Results Summary
What was the research about?
Juvenile idiopathic arthritis, or JIA, causes joint stiffness, swelling, and damage. More than half of children with JIA have a form of the disease called polyarticular JIA, or pJIA, which causes swelling in five or more joints. Having pJIA can make it hard to move around, walk, or do daily activities such as getting dressed or brushing teeth.
Although pJIA doesn’t have a cure, treatments can limit inflammation and reduce damage in the joints. Medicines used to treat pJIA are called disease-modifying antirheumatic drugs, or DMARDs. The newest types of DMARDs are called biologics. Doctors don’t know the best time to start treatment with biologic medicines.
In this study, patients and their families worked with their doctors to choose from three treatment options:
- Step Up: Patients started treatment with a non-biologic DMARD medicine only.
- Early Combination: Patients started treatment with both a non-biologic DMARD and a biologic medicine.
- Biologic First: Patients started treatment with a biologic medicine only.
If patients’ symptoms didn’t improve after three months, they could change or add medicines.
What were the results?
Of patients in the study, 64 percent chose Step Up, 25 percent chose Early Combination, and 11 percent chose Biologic First as their treatment option.
After a year, the three options worked about the same in limiting inflammation and helping patients become symptom free. Less than half of patients became symptom free using any of the options.
For all three options, patients had less pain and were better able to move after one year. The three options didn’t differ in improving pain and ability to move.
Who was in the study?
The study included 400 children with pJIA. Of these, 73 percent were white, 8 percent were black, and 20 percent were another race. The average age was 10, and 73 percent were girls. All received care at one of 56 clinics across the United States.
What did the research team do?
The research team enrolled patients with pJIA before they started treatment. After starting treatment, patients went to the doctor every three months for a year. At these visits, patients or caregivers filled out surveys about their health. Doctors also filled out forms in patients’ health records at these visits.
Patients with pJIA, caregivers, an arthritis foundation staff member, a nurse, a research coordinator, and an insurer gave input during the study.
What were the limits of the study?
Fewer patients started treatment with both medicines or with the biologic medicine alone than planned, making it hard to detect differences between treatments. For 18 percent of patients, data to determine whether they were symptom free and had reduced inflammation were missing. Results may have differed if the team had complete data on more patients. Over time, fewer patients answered the surveys, which could also make the results less accurate.
Future research could look at other ways to help patients with pJIA become symptom free.
How can people use the results?
Patients with pJIA and their doctors can use these results when discussing treatment options.
Professional Abstract
Objective
To compare the effectiveness of varying the initial timing of biologic therapy on producing clinically inactive disease in patients with polyarticular juvenile idiopathic arthritis (pJIA)
Study Design
| Design Element | Description |
|---|---|
| Design | Observational: cohort study |
| Population | 400 children and adolescents under age 19 who were diagnosed with pJIA and who had no prior treatment |
| Interventions/ Comparators |
|
| Outcomes |
Primary: clinically inactive disease without use of glucocorticoids Secondary: patient-reported measures of pain interference and mobility |
| Timeframe | 1-year follow-up for primary outcome |
This prospective, observational cohort study compared the effectiveness of three treatment approaches on achieving inactive disease and improving pain and mobility in children with pJIA.
Patients had clinic visits every three months for one year. During the first visit, caregivers, children, and doctors chose one of the following treatment approaches:
- Step Up: Patients started treatment with a non-biologic disease-modifying antirheumatic drug (DMARD) only.
- Early Combination: Patients started treatment with a DMARD and a biologic medication together.
- Biologic First: Patients started with a biologic medication only.
For all three approaches, every three months, doctors could add or change medications if the patient’s condition worsened or did not improve.
The study included 400 children diagnosed with pJIA who were receiving care at one of 56 clinical sites across the United States. Of these, 73% were white, 8% were black, and 20% were other races. The average age was 10, and 73% were girls.
Every three months, patients or caregivers completed surveys at doctor’s visits. Staff from the doctor’s office entered medical record data in the CARRA Registry, a multicenter registry of patients with pediatric rheumatic diseases. Researchers used propensity weighting to control for the influence of patient characteristics on study outcomes and multiple imputation to account for missing data.
Several patients with pJIA, caregivers, an arthritis foundation staff member, a nurse, a research coordinator, and an insurer provided input during the study.
Results
Of the participants, 64% chose Step Up, 25% chose Early Combination, and 11% chose Biologic First as their treatment approach.
After one year, the three approaches did not differ significantly in achieving clinically inactive disease without use of glucocorticoids. With all three approaches, less than half of patients achieved clinically inactive disease (Step up: 38%; Early Combination: 47%; Biologic First: 34%).
With all three approaches, patients reported decreased pain interference levels and increased mobility after one year (both p<0.0001), but the approaches did not differ significantly in achievement of these outcomes.
Limitations
Fewer patients than anticipated enrolled in the Early Combination and Biologic First treatment approaches, making it difficult to detect differences between treatments. At the end of the study, data necessary to determine clinically inactive disease were missing for 18% of patients. Results may have differed if data for clinically inactive disease were available for more patients. Response rates for patient-reported outcomes decreased with longer follow-up, leading to missing data that could also bias results.
Conclusions and Relevance
In this study, the three treatment approaches did not differ in achieving clinically inactive disease, decreased pain, or improved mobility at one year. Patients reported improvements in pain and mobility regardless of treatment approach.
Future Research Needs
Future research could assess relevant longer-term outcomes and explore additional ways to achieve clinically inactive disease among patients with pJIA.
Final Research Report
View this project's final research report.
Journal Citations
Article Highlight: Clinicians and patients have been uncertain about the optimal time to start biologics for polyarticular juvenile idiopathic arthritis (pJIA). However, results from this PCORI-funded study, published in Arthritis & Rheumatology, provide useful new evidence about first-line treatment options for this disease. Researchers compared the effectiveness of three treatment approaches among 400 children with pJIA. The approaches included starting patients on a non-biologic disease-modifying antirheumatic drug (DMARD) only, starting patients with a DMARD and a biologic medication, and starting patients with a biologic only. After one year, the approaches did not differ significantly in achieving clinically inactive disease without the use of glucorticoids, and patients reported decreased pain interference levels and increased mobility with all three approaches.
Results of This Project
Related Journal Citations
Peer-Review Summary
Peer review of PCORI-funded research helps make sure the report presents complete, balanced, and useful information about the research. It also assesses how the project addressed PCORI’s Methodology Standards. During peer review, experts read a draft report of the research and provide comments about the report. These experts may include a scientist focused on the research topic, a specialist in research methods, a patient or caregiver, and a healthcare professional. These reviewers cannot have conflicts of interest with the study.
The peer reviewers point out where the draft report may need revision. For example, they may suggest ways to improve descriptions of the conduct of the study or to clarify the connection between results and conclusions. Sometimes, awardees revise their draft reports twice or more to address all of the reviewers’ comments.
Peer reviewers commented and the researchers made changes or provided responses. Those comments and responses included the following:
- The reviewers questioned the researchers’ handling of missing data. They posited that the problem of missing data was greater in this observational study compared to what would be expected in a randomized controlled trial (RCT) because of limited active follow-up in the Childhood Arthritis and Rheumatology Research Alliance (CARRA) registry as opposed to a more robust follow-up plan that would be expected in an RCT. The researchers acknowledged this point and noted it in their study limitations section. They also noted that they did not report on recommended multiple imputation techniques to account for missing data for patient-reported outcomes, where the missing data rate was especially high. The researchers explained that multiple imputation requires the assumption that the patient-reported outcome data are missing at random, but the researchers could not verify this fact in their dataset.
- The reviewers asked why the researchers stated that RCTs were less feasible than cohort studies for a rare condition. The researchers explained that their statement was specifically about the resources available to them for studying juvenile idiopathic arthritis (JIA) through the CARRA Registry. Given the registry’s existing agreements with patients, no additional institutional review board (IRB) approval was needed to conduct this study, so the study could be rolled out quickly across all CARRA sites. An RCT would have required additional IRB approval and contracting, which would have made the study no longer feasible in the time frame required, particularly given the low incidence of JIA. Also, not all insurance plans would allow the use of all the medications tested, further reducing the pool of eligible participants for an RCT. Finally, the researchers noted, families and physicians may not have been as likely to participate in the study if patients were randomly assigned medications rather than allowed to choose their own treatment regimens in consultation with their physicians.
- The reviewers said it would have been helpful if the researchers had reported results according to more specific diagnoses than JIA, since that is a broad category of conditions. The researchers agreed that it is reasonable to ask whether specific treatments work better for specific diagnoses, but the study did not have enough statistical power to detect differences among different categories of JIA. However, the researchers said that exploratory analyses are underway to assess whether individual patient trajectories demonstrate specific phenotypes that could predict treatment outcomes.
- The reviewers asked whether the researchers looked at outcomes based on the actual treatments that patients received, rather than only comparing patient outcomes based on the consensus treatment plans. The researchers said they did follow up on the treatments that patients received and grouped patients into three categories: patient received treatment as planned, patient did not receive treatment as planned, and patient treatment status was uncertain. The researchers populated these categories based on a review of whether and when patients received biologics or disease-modifying antirheumatic drugs, or both treatments. This review allowed the researchers to determine which consensus treatment plan each patient actually received and rerun analyses based on this information. The researchers noted that because there was very close alignment between the planned treatment and the executed treatment, the results of the as-treated analyses were similar to the results of the intent-to-treat analyses.