Personalized Estimates of Response and Severity Outcomes in Newly-diagnosed JIA

Overview

The PERSON-JIA Trial is a cluster-randomized trial testing the use of Shared Decision Making (SDM) with families for treatment of children with arthritis. The intervention is a discussion between physicians and families at the time of diagnosis that uses computer-generated personalized outcome reports generated by previously developed prediction algorithms. By using information provided by thousands of families, the investigators have developed a way of providing answers to common questions asked by patients and their families at diagnosis. We will test whether a structured discussion and shared decision between families and doctors (guided by the patient's personal report) will improve the tailoring of treatment to the child and control of their disease. The personal report is called the PERSON-JIA report and presents the child's expected disease severity, the likelihood the child will be arthritis free by age 18 and the chance treatments will be effective and/or have side effects. This way, answers to these questions can be shared by physicians and families to weigh potential benefits and harms according to family values and preferences. The investigators expect that using the personalized report in a frank and thoughtful discussion will help physicians and families make better decisions about managing the child's disease. This in turn will result in better disease control, greater family engagement and satisfaction with care and better-tailored treatment. If so, this will be a ground-breaking way of using information provided by families and doctors to improve the care provided to and the outcomes of children with arthritis in Canada.

The PERSON-JIA (Personalized Estimates of Response and Severity Outcomes in Newly-diagnosed Juvenile Idiopathic Arthritis) trial, will test an innovative shared decision making (SDM) intervention - a structured discussion, between physician and family, of the likely outcomes/response to treatments, as predicted by Personalized Prediction Algorithms (PPA), presented in a graphical summary. Juvenile Idiopathic Arthritis (JIA) is chronic arthritis of unknown cause starting before the 16th birthday. At present, evidence does not suggest that any single treatment approach for treating JIA is clearly superior, and current practice guidelines recommend Shared Decision Making (SDM) with families to select treatment. SDM requires a realistic appreciation of disease severity along with the likely benefits vs. harms associated with treatment options, and to then weigh these factors according to the family's values and preferences. To facilitate SDM, the investigators have built and validated PPA using data from previous Canadian JIA cohorts. As compared to treatment algorithms/flowcharts that tell physicians what to do, PPA are mathematical models that calculate the probability of outcomes for a child, based on their clinical presentation. The investigators have also established a Canadian Alliance of Pediatric Rheumatology Investigators (CAPRI) JIA Registry, which supports point of care data entry using smartphones/tablets. After data is entered in the registry for a newly diagnosed patient, PPA produce a graphical report depicting that child's probability of severe disease, remission in adulthood, and response and side effects with different treatments. This randomized controlled trial (RCT) will test the impact on JIA outcome of a SDM intervention that uses this graphical report to better tailor treatment to the child and family. The investigators hypothesize that compared to current standard care, use of this SDM intervention will lead to improved JIA disease control at 6 months via increased family engagement, greater satisfaction with treatment decisions, optimal tailoring of treatment to the child, and consistent adherence to the treatment plan. PERSON-JIA is a registry-based, pragmatic, cluster-randomized (by physician) adaptive trial. Pediatric rheumatologists from all pediatric rheumatology sites in Canada who agree to participate will be matched with another rheumatologist (by practice size, province and years in practice). There will be a one to one randomization of physicians to the intervention group or control group. Patients will be placed into the intervention or control group dependent on which group their pediatric rheumatologist has been randomized to. Control pediatric rheumatologists will practice standard care without the PPA or shared decision making discussion, but will still enroll all their new JIA patients into the CAPRI JIA Registry. Pediatric rheumatologists in the intervention group will have structured SDM discussions with the patient and family using the graphical PERSON-JIA report generated by PPA, at the time of diagnosis. The intervention will occur once, at the time of initial treatment decision making. Patients in the intervention arm of the study may have slightly longer clinic visits to account for a tailored, shared decision making discussion regarding their disease course and treatment options. The primary outcome will be inactive or minimally active disease 6 months after enrolment, defined using the clinical Juvenile Arthritis Disease Activity Score 10 (cJADAS10). CJADAS10 scores go from 0=best to 30=worst. Inactive or minimally active disease is 2.5 or less for patients with \>4 joints involved and 1.5 or less for patients with 1-4 joints. Secondary outcomes include inactive or minimally active disease 12 months after enrolment; quality of life and function; implementation of treatment plan; parent and doctor assessments of how much decisions were shared; parent's knowledge, value alignment, uncertainty and satisfaction with the decision; and intervention fidelity (analysis of medical encounter audio recordings). The primary statistic will be the odds ratio of patients achieving inactive or minimally active disease at 6mos, adjusted for clustering effects and baseline cJADAS10 scores.