Appendix Table B1Changes to elements of the nominated topic with rationale for refinements

Original ElementSource of InputCommentDecisionChangeRationale
Nominated KQLocal expert, literature scanThere are at least six subtypes of JIA, with distinct clinical characteristics and different treatment approaches. The amount of published literature for each subtype varies substantially.Specify in the KQ that subtypes of JIA exist and that the population of interest will include children with any subtype.

No change in PICO.


KQ 3 was added about possible variations in effectiveness and safety of DMARDs between subtypes.

Added detail about subtypes makes the key questions more specific, and improves the accuracy and research feasibility of the SR. Inclusion and analysis by JIA subtypes might expand the scope and heterogeneity of the SR; however the literature predominately addresses two subtypes and reduces this concern.
PICO (Intervention): Corticosteroids; Synthetic disease-modifying anti-rheumatic drugs (DMARDs); Biologic DMARDsLiterature scan, Key InformantCorticosteroids are commonly used as first-line treatment for most cases of JIA.Remove as a intervention, and include as a comparatorIntervention: DMARDsThis change reflects the standard of care and the literature. This does not significantly compromise fidelity to the original nomination. The principal dilemma relates to DMARDs and not corticosteroids; this makes them better suited as a comparator for DMARDs.
PICO (Outcome): Outcomes include looking at potential harms and benefits of various treatmentsLiterature scan, Key Informants, Local ExpertsSpecific outcomes are not includedInclude relevant outcomes, and specify them in the key questions and PICO

See refined KQs


Outcome: Patient-centered outcomes (such as pain control, clinical remission, and quality of life); intermediate outcomes (laboratory measure of inflammation, number of joints with limited range of motion); and adverse effects of treatment.

Distinguishing between patient-centered outcomes and intermediate outcomes elucidates the underlying relationship of the outcomes and the logic of the SR
Nominated KQLiterature scan, key informant, local expertsThe outcomes listed do not reflect the clinical logic typically seen in AFs and refined KQs. The nominated topic places patient-centered outcomes (e.g., patient functioning) and intermediate outcomes (e.g., radiographic joint damage) in the same key question.Formulate key questions specific to the outcome categories (patient-centered outcome; intermediate outcome).

KQ: See refined KQ 1 (patient-centered outcomes) and KQ 2 (intermediate outcomes).


AF: The relationship of the outcome categories is represented in the AF

Accuracy and research feasibility are improved by including specific outcomes in the KQ.

Distinguishing patient-centered outcomes from intermediate outcomes elucidates the underlying relationship of the outcomes and the logic of the SR.
Nominated KQLiterature scanMany studies use ACR Pediatric 30, a validated composite measure of improvement of JIA. It includes patient –centered outcomes and intermediate measures. Some measures of the Peds 30 were included in the nominated materials.Include mention of Peds 30 measure in the AF.In the AF, asterisks (*) have been added to the outcomes that are constituents of the Peds 30 measure.The literature scan provided added detail about relevant outcomes, including that part of the ACR Pediatric 30. This improves the accuracy and research feasibility of the review.

From: Appendix B, Example of Selected Aspects of a Topic Refinement

Cover of The Refinement of Topics for Systematic Reviews
The Refinement of Topics for Systematic Reviews: Lessons and Recommendations From the Effective Health Care Program [Internet].
Buckley DI, Ansari M, Butler M, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.