Health related quality of life measure in systemic pediatric rheumatic diseases and its translation to different languages: an international collaboration

Abstract

Background

Rheumatic diseases in children are associated with significant morbidity andpoor health-related quality of life (HRQOL). There is no health-relatedquality of life (HRQOL) scale available specifically for children with lesscommon rheumatic diseases. These diseases share several features withsystemic lupus erythematosus (SLE) such as their chronic episodic nature,multi-systemic involvement, and the need for immunosuppressive medications.HRQOL scale developed for pediatric SLE will likely be applicable tochildren with systemic inflammatory diseases.

Findings

We adapted Simple Measure of Impact of Lupus Erythematosus in Youngsters(SMILEY©) to Simple Measure of Impact of Illness in Youngsters(SMILY©-Illness) and had it reviewed by pediatric rheumatologists forits appropriateness and cultural suitability. We tested SMILY©-Illnessin patients with inflammatory rheumatic diseases and then translated it into28 languages.

Nineteen children (79% female, n=15) and 17 parents participated. The meanage was 12±4 years, with median disease duration of 21 months (1-172months). We translated SMILY©-Illness into the following 28 languages:Danish, Dutch, French (France), English (UK), German (Germany), German(Austria), German (Switzerland), Hebrew, Italian, Portuguese (Brazil),Slovene, Spanish (USA and Puerto Rico), Spanish (Spain), Spanish(Argentina), Spanish (Mexico), Spanish (Venezuela), Turkish, Afrikaans,Arabic (Saudi Arabia), Arabic (Egypt), Czech, Greek, Hindi, Hungarian,Japanese, Romanian, Serbian and Xhosa.

Conclusion

SMILY©-Illness is a brief, easy to administer and score HRQOL scale forchildren with systemic rheumatic diseases. It is suitable for use acrossdifferent age groups and literacy levels. SMILY©-Illness with itsavailable translations may be used as useful adjuncts to clinical practiceand research.

Findings

Introduction

In children, chronic rheumatic diseases are associated with significant disease-and treatment-related morbidity, thus impacting their health-related quality oflife (HRQOL). There are generic scales available to assess HRQOL in childrenwith rheumatic diseases such as the Pediatric Quality of Life Inventory(PedsQL)-Rheumatology module [1]. But there is no specific health-related quality of life (HRQOL)scale that addresses the impact of the less common rheumatic diseases such asmixed connective tissue disease (MCTD), juvenile dermatomyositis (JDM), systemicsclerosis (SS), Sjögren’s syndrome, vasculitides, Behçets,sarcoidosis or systemic arthritis (SJIA).

Simple Measure of the Impact of Lupus Erythematosus in Youngsters©" (SMILEY)is valid in US-English [2] and in Portuguese-for Brazil [3]. SMILEY-US English was validated through a multicenter study in theUS [2]. Subjects with SLE completed other gold standards and SLE statusmeasures and psychometric properties were determined [2]. Relationship of HRQOL and changes in disease activity were measuredover time [4]. SMILEY US English was further translated and adapted into severallanguages [2, 5, 6].

Hypothesis

Systemic Lupus Erythematosus (SLE) and systemic inflammatory diseases shareseveral features such as their chronic episodic nature, multi-systemicinvolvement, and the need for immunosuppressive medications. HRQOL scaledeveloped for pediatric SLE will be applicable to children with systemicinflammatory diseases. We decided to adapt a tool that is valid in SLE, titled,SMILEY [2]. We will report the: (i) adaptation of SMILEY© to Simple Measureof Impact of Illness in Youngsters (SMILY©-Illness) for use in childrenwith systemic inflammatory diseases such as MCTD, JDM, SS, Sjögren’ssyndrome, vasculitis and SJIA and preliminary testing in patients; and (ii)translation into different languages. We think this is very important since thesystemic rheumatic diseases mentioned above can lead to significant disabilitywhich impact HRQOL.

Methods used

Overview in brief

(i) We adapted SMILEY to SMILY©- Illness and had it reviewed by pediatric rheumatologists from two centers (RWJMS, HSS) for its appropriateness and cultural suitability, and tested SMILY©- Illness in as small sample. We examined time taken to complete questionnaire, feasibility, and also collected demographic and disease-related data. We subsequently translated it into the following 28 languages using professional translators: Danish, Dutch, French (France), English (UK), German (Germany), German (Austria), German (Switzerland), Hebrew, Italian, Portuguese(Brazil), Slovene, Spanish (USA and Puerto Rico), Spanish (Spain), Spanish (Argentina), Spanish (Mexico), Spanish (Venezuela ), Turkish, Afrikaans, Arabic (Saudi Arabia), Arabic (Egypt), Czech, Greek, Hindi, Hungarian, Japanese, Romanian, Serbian and Xhosa. Each translation was reviewed by the pediatric rheumatologist(s) from that country for its applicability and cultural suitability in order to be approved.

Subjects and settings

Children ≤18 years of age diagnosed with the following systemic chronicrheumatic diseases were included: MCTD, JDM, Sjögren’s syndrome,Systemic sclerosis/CREST, Behçets, sarcoidosis and SJIA. The patients whohad to have been followed for at least one month, and able to participate in thestudy as determined by the pediatric rheumatologist, and their parents (orguardians) were recruited from two US pediatric rheumatologypractices^a. Children were excluded if they were unable tocomplete the questionnaires, or had a significant co-morbid condition likely toimpact HRQOL exclusive of their rheumatic disease (such as an infectious,endocrine, psychiatric, congenital, genetic, neurodegenerative or an oncologicalprocess).

Measures used

The 26-item SMILY©-Illness for children <19 years features parallel childself reports and parent reports with responses in the form of a five-step scalewith different facial expressions with 5^th grade reading level. Thefour domains are similar to that of SMILEY and are: Effect on self (5 items),Limitations (8 items), Social (4 items) and Burden of Illness (7 items). Scoringis also similar to SMILEY, where each item score ranges from 1 to 5 and thetotal score is transformed to a 1 to 100 scale. Higher scores indicate betterHRQOL. If >12 questions are not answered, the SMILY©-Illness cannot bescored. The first two items on current illness status and HRQOL assessment arenot included in the domains or calculating the final score. The remainingquestions refer the respondent to the previous month.

Additional data

We examined self-esteem using the Piers Harris Self concept scale (SCS) [7, 8], entitled, “The Way I Feel About Myself.” Average scoresusually range from 46–60 with higher scores corresponding to betterself-concept. We collected data on demographics, ethnicity, co-morbidity,insurance, education; and impact of disease using the PedsQL-Family informationform. We recorded the date of disease onset, and the current/prior use of allmedication(s). The Hollingshead Socioeconomic scale (SES) score, which takesinto account the educational and occupational status of the family members, wascalculated using the educational and occupational status of the parents [9]. The scores range from 8 to 66, with higher scores indicating ahigher socioeconomic status [9].

Procedure

Appropriate Institutional Review Board approval was obtained at both sites.Potential subjects were identified at each center through the clinic appointmentschedule or during in-patient admissions. Children and parents completedcorresponding versions of the 26-item SMILY©-Illness and the SCS.The investigator was available at all times to respond to queries posed by studyrespondents.

Methods and statistical analysis

Using the SPSS statistical package for Windows (SPSS Inc, Chicago, Illinoisversions 20), we performed descriptive analyses on all variables and examineddata distribution, and examined instrument scores for ceiling and floor effects.Minimal missing data were handled in accordance with rules for scoring eachquestionnaire. Feasibility was determined from the percentage of missing valuesfor each item and the distribution of item responses [10]. Spearman’s rho correlation was used.

Results

Nineteen children (79% female, n=15) and 17 parents (16 mothers) participated in thestudy. The mean age was 12±4 years (3–18 years) with median diseaseduration of 21 months (1–172 months), and mean self -concept of 50±8(36–69). Hollingshead socioeconomic score was 47±11 (26–61).Subjects had the following diagnosis: SJIA (n=5, 26%), dermatomyositis (n=4, 21%),systemic sclerosis/CREST syndrome (n=5, 26%), mixed connective tissue disease (n=2,11%), Behçet’s disease (n=1, 5%), sarcoidosis (n=1, 5%), andSjögren’s syndrome (n=1, 5%). Seventeen patients used the Englishtranslation and two patients used the Spanish translation. They were of thefollowing ethnicities: White (n=9, 47%), Black (n=3, 16%), Mexican/Latino (n=6,32%), and Asian (n=1, 5%). The following had major life events (injury/illness-2,change of job-1, unable to pay bills-1, >/=2 events −4). Six children werein preschool-5^th grade, 11 from 6^th grade–11^th grade and 1 was in college. They had the followinginsurance to cover their standard clinical care: private (n=12, 63%), Medicaid (n=5,26%), and other (n=1). The subjects were either currently using the followingmedications or discontinued them: corticosteroids (14/19, 74%), mycophoenolatemofetil (2/19, 11%), cyclosporine (5/19, 26%), cyclophosphamide and/or rituximab(3/19, 16%), hydroxychloroquine (9/19, 47%), azathioprine (1/19), methotrexate(4/19, 21%), and thalidomide (1/19).

Seventeen parents stated that their child had a health condition. Fourteen patientshad an emergency room/urgent care visit in the last year. Parents reported a mean of2.5 ±4, median 2.5 missed work-days in the past 30 days. Parents perceived theimpact of child’s illness on daily routine at work (sometimes, often or almostalways) in 10/14 cases, and ability to concentrate at work (sometimes, often oralmost always) in 12/14 cases). The conditions (other than rheumatic diseases)mentioned by the subjects were: neurocardiogenic syncope (n=1), and celiac disease(n=1).

Child and parent SMILY scores were highly correlated (Spearman rho 0.7, p <0.05,n=17). Child SMILY score correlated with duration illness (Spearman rho =0.4, NS).We examined the HRQOL scores of patients who had ever used disease modifyinganti-rheumatic drugs (DMARDS) versus those had never used DMARDS. The mean SMILYscores were 71± 49 (child, n=15) and 66± 15 (parent, n=12) forthose had had ever used DMARDS. The mean SMILY scores were 49± 13 (child,n=2) and 53± 20 (parent) for those had had ever used DMARDS.

Feasibility

17 child and 15 parent subjects completed the corresponding reports of SMILY©-illness. Subjects completed SMILY©-Illness in ≤10 minutes and scoring each questionnaire took ≤ 10 minutes. For the child report of SMILY©-Illness, 5 items were omitted out of a total of 442 items (26 items × 17 children) with mean number of items omitted =0.3±0.7 (range 0–2). Two children did not complete any forms. For the parent report of SMILY©-Illness, 19 items were omitted out of a total of 390 items (26 items × 15 parents who completed the scale). Mean number of items omitted =1.3±2.4 (range 0–8). Maximum number omitted was 8 items by one parent.

Means, standard deviations and response range of SMILY©-Illness andother questionnaires

Scores and distribution of SMILY©-Illness, are provided in Table 1. All the reviewers of SMILY©-Illness approved thecontent, found it to be valid and relevant, easy to understand and especiallyliked the responses in the form of faces. The questionnaire has face and contentvalidity (Table 2). Due to small sample size we didnot perform calculation for psychometric properties.

Table 1 Scale descriptives for child and parent reports of measures ofSMILY-Illness

Table 2 Translation and adaptation for cultural suitability of US EnglishSMILY-Illness

Translation process

We had already described the rigorous translation process of SMILEY in previousmanuscripts [5, 6]. All the SMILEY translations were adapted toSMILY©-Illness using a professional translation company.Collaborative relationships with the different centers across the world werealready set up. The review process was similar to the process we followed forSMILEY translations [5, 6]. From each country, pediatric rheumatologists reviewed thetranslation and approved them for content and cultural appropriateness for theirpopulation. Table 2 details the entire adaptationprocess of 28 languages and all the translations are enclosed at the end of thisbrief report as Additional file 1.

Conclusion

SMILY©-Illness is a brief, easy to administer and score HRQOL scale for childrenwith systemic rheumatic diseases. SMILY©-Illness is suitable for use acrossdifferent age groups and literacy levels. SMILY©-Illness has good faceand content validity based on its process of adaptation, review by multiplepediatric rheumatologists and initial testing. However, further validation in eachcountry is required for the translated and adapted versions. In our population, asignificant percentage of children were on immunosuppressive/immunomodulatorymedications. Parents appeared to feel the impact of their child’s illness on adaily basis. The children’s self-concept was only average. The mean totalSMILY illness scores were similar in the range of what we found for SMILEY scores inpatients with SLE [2]. The lowest scores were found in the domain of “burden ofillness” and the highest score indicating better HRQOL was found in the socialdomain as reported in other studies [2]. As found in the literature, children had higher scores compared toparents [2].

The number of subjects is very small and it would be ideal if the disease types werewell distributed. Unfortunately in this sample they are not due to referral bias atthe time of the study. Due to the small sample size, we cannot make any definitiveconclusions regarding the correlations. Another limitation is that we do not haveinformation regarding the duration of disease prior to diagnosis.

The availability of translations will make recruitment for validation easier sincethese diseases are rare. SMILY©-Illness with its available translations may beused as useful adjuncts to clinical practice and research, providing valuableinsight to the impact of disease on the overall HRQOL of the child.

Endnote

^aRobert Wood Johnson Medical School, New Brunswick, NJ; and Hospital forSpecial Surgery, New York, NY.