Achieving tight control in rheumatoid arthritis

Has the gap between evidence and clinical practice been bridged? The treatment of RA has changed dramatically in the past three decades. In the 1980s, rheumatologists were advised to go low and slow. Nowadays, they are recommended to achieve and sustain a state of low disease activity or even remission as soon and for as long as possible using the widely endorsed tight control approach. New strategies for optimum treatment cannot be implemented because of difficulties with introducing new evidence into a clinical-practice setting. Moreover, adhering to tight control, includes the measurement of a valid composite measure, is still some way in bridging this gap between evidence and clinical practice. Until the late 1980s, rheumatologists used many clinical and laboratory variables to assess disease activity. For instance, many joint counts existed depending on the number of joints that need to be assessed and graded or not. In addition morning stiffness, duration of fatigue and multiple laboratory variables were also used. As multiple measures were considered, comparing the effects of anti-rheumatic drugs was difficult, especially in clinical trials. However, the evaluation and comparison of treatment response in trials became much easier with the development of the ACR core set [1]. Furthermore, composite measures such as the DAS and the more simplified DAS using 28 joints (DAS-28) were developed [2]. These measures enabled evaluation of disease activity by using multiple single outcomes at one time, resulting in less methodological problems for trials. In the past decade, several controlled studies showed that systematic monitoring of the disease using a composite measure, along with treatment adjustments to achieve the outcome, leads to lower disease activity in RA. Therefore, composite measures are useful not only for trials but also in a clinical practice setting. The meta-analyses by Schipper et al. [3] showed that when treatment adjustments had a set protocol, it resulted in better clinical outcomes than when treatment decisions were made by the rheumatologist. In four of the five controlled studies, the DAS-28 or DAS was used as the outcome measure. All the trials used different medication strategies until now and therefore a lack of evidence about which medication regimen is the most effective in RA. Consequently, the EULAR task force does not recommend adhering to a specific regimen, but instead suggests treating to target using a tight control approach [4]. Despite DAS-28 being the most commonly used and validated measure, it is often criticized as being far too complex to calculate and too time consuming for use in clinical practice. Due to these constraints, other groups developed other composite measures that may be equal to or better than the DAS-28 and easier to calculate [5]. The most well-known measures are probably the simplified disease activity index (SDAI) and the clinical disease activity index (CDAI) [6]. Pincus et al. [7] reported that 14% of the 600 rheumatologists surveyed performed joint counts, one of the main components of all composite measures, at each visit. Schoels et al. [8] surveyed 900 rheumatologists and found that 46% of them measured a patient's diseases status with the DAS-28 every 3 months, while 16% assessed the CDAI or SDAI every 3 months. Importantly, these studies describe rheumatologists’ self-reported care. So these percentages could be even lower in the real world as found in our own medical chart reviews. A detailed examination of almost 3000 visits revealed that the DAS-28 was measured in only 16% of these visits, whereas none of the other composite measures was used [9]. What about the value of future techniques in detecting inflammation? US and MRI are hot topics in RA research because the current cut-off points of the composite measures might not be sensitive and reliable enough to detect inflammation that is clinically relevant [10]. Although the development of new techniques is very important, the same or new problems may arise when implementing them in clinical practice. Aspects such as validity, reliability, the number of joints and the use of standardized scoring methods, including the cut-off points for medication changes, are important issues that need to be addressed before using them in the future. The assessment of a patient's disease activity using a composite measure is only beneficial when followed by appropriate action: i.e. adjustment of the anti-rheumatic drugs. According to our chart review, medication was changed in 33% of the visits in case of active disease [9]. A comparable percentage was found in a cluster randomized study. In this trial, rheumatologists were randomly assigned to usual care (three rheumatologists, n = 104 patients) or the intervention group [11]. A nurse measured the DAS-28 before the rheumatologist's visit, in order to save time (four rheumatologists, n = 144 patients). After 18 months, DAS-28 marginally reduced by −0.69 and −0.66 (P = 0.70) in the intervention and usual care group, respectively. In the intervention group, medication was changed by rheumatologists in 35% of the visits when clinically indicated by a DAS-28 > 3.2, while it was changed in 33% in the usual care group, which explains the comparable DAS-28 decrease. Time constraints are not the only problem. It is possible to implement strategies in the real-world setting based on evidence, but this change requires comprehensive approaches at various levels that are tailored to target groups and to address barriers faced by them. Therefore, a better understanding of the barriers faced by rheumatologists and their patients is necessary [12]. We also explored these barriers with composite measures in daily clinical practice via semi-structured interviews (Fig. 1). Rheumatologists were doubtful of the reliability and validity of composite scores; were doubtful of whether assessment of the composite score would lead to better outcomes compared with their own gut feeling, mentioned that patients were not familiar with disease activity measures (which was recognized by the patient as well), experienced organization-related barriers (e.g. lack of time), mentioned that there was a lack of transparency; and finally, were concerned that health-care insurers would use the outcome in reimbursement negotiations [13]. Barriers concerning the DAS-28. Fransen et al. [14], showed that rheumatologists were often reluctant to change medication because the disease activity had improved compared with the previous visit (85%). In another experiment, rheumatologists gave high importance to DAS (30.7%) in their decision to change treatment. However, age of the patient (21.5%) and erosions (20.5%) were rated as equally important in this decision [15]. From the patients’ perspective, a survey of 6135 RA patients showed that most of them were satisfied with their therapy, even if their RA was still active. Fear of loss of control (68%) and side effects (73%) were reported as major concerns [16]. The initiatives to improve quality of care should target the barriers faced to achieve tight control of RA. Rheumatologists’ attitude towards composite scores in general needs to be improved, as well as patients’ general knowledge about these measures and the importance of tight control. Last but not the least, initiatives to improve quality of care should also target the different considerations that are made by patients and rheumatologist when deciding whether or not to change the treatment. Disclosure statement: The authors have declared no conflicts of interest.