You are here

Quality in rheumatoid arthritis care

Best Practice & Research Clinical Rheumatology


While most rheumatology practices are characterized by strong commitment to quality of care and continuous improvement to limit disability and optimize quality of life for patients and their families, the actual step toward improvement is often difficult. This is because there are still barriers to be addressed and facilitators to be captured before a satisfying and cost-effective practice management is installed. Therefore, this review aims to assist practicing rheumatologists with quality improvement of their daily practice, focusing on care for rheumatoid arthritis (RA) patients.

First we define quality of care as “the degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge”. Often quality is determined by the interplay between structure, processes, and outcomes of care, which is also reflected in the corresponding indicators to measure quality of care. Next, a brief overview is given of the current treatment strategies used in RA, focusing on the tight control strategy, since this strategy forms the basis of international treatment guidelines. Adherence to tight control strategies leads, also in daily practice, to better outcomes in patients with regard to disease control, functional status, and work productivity. Despite evidence in favor of tight control strategies, adherence in daily practice is often challenging. Therefore, the next part of the review focuses on possible barriers and facilitators of adherence, and potential interventions to improve quality of care. Many different barriers and facilitators are known and targeting these can be effective in changing care, but these effects are rather small to moderate. With regard to RA, few studies have tried to improve care, such as a study aiming to increase the number of disease activity measures done by a combination of education and feedback. Two out of the three studies showed markedly positive effects of their interventions, suggesting that change is possible. Finally, a simple step-by-step plan is described, which could be used by rheumatologists in daily practice wanting to improve their RA patient care.

Keywords: Quality of care, Quality indicators, Best practice, Barriers, Facilitators, Interventions.


Musculoskeletal disorders such as gout, osteoarthritis, and rheumatoid arthritis (RA) are considered to be among the most burdensome medical conditions [1]. This has led to the execution of many randomized controlled trials (RCTs) that have provided evidence for the best therapeutic interventions for these diseases. Despite this constant stream of evidence-based recommendations, the translation into daily practice is often suboptimal [1].

While many practicing rheumatologists will agree that quality of care is an important aspect in rheumatology, the actual step to improve quality of care is often difficult, since rheumatologists do not know where and how to start, and there are no clear strategies available how to approach improvement of quality of care in their clinical practice. This review, with the goal of assisting practicing rheumatologists with their own quality improvement of care, aims to fill this gap. It starts with a brief general introduction on quality of care and its measurement methods. Thereafter, the focus will shift to RA and we will discuss what optimal RA care is, how we can measure whether quality demands are met or not, and how this could be improved. In the latter part, two case descriptions of successful quality improvement projects in RA will be discussed. Finally, we will give practical recommendations to rheumatologists who want to further improve their own performance.

A. What is quality of care and how can you measure it?

Quality of care in itself is a rather abstract term, but more practical descriptions do exist. One of the most commonly used descriptions, developed around 1980 by Donabedian, distinguishes structures, processes, and outcomes of care [2]. The structure of care describes aspects of the setting in which care is delivered, such as the number of rheumatologists or the presence of a treatment protocol. Next, the process of care describes the actions of the health-care professionals, for example, whether the protocol is indeed followed. Finally, the outcome reflects the effect of the given care in terms of mortality, morbidity, and health status. It is believed that more desirable outcomes are obtained if the structure of care provides the opportunity to deliver the most optimal care processes (Fig. 1).

[locator:gr1] (link type not set)

Fig. 1

The Donabedian Triad. Donabedian hypothesized that all elements are linked to each other [3].

Around 1990, the Institute of Medicine (IOM) defined quality of care as “the degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge.” Furthermore, the IOM formulated the following six criteria that pertain to quality of care: Care should be i) safe, ii) effective, iii) patient-centered, iv) timely, v) efficient, and vi) equitable [4]. When using these criteria, it is important to take into account the different perspectives of the stakeholders (e.g., patients or health insurers) [5].

Knowing how to describe quality of care is a prerequisite for its measurement. Often quality indicators are used to assess quality of care. A quality indicator is “a measurable element of practice performance for which there is evidence or consensus that it can be used to assess the quality, and hence change the quality of care provided.” [6] Quality indicators are often grouped using the abovementioned quality definition by Donabedian, thus providing structure, process, and outcome indicators. Outcome indicators reflect the result of the care that was provided by the health-care provider, while process indicators reflect the actual care given to patients (“what is done”). Structure indicators, on the other hand, describe organizational aspects (“what is available”) [7]. How these indicators are used within rheumatology will be described later in this review; we will now first describe what optimal care in RA is.

B. What is optimal RA care?

The treatment of RA has substantially improved during the last two decades. Until 1990, the use of disease-modifying anti-rheumatic drugs (DMARDs) was limited, due to the belief that DMARDs were too toxic to use for a non-life-threatening disease such as RA [8], [9], and [10]. However, these assumptions have changed and the use of DMARDs, both synthetic and biological, in the management of RA has gained prominence. Treatment methods such as a step-up approach, combination therapy and treat-to-target strategies have been developed [8], [9], and [10].

Improvements in care for RA patients have not only been the result of an increase in the number of effective therapeutic options, but also because of the broader insight into the course of the disease and its prognosis. For example, it is evident that active RA is associated with a high risk of cardiovascular morbidity and mortality. Furthermore, it was found that RA treatment should be started at the earliest (in the so-called “window of opportunity”) in order to prevent the occurrence of irreversible joint damage or at least to halt progression of the disease [11] and [12] Another terminology used in this context is “hit hard, hit early” (intensive treatment early in the disease course) and “tight control”. Tight control, though being the mainstay of optimal clinical RA care, is not the only facet of good RA care. Shared care with specialized nurses or physician assistants, cardiovascular risk management and the management of comorbidities are some other important aspects of RA care. However, all these aspects cannot be elucidated in detail. Hence, we focus on the tight control principle that currently forms the basis of major treatment guidelines [13] and [14].

Tight control, also called ‘treat to target’, can be defined as “frequent assessment of disease activity combined with an objective structured protocol to make treatment changes that maintain low disease activity or remission at an agreed target.” [15] Recently, an international task force provided an update of the 2010 treat-to-target recommendations [14]. These recommendations describe a generic principle or strategy, not necessarily advocating a particular type of intervention, that should be adhered to in order to achieve disease remission or low disease activity in RA patients. The four overarching principles and 10 recommendations focus around shared decision making, the importance of setting a treatment target, measuring disease activity, changing treatment until the desired goal is reached, and maintaining the treatment goal thereafter [14].

Various studies have proven the effectiveness of the tight control regime, with the Tight Control of Rheumatoid Arthritis (TICORA) study being one of the first to show the beneficial effects of tight control. In the TICORA study, patients in the tight control group had a significantly better disease outcome after 18 months as compared to the control group with regard to the European League Against Rheumatism (EULAR) good response criteria (82% vs. 44%, p < 0.0001) and the mean decrease in disease activity score (DAS; −3.5 vs. −1.5, p < 0.0001) [11].

After the TICORA study, several studies have replicated these findings, and in 2010 a meta-analysis on the effects of tight control was published. This meta-analysis concluded that patients treated according to the tight control principles had significantly better DAS-28 responses as compared to patients treated with usual care (mean difference = 0.59; p < 0.001) [16]. In addition, they also compared tight control with and without protocolized treatment adjustments. These comparisons showed a beneficial effect of protocolized treatment adjustments, with a 0.66-point decrease in the DAS-28 response (95% CI: 0.72–1.11; P < 0.0001) if a specific treatment protocol was used. Besides, there was an improvement in functionality and a decrease in joint damage [16].

Although tight control studies so far have focused on reaching remission or low disease activity, secondary analyses have shown that lower disease activity is also associated with improved work productivity, less comorbidity, and lower cardiovascular risk [17]. This may imply that applying tight control in daily practice benefits RA patients with regard to disease control and other important aspects of their lives such as work.

In summary, due to the complexity of RA and the increasing treatment arsenal, it can be difficult for rheumatologists to provide optimal RA care in all patients. However, it seems that using tight control-based treatment strategies could assist rheumatologists in achieving low disease activity or remission in the majority of their patients, ensuring better clinical outcomes, and promoting better work productivity, less comorbidity, and lower cardiovascular risk [17].

C. How can we measure whether optimal RA care is provided?

As mentioned in the first section, quality of care can be assessed using predefined quality indicators for the structure, processes, and outcomes of care [6] and [7]. With regard to RA, a broadly accepted set of quality indicators is lacking. However, several groups around the world have made an attempt to develop sets of RA quality indicators. Some of these indicator sets are as follows:

Dutch researchers have described one of the first sets, designed to monitor RA disease course in the Dutch Rheumatoid Arthritis Monitoring (DREAM) cohort. This indicator set consists of 10 process, five structure and three outcome indicators and is divided into different subcategories. These subcategories are the measurement of disease activity, structural damage, functionality, follow-up frequency, intensification of pharmacological therapy, prerequisites for measuring disease activity, and patients׳ disease activity (e.g., the percentage of RA patients in remission a year after diagnosis) [18].

Two other groups in Europe have also developed sets of quality indicators. Firstly, the National Health Service in England (NHS) has developed quality indicators for RA, along with indicators for other diseases, in order to standardize improvements in the delivery of primary care [19]. Management of RA in primary care may include examining and assessing cardiovascular risk and blood pressure, the risk for osteoporosis, and signs of depression. During an annual meeting on primary care, the effects of the disease upon a person׳s life can be assessed, for example, by monitoring the side effects of medication or assessing the psychological situation of the patient. The NHS indicator sets reflect this care and comprise one structural, one outcome, and two process indicators, subdivided in two domains. The first domain is “records” [19]. This domain documents whether the primary-care physician establishes and maintains a register for patients of 16 years and older with RA. The second domain is “ongoing management”. This domain includes the documentation of the percentage of patient who had an annual face-to-face meeting with general practitioner in the preceding 12 months [19].

The second European indicator set is developed by the European Musculoskeletal Conditions Surveillance and Information Network (EUMUSC.NET) and contains 14 indicators (one outcome, two structural, and 11 process indicators) [20]. To our knowledge, the EUMUSC.NET has not divided these indicators in domains. Therefore, we decided to divide this extensive list of indicators into six domains, namely organization, screening, pharmacological treatment, and non-pharmacological treatment, monitoring, and outcome. These indicators have also been stated in Table 1, and compared with the American College of Rheumatology (ACR) indicators.

Table 1

Indicators from the ACR and subdivided into domains.

SubjectIndicator aSource a
OrganizationRheumatology practices should provide information (written or website) on how a patient can contact the practitioner for urgent consultations (in case of flares/worsening of the disease, serious side effects)
If a patient is referred to a physician for a new diagnosis of rheumatoid arthritis, then the patient should be seen by the physician within 3
If a patient presents with suspected rheumatoid arthritis, then he/she should be referred to and seen by a specialist (preferably a rheumatologist) for confirmation of diagnosis within 6 weeks after the onset of
If a patient is diagnosed with RA and there are joint damage/soft tissue problems that may be solved by surgery, then the patient should be assessed by an orthopedic surgeon within 3
EducationIf a patient is newly diagnosed with RA, then he or she should be given individually tailored education by relevant health professionals about the natural history, treatment, and self-management of the disease within 3
ScreeningPercentage of patients 18 years and older with a diagnosis of rheumatoid arthritis who have documentation of a tuberculosis (TB) screening performed and results interpreted within 6 months prior to receiving a first course of therapy using a biologic disease-modifying anti-rheumatic drug (; ACR
Non-pharmacological treatmentIf a patient is newly diagnosed with RA, then a referral to a relevant health professional for instruction on an individualized exercise programme including advice for physical activity, range of motion, muscle strengthening, and aerobic exercises should be provided within 3
Pharmacological treatmentIf a patient has rheumatoid arthritis and is being treated with a DMARD and there is evidence of increased disease activity or there is evidence of progression of RA bony damage over a 6-month period of time, then one of the following should be done: change DMARD dose or route of administration, change DMARD, add an additional DMARD, start or increase dose of glucocorticoids or provide local glucocorticoid injection(s), unless the patient refuses or all of the above are contraindicatedACR
A rheumatologist should intensify disease-modifying medication when disease activity is moderate or high (according to EULAR recommendations)
If a patient has RA, then he/she should have a treatment plan developed between him/her and his/her clinician/health professionals at each
MonitoringIf a patient is diagnosed with RA, then a rheumatologist and/or relevant health professionals from the multidisciplinary team should assess and document the following variables: 1) a measure of disease activity such as composite scores like DAS-28 or any of its variants CDAI or S-DAI, 2) structural damage (using the best available method, e.g., X-ray, MRI, ultrasound), 3) functional status (e.g., HAQ), and 4) labor force participation. The assessment and documentation should occur at baseline and thereafter at appropriate time intervals, at least annually for first, third, and fourth
If a patient has a diagnosis of rheumatoid arthritis, then baseline radiographs of the hands or feet should be performed within 3 months of the initial diagnosis and every 3
Percentage of patients 18 years and older with a diagnosis of rheumatoid arthritis who have an assessment and classification of disease activity at least once within 12 monthsACR
Percentage of patients 18 years and older with a diagnosis of rheumatoid arthritis who have an assessment and classification of disease prognosis at least once within 12 months.ACR
Percentage of patients 18 years and older with a diagnosis of rheumatoid arthritis who have been assessed for glucocorticoid use and, for those on prolonged doses of prednisone >10 mg daily (or equivalent) with improvement or no change in disease activity, documentation of glucocorticoid management plan within 12 months.ACR
Percentage of patients 18 years and older with a diagnosis of rheumatoid arthritis for whom a functional status assessment was performed at least once within 12 months.ACR
OutcomeIf a patient is diagnosed with active RA (i.e., DAS-28 over 3.2), then the disease activity should be low (i.e., DAS-28 below 3.2), 6 months after treatment has
Follow upIf a patient is diagnosed with RA and the target (remission or low disease activity) is not attained then follow up visit should be scheduled by a rheumatologist within 3 months and when the target is attained a rheumatologist or a specialized nurse in rheumatology should schedule follow up visits at least once a

a Indicators are reported precisely as used by corresponding authors of each published set.

In the United States, the Arthritis Foundation and the ACR have also developed sets of indicators. The extensive set from the Arthritis Foundation comprises 27 process indicators and they can be divided into 17 domains such as time to referral, history and examination, regular follow-up, radiographs of hand and feet, radiographs of cervical spine, DMARDs, folic acid with methotrexate (MTX), osteoporosis prophylaxis, use of glucocorticoids, exercise, assistive devices, surgery, baseline and follow-up studies, methotrexate transminitis (increase in aminotransferases), informing patients about risks (such as risks regarding the use of non-steroidal anti-inflammatory drugs (NSAIDs), DMARDs, glucocorticoids and narcotics), reproductive issues, and finally vaccines [21]. The set developed by the ACR includes five process indicators and one outcome indicator. We have grouped these indicators in the following domains: screening, pharmacological treatment, and monitoring; to our understanding, no domains were proposed by the research team. The indicators from this set are also stated in Table 1 for illustrative purposes [22].

An international task force developed a set of 10 quality indicators, using the Measurement of Efficacy of Treatment in the Era of Outcome in Rheumatology (METEOR) database. This set consists of seven process indicators and three outcome indicators: time to diagnosis, antibodies and radiographic assessment, frequency of visits, disease activity assessment, functional status assessment, remission of disease activity (clinical remission), low disease activity, level of functional limitation, time to first DMARD, and type of first DMARD [23].

Finally, the Australian Rheumatoid Association has proposed a set of three process indicators. These indicators cover measurement of disease activity and comorbidities [24].

It is evident from these seven sets that the majority of these indicators include process measures. Of the 82 indicators, only nine were outcome indicators [18], [19], [20], [21], [22], [23], and [24]. The majority report the “number of times” a certain outcome is measured (process), rather than the actual outcomes themselves. Referring to the triad suggested by Donabedian, which links process, structure, and outcome of care to each other, an imbalance between the different types of indicators in the current sets is quite apparent [2] and [7]. Furthermore, the availability of many sets for selection may further jeopardize the implementation in daily practice. In conclusion, a better-balanced (more outcome, less process) and more widely accepted indicator set would be instrumental in achieving a uniform measurement of RA care. In the meantime, a rheumatologist willing to measure the quality of his own practice should choose one of the available indicator sets which best reflects what one wants to measure.

D. Is optimal care delivered to RA patients?

As described in this review, the use of tight control strategies is beneficial to RA patients and major treatment guidelines have embraced the tight control strategy [11], [13], and [15]. Unfortunately, the existence of these guidelines and the underlying evidence for their efficacy seem to be insufficient to ensure application of tight control in daily practice. This issue has been addressed by several studies, with rather underwhelming results. Here we will briefly summarize some of these studies.

Benhamou et al. assessed the potential gap between daily practice and recommendations on first DMARD prescription in RA in the French multicenter ESPOIR cohort [25]. This cohort included early RA patients between 2002 and 2005, and during this period two guidelines on DMARD treatment in early RA were introduced: the national guideline by the French Society of Rheumatologists (STPR guideline) and the international set of management recommendations by the EULAR. Benhamou et al. observed that the first DMARD prescriptions in early RA were followed according to STPR recommendations in 58% of the patients, while 54% of the prescriptions adhered to the EULAR guidelines. As both guidelines were presented at international conferences at the end of the ESPOIR inclusion period, the authors concluded that the potential gap between evidence and practice was substantial [25].

Around the same time, another European study assessed treatment patterns in early RA patients (ERAN cohort) [15]. In this cohort, 97% of the patients were prescribed a DMARD; however, median time between onset of the RA symptoms and DMARD prescription was 8 months. Most often, the first DMARD was prescribed as monotherapy (91%) and the addition of a second DMARD later in the treatment course was observed in 48%. Despite the high percentage of DMARD users, only 33% of the patients met the DAS-28 remission criteria after 3 years [15].

Furthermore, in the United States, the prescribing practices of rheumatologists were assessed. In contrast to the ESPOIR and ERAN cohort, this study also included biologic DMARD (bDMARD) prescriptions and compared adherence before and after the publication of the ACR treatment recommendations [26]. In this study, 43% and 51% of the MTX monotherapy users with moderate disease activity as well as poor prognosis and high disease activity, respectively, received care according to the ACR guideline. In patients using multiple conventional synthetic DMARDs (csDMARDs), 43% and 51% of those with moderate and high disease activity, respectively, received care consistent with the ACR guideline. Interestingly, the publication of the ACR guidelines did not result in improved guideline adherence in patients with active disease [26].

The results of these three studies may seem rather disappointing. However, a Dutch study on guideline adherence in the DREAM remission induction cohort yielded more positive results [27]. In this early RA cohort, adherence to adequate monitoring of disease activity (DAS-28 assessed at least every 3 months) and a predefined treatment protocol was assessed. The researchers observed that adequate monitoring of disease activity took place in 88% of the visits and the rheumatologist adhered to the treatment recommendations in 69%. According to the authors, these results point to the feasibility of using a tight control strategy in daily practice [27].

Finally, in another study using data from the ESPOIR cohort, it was observed that adherence to tight control strategies in daily practice may have real benefits for patients [28]. It was found that early-RA patients who were not treated according to the 2007 EULAR recommendations on early RA were at an increased risk of radiographic progression at 1 year and functional impairment at 2 years (odds ratio (OR) 1.98; 1.08–3.62) and (OR 2.36; 1.17–4.67 respectively) [28]. In addition, two studies presented at the 2014 ACR annual meeting concluded that RA disease control was better in patients in whom a tight control strategy was actively applied than in patients in whom this was not the case. [29] and [30].

From the above-described literature data, we can conclude that application of a tight control strategy in daily practice is feasible, but general adherence is not yet optimal [14], [24], [25], [26], [27], [28], and [29]. In addition, it was observed that suboptimal adherence may have negative consequences for patients with regard to disease control, radiographic progression, and functional status.

E. How can you improve RA care?

We have learned from various studies described in the previous section that optimal RA care is not always delivered to patients. The main issue now lies in the methods to improve RA care, for which we need to consider the field of implementation science, an area specifically focusing on bridging the gap between evidence and clinical practice. First, the factors that influence the successful uptake of evidence by physicians (also called “barriers and facilitators” to implementation) are explained, followed by the potential interventions to change clinical practice.

Factors influencing adherence to evidence-based recommendations

Many studies have made an attempt to understand barriers and facilitators associated with implementing change and consequently, different checklists, frameworks, and taxonomies have been developed. In 2013, a systematic review by Flottorp et al. was published, describing the development of a comprehensive checklist (TICD checklist), integrating previous checklists, frameworks, and taxonomies [31]. In this review, “barriers and facilitators to implementation” are called “determinants of practice”, and the developed checklist consists of 54 of such determinants. The determinants are subdivided into seven domains, namely guideline factors; health professional factors; patient factors; professional interactions; incentives, and recourses; capacity for organizational change; and social, political and legal factors (Table 2). According to the authors, this checklist can be used as a screening tool to identify determinants that need further investigation before a particular change can be implemented. The underlying idea is that this can facilitate the development and evaluation of interventions tailored to specific determinants [31]. This hypothesis will be discussed further when potential interventions to change practice are described.

Table 2

Overview of the domains and determinants of the TICD checklist.

Domain 1: Guideline factors
RecommendationQuality of evidence supporting the recommendation
Strength of recommendation
Cultural appropriateness
Accessibility of the recommendation
Source of the recommendation
Consistency with other guidelines
Recommended clinical interventionFeasibility
Accessibility of the intervention
Recommended behaviorCompatibility
Domain 2: Individual health professional factors
Knowledge and skillsDomain knowledge
Awareness and familiarity with the recommendation
Knowledge about own practice
Skills needed to adhere
Cognitions (including attitudes)Agreement with the recommendation
Attitudes toward guidelines in general
Expected outcome
Intention and motivation
Learning style
Professional behaviorNature of the behavior
Capacity to plan change
Self-monitoring or feedback
Domain 3: Patient factors
n/aPatient needs
Patient beliefs and knowledge
Patient preferences
Patient motivation
Patient behavior
Domain 4: Professional interactions
n/aCommunication and influence
Team processes
Referral processes
Domain 5: Incentives and resources
n/aAvailability of necessary resources
Financial incentives and disincentives
Nonfinancial incentives and disincentives
Information system
Quality assurance and patient safety systems
Continuing education system
Assistance for clinicians
Domain 6: Capacity for organizational change
n/aMandate, authority, accountability
Capable leadership
Relative strength of supporters and opponents
Regulations, rules, policies
Priority of necessary change
Monitoring and feedback
Assistance for organizational changes
Domain 7: Social, political and legal factors
n/aEconomic constraints on the health-care budget
Payer or funder policies
Malpractice liability
Influential people
Political stability

As described in Table 2, many factors can influence the translation of research into practice, with the diffusion and dissemination of innovations facilitated or hindered by these factors. The TICD provides a comprehensive overview of these factors, but for a practicing rheumatologist in search of feasible tools to improve his own practice, it is difficult to implement this checklist. In our opinion, and partly based on personal experience, we would suggest that successful implementation of change in daily practice relies mainly on three domains: 1) knowledge about the desired change, 2) motivation to realize the change, and 3) being able to apply the new behavior.

Potential interventions to improve care

Recommendations or guidelines can assist rheumatologists in providing optimal clinical care to RA patients, but we have also seen that implementation of guidelines is often difficult [1], [14], [24], and [25]. In the previous section, we have discussed studies that have identified many factors that may influence the successful implementation of change. Now the question remains as to what types of interventions could be applied to improve care and their effectiveness.

Before answering this question, we will first summarize the types of interventions that do exist. The Effective Practice and Organization of Care (EPOC) group is a Cochrane review group specialized in undertaking reviews on all types of interventions that aim to improve health professional practice ( To aid researchers and clinicians in classifying their interventions, the EPOC has developed a taxonomy which can be used as a framework for characterizing interventions [32]. Interventions are first divided into three different categories reflecting the health-care system: interventions targeted at i) health-care organizations; ii) health-care workers; or iii) specific types of practice. Each of these categories can be further divided into different subcategories, but since this review focuses specifically on rheumatologists, we will only discuss the second category, that is, the interventions targeted at health-care workers. Audit and feedback, educational meetings, and reminders are a few examples of these interventions [31]. A full overview of the EPOC taxonomy on interventions targeted at health-care workers is given in Table 3.

Table 3

EPOC taxonomy of interventions targeted at health-care workers.

InterventionDescription a
Audit and feedbackA written, electronic or verbal summary given to health-care providers, regarding their performance over a specified period of time
Clinical incident reportingA system where critical incidents can be reported
Monitoring the performance of the delivery of health careMonitoring of services provided by individuals/health-care organizations
Communities of practiceA group with common interest who aim at gaining knowledge in this area by interacting on an ongoing basis
Continuous quality improvementAn ongoing process to review and improve care including involvement of health-care teams, analysis of a process or system, a structured process improvement method or problem-solving approach, and use of data analysis to assess changes.
Educational gamesIn order to improve standards of care, games can be used as an educational strategy.
Educational materialsDistribution of educational materials to support clinical care.
Educational meetingsEducational meetings such as courses, conferences or workshops
Educational outreach visits, or academic detailingProviding information by a trained person, during personal visits, with the aim of changing practice
Clinical practice guidelinesSystematically developed statements to assist health-care providers and patients
Inter-professional educationContinuing education for health professionals, involving participants from different professions.
Local consensus processesFormal or informal consensus processes, for example, agreeing to a clinical protocol or promoting the implementation of guidelines.
Local opinion leadersPromoting good clinical care by identifying and using local opinion leaders
Managerial supervisionRoutine supervision visits by health staff
Patient-mediated interventionsEngage input from patients to change professional practice
Public release of performance dataRelease of performance data, to inform the public about the practice of health-care providers
RemindersManual of computerized reminders prompting health-care providers to perform an action
Routine patient-reported outcome measuresReporting and administration of patient- reported outcome measures
Tailored interventionsDevelopment of interventions to change practice, based on assessment of barriers for a specific clinical setting

a The description is derived from the EPOC taxonomy of interventions targeted at health-care workers.

As can be seen in Table 3, many different types of interventions could be used when implementing, for example, a new RA treatment guideline. As a consequence, many studies have been conducted on the effectiveness of these interventions. An in-depth discussion regarding all these studies is beyond the scope of this article. But in a systematic review published in 2004, the available evidence was summarized [32]. This review concluded that the overall quality of the studies was poor and that the intervention effects varied considerably both across and within interventions, making it impossible to give evidence-based recommendations on when to use which type of intervention [33]. Since the publication of this review, many additional studies have been conducted and new systematic reviews were published as well. We will describe the results of the reviews on the three most performed interventions: educational meetings, audit and feedback, and reminders.

Educational meetings

The EPOC systematic review on educational meetings was published in 2006 and included 81 trials with over 11,000 included health professionals [34]. When the researchers compared “no intervention” with “any intervention including educational meetings as a component”, the risk difference (RD) for compliance with the desired practice was 6% (interquartile range: 1.8–15.9). For interventions with educational meetings alone, similar results were obtained. In addition, analyses were done on determinants of success. It was observed that higher attendance and a mix between interactive and didactic educational meetings were associated with a higher RD. In contrast, educational meetings targeted at complex behavior or at perceived less serious outcomes seemed to be less effective [34].

Audit and feedback

Another EPOC review focused on audit and feedback and this review included 140 different studies [35]. Comparing “interventions with audit and feedback” as an essential component to “no intervention” resulted in a 4.3% increase in the desired behavior (interquartile range: 0.5%–16%). Determinants of success were low baseline performance; the audit and feedback being provided by a colleague, being provided more than once (“booster” session, or regular feedback), or being delivered in verbal and written format; and finally when it includes specific targets or action plans [35].


On-screen computer reminders are often used as an intervention and their effectiveness has also been evaluated by EPOC [36]. Reminders resulted in a median improvement of 4.2% (interquartile range: 0.8%–18.8%) when all process outcomes were taken together. Considering the separate outcomes, the median improvement in adherence was 3.3% for medication ordering, 3.8% for vaccinations, and 3.8% for test ordering. A few studies also reported clinical outcomes (e.g., blood pressure) and the median absolute improvement for these outcomes was 2.5%. In this review, none of the specific reminders or contextual factors were significantly associated with the magnitude of the observed effects [36].

Tailored interventions

Taking into account the abovementioned reviews and the determinants of successful implementation, one might wonder if interventions specifically tailored to facilitators or barriers of implementation are more effective than non-tailored interventions. The EPOC tried to answer this question as well, but studies directly comparing “tailored interventions” to “non-tailored interventions” or “no intervention at all” were scarce, and a definitive conclusion was not possible. In spite of this scarcity, the conclusion was that tailored interventions can be effective, but that their effect is variable and tends to be small to moderate at best [37].

In summary, this section described different interventions that could be used when trying to improve quality in clinical practice. The effects of any one of these different interventions are often small to moderate, but some evidence exist that a combination of different interventions is more effective [33]. Despite the lack of evidence on which specific intervention should be used in which situation, the reviews may provide some guidance as to how to choose an intervention for a desired change in daily practice. Based on the information in this section, we would advise to practicing rheumatologists willing to change their practice that, after measuring a chosen indicator set, a simple barrier analysis is done using the before-mentioned themes (knowledge, motivation, and being able to change) followed by a intervention tailored to these barriers. If possible, a combination of different intervention types is preferred above a single component intervention.

F. What initiatives exist to specifically improve RA care?

Several studies have tried to specifically improve RA care by introducing different types of interventions. To our knowledge, there is no complete overview of these studies available, so we will discuss three different studies that describe some kind of intervention designed to improve care.

The first example is a pilot study aiming at improvement of disease activity and medication prescription in RA patients by implementing nurse-led DAS-28 measurements [38]. As stated in this study, the assessment of a combined disease activity index like the DAS-28 is often perceived as too complex to collect and calculate regularly in a busy day-to-day clinical practice. Therefore, the researchers hypothesized that delegating DAS-28 measurement to a well-trained specialized nurse could be of value, since this would save time for the rheumatologist and enhances adherence [38]. In this study, three rheumatologists were randomized to the control group and four to the intervention group. In both groups, the DAS-28 was performed and calculated by a specialized nurse, but only in the intervention group the DAS-28 score was provided to the rheumatologist (before the patient visited the rheumatologist). This was combined with a general advice to the rheumatologists to adjust DMARD therapy in case of active disease (DAS-28 ≥ 3.2). In order to investigate whether this intervention was effective, the change in DAS-28 score and the number of medication changes were used as primary outcome measures. After 18 months, a decrease of 0.66 and 0.69 DAS-28 points was found in the usual care group and in the intervention group, respectively (p = 0.7). In addition, no significant differences in number of medication changes were observed between the two groups (33% in the usual care group vs. 35% in the intervention group; p = 0.99) [38]. According to the authors, one of the reasons for the failure of this intervention was the absence of a strict treatment protocol. Therefore, they concluded that nurse-led care may be useful in making DAS-28 assessments more feasible for use in daily practice, but that rheumatologists should be encouraged to change medication when necessary, using the individual DAS-28 values provided to them [38].

Another example of an attempt to change clinical practice is the Metrix study [39]. This study assessed the effect of an educational intervention on rheumatologists׳ practice behavior in RA patients. In this study, 20 rheumatologists participated, and they all had to perform a prospective chart audit of 50 consecutive RA patients at study start. During this chart audit, they had to collect information on patient demographics, current and previous DMARD use, and measurements done during the visit such as tender and swollen joint counts. After the chart audit, the rheumatologists were randomized in an intervention group and a control group (10 in both groups) [39]. Rheumatologists in the control group received no further interventions and also did not receive their results from the chart audit. In contrast, rheumatologists in the intervention group received the chart audit results as feedback, and had to attend monthly web-based conferences and a journal club. During the web-based conferences and journal club sessions, topics such as the value of systematic assessment in RA and the value of using tight control strategies were discussed. After 6 months, the intervention group collected more global assessments (a 13% increase post-intervention for patient global and an increase of 9% in the physician global; p < 0.05) and health assessment questionnaires (increase of 5%, p < 0.05), whereas the control group did not show any changes. Furthermore, a significant increase of 32% was seen in the presence of calculable composite scores (any version of CDAI, SDAI, or DAS); no change in the control group was observed [39]. In patients with active disease (either defined by SDAI or by DAS), therapy was changed more often in the intervention group than in the control group (66% vs. 36% for patients with SDAI between 3.3 and 11; 57% vs. 38% for patients with a DAS between 2.4 and 3.6, respectively) [39]. The results of this study show that the combination of feedback and educational meetings can improve daily clinical practice. Interestingly, the rheumatologists in the control group, who also reviewed 50 of their own charts, did not change at all. This implies that in order to change practice, only reviewing your own work, without receiving its results and additional education, is not enough [39].

In the third and final study, Ledwich and colleagues determined the effects of an Electronic Health Record (EHR) best practice alert (BPA) on vaccination rates in patients with a rheumatic disease using an immunosuppressive drug [40]. In this study, the BPA was a clinical reminder on influenza and pneumococcal vaccinations. When an adult patient using immunosuppressive drugs visited the clinic during the influenza season, the BPA appeared prominently on screen to remind the rheumatologist about an influenza vaccination. For pneumococcal vaccinations, the BPA only appeared in patients using an immunosuppressive drug if the patient did not received a vaccination before. With both BPAs, rheumatologists were also able to document why a vaccination was not given. After implementation of the BPA, vaccination rates increased significantly with influenza vaccination rates from 47% to 65% and pneumococcal vaccinations from 19% to 41%. Based on these results, the authors stated that the BPA is an effective tool for improving quality of care for patients receiving immunosuppressive drugs [40].

Together these three studies show that initiatives are started within rheumatology evaluating strategies to improve quality of care for patients with RA or rheumatic diseases in general. Of these three studies, the latter two have demonstrated beneficial effects with regard to improving care, whereas the first study suggests that monitoring alone without a strict treatment protocol is not effective enough to truly change practice behavior. It is evident from the above studies that efforts have been made to improve the quality of RA care, but the goal has not been achieved yet.

Our conclusions from the previous sections are based on results of published studies. However, it is likely that many initiatives made to improve quality of care have not been detailed in international peer-reviewed journals (publication bias). Two of such quality improvement initiatives previously described only in the Dutch literature will be discussed here.

Both of the described initiatives have been taken from the Sint Maartenskliniek (SMK) in the Netherlands. In this specialized clinic for rheumatology, orthopedic surgery, and rehabilitation medicine, the rheumatology department has implemented nurse-led DAS-28 assessments for all RA patients visiting the outpatient clinic, starting in 2010. Since then, RA patients arrive at the clinic 1 h before their visit with the rheumatologist. Upon arrival, blood is drawn for routine laboratory testing and the patient is seen by a specialized nurse. During this visit, the DAS-28 and HAQ are performed, and the current medication of the patient is discussed in order to identify any side effects or changes that have occurred since the last visit. All the information gathered during the visit with the nurse is provided to the rheumatologist prior to his or her consultation with the patient, increasing the efficiency [41]. At the same time, a locally developed RA treatment protocol also became available to the rheumatologists. This protocol included a strict, tight control-based, set of treatment recommendations, explicitly stating which DMARD should be given in what order and when. For example, according to this protocol, patients with newly diagnosed RA should start with methotrexate and hydroxychloroquine combination therapy. When this first combination fails, etanercept and methotrexate should be given. Till now, 82% of the RA patients combine their visit with the rheumatologist with a nurse-led DAS-28 assessment. This has resulted in DAS-28 measurements being available in 85% of the visits. Notably, 72% of all RA patients (irrespective of disease duration) have a DAS-28-CRP <2.9, and 60% < 2.4.

After the implementation of nurse-led DAS-28 assessments in the department, it became evident that to obtain reliable DAS-28 scores across the different health-care professionals (nurses, physician assistants, residents, and rheumatologists) who performed the DAS-28, an acceptable level of agreement in the DAS-28 scores is mandatory. Furthermore, as rheumatologists did not perform the DAS-28 themselves anymore, they had to learn to rely on the nurses. Therefore, an interactive and competitive DAS-28 training was designed to increase inter-observer agreement and to improve mutual confidence between rheumatologists and nurses. In this so-called “DAS-28 battle” using elements of serious gaming, rheumatologists and nurses were first trained by an experienced rheumatologist to perform the joint counts needed to calculate the DAS-28. Next, the participants were divided into small groups and were asked to perform a tender and swollen joint count in four different patients. In every group of patients, one “fake patient” was present (usually a partner of a real RA patient) and these persons served as “healthy controls”. This extra twist was added to assess the number of false-positive joints (joint scored as swollen in a healthy control). Blood tests were available, so DAS-28 scores could be calculated and compared immediately. Measurement error and the number of false-positive joints were calculated per team, and the team with the best score on both items was awarded the “Golden Hand” [42]. Since 2009, two DAS-28 battles have been organized, and during the first battle the measurement error ranged between 0.16 and 0.44 (mean of 0.31), and the number of false-positive joints varied between 0 and 4 per team. Three years later, the battle was repeated, giving a similar mean measurement error but the variation between the teams had decreased (measurement error between 0.29 and 0.36). Due to changes in the health-care professionals working at the SMK, the positive effects of the battle on both inter-observer agreement and trust, and the enthusiasm of the participants, the DAS-28 battle will be organized again in the near future.

G. Practical implications

In this review, we have provided an overview of the current status of quality of care in RA. In the first section, we saw that defining and measuring quality of care can be challenging and that different types of quality indicators exist. Next, we described current treatment strategies used in RA, which are based on tight control. Different sets of indicators to measure RA care were discussed thereafter, which was followed by describing different studies assessing adherence to tight control recommendations. Unfortunately, adherence turned out to be suboptimal in most cases. In the subsequent section, we have introduced the field of implementation science that has addressed this issue before, and examples of effective interventions were given. Finally, some of these interventions that are already applied in RA and have led to improvements in care provided to patients were described.

In our opinion, these data show that evidence for the most effective RA treatment is available and that rheumatologists are willing to use this evidence in order to treat their patients to the best they can, but they need to be assisted in doing so. So, how could rheumatologists be assisted in improving their own RA care? In this final section, we will address this theme from two different points of view: the researcher׳s view and the practicing rheumatologist׳s view.

The researcher׳s view

We have seen that different groups around the world have developed RA quality indicators. However, none of these indicators sets are universally accepted, making it difficult to use them in research. While the indicator set formed by the international taskforce of METEOR attempts a more international approach, much more still needs to be done [23]. Therefore, we need a set of clear and internationally accepted indicators to gain more insight into the processes and outcomes of RA care. In addition, this indicator set should also incorporate outcome indicators next to process indicators. For the development of a universally applicable indicator set, based on international consensus, a joint, international taskforce under the auspices of international societies (such as ACR and EULAR) is needed. This indicator set may not only help in measuring care, but also in improving care and in the evaluation of improvement interventions.

In addition, far more attention should be given to the translation of evidence into practice. Here we refer to Buchbinder et al., who have recently stated that “investment in discovery research is essentially wasted if implementation research is ignored” [1]. Therefore, we would make an urgent call to all stakeholders involved in rheumatology research to invest in studies trying to find effective interventions that improve the quality of care for RA patients.

Finally, policy makers and developers of guidelines or practice recommendations should be more aware of the fact that only disseminating guidelines does not suffice to ensure uptake of recommendations in clinical practice. Therefore, any new or updated version of a guideline should be accompanied by an implementation plan or at least some recommendations on how to implement the guideline in daily clinical practice. The AGREE (Appraisal of Guidelines Research & Evaluation) tool is a helpful aid when developing a guideline as the AGREE gives recommendations to develop a high quality guideline. Besides recommendations on topics such as clarity and presentation of the guideline, AGREE also stresses the need for an implementation plan as a supplement with a guideline. In addition, additional materials could be useful, such as a summary document, educational tools or computer support. These additional materials should be provided with the guidelines in order to enhance their use [43]. Furthermore, it might be needed that financial incentives are available to facilitate implementation of change.

The Rheumatologist׳s view

In this review, we have tried to answer the question “how can a rheumatologist improve his or her own practice”. Unfortunately, the literature is inconclusive and research in rheumatology is scarce. This problem has been addressed in the previous sections, and we will now propose some simple steps that rheumatology practices could use to improve their quality of care.

When changing current practice, a first step would be to define a manageable goal. For example, “treat your RA patients in such a way that you achieve low disease activity in 60% of all your RA patients after one year”. Of note, the goal of 60% in this example is arbitrary and not based on evidence. Unfortunately, we do not know whether 60% is optimal; however, based on clinical trials this could be a feasible goal to start implementation with.

Next, it is necessary to check the availability of resources to reach the preset goal. For example, see if an up-to-date local RA treatment guideline is available and, if not, try to see that such a guideline will become available. In our experience, guidelines are easier to use in daily practice if they include brief and specific descriptions of what to do in specific situations rather than elaborating on the underlying evidence. For example, provide a step-by-step description on what to do in a patient with active disease. When such a new or updated guideline is finished, all relevant stakeholders (nurses, residents, rheumatologists, pharmacists, etc.) should be informed about this. If necessary, additional actions such as an educational meeting may be needed to improve implementation of the guideline. Apart from a clear treatment guideline, additional resources might be needed. As observed in the TICD checklist by Flottorp, many potential barriers to adherence exist [31]. However, for daily practice, it is not feasible to do a full barrier analysis. Therefore, we recommend to only focus on issues that have a potentially large impact on the results and that can relatively easy be implemented. For example, if a specialized nurse or physician assistant is available, “shared care” between them and the rheumatologists might be an option.

After all necessary actions are implemented, the next step would be to check at specified intervals if the preset goal has been met. Often, appropriate information for this check is not readily available from the existing systems such as the EHR. One solution could be to conduct a chart review after, for example, 6 months (medical audit), collecting data on disease activity and on what has been done in response. Such a chart review can be very labor intensive if done in many patients, but for feedback purposes a sample from the total patient population is often enough. Using the local treatment protocol, a few aspects of RA care could be checked for (“Is disease activity measured during every routine visit?”; “Is DMARD medication changed in response to active disease?’ and ‘Is low disease activity present?”). Individual data from the patient׳s charts can then be aggregated in order to see if the preset goal has been reached. This chart review would be most useful if all rheumatologists in one practice are involved and individual results are compared. Of note, a safe learning environment is critical when comparing non-anonymized performances among rheumatologists.

Almost always such a chart review will reveal that not all the care is in accordance with the guidelines or the preset goal. Additional measures may be needed to further improve the quality of care and reach the preset goal. When finally those additional measures have resulted in meeting your goal, a new cycle starts and continuous evaluation will be necessary to maintain quality improvement.

The above-mentioned steps are also known as the Plan, Do, Check, Act (PDCA) cycle or the Define, Measure, Analyze, Improve, Control (DMAIC) cycle [44] and closely resemble the tight control strategy used in RA treatment itself, as this cycle also involves goal setting, measuring, and acting if the goals have not yet been reached. The application of such strategies in both patients and rheumatologists may result in a better translation of evidence into practice, and consequently guarantee the best possible care we could provide to our patients.

Practice points
  • The translation of scientific evidence on optimal care for patients with RA from clinical trials into clinical practice is difficult, and often suboptimal.
  • To bridge this gap, we need a set of clear and internationally accepted indicators to measure the current quality of care, more research should be done to identify effective interventions and implementation of guidelines should be an integral part of guideline development.
  • In the meantime, rheumatologists wanting to improve their own RA care should start with selecting or developing a local, tight control-based RA treatment protocol as this has shown to have beneficial effects on patient outcomes.
  • When implementing such a protocol in daily practice, a “tight control” strategy can be used comprising three steps: 1) setting a goal and assessing whether this goal is already met, 2) implementing changes if the goal is not yet reached and keep adjusting this until the goal has been reached, and 3) continuously evaluating such preset goals in order to maintain optimal quality of care.

Conflicts of interest

The authors have no conflicts of interest to declare.


  • [1] R. Buchbinder, C. Maher, I.A. Harris. Setting the research agenda for improving health care in musculoskeletal disorders. Nature Reviews Rheumatology. 2015;11:597-605
  • *[2] A. Donabedian. Quality of care. How can it be assessed?. JAMA. 1988;260(12)
  • [3] D.C.J. Nash, A. Skoufalos, M. Horowitz. Conceptualizations and definitions of quality. Health care quality: the clinician׳s primer: American college of physician executives. (, 2012) 412
  • [4] Medicine Io. Crossing the quality chasm: a new health system for the 21st century. (, 2001) Washington (DC)
  • [5] J. Markhorst, L. Martirosyan, H. Calsbeek, et al. Stakeholders׳ perspectives on quality indicators for diabetes care: a qualitative study. Quality in Primary Care. 2012;20(4):253-261
  • [6] M. Lawrence, F. Olesen. Indicators of quality in health care. European Journal of General Practice. 1997;3:103-108
  • [7] A. Donabedian. Explorations in quality assessment and monitoring. (Health Administration Press, Ann Arbor, Michigan, 1980)
  • [8] S.J. Lee, A. Kavanaugh. Pharmacological treatment of established rheumatoid arthritis. Best Practice & Research Clinical Rheumatology. 2003;17(5):811-829
  • [9] J.F. Fries. Current treatment paradigms in rheumatoid arthritis. Rheumatology. 2000;39(Suppl. 1):30-35
  • [10] L. Abasolo Alcazar. Triple therapy in rheumatoid arthritis. Reumatologia Clinica. 2014;10(5):275-277
  • *[11] M.F. Bakker, J.W. Jacobs, S.M. Verstappen, et al. Tight control in the treatment of rheumatoid arthritis: efficacy and feasibility. Annals of the Rheumatic Diseases. 2007;66(Suppl. 3):iii56-60
  • [12] J.A. van Nies, A. Krabben, J.W. Schoones, et al. What is the evidence for the presence of a therapeutic window of opportunity in rheumatoid arthritis? A systematic literature review. Annals of the Rheumatic Diseases. 2014;73(5):861-870
  • [13] J.S. Smolen, D. Aletaha, J.W. Bijlsma, et al. Treating rheumatoid arthritis to target: recommendations of an international task force. Annals of the Rheumatic Diseases. 2010;69(4):631-637
  • *[14] J.S. Smolen, F.C. Breedveld, G.R. Burmester, et al. Treating rheumatoid arthritis to target: 2014 update of the recommendations of an international task force. Annals of the Rheumatic Diseases. 2015;:1-13
  • [15] P. Kiely, R. Williams, D. Walsh, et al. Early rheumatoid Arthritis N. Contemporary patterns of care and disease activity outcome in early rheumatoid arthritis: the ERAN cohort. Rheumatology. 2009;48(1):57-60
  • *[16] L.G. Schipper, L.T. van Hulst, R. Grol, et al. Meta-analysis of tight control strategies in rheumatoid arthritis: protocolized treatment has additional value with respect to the clinical outcome. Rheumatology. 2010;49(11):2154-2164
  • [17] M.A. Stoffer, M.M. Schoels, J.S. Smolen, et al. Evidence for treating rheumatoid arthritis to target: results of a systematic literature search update. Annals of the Rheumatic Diseases. 2015;:1-7
  • [18] L.T. van Hulst, J. Fransen, A.A. den Broeder, et al. Development of quality indicators for monitoring of the disease course in rheumatoid arthritis. Annals of the Rheumatic Diseases. 2009;68(12):1805-1810
  • [19] N. Employers. 2014/15 general medical services (GMS) contract quality and outcomes framework (QOF). (, 2014) 120-122 NHS England Gateway reference: 01264
  • [20] I.F. Petersson, B. Strombeck, L. Andersen, et al. Development of healthcare quality indicators for rheumatoid arthritis in Europe: the project. Annals of the Rheumatic Diseases. 2014;73(5):906-908
  • [21] C.H. MacLean, K.G. Saag, D.H. Solomon, et al. Measuring quality in arthritis care: methods for developing the arthritis foundation׳s quality indicator set. Arthritis and Rheumatism. 2004;51(2):193-202
  • [22] National Committee for Quality Assurance PCfPI. American college of rheumatology. Rheumatoid arthritis physician performance measurement set American medical association and national committee for quality assurance. (, 2008)
  • [23] V. Navarro-Compan, J.S. Smolen, T.W. Huizinga, et al. Quality indicators in rheumatoid arthritis: results from the METEOR database. Rheumatology. 2015;54(9):1630-1639
  • [24] Conditions NCfMAaM. National indicators for monitoring osteoarthritis, rheumatoid arthritis and osteoporosis. (Australian Institute of Health and Welfare, Canberra, 2006)
  • [25] M. Benhamou, N. Rincheval, C. Roy, et al. The gap between practice and guidelines in the choice of first-line disease modifying antirheumatic drug in early rheumatoid arthritis: results from the ESPOIR cohort. The Journal of Rheumatology. 2009;36(5):934-942
  • [26] L.R. Harrold, J.T. Harrington, J.R. Curtis, et al. Prescribing practices in a US cohort of rheumatoid arthritis patients before and after publication of the American college of rheumatology treatment recommendations. Arthritis and Rheumatism. 2012;64(3):630-638
  • *[27] M. Vermeer, H.H. Kuper, H.J. Bernelot Moens, et al. Adherence to a treat-to-target strategy in early rheumatoid arthritis: results of the DREAM remission induction cohort. Arthritis Research and Therapy. 2012;14(6):R254
  • [28] C. Escalas, M. Dalichampt, B. Combe, et al. Effect of adherence to European treatment recommendations on early arthritis outcome: data from the ESPOIR cohort. Annals of the Rheumatic Diseases. 2012;71(11):1803-1808
  • [29] W.ø.M. Maksymowych, O. Elkayam, R. Landewé, et al. Impact of failure to Adhere to treat-to-target of rheumatoid arthritis in real world practice: data from the international rheumatoid arthritis biomarker program. (, 2014) Abstract Number: 2912
  • [30] L. Kuusalo, K. Puolakka, H. Kautiainen, et al. Impact of physicians׳ adherence to treat-to-target strategy on outcomes in early rheumatoid arthritis in the NEO-RACo trial. Scandinavian Journal of Rheumatology. 2015;:1-7
  • *[31] S.A. Flottorp, A.D. Oxman, J. Krause, et al. A checklist for identifying determinants of practice: a systematic review and synthesis of frameworks and taxonomies of factors that prevent or enable improvements in healthcare professional practice. Implementation Science. 2013;8:35
  • *[32] Collaboration C. Effective practice and organisation of care (EPOC). EPOC Taxonomy. 2015;:1-11 Cochrane Collaboration, 2015
  • *[33] J.M. Grimshaw, R.E. Thomas, G. MacLennan, et al. Effectiveness and efficiency of guideline dissemination and implementation strategies. Health Technology Assessment. 2004;8(6):1-72 iii-iv
  • [34] L. Forsetlund, A. Bjorndal, A. Rashidian, et al. Continuing education meetings and workshops: effects on professional practice and health care outcomes. The Cochrane Database of Systematic Reviews. 2009;2:CD003030
  • [35] N. Ivers, G. Jamtvedt, S. Flottorp, et al. Audit and feedback: effects on professional practice and healthcare outcomes. The Cochrane Database of Systematic Reviews. 2012;6:CD000259
  • [36] K.G. Shojania, A. Jennings, A. Mayhew, et al. The effects of on-screen, point of care computer reminders on processes and outcomes of care. The Cochrane Database of Systematic Reviews. 2009;3:CD001096
  • [37] R. Baker, J. Camosso-Stefinovic, C. Gillies, et al. Tailored interventions to address determinants of practice. The Cochrane Database of Systematic Reviews. 2015;4:CD005470
  • [38] L.T. van Hulst, M.C. Creemers, J. Fransen, et al. How to improve DAS28 use in daily clinical practice?–a pilot study of a nurse-led intervention. Rheumatology. 2010;49(4):741-748
  • [39] J. Pope, C. Thorne, A. Cividino, et al. Effect of rheumatologist education on systematic measurements and treatment decisions in rheumatoid arthritis: the metrix study. The Journal of Rheumatology. 2012;39(12):2247-2252
  • [40] L.J. Ledwich, T.M. Harrington, W.T. Ayoub, et al. Improved influenza and pneumococcal vaccination in rheumatology patients taking immunosuppressants using an electronic health record best practice alert. Arthritis and Rheumatism. 2009;61(11):1505-1510
  • [41] A. den Broeder, E. van den Ende, I. Schasfoort, et al. Implementatie van klinimetrieondersteuning in de reumatologische praktijk. Nederlands Tijdschrift voor Reumatologie. 2011;:26-31
  • [42] V.L.N. Straten, A. den Broeder. Naar een eenduidige uitvoering van de DAS28: de ׳DAS28-battle׳. Nederlands Tijdschrift voor Reumatologie. 2012;:10-13
  • *[43] Collaboration TA. Appraisal of guidelines for research & evaluation. (St George’s Hospital Medical School, London, 2001)
  • *[44] C.R. Nicolay, S. Purkayastha, A. Greenhalgh, et al. Systematic review of the application of quality improvement methodologies from the manufacturing industry to surgical healthcare. The British Journal of Surgery. 2012;99(3):324-335


a Amsterdam Rheumatology and Immunology Center, VU University Medical Center, Department of Rheumatology, Room ZH 3A-58, De boelelaan 1117, 1081, HV, Amsterdam, The Netherlands

b Sint Maartenskliniek, Department of Rheumatology, Hengstdal 3, 6500, GM, Nijmegen, The Netherlands

c Amsterdam Rheumatology and Immunology Center, VU University Medical Center, Department of Rheumatology, Room ZH 3A-56, De boelelaan 1117, 1081, HV, Amsterdam, The Netherlands

d Radboud University Medical Center, Radboud Institute for Health Sciences, IQ Healthcare, P.O. Box 9101, 6500, HB, Nijmegen, The Netherlands

e Amsterdam Rheumatology and Immunology Center, Academic Medical Center, Meibergdreef 9, The Netherlands

Corresponding author. Tel.: +31 20 4445116.

1 Tel.: +31 20 44 44725.

Stay informed

Subscribe to our E-Alert to keep up to date with the new items in the Resource Centre.

Clinical Rheumatology

Journal cover clinical Rheumatoloy - Elsevier Resource Centre

Spread the word

If you like this resource centre, please share it.