SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgments
  9. REFERENCES

Objective

The PedsQL rheumatology module is currently the only available measure of disease-specific quality of life for children and adolescents with juvenile fibromyalgia (FM), but limited information has been published about the psychometric properties of the instrument, specifically in juvenile FM. The objective of this study was to assess the reliability, validity, and sensitivity to change of the 5 scales (pain and hurt, daily activities, treatment, worry, and communication) of the patient and parent proxy versions of the PedsQL rheumatology module in the context of a randomized controlled trial in juvenile FM.

Methods

The entire PedsQL rheumatology module was administered as a supplementary outcome measure at baseline, posttreatment, and 6-month followup assessments of 114 children and adolescents with juvenile FM enrolled in a trial testing the efficacy of cognitive–behavioral therapy.

Results

Internal consistency reliabilities for the scales were adequate to strong (Cronbach's α = 0.68–0.86). Parent proxy and child reports on most scales (except for daily activities and communication) showed moderate correlations (Spearman's r = 0.33–0.45). Support for construct validity was found by comparing child and parent reports with other related measures of pain and functioning (visual analog scale pain ratings and the Functional Disability Inventory). Finally, sensitivity to change was demonstrated by significant changes in 4 of the 5 scales (excluding the daily activities scale) after treatment.

Conclusion

The PedsQL rheumatology module generally appears to have good utility for use in juvenile FM patients, but there are some caveats to the interpretation of specific scales in this population.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgments
  9. REFERENCES

The PedsQL rheumatology module is a disease-specific quality of life (QOL) measure that has been validated for use in children and adolescents with a variety of rheumatologic conditions, including juvenile idiopathic arthritis (JIA), systemic lupus erythematosus (SLE), juvenile dermatomyositis (DM), and others ([1]). A small number of patients with juvenile fibromyalgia (FM; n = 35, 12.9% of the sample) were also included in the original validation sample for the PedsQL rheumatology module. Juvenile FM, a noninflammatory pain condition, varies from other rheumatic diseases such as JIA in many ways (i.e., diffuse, widespread, and ongoing pain in juvenile FM compared with localized joint pain associated with inactivity stiffness in JIA). Juvenile FM also differs from other rheumatic diseases such as SLE and juvenile DM, which are multisystem diseases that require frequent laboratory testing and management with multiple medications. Given the differences in disease processes and treatment regimens between juvenile FM and other rheumatic diseases, it might be expected that the effects of symptoms on QOL also vary. Furthermore, juvenile FM is not objectively apparent compared with most rheumatic diseases and this may present some unique challenges in the communication between patients, caregivers, and health care providers about the illness. It should be noted that juvenile FM can coexist with other rheumatic diseases, particularly when they are longstanding, but as an initial step, the goal of this investigation was to understand the impact of primary juvenile FM without a known underlying inflammatory component.

The initial validation study of the PedsQL rheumatology module indicated that juvenile FM patients formed a distinct subgroup that scored lower on all aspects of QOL compared with other rheumatic diseases. The study also found that the measure discriminated between healthy children and those with rheumatic diseases ([1]) and demonstrated acceptable reliability (α = 0.75–0.86 for child self-report and α = 0.82–0.91 for parent proxy report). Of the 5 scales, the pain and hurt scale has received the most attention for juvenile FM patients ([2]), on which juvenile FM patients report the lowest scores (more difficulties) compared with all rheumatic disease groups. Given that the remaining scales in the measure were primarily designed for patients with rheumatic diseases such as JIA, SLE, and juvenile DM, further study of whether the entire measure has clinical and research utility specifically for youths with juvenile FM is necessary.

Our research group recently completed a multisite randomized clinical trial of cognitive–behavioral therapy (CBT) compared with FM education (FE; attention control group) among a large sample of children and adolescents with primary juvenile FM (n = 114), in which we included the PedsQL rheumatology module (all 5 scales) as a supplementary outcome measure. The repeated administrations of the measure in this controlled trial (at baseline, posttreatment, and followup) and the availability of other related measures of pain and functioning by both patient and parent proxy reports provided us with a unique opportunity to more thoroughly evaluate the psychometric properties of the PedsQL rheumatology module for juvenile FM. The results of the trial indicated that CBT was significantly better than FE in reducing functional disability in patients with juvenile FM and that both groups experienced significant reductions in depressive symptoms ([3]).

The present study is a secondary analysis in which we aimed to 1) examine the psychometric properties of the PedsQL rheumatology module in adolescents with juvenile FM, including internal consistency reliability and construct validity (by comparing child and parent proxy reports of QOL with other related measures of pain and functioning), and 2) assess the PedsQL rheumatology module's sensitivity to change in the context of the randomized controlled trial. We hypothesized that the measure would demonstrate adequate reliability and validity. Given the positive results of the completed clinical trial, we expected juvenile FM patients' scores on the PedsQL rheumatology module would significantly improve over time in both treatment groups (from baseline to posttreatment and followup). No hypotheses were made regarding group differences (CBT versus FE) due to the absence of prior research using the PedsQL rheumatology module as an outcome measure in juvenile FM clinical trials.

Box 1. Significance & Innovations

  • This is the first study to assess the utility of the PedsQL rheumatology module for use in youths with juvenile fibromyalgia (FM).
  • Specific recommendations for the interpretation of patient and parent proxy reports on the 5 scales of the PedsQL rheumatology module in youths with juvenile FM are presented.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgments
  9. REFERENCES

Participants

The participants were 114 children and adolescents (ages 11–18 years) with primary juvenile FM and their parents who participated in a clinical trial on CBT for juvenile FM ([3]). The participants were recruited across 4 pediatric rheumatology centers in the Midwestern US. All patients met the classification criteria for juvenile FM ([4]), were stable on medications, and reported an average pain severity of ≥4 on a 0–10-cm visual analog scale (VAS). Adolescents were excluded if they had any other rheumatic disease, developmental delay, or current major psychiatric illness. Prior to study participation, written informed consent was obtained from the parents and written assent was obtained from the adolescents. The research protocol was approved by the institutional review board at each site.

Measures

The measures included in the analysis of the present study were a subset of the measures administered as part of the comprehensive assessments in the trial and were completed at baseline, posttreatment (week 9), and followup (6 months).

Disease-specific QOL

Disease-specific QOL was assessed using the PedsQL rheumatology module, a 22-item measure consisting of 5 scales (pain and hurt, daily activities, treatment, worry, and communication). The PedsQL includes both child and adolescent report forms (for ages 8–12 years and 13–18 years, respectively) and parent proxy report forms (completed by a parent or caregiver) ([1, 2]). The pain and hurt scale assesses pain, stiffness, and disrupted sleep. The daily activities scale assesses difficulty with tasks such as turning handles and carrying school books. The treatment scale examines the physical and emotional impact of treatment. The worry scale assesses worry related to medical treatment and the illness. The communication scale evaluates how hard it is to ask medical staff questions and discuss the illness with others. The items are rated on a 5-point Likert scale, where 0 = never, 1 = almost never, 2 = sometimes, 3 = often, and 4 = almost always. The possible range of scores on each scale is 0–100, with higher scores showing better QOL. In the initial validation study, scores for children with rheumatic diseases (including JIA, SLE, and juvenile DM) ranged from 69.55–88.33 for the 5 domains and from 33.64–67.16 for children with juvenile FM ([1]).

Pain intensity

Pain intensity was assessed using a 0–10-cm VAS (with the anchors 0 = no pain and 10 = worst pain possible) of average pain, completed each day for 1 week ([5, 6]). The average pain score was the mean of the 7 daily ratings. Parents also rated their child's pain in the past week using a single 0–10 VAS rating. In this study, the pain VAS was included for validation purposes and average pain intensity was expected to correlate most strongly with the pain and hurt scale of the PedsQL.

Functional disability

Functional disability was assessed using the Functional Disability Inventory (FDI) ([7]), a well-validated measure recommended for use in pediatric chronic pain ([8]). The FDI assesses the patient's ability to function in daily physical, social, and recreational activities. Both child and parent report versions were used and were expected to correlate most strongly with the daily activities scale.

Procedures

Patients were randomly assigned to either a CBT or FE group. Both groups attended individual sessions for 8 weeks, followed by 2 additional booster sessions. The CBT group focused on a variety of aspects of behavioral pain management and coping skills training, whereas the FE group served as an attention control condition and included support and education without any instruction for behavior change. The PedsQL rheumatology module was administered along with the other measures at baseline, posttreatment, and followup. Additional details regarding the study procedure and the main outcomes of the trial have been previously published ([3]).

Statistical analyses

All data were analyzed using SPSS, version 20 software. Descriptive statistics were first computed for the PedsQL rheumatology module scales at baseline. Cronbach's alpha was then computed for each scale to determine internal consistency reliability. Cronbach's α ≥0.70 was used as the benchmark for acceptability ([9]). Spearman's rank correlations between child and parent proxy report scores were computed to examine interrater reliability. In order to examine the validity of the measure, correlation coefficients were also calculated for child and parent proxy report versions of the PedsQL rheumatology module scales with the FDI and VAS pain intensity scores. The magnitude of these coefficients was discussed as follows: ∼0.10 = small, ∼0.30 = medium, and ∼0.50 = large ([10]). Finally, a mixed between-/within-subject repeated-measures analysis of variance (ANOVA) was conducted to examine the sensitivity of the PedsQL rheumatology module scales (child report only) by assessing the change in scores across 3 time periods (baseline, posttreatment, and followup) for the CBT and FE study arms. Finally, the standardized response mean was used to examine the magnitude of change scores and assess whether changes were maintained over time.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgments
  9. REFERENCES

Although the Consolidated Standards of Reporting Trials flow chart and sample characteristics were previously published ([3]), they are briefly described here to provide a context for the interpretation of the current findings. A total of 114 children and adolescents diagnosed with juvenile FM were randomized to either a CBT group (n = 57) or an FE group (n = 57). Two patients withdrew from the study before attending any sessions and 11 dropped out during treatment. This resulted in a sample of 105 participants (52 CBT and 53 FE) at posttreatment and 101 participants (51 CBT and 50 FE) at followup. Participants ranged from ages 11–18 years (mean ± SD 15.02 ± 1.75 years). The sample was 93% female and 90% white. There were no significant demographic or baseline differences between the participants in the 2 groups.

Descriptive analyses of the PedsQL rheumatology module items and scales

Table 1 shows the item-level data for the PedsQL rheumatology module at each assessment time point. In general, all items were endorsed by at least some patients, indicating that most of the items were perceived by the participants as applicable. Items on the pain and hurt scale were the most frequently endorsed, and items that were the least likely to be reported were related to daily activities involving fine motor tasks (eating with utensils and turning door handles and faucets). On the treatment scale, relatively low scores were obtained for medications making the participants feel sick or being scared to go to the doctor. On the communication scale, finding it hard to explain their illness to other people was the most problematic for the participants.

Table 1. PedsQL rheumatology module individual-item mean ± SD for patient self-report*
GroupBaselinePosttreatmentFollowup
FE (n = 57)CBT (n = 57)FE (n = 53)CBT (n = 52)FE (n = 50)CBT (n = 51)
  1. Item responses ranged from 0–4, where 0 = never, 1 = almost never, 2 = sometimes, 3 = often, and 4 = almost always. FE = fibromyalgia education; CBT = cognitive–behavioral therapy.

Pain and hurt      
1. I ache or hurt in my joints and/or muscles3.38 ± 0.683.51 ± 0.633.06 ± 0.823.10 ± 0.762.78 ± 0.932.65 ± 0.89
2. I hurt a lot3.15 ± 0.763.26 ± 0.772.78 ± 0.952.69 ± 0.882.34 ± 1.082.24 ± 1.01
3. I have trouble sleeping because of pain or aching in my joints and/or muscles2.55 ± 1.052.65 ± 0.972.42 ± 1.032.24 ± 0.971.92 ± 0.991.96 ± 1.10
4. I feel stiff in the morning or when I sit too long2.82 ± 1.122.86 ± 1.012.5 ± 1.062.96 ± 1.002.60 ± 1.052.39 ± 1.02
Daily activities      
1. It is hard to turn on water faucets0.35 ± 0.620.49 ± 0.740.38 ± 0.570.39 ± 0.640.28 ± 0.570.22 ± 0.54
2. It is hard to turn door handles0.47 ± 0.660.60 ± 0.750.48 ± 0.680.37 ± 0.660.38 ± 0.670.24 ± 0.56
3. I have trouble eating with a fork and knife0.38 ± 0.710.51 ± 0.780.38 ± 0.830.27 ± 0.700.28 ± 0.610.31 ± 0.62
4. It is hard to write or draw with a pen and pencil0.96 ± 1.151.02 ± 1.041.14 ± 1.230.80 ± 1.020.96 ± 1.280.75 ± 1.04
5. I have trouble carrying my school books1.82 ± 1.171.77 ± 1.271.74 ± 1.341.53 ± 1.171.48 ± 1.251.29 ± 1.29
Treatment      
1. My medicines make me feel sick0.70 ± 0.920.93 ± 1.100.66 ± 0.880.77 ± 0.980.76 ± 0.900.67 ± 0.89
2. My physical therapy or daily exercise hurts1.93 ± 1.142.02 ± 1.231.74 ± 1.181.48 ± 1.001.40 ± 1.091.18 ± 0.97
3. It is hard to be responsible for my medicines or physical therapy1.28 ± 1.091.09 ± 0.991.13 ± 1.210.94 ± 0.971.10 ± 1.230.82 ± 0.93
4. It is hard to manage my illness1.87 ± 1.022.18 ± 0.941.60 ± 1.061.40 ± 0.911.38 ± 0.951.31 ± 0.97
5. I get scared when I have to have blood tests1.40 ± 1.571.33 ± 1.561.09 ± 1.421.23 ± 1.460.90 ± 1.401.12 ± 1.44
6. I get scared about having needle sticks/shots1.45 ± 1.541.47 ± 1.541.25 ± 1.521.29 ± 1.491.00 ± 1.431.16 ± 1.43
7. I get scared when I have to go to the doctor0.69 ± 1.030.88 ± 0.970.60 ± 0.860.69 ± 0.920.44 ± 0.730.57 ± 0.86
Worry      
1. I worry about the side effects from medicines1.09 ± 0.971.09 ± 1.071.25 ± 1.071.10 ± 1.210.82 ± 0.921.08 ± 1.15
2. I worry about whether or not my medicines are working1.18 ± 1.001.32 ± 1.201.34 ± 1.141.15 ± 1.161.02 ± 0.981.02 ± 1.91
3. I worry about my illness1.78 ± 1.102.07 ± 1.101.74 ± 1.231.50 ± 1.001.36 ± 1.031.31 ± 1.10
Communication      
1. It is hard for me to tell the doctors and nurses how I feel1.56 ± 1.141.72 ± 1.181.30 ± 1.121.54 ± 1.091.28 ± 1.111.25 ± 1.00
2. It is hard for me to ask the doctors and nurses questions1.20 ± 1.011.39 ± 1.611.02 ± 1.081.17 ± 1.801.00 ± 1.071.08 ± 1.00
3. It is hard for me to explain my illness to other people2.58 ± 1.302.42 ± 1.122.49 ± 1.152.06 ± 1.271.86 ± 1.312.10 ± 1.27

Table 2 shows the mean scores for each PedsQL rheumatology module scale as reported by patients and parents. In general, patient and parent proxy scores were similar, with the exception of the communication scale, in which patients rated themselves as having more trouble communicating about their illness than their parents had rated them. Both patient and parent proxy scores indicated that items addressed by the daily activities scale were the least problematic and that the pain and hurt items were the most problematic.

Table 2. PedsQL rheumatology module scales, FDI, and VAS scores for patient self-report and parent proxy report at baseline (n = 114)*
MeasureSelf-reportProxy report
  1. Values are the mean ± SD. FDI = Functional Disability Inventory; VAS = visual analog scale.

  2. a

    Lower scores on all PedsQL scales indicate more pain or lower functioning.

  3. b

    Higher scores on the VAS and FDI indicate more pain or impairment.

Pain and hurt (range 0–100)a24.40 ± 15.4422.46 ± 17.57
Daily activities (range 0–100)a79.39 ± 17.1281.06 ± 15.62
Treatment (range 0–100)a65.34 ± 18.4965.00 ± 18.38
Worry (range 0–100)a63.96 ± 22.5267.77 ± 21.67
Communication (range 0–100)a54.46 ± 23.8963.13 ± 24.75
FDI (range 0–60)b20.26 ± 8.7820.18 ± 9.59
VAS (range 0–10)b5.74 ± 1.366.06 ± 1.76

Reliability

Cronbach's alpha estimates showed that internal consistency reliability for the PedsQL rheumatology module scales for both child/adolescent and parent proxy versions was in the acceptable range (α ≥ 0.70) ([11]) for communication (child report α = 0.86, parent report α = 0.86), treatment (child report α = 0.73, parent report α = 0.73), worry (child report α = 0.76, parent report α = 0.78), daily activities (child report α = 0.81, parent report α = 0.79), and parent-reported pain and hurt (α = 0.83). Only the child-reported pain and hurt scale α was below 0.70 (α = 0.68), but was acceptable (α increased to 0.72) if item 4, feeling stiff in the morning, was excluded.

Moderate correlations were seen between child/ adolescent and parent proxy versions of the PedsQL rheumatology module on the pain and hurt (r = 0.40, P < 0.01), treatment (r = 0.45, P < 0.01), and worry (r = 0.33, P < 0.01) scales. The correlation between child and parent reports on the communication scale (r = 0.05, P = 0.59) and daily activities scale (r = 0.25, P < 0.01) was small. This suggested lower interrater reliability for the communication and daily activities scales, particularly the communication scale.

Validity

The relationship between the FDI scores and daily activities scale scores showed moderate negative correlations for both child/adolescent report (r = −0.44, P < 0.01) and parent proxy report (r = −0.42, P < 0.01), which was expected based on the scoring methods of the 2 scales (i.e., higher FDI scores indicate more impairment and lower daily activities scores indicate more impairment) (Table 3). Correlations between the VAS pain scores and the pain and hurt scale showed a large correlation for the parent proxy report (r = −0.52, P < 0.01; higher VAS and lower pain and hurt scores correspond to more pain) but showed a small negative correlation for the child/adolescent report (r = −0.28, P < 0.05). Generally, this pattern of correlations provided support for the convergent validity of the daily activities scale and pain and hurt scale. However, the magnitude of the child-reported correlations between the pain and hurt scale and the VAS was quite low.

Table 3. Correlations between parent proxy and patient self-report of the PedsQL rheumatology module subscale scores, functional disability, and pain intensity*
ScalePPHPDAPTPWPCPFDIPVASCPHCDACTCWCCCFDI
  1. PPH = pain and hurt parent proxy report; PDA = daily activities parent proxy report; PT = treatment parent proxy report; PW = worry parent proxy report; PC = communication parent proxy report; PFDI = Functional Disability Inventory parent proxy report; PVAS = visual analog scale parent proxy report; CPH = pain and hurt child/adolescent report; CDA = daily activities child/adolescent report; CT = treatment child/adolescent report; CW = worry child/adolescent report; CC = communication child/adolescent report; CFDI = Functional Disability Index child/adolescent report; CVAS = visual analog scale child/adolescent report.

  2. a

    P < 0.01.

  3. b

    P < 0.05.

  4. c

    Higher scores are indicative of poorer functioning or more pain.

Parent report             
PedsQL rheumatology module (range 0–100)             
PPH             
PDA0.495a            
PT0.287a0.298a           
PW0.398a0.427a0.525a          
PC0.0500.212b0.297a0.205b         
PFDI (range 0–60)c−0.494a−0.421a−0.391a−0.289a−0.183        
PVAS (range 0–10)c−0.522a−0.408a−0.353a−0.289a−0.1180.556a       
Child report             
PedsQL rheumatology module (range 0–100)             
CPH0.396a0.246a0.1440.242a−0.091−0.245a−0.309a      
CDA0.1230.247a0.0200.0110.052−0.136−0.1590.204b     
CT0.0040.0390.449a0.226b0.041−0.054−0.220b0.1160.190b    
CW0.0580.0260.1750.325a0.147−0.102−0.1110.1370.276a0.382a   
CC−0.134−0.1080.1100.0070.0520.003−0.0270.0870.1550.373a0.326a  
CFDI (range 0–60)c−0.337a−0.332a−0.151−0.227a0.0080.467a0.376a−0.414a−0.439a−0.182−0.338a−0.159 
CVAS (range 0–10)c−0.231b−0.150−0.147−0.160−0.1180.287a0.375a−0.277a−0.048−0.124−0.1060.0950.206b

When examining relationships of the child/adolescent report on other scales of the PedsQL rheumatology module with the FDI and VAS, the results showed that the FDI was moderately negatively correlated with only the pain and hurt (r = −0.41, P < 0.01) and worry (r = −0.34, P < 0.01) scales. On the basis of the scoring system for the PedsQL rheumatology module scales, this would mean greater impairment on the FDI was associated with more difficulties with pain and greater worry. The VAS showed only small correlations with all of the PedsQL rheumatology module scales other than the pain and hurt scale. These results provide support for the discriminant validity of the measure.

Sensitivity

The results of the mixed between/within-subject ANOVAs showed that there was a significant main effect of time for the communication, treatment, pain and hurt, and worry scales, with both the CBT and FE groups improving over the course of the study (Figure 1). No significant interaction effects that might suggest differences in outcome by treatment type (CBT versus FE) were found. For the daily activities scale, the results showed no main effects for time or any significant interaction effects between group and time (Table 4). For the total sample, the standardized response means from baseline to posttreatment ranged from 0.10–0.40 and from posttreatment to followup ranged from 0.15–0.30. These effect size ranges were consistent with the analysis results that showed the treatment effect was well maintained at followup.

image

Figure 1. Mean scale scores on the PedsQL rheumatology module scales for all juvenile fibromyalgia participants. Higher scores indicate improvement. Significant improvement was observed in all domains other than daily activities. * = P < 0.005; + = P < 0.05.

Download figure to PowerPoint

Table 4. Results of ANOVA comparing PedsQL rheumatology module scales at baseline, posttreatment, and followup (time) for the CBT and FE groups (group)*
 Sum of squaresdfMean squareFP
  1. ANOVA = analysis of variance; CBT = cognitive–behavioral therapy; FE = fibromyalgia education.

  2. a

    P ≤ 0.001.

Communication     
Time5,447.691.862,936.6810.190.000a
Group144.681144.680.120.731
Time × group214.351.86115.550.400.655
Treatment     
Time3,563.061.861,920.0317.900.000a
Group48.86148.860.080.784
Time × group235.871.86127.101.190.306
Pain and hurt     
Time12,462.861.796,945.5619.200.000a
Group12.25112.250.030.854
Time × group331.791.79184.900.510.581
Daily activities     
Time794.431.59501.331.870.165
Group99.16199.160.230.630
Time × group302.071.59190.620.710.461
Worry     
Time2,568.521.991,291.277.560.001a
Group8.3318.330.010.934
Time × group716.671.99360.292.110.124

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgments
  9. REFERENCES

This study provided a thorough analysis of the PedsQL rheumatology module and its component scales for use in the assessment of disease-specific QOL in children and adolescents with juvenile FM. Overall, the instrument showed strong reliability and adequate construct validity. The measure also showed sensitivity to change in the context of a clinical trial. A review of the descriptive data at the item level showed that all items on the measure were endorsed by at least some patients, with the most frequently endorsed items being those related to pain symptoms and sleep difficulties and the least frequently endorsed items related to difficulties with daily activities requiring fine motor coordination activities (such as turning water faucets or door handles). This was not unexpected because juvenile FM patients mostly report widespread pain, fatigue, and sleep problems along with more global impairments in daily function (attending school, participating in sports, etc.). These patients do not typically have difficulties performing fine motor activities, which are more often reported by patients with inflammatory rheumatic diseases.

The internal consistency reliability of the 5 scales was generally strong for both the patient and parent proxy reports, except for the child/adolescent version of the pain and hurt scale. The lower consistency on this scale was driven by a single item on stiffness, which may be less of a concern in juvenile FM (although anecdotally we have found that several juvenile FM patients do experience stiffness, so deletion of the item from the measure is not warranted). Parent and child/adolescent reports for the PedsQL rheumatology module pain and hurt, worry, and treatment scales were moderately correlated, which indicates that the majority of the scales have acceptable (although modest) interrater reliability. However, correlations between the parent and patient reports for the communication and daily activities scales were small, indicating lower congruence. The communication scale in particular showed small correlation effects between reporters. This could be because items on this scale are more focused on reports of internal states (i.e., how well the child can communicate about their illness with others). Clearly, parents and children did not agree in their perspectives, with parents reporting higher communication scores than their children. This discrepancy in parent and child reports of less observable internal states has previously been noted ([12, 13]). Given the relatively moderate interrater agreement overall and the poor relationship between parent and patient reports on the communication scale, it may be prudent to administer both parent and child versions of the measure rather than relying solely on either the patient (child-reported) or parent version.

With regard to validity of the PedsQL rheumatology module, we found that the FDI was moderately correlated with the daily activities scale for both the child/adolescent and parent proxy reports. The moderate size of the correlations indicated that the PedsQL scales were measuring separate but related constructs, particularly because the FDI tends to capture gross motor activities and the daily activities scale tends to focus on fine motor movements. The VAS pain ratings showed large correlations with the pain and hurt scale for the parent report but smaller relationships with the child/adolescent report. Overall, the results were in the expected direction and showed support for construct validity.

With respect to the sensitivity of the PedsQL rheumatology module scales to treatment effects, we found that 4 of the 5 scales did show sensitivity to change and were equally responsive to both FE and CBT, which could be attributed to the effects of support and reassurance from health care providers and increased knowledge about juvenile FM in both groups. The daily activities scale did not change significantly in either group over the course of treatment and followup. The lack of significant improvement in daily activities scores on the PedsQL rheumatology module is in contrast to findings from our clinical trial that demonstrated significant reductions in physical impairment (measured by the FDI) in both groups, with the CBT group showing markedly greater improvement. The reason for the lack of sensitivity of the daily activities scale to treatment effects is likely due to the nature of the items, which focus mainly on activities requiring fine motor dexterity (turning knobs and using utensils), and these activities are not typically impacted in individuals with juvenile FM or specifically targeted in CBT.

When comparing the mean scores of the juvenile FM patients in the current study to the small group of juvenile FM patients in the initial validation study of the PedsQL rheumatology module ([1]), we found that the scores were similar for the daily activities, treatment, and worry scales. The juvenile FM sample in the current study scored lower on the pain and hurt scale (24.40 versus 33.64 in the prior study) and the communication scale (54.46 versus 67.16 in the prior study). This difference could be explained by the current sample potentially having more severely affected patients due to the eligibility criteria for enrolling in the trial. The juvenile FM patients (regardless of the sample) reported markedly lower QOL on all PedsQL rheumatology module scales compared to children with JIA, SLE, or juvenile DM in the initial validation study. The reason for this difference in the impact of symptoms on disease-specific QOL is currently unknown. One potential link may be found in the higher levels of emotional distress present in children with chronic pain in general ([14-16]) and in patients with juvenile FM in particular ([17]). The lack of a clear medical etiology and treatments for juvenile FM may also contribute to greater uncertainty, worry, and difficulties in communication surrounding a juvenile FM diagnosis.

In summary, the PedsQL rheumatology module demonstrated good clinical utility for use in juvenile FM, but there are a few caveats to keep in mind while interpreting the results. First, the daily activities scale probably does not provide an accurate measure of daily physical functioning. Therefore, use of a separate measure of functional impairment, like the FDI, is recommended if this domain is to be measured. Second, parents and children may have different perspectives on the communication items and therefore it is advisable to use both child and parent proxy reports to capture these discrepancies. In clinical settings, obtaining both perspectives might be very useful in implementing appropriate interventions to address difficulties in communication with health care providers.

The strengths of this study are the relatively large sample of juvenile FM patients recruited from 4 different sites and the clinical trial design that allowed an examination of sensitivity. One limitation was the lack of contemporaneous comparison groups of children with other rheumatic diseases or healthy controls; therefore, generalizability is limited to clinical samples of primary juvenile FM patients ([18]). It would be particularly interesting in future research to examine the utility of the measure in those who have secondary juvenile FM along with a rheumatic disease and changes in QOL as it relates to inflammatory joint activity. Further studies of why disease-specific QOL in juvenile FM appears to be markedly lower than other rheumatic diseases are also needed, because many of these diseases have multisystemic disease burdens and more complex treatment and monitoring regimens. In general, we would recommend that the PedsQL rheumatology module be used as a supplemental measure of disease-related QOL in clinical trials and not as a specific measure for pain and function in juvenile FM, for which more appropriate instruments such as the VAS and the FDI are available. Nevertheless, the instrument does provide unique information about the difficulties that patients experience in communicating about their condition, their worries about their illness, and treatment concerns, which can be useful for health care professionals treating children and adolescents with juvenile FM.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgments
  9. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Joffe had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Lovell, Powers, Kashikar-Zuck.

Acquisition of data. Lynch-Jordan, Ting, Hashkes, Lovell, Passo, Schikler, Kashikar-Zuck.

Analysis and interpretation of data. Joffe, Arnold, Powers, Kashikar-Zuck.

Acknowledgments

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgments
  9. REFERENCES

We would like to thank additional members of the Juvenile Fibromyalgia Study Team who assisted with subject recruitment, medical evaluations, and treatment delivery: Dr. T. Brent Graham (Vanderbilt University School of Medicine) and Drs. Steven Spalding, Margaret Richards, and Gerard Banez (Cleveland Clinic Lerner School of Medicine). We would also like to thank the research coordinators who were responsible for the data collection, data management, and assistance with regulatory aspects of the trial: Sonya Crook, RN (Cleveland Clinic Lerner School of Medicine) and Megan Johnston, BA, Emily Verkamp, MSW, Daniel Strotman, BA, and Kimberly Barnett, BA (Cincinnati Children's Hospital Medical Center).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgments
  9. REFERENCES