Date Published: September 10, 2015
Publisher: Public Library of Science
Author(s): Christine A. M. Smith, Karine Toupin-April, Jeffrey W. Jutai, Ciarán M. Duffy, Prinon Rahman, Sabrina Cavallo, Lucie Brosseau, Michael Nurmohamed.
The objectives of this review are to: 1) appraise the methodological quality of clinical practice guidelines (CPGs) in juvenile idiopathic arthritis (JIA) providing pharmacological and/or non-pharmacological intervention recommendations, and 2) summarize the recommendations provided by the included CPGs and compare them where possible.
A systematic search was performed. Three trained appraisers independently evaluated the methodological quality of the CPGs using a validated and reliable instrument, the Appraisal of Guidelines in Research and Evaluation II. Six domains were considered: 1) score and purpose; 2) stakeholder involvement; 3) rigor of development; 4) clarity of presentation; 5) applicability; and 6) editorial independence. The domains consist of a total of 23 items each scored on a 7-point scale. High quality CPGs were identified if they had a domain score above 60% in rigor of development, and two other domains.
Of the three included CPGs, the Royal Australian College of General Practitioners (RACGP) and American College of Rheumatology (ACR) CPGs were considered to be of high quality, but the German Society for Pediatric Rheumatology was of lower quality. Domains one to four had high domain scores across the guidelines (mean (standard deviation)): 72.76 (13.80); 66.67 (9.81); 64.67 (7.77); and 87.00 (9.64), respectively. Lower scores were obtained for applicability (14.00 (5.57)) and editorial independence (43.44 (7.02)). Recommendations varied across CPGs due to differences in context, target audience (general practitioners, rheumatologists, and other multidisciplinary healthcare professionals) and patients’ disease presentations. Despite this variability, progression of pharmacological treatment did not conflict between CPGs. Recommendations for non-pharmacological interventions were vague and the interventions considered varied between CPGs.
Overall, recommendations were based on a paucity of evidence and weak study designs. Further research is needed on interventions in JIA, as well as higher quality CPGs to facilitate implementation of the best evidence-based recommendations in clinical practice.
Juvenile idiopathic arthritis (JIA) is the most common rheumatic disease of childhood, with the precise etiology unknown and an incidence of 1 in 10,000 in children under 16 years of age [1,2]. A diagnosis requires that arthritis be present for a minimum of six weeks in patients younger than 16 years old [3,4]. There are seven onset types of JIA: systemic arthritis, oligoarthritis, polyarthritis (rheumatoid factor negative), polyarthritis (rheumatoid factor positive), psoriatic arthritis, enthesitis related arthritis, and undifferentiated arthritis [3,4]. These categories are mutually exclusive and differ based on the number of joints affected by arthritis (4 or fewer joints for oligoarthritis and 5 or more joints for polyarthritis), the presence of serological markers (e.g. rheumatoid factor positive or negative polyarthritis), or the area of the body affected (e.g. tenderness of the sacroiliac joint in enthesitis related arthritis). Symptoms of JIA include joint symptoms such as joint pain, swelling, and stiffness, for all onset types but also systemic symptoms such as fever and rash for those with systemic arthritis [2,5–8].
The systematic review of CPGs used the Cochrane Methodology  to identify, select and analyze the data and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement to guide the reporting of the systematic review (S1 Appendix). Ethics approval was not required, as this work was based on a systematic literature review.
Of the included CPGs for this critical appraisal, RACGP  and ACR [29,30] were found to be of high quality and were recommended with or without modifications by all three appraisers, but GKJR  was considered of lower quality. Clinicians must therefore use their discretion in applying the recommendations from GKJR. Domains 1–4 generally had high scores for all three CPGs; they concern scope and purpose, stakeholder involvement, rigor of development, and clarity of presentation, respectfully. On the other hand, domains 5 and 6, which address applicability and editorial independence, received the lowest scores as they were not well addressed in any of the CPGs. The differences in the CPGs in terms of context, target audience, patient population and clinical scenarios made it challenging to compare the recommended interventions. Furthermore, each CPG provided their own criteria for grading (i.e. strength of recommendations and level of evidence) and used different scales (e.g. a 3-level or 5-level grading scale).
The ACR [29,30] and RACGP  CPGs were found to be of higher quality using the AGREE II instrument while the GKJR  CPG was considered to be of lower quality. Future CPGs should focus on improving the methodological quality, particularly for monitoring the implementation of recommendations and assessing the CPG’s impact. Recommendations varied across the CPGs due to differences in context, target audience and patients’ clinical presentation of disease. Even though there was variability in CPGs, the progression of pharmacological treatment did not seem to be in conflict between CPGs. RACGP and GKJR covered different non-pharmacological interventions and the recommendations were vague. Overall, both pharmacological and non-pharmacological recommendations were based on a paucity of evidence and studies of weaker design. Therefore, there is a need for further research on interventions for JIA in order to have higher quality evidence to ensure greater confidence for clinicians to implement the recommendations of a particular CPG in rheumatology practice.