVS (n = 26; 65%), including amitriptyline in 17 (42%), propranolol in 11 (27%), L-carnitine and/or coenzyme Q10 in 10 (25%), cyproheptadine in 4 (10%), and others in 2 (5%). One subject was using marijuana for symptom control. Twenty-six children (65%) were prescribed an abortive medication (eg, antiemetic, sedative, analgesic) to Tarbell and Li
ORIGINAL ARTICLES
September 2015 take at the onset of a CVS attack. Six children (15%) were receiving no medications for CVS. Anxiety Symptoms The SCARED demonstrated excellent internal reliability for children with CVS (Cronbach a = 0.92) and for parentproxy reports (Cronbach a = 0.92). There was good agreement between parent and child ratings of the child’s anxiety symptoms on the SCARED (ICC = 0.68; 95% CI, 0.40-0.83; P < .001). The children’s mean score on the SCARED was 18.2 12.7, with 11 children (27%) meeting the clinical cutoff for anxiety symptoms (ie, total score of $25). The parents’ mean SCARED score was 14.2 11.3, and 6 children by parent report (15%) met the clinical cutoff for anxiety symptoms. Child-Reported HRQoL: Comparison with Published Data on Healthy Children The PedsQL demonstrated excellent internal consistency for child and adolescent self-reports (Cronbach a = 0.91-0.92). Overall, the lowest score for children with CVS was in the school domain, and the highest was in the social domain (Table II). Children with CVS reported significantly lower HRQoL than published normative data from healthy controls for the PedsQL total score and the physical, emotional, and psychosocial composite scores.17 Effect sizes for these differences ranged from small to large (Cohen d = 0.39-1.03). The social domain was not significantly different between children with CVS and healthy children (Table II). Parent-Proxy–Reported HRQoL: Comparison with Child-Reported HRQoL and Published Data Internal consistency for the parent-proxy–reported PedsQL for children and adolescents was excellent (Cronbach a = 0.90 for children, 0.92 for adolescents). As with the child’s self-report, the lowest scores were in the school domain, and the highest scores were in the social domain (Table II). Intraclass correlations were performed to evaluate agreement between parents and children on the PedsQL. Agreement was moderate to good across the subscale scores as well as for the PedsQL total score (ICC = 0.65-0.85; all significant at P < .01). Parent-proxy reported HRQoL for the children with CVS was significantly lower than that of healthy controls for the PedsQL total score and the physical, emotional, and psychosocial composite scales. Effect sizes for these differences ranged from small to large (Cohen d = 0.320.92). The social domain was not significantly different between parents of children with CVS and parents of healthy children (Table II). Relationship of CVS Characteristics and Anxiety Symptoms to Child- and Parent-Reported HRQoL Stepwise linear regression analyses were performed to determine the extent to which CVS characteristics (episode frequency, episode duration, chronicity, delay in diag-
Table II. PedsQL scale scores for child self-report and parent-proxy report compared with healthy group norms CVS Scale Child self-report Total Physical Psychosocial Emotional Social School Parent-proxy report Total Physical Psychosocial Emotional Social School
Healthy
n
Mean
SD
40 40 40 40 40 40
74.26 78.50 72.00 70.87 82.62 62.50
15.18 16.10 16.46 18.07 20.79 19.58
39 39 40 40 40 40
72.08 76.94 69.26 64.59 78.87 64.31
14.62 16.71 15.46 17.92 18.72 19.68
Mean
SD
Effect size; Cohen d
5972 5962 5963 5961 5948 5908
82.87* 86.86* 80.73* 78.21† 84.04 79.92*
13.16 13.88 14.70 18.64 17.43 16.93
0.65 0.60 0.59 0.39 0.08 1.03
10 070 10 050 10 071 10 044 10 036 8466
81.34* 83.26† 80.22* 80.28* 82.15 76.91*
15.92 19.98 15.84 16.99 20.08 20.16
0.58 0.32 0.69 0.92 0.16 0.63
n
Effect sizes are for comparisons between the CVS and healthy samples. The Cohen d was used to calculate effect sizes, with 0.20-0.30 designated as a small effect, 0.50 as a moderate effect, and 0.80 as a large effect. The t test was used to examine differences between children with CVS and the healthy comparison group. Healthy comparison group data are reprinted with permission from Varni et al.17 Significant differences between the CVS and healthy comparison groups: *P < .001. †P < .01.
nosis) and anxiety symptoms (total SCARED score) can predict the PedsQL total score. Separate analyses were conducted for the child and the parent data. A strong positive correlation was found between the child-rated total SCARED score and the self-reported PedsQL total score (r = 0.71), and the regression model predicted 49% of the variance (adjusted R2 = 0.49) (Table III). Only the SCARED total score entered the regression model as a significant predictor of the PedsQL total score, and the model was significant (F = 38.65; df = 1, 38; P < .001). Similarly, a positive correlation was found between the parent-proxy–rated total SCARED score and the parentproxy–rated PedsQL total score (r = 0.44) (Table IV). The regression model predicted 17% of the variance in the PedsQL total score (adjusted R2 = 0.17). Again, only the SCARED total score entered the regression model as a significant predictor of the total parent-proxy–rated PedsQL total score, and the model was significant (F = 8.89; df = 1, 37; P = .005). The regression analysis was repeated for the child data with the emotion domain scale removed from the PedsQL total score to control for possible confounding of the predictor variable, SCARED total score, with the dependent variable, PedsQL total score. The CVS characteristics and SCARED total score were used as predictors of this modified PedsQL total score. Again, the SCARED total score was the sole variable that entered the regression model as a significant predictor of this modified PedsQL total score (r = 0.68). The regression model predicted 45% of the variance in the modified PedsQL total score (adjusted R2 = 0.45). The model was significant (F = 32.76; df = 1, 38; P < .001).
Anxiety Measures Predict Health-Related Quality of Life in Children and Adolescents with Cyclic Vomiting Syndrome
635
THE JOURNAL OF PEDIATRICS
www.jpeds.com
Vol. 167, No. 3
Table III. Predictors of HRQoL in children with CVS by self-report Stepwise multiple regression model summaries for child predictors of HRQoL, child self-report Model 1
r
R2
Adjusted R2
SE of the estimate
R2 change
F change
df1
df2
Significance of F change
0.71*
0.50
0.49
10.83
0.50
38.65
1
38
<.001
Excluded variables Model 1, variable included: child SCARED total score
Beta in
Delay in diagnosis (mo) Duration of CVS episode (d) Episode frequency (annual) Chronicity (mo since diagnosis)
t
0.02 0.11 0.012 0.131
Significance
0.14 0.99 0.10 1.15
.89 .33 .92 .25
*Dependent variable: child-reported PedsQL total score.
Discussion Anxiety symptoms were significantly associated with HRQoL in the children with CVS. Quantitative disease characteristics, including illness chronicity, delay in CVS diagnosis, and frequency and duration of CVS episodes, did not contribute significantly to the amount of variance in HRQoL beyond that accounted for by anxiety symptoms. This was true even when the emotional subscale was removed from the PedsQL total score to control for possible confounding of the predictor and outcome variables; anxiety symptoms still accounted for a large portion of the variance in HRQoL. The strong association between anxiety and HRQoL also has been found in a study of children with epilepsy, in which anxiety, depression, and school achievement accounted for 51% of the variance in HRQoL,12 similar to the findings of the present study. This strong relationship between psychological symptoms and HRQoL underscores the importance of screening for psychiatric symptoms as part of the medical evaluation of children with CVS. The high rate of anxiety identified in these children (with 27% evidencing clinically significant anxiety symptoms by self-report) is consistent with findings from studies of children with other functional gastrointestinal disorders, especially functional abdominal pain.7,18 The relationship between anxiety and functional gastrointestinal disorders
is not well understood. Several explanations are offered. It is known that individuals with anxiety are hypervigilant for and differentially attend to interoceptive sensory information, allowing an augmented awareness of gastrointestinal discomfort.19 The heightened sympathetic arousal associated with anxiety states20 could lead to unpleasant gastrointestinal symptoms, including nausea and abdominal pain. CVS also is hypothesized to be a migraine variant for as many as 82% of children.21,22 The strong associations of migraine, anxiety, and mood disorders23,24 suggests a possible common underlying pathophysiology.24 Symptoms of CVS are very distressing to the affected children and their families, possibly leading to anticipatory anxiety related to the often-unpredictable nature of these severe vomiting attacks. The association of CVS with anxiety symptoms likely is multifactorial, given the numerous possible mechanisms through which anxiety symptoms and CVS could be associated. Further research is needed to better elaborate shared mechanisms from a genetic as well as metabolic, autonomic, neurocognitive, and behavioral standpoints. There is preliminary evidence of mutations in the mitochondrial DNA control region in children with CVS whose families share an increased incidence of other functional disorders, such as chronic fatigue, fibromyalgia, and irritable bowel syndrome.25,26 In addition, there is evidence of dysautonomia in children with CVS,27,28 functional abdominal pain,29,30 and anxiety disorders.20,31
Table IV. Predictors of HRQoL in children with CVS by parent-proxy report Stepwise multiple regression model summary for parent predictors of HRQoL, parent-proxy report Model 1
r
R2
Adjusted R2
SE of the estimate
R2 change
F change
df1
df2
Significance of F change
0.44*
0.19
0.17
13.30
0.17
8.89
1
37
.005
Excluded variables Model 1, variable included: parent-rated SCARED total score Delay in diagnosis (mo) Duration of CVS episode (d) Episode frequency (annual) Chronicity (mo since diagnosis)
Beta in 0.04 0.02 0.21 0.16
t 0.27 0.12 1.40 1.06
Significance .79 .90 .17 .29
*Dependent variable: parent-reported PedsQL total score.
636
Tarbell and Li
September 2015 Selective attention to threat-related information is a cognitive bias seen in children with anxiety disorders32 that also has been found in children with functional abdominal pain.33 Studies are currently underway to evaluate this tendency in youths with CVS. If such a cognitive bias is found, then attention bias modification, an emerging treatment for children with anxiety,34,35 also may be evaluated in children with CVS. Finally, there is a substantial literature supporting the use of cognitive behavioral therapy (CBT) for the treatment of anxiety in children,36 and emerging evidence suggesting that CBT can be effective in treating children with functional abdominal pain.37,38 There is a case report of successful symptom reduction in a child with CVS using CBT and heart rate variability biofeedback.39 These behavioral approaches warrant further study in CVS and its associated comorbidities, especially anxiety. Overall, the total HRQoL reported by the children and their parents was significantly lower than healthy norms as we found in a larger sample of children aged 2-18 years in a previous study.3 However, unlike in previous studies where the duration of the CVS episode was significantly associated with lower child-reported HRQoL, in the present study disease characteristics were not significantly associated with total HRQoL once entered into a predictor equation alongside anxiety symptoms. This may be attributable to the robust association between anxiety symptoms and HRQoL that was not investigated in the previous study. The lowest HRQoL subscore for children with CVS was in the school domain, whereas the highest score was in the social domain and did not differ from healthy norms by both child and parent report. The low school HRQoL may be attributable to the high average number of school days missed due to CVS (12; range, 0-60). Parents and children converged in their reports of HRQoL, supporting the reliability of these findings. This moderate-to-good agreement between parents’ and children’s ratings of HRQoL may reflect the objective nature of the characteristic vomiting attacks, where the parents are actively involved in symptom management. This study has several limitations. Even though we were able to recruit 85% of eligible participants for the study, the final sample was truncated owing to incomplete parent and/or child questionnaires. Although the use of published normative group data did not allow for precise demographic matching, the published data represented a large, heterogeneous population and provided reliable reference data for our study sample. Our study sample had limited ethnic diversity, being primarily Caucasian, in agreement with previous reports indicating that CVS may be more common in Caucasians.3,40,41 Furthermore, the participating families seen in a clinic specific for CVS may be more severely affected than the typical pediatric patients with CVS, leading them to seek out subspecialty care. Given the lack of studies of the severity of CVS episodes in children from a general pediatric practice compared with those from a subspecialty practice, we cannot judge the representativeness of our sample in terms of the larger affected population. Finally, owing to the study’s cross-sectional design, we are not able to
ORIGINAL ARTICLES comment on the impact of anxiety symptoms on HRQoL over time. Research on functional abdominal pain associated with anxiety has shown differences in the trajectory of the course of the illness,7 and thus a longitudinal evaluation of our CVS population could provide insight into how the treatment of anxiety may affect HRQoL over time. In summary, our study identifies anxiety symptoms as the strongest predictor of HRQoL in children and adolescents with CVS by both self-report and parent-proxy report. Because lower ratings of HRQoL have been linked to increased pediatric health care utilization and costs,42 assessment of HRQoL can be a tool for both monitoring and improving patient care. Most importantly, these data point toward potentially important clinical impacts—first that a biopsychosocial approach, including assessment of anxiety, needs to be part of the standard of care of children with CVS, and second, that identification and successful treatment of anxiety comorbidity could play a key role in improving clinical outcomes and reducing the substantial impact of this functional gastrointestinal disorder on the child, the family, and the health care system. There are several brief anxiety screeners that can be completed by parents and children and scored in a few minutes (http://psychiatry.pitt. edu/research/tools-research/assessment-instruments; http:// www.childfirst.ucla.edu/Resources.html; www.scaswebsite. com).13,43-45 The fact that these scales are available online at no cost and can be administered by clinic personnel with oversight by a mental health professional supports the feasibility of such screening. n Submitted for publication Feb 27, 2015; last revision received Apr 15, 2015; accepted May 19, 2015. Reprint requests: Sally E. Tarbell, PhD, Department of Psychiatry and Behavioral Sciences, Children’s Hospital Colorado, 13123 East 16th Ave, B130, Aurora, CO 80045. E-mail: [email protected]
References 1. Li BUK, Balint JP. Cyclic vomiting syndrome: evolution of our understanding of a brain-gut disorder. Adv Pediatr 2000;47:117-60. 2. Li BUK, Lefevre F, Chelimsky G, Boles RG, Nelson SP, Lewis DW, et al. North American Society for Pediatric Gastroenterology, Hepatology and Nutrition consensus statement on the diagnosis and management of cyclic vomiting syndrome. J Pediatr Gastroenterol Nutr 2008;47: 379-93. 3. Tarbell S, Li BUK. Health-related quality of life in children and adolescents with cyclic vomiting syndrome: a comparison with published data on youth with irritable bowel syndrome and organic gastrointestinal disorders. J Pediatr 2013;163:493-7. 4. Youssef N, Murphy T, Langseder A, Rosh J. Quality of life for children with functional abdominal pain: a comparison study of patients’ and parents’ perspectives. Pediatrics 2006;117:54-9. 5. Tarbell S, Li BUK. Psychiatric symptoms in children and adolescents with cyclic vomiting syndrome and their parents. Headache 2008;48: 259-66. 6. Forbes D, Withers G, Silburn S, McKelvey R. Psychological and social characteristics and precipitants of vomiting in children with cyclic vomiting syndrome. Dig Dis Sci 1999;44:19S-22S. 7. Shelby G, Shirkey K, Sherman A, Beck J, Haman K, Shears A, et al. Functional abdominal pain in childhood and long-term vulnerability to anxiety disorders. Pediatrics 2013;132:475-82.
Anxiety Measures Predict Health-Related Quality of Life in Children and Adolescents with Cyclic Vomiting Syndrome
637
THE JOURNAL OF PEDIATRICS
www.jpeds.com
8. Yacob D, DiLorenzo C, Bridge J, Rosenstein P, Onorato M, Bravender T, et al. Prevalence of pain-predominant functional gastrointestinal disorders and somatic symptoms in patients with anxiety or depression disorders. J Pediatr 2013;163:767-70. 9. Dufton L, Dunn M, Compas B. Anxiety and somatic complaints in children with recurrent abdominal pain and anxiety disorders. J Pediatr Psychol 2009;34:176-86. 10. Tarbell S, Shaltout H, Wagoner A, Diz D, Fortunato J. Relationship among nausea, anxiety and orthostatic symptoms in pediatric patients with chronic unexplained nausea. Exp Brain Res 2014;232: 2645-50. 11. Mahrer N, Monta~ no Z, Gold J. Relations between anxiety sensitivity, somatization and health-related quality of life in children with chronic pain. J Pediatr Psychol 2012;37:806-16. 12. Stevanovic D, Jancic J, Lakic A. The impact of depression and anxiety symptoms on the health-related quality of life of children and adolescents with epilepsy. Epilepsia 2011;52:e75-8. 13. Birmaher B, Brent D, Chiappetta L, Bridge J, Monga S, Baugher M. Psychometric properties of the Screen for Child Anxiety Related Emotional Disorders (SCARED): a replication study. J Am Acad Child Adolesc Psychiatry 1999;38:1230-6. 14. Varni JW, Seid M, Knight T, Uzark K, Szer I. The PedsQL 4.0 Generic Core Scales: sensitivity, responsiveness, and impact on clinical decision-making. J Behav Med 2002;25:175-93. 15. Varni J, Limbers C, Burwinkle T. How young can children reliably and validly self-report their health-related quality of life?: an analysis of 8,591 children across age subgroups with the PedsQL 4.0 Generic Core Scales. Health Qual Life Outcomes 2007;5:1-13. 16. Varni J, Limbers C, Burwinkle T. Parent proxy-report of their children’s health-related quality of life: an analysis of 13,878 parents’ reliability and validity across ages subgroups using the PedsQL 4.0 Generic Core Scales. Health Qual Life Outcomes 2007;5:1-10. 17. Varni J, Burwinkle T, Rapoli M, Seid M, Skarr D. The PedsQL 4.0 as a pediatric population health measure: feasibility, reliability, and validity. Ambul Pediatr 2003;3:329-41. 18. Campo J, Bridge J, Ehmann M. Recurrent abdominal pain, anxiety and depression in primary care. JAMA Pediatr 2004;113:817-24. 19. Barlow D. Anxiety and its disorders: The nature and treatment of anxiety and panic. New York: Guilford Press; 2004. 20. Cohen H, Benjamin J. Power spectrum analysis and cardiovascular morbidity in anxiety disorders. Auton Neurosci 2006;128:1-8. 21. Li BUK, Murray RD, Heitlinger LA, Robbins JL, Hayes JR. Is cyclic vomiting syndrome related to migraine? J Pediatr 1999;134:567-72. 22. Stickler GB. Relationship between cyclic vomiting syndrome and migraine. Clin Pediatr 2005;44:505-8. 23. Hamelsky S, Lipton R. Psychiatric comorbidity of migraine. Headache 2006;46:1327-33. 24. Merikangas KR, Stevens DE. Comorbidity of migraine and psychiatric disorders. Neurol Clin 1997;15:115-23. 25. Boles RG, Adams K, Li BUK. Maternal inheritance in cyclic vomiting syndrome. Am J Med Genet A 2005;133:71-7. 26. Boles RG, Powers ALR, Adams K. Cyclic vomiting syndrome plus. J Child Neurol 2006;21:182-8.
638
Vol. 167, No. 3 27. To J, Issenman RM, Kamath MV. Evaluation of neurocardiac signals in pediatric patients with cyclic vomiting syndrome through power spectral analysis of heart rate variability. J Pediatr 1999;135:363-6. 28. Chelimsky TC, Chelimsky GC. Autonomic abnormalities in cyclic vomiting syndrome. J Pediatr Gastroenterol Nutr 2007;44:326-30. 29. Sowder E, Gevirtz R, Shapiro W, Ebert C. Restoration of vagal tone: a possible mechanism for functional abdominal pain. Appl Psychophysiol Biofeedback 2010;35:199-206. 30. Chelimsky G, Boyle J, Tusing L, Chelimsky T. Autonomic abnormalities in children with functional abdominal pain: coincidence or etiology? J Pediatr Gastroenterol Nutr 2001;33:47-53. 31. Boyce W, Quas J, Alkon A, Smider N, Essex M, Kupfer D. Autonomic reactivity and psychopathology in middle childhood. Br J Psychiatry 2001;179:144-50. 32. Muris P, Rapee R, Meesters C, Schouten E, Geers M. Threat perception abnormalities in children: the role of anxiety disorder symptoms, chronic anxiety, and state anxiety. J Anxiety Disord 2003;17:271-87. 33. Boyer MC, Compas BE, Stanger C, Colletti RB, Konik BS, Morrow SB, et al. Attentional biases to pain and social threat in children with recurrent abdominal pain. J Pediatr Psychol 2006;31:209-20. 34. Eldar S, Apter A, Lotan D, Edgar K, Naim R, Fox N, et al. Attention bias modification treatment for pediatric anxiety disorders: a randomized controlled trial. Am J Psychiatry 2012;169:213-30. 35. Hakamata Y, Lissek S, Bar-Haim Y, Britton J, Fox N, Leibenluft E, et al. Attention bias modification treatment: a meta-analysis towards the establishment of a novel treatment for anxiety. Biol Psychiatry 2010; 68:982-90. 36. Silverman W, Pina A, Viswesvaran C. Evidence-based psychosocial treatments for phobic and anxiety disorders in children and adolescents. J Clin Child Adolesc Psychol 2008;37:105-30. 37. Brent M, Lobato D, LeLeiko N. Psychological treatments for pediatric functional gastrointestional disorders. J Pediatr Gastroenterol Nutr 2008;48:13-21. 38. Fisher E, Heathcote L, Palermo T, de C Williams AC, Lau J, Eccleston C. Systematic review and meta-analysis: psychological therapies for children with chronic pain. J Pediatr Psychol 2014;39:866-86. 39. Slutsker B, Konichezky A, Gothelf D. Breaking the cycle: cognitive behavioral therapy and biofeedback training in a case of cyclic vomiting syndrome. Psychol Health Med 2010;15:625-31. 40. Li BUK, Misiewicz L. Cyclic vomiting syndrome: a brain-gut disorder. Gastroenterol Clin North Am 2003;32:997-1019. 41. Pareek N, Fleisher DR, Abell TA. Cyclic vomiting syndrome: what a gastroenterologist needs to know. Am J Gastroenterol 2007;102:2832-40. 42. Seid M, Varni J, Segall D, Kurtin P. Health-related quality of life as a predictor of pediatric healthcare costs: a two-year prospective cohort analysis. Health Qual Life Outcomes 2004;2:1-10. 43. Chorpita B, Moffitt C, Gray J. Psychometric properties of the Revised Child Anxiety and Depression Scale in a clinical sample. Behav Res Ther 2005;43:309-22. 44. Spence S. Structure of anxiety symptoms among children: a confirmatory factor-analysis. J Abnorm Psychol 1997;1997:280-97. 45. Spence S, Barrett P, Turner C. Psychometric properties of the Spence Children’s Anxiety Scale. J Anxiety Disord 2003;17:605-25.
Tarbell and Li
ORIGINAL ARTICLES
September 2015
Figure. Subject enrollment.
Table I. Patient characteristics Characteristics
Value
Age, y, mean (SD) Sex, n (%) Male Female Ethnic minority, n (%) Annual number of CVS attacks, mean (SD) Duration of CVS episodes, d, median; mean (SD) Annual emergency department visits for CVS, range; mean (SD) Chronicity (onset of symptoms to visit), y, mean (SD) Delay in diagnosis (symptom onset to diagnosis of CVS), y, mean (SD) School days missed per year, range; mean (SD)
12.5 (2.7) 20 (50) 20 (50) 2 (5) 8.2 (15.3) 2; 3 (2.4) 0-12; 1.3 (2.5) 5.8 (3.4) 2.4 (1.9) 0-60; 12.3 (14.9)
Anxiety Measures Predict Health-Related Quality of Life in Children and Adolescents with Cyclic Vomiting Syndrome
638.e1