In addition to reviewing core outcome measures for assessing treatment efficacy, the panelists also share their thoughts on issues that impact patient adherence.

       Dr. Goldsmith: There have been three or four eras of treatment of JIA. Initially, salicylates, gold and systemic corticosteroids were the mainstays of management. Then NSAIDs without significant hepatotoxicity or associated Reye’s syndrome largely supplanted aspirin. Methotrexate became available in the late 1980s and quickly replaced gold as the primary remissive medication.
       Selective Cox-2 inhibitor NSAIDs were developed and one of these agents, rofecoxib, was approved for use in JIA. However, accumulated evidence of significant cardiovascular events in adults lead to the removal of rofecoxib from the market just four weeks after the JIA indication. Subsequently, another selective agent, celecoxib, has been approved for JIA.
       Intraarticular corticosteroid therapy has just about become the standard of care for oligoarticular JIA and systemic corticosteroids are now reserved solely for use in narrow specific clinical circumstances. Starting in the late 1990s, the emergence of biologic therapies has brought us closer to the goal of achieving complete clinical remission and even reversal of radiographic changes in many children.
       Dr. Lovell, since you have been at the forefront of so many of these recent clinical trials, I would like to see if you can help us understand how researchers measure improvement in these studies and if you are currently satisfied with these outcome measures.
       Dr. Lovell: The six core set parameters we use in clinical trials are: MD global assessment of disease activity; patient global assessment of overall disease impact; the number of joints with active arthritis; the number of joints with loss of motion; a laboratory measure of inflammation such as the ESR or CRP; and a standardized measure of daily physical functional activity. The tool that has generally been used in the JIA trials is the Childhood Health Assessment Questionnaire (CHAQ).
       The standard outcome measure we have used for JIA trials is the American College of Rheumatology pediatric 30 (ACR Pedi 30) level of response. In other words, patients must improve by 30 percent or more in at least three of the six core set parameters, and no more than one of the parameters shows a 30 percent decrease.
       One of the very gratifying things we have seen in trials of anti-tumor necrosis factor (anti-TNF) agents is that the vast majority of patients meet the ACR Pedi 30 level of response. Now we use more demanding definitions. Now there is an ACR pediatric 50 (ACR Pedi 50) level of response, for example, that requires 50 percent improvement in at least three of the six core set parameters, and no more than one of the parameters can get worse by 30 percent. We have also used ACR Pedi 70, ACR Pedi 90 and, in one trial, we even used an ACR pediatric 100 (ACR Pedi 100) level of response. This would be 100 percent improvement in at least three of the six core set parameters and no more than one parameter getting worse by 30 percent.
       The most remarkable thing is anti-TNF therapies are extraordinarily effective in controlling the disease in some patients. In many patients, very little, if any, manifestations are present when they are taking anti-TNF agents. We have had to increase the level of our responses in trials based on the effectiveness of these agents.
       Dr. Goldsmith: Three years ago, Wallace et al., reported the results of an international consensus conference in describing the definition of inactive disease in patients treated for JIA.⁷ This definition has five components:
• no joints with active arthritis (with active arthritis being defined as either swelling or limitation of range of motion plus pain and tenderness);
• no fever, rash, serositis, splenomegaly or generalized lymphadenopathy attributable to JIA;
• no active uveitis;
• normal ESR and/or CRP; and
• a best possible score on the physician’s global assessment of disease activity.
       The authors also developed definitions for clinical remission. While the child is still on medication, there should be no active disease for a minimum of six months. When a child is no longer on medication, there should be no active disease for a minimum of 12 months.
       These are all critically important definitions. All parents of children with JIA need us to be precise as possible in our clinical assessment and our subsequent recommendations to both start and stop medications, particularly those without a record of long-term clinical use in JIA.
       Without adequate development of these clinical guidelines, many children will not receive the treatment that you believe to be best. Many parents are worried about the possible consequences of long-term therapy. While we, of course, cannot give complete assurances, real trust develops not only with optimum communication but also with the clearest therapeutic roadmap that you can project.
       Taylor: That is an excellent point. Meta-analyses have shown that collaborative communication promotes patient and family-centered health care, and improves outcomes.^8,9 In addition, the preliminary criteria you mentioned from Wallace and colleagues is in the process of being validated both retrospectively and prospectively.⁷ They developed those definitions of clinical remission through consensus testing, delphi technique and nominal group technique. The group felt there needed to be at least a 20 percent or more chance of clinical remission being predictive of whether there would be remission of the disease in the next five years. This would allow practitioners to provide more accurate estimates of prognosis.

Current Concepts With Patient Adherence
       Dr. Goldsmith: Before we turn to the discussion of current approaches to the management of JIA, we should explore the issue of adherence to medications. Ms. Taylor, would you share with us some of the concepts of adherence and how you try to employ them in the clinical setting?
       Taylor: Adherence is the new nomenclature that has replaced compliance. Adherence also implies there is more of a patient and family role in participating in the ongoing care. Numerous studies have looked at adherence and various methods of assessing adherence. Rapoff and Lindsley found that about 50 to 55 percent of patients were non-adherent with medications.¹⁰ That number increased when they assessed adherence with home exercise programs and the use of splints. Some of this data is a little bit older because we do not use splints quite as often today as we used to. As we have moved into the biologics and other treatments, splints have not been as necessary. Another avenue that researchers have studied is the use of behavioral incentives and reward systems to increase adherence. Games, gift certificates or monetary rewards have been shown to improve treatment adherence in children with chronic illness.¹¹
       In addition, studies have also examined the use of selfmonitoring which would be similar to keeping records or logs. Perhaps one of the most well known studies on selfmonitoring is in the diabetes literature in which researchers required patients to do self-monitoring for their hemoglobin A1C counts. These types of interventions have been shown to increase adherence and improve selfmanagement.^12-15 As this relates to rheumatology, one could track self-monitoring for taking medications or for exercising.
       One of the difficult components of adherene is trying to identify early on who is at highest risk for nonadherence and how and what health care resources do we utilize across our busy health care system to improve this recognition. There are studies that have shown the level of family conflict and psychosocial distress within families is predictive of non-adherence.¹⁶ The more family conflict there is, the higher the level of nonadherence. Those patients probably need closer monitoring and closer care management than those who do not have as much family conflict.
       Researchers have actually utilized pill counts in drug trials to look at adherence. In addition, they have utilized more sophisticated approaches, which involve the use of a computer chip and monitor in the caps of pill bottles so one can measure adherence with medication regimens. This was done in a trial on the effects of calcium administration on bone mineralization.17 This device measures every time patients open and close the bottle.
       Dr. Lovell: In the particular clinical trial looking at the effect of calcium supplementation on bone mineralization, we made a huge emphasis on assisting families with adherence and measuring adherence. Close to 90 percent of the patients took their medicine as prescribed in that interventional trial.¹⁷ However, there was a very different emphasis and level of sophistication in monitoring and assisting families with adherence in this trial than we normally do in clinical day-to-day care.
       These microelectronic monitoring systems for pill caps are available and have been used in certain clinical circumstances. Researchers have shown that the use of these monitoring systems for pill caps increase compliance by anywhere from 20 to 40 percent.^18-20 They also allow you to get very quantitative chronologic data on a particular patient. Accordingly, you can work with a family and note the times when the pill was not taken. Then you can help determine what might be occurring at those times and work with the family collaboratively to improve the patient’s ability to take the medication.
       However, the microelectronic pill caps are fairly expensive and require some techniques to download the data. While they are not commonly used in clinical care situations, they do represent an option for a family in whom compliance is a big problem and is getting in the way of effective treatment of the child.

Emphasizing Communication And Trust With The Patients And Families
       Dr. Lovell: One needs to be open and candid with families, and say it is just a fact of life that there are going to be times when you cannot get the medicine in the child as effectively as you can at other times. There will be times when you are going to go on a trip and forget to take the medicines. Those things occur. I believe a sympathetic and open dialogue will facilitate discussions around adherence so all involved know about the issues and can then figure ways to more effectively address them.
       One of the givens is that if you are taking care of adolescents, adherence is going to be an issue. Almost all teenagers with a chronic illness at some point during the course of their disease feel the need to stop their treatment. Most of the time, it may be related to the fact they are tired of taking it. If you have an open relationship with the patient and the family, you can deal with this in a more productive fashion. Children and adolescents are more likely to tell you about it when you have discussions if, in fact, you have established this relationship of trust and understanding about the difficulty of long-term adherence for everybody.

Exploring Avenues Of Self-Management Education On The Internet
       Dr. Goldsmith: Stinson et al., discussed Web-based interventions as a methodology to augment self-management strategies in adolescent patients with JIA.²¹ Effective self-management behaviors are integral to enhanced medication adherence. Better understanding of their disease through social support and acquiring knowledge through two-way communications strengthened self-management behaviors. This paper again illustrates the continuous need for effective interactions with our patients and their families so the remarkable array of medications now available may be optimally put to use.
       Taylor: Some studies have looked at interactive health care communications via the Web or other technologies. Computer-based information packages have shown a positive effect on increasing knowledge, promoting healthy behaviors, monitoring behaviors and adherence, and increasing social support. In addition, computer-assisted assessments that provide immediate feedback on key areas of self-management have been found to improve health behavioral change and outcomes.²²