Article Text

FRI0427 Getting a grip on tight control: Predictors of medication escalation in RA in practice
  1. J. Hendrikx,
  2. W. Kievit,
  3. J. Fransen,
  4. P. van Riel
  1. Department of Rheumatology, Radboud University Medical Centre Nijmegen, Nijmegen, Netherlands


Background Trials of tight control strategies have shown additional effectiveness over usual care. Despite this, in daily clinical practice medication is not always escalated according to guidelines. It is hypothesized that, additional to attained disease activity level, other clinical factors or patient related factors might explain divergence from tight control.

Objectives To investigate factors predicting escalation of medication in rheumatoid arthritis (RA) patients.

Methods All RA patients with regular monitoring visits to the Nijmegen outpatient clinic received standard clinical assessment including DAS28 assessment and registration of medication changes. In addition, patients were asked 4 questions regarding: perceived health state transition, satisfaction with their current health state, willingness to change therapy and expected health state transition in the near future. To investigate explanatory factors, three models predicting escalation of medication were compared by means of ROC analyses. Predictors for the respective models were: model 1 (reflecting tight control) DAS28>3.2 (yes/no), model 2 gender, age, rheumatoid factor, disease duration, DAS28 response since last visit + attained DAS28 level, model 3 predictors in model 2 + the above mentioned patient perception parameters. Escalation of medication was defined as an increase in frequency and/or dose of medication, or starting a new drug, between the current and next visit.

Results In total 337 RA patients; 65.9% female, 71.5% rheumatoid factor positive were included in the analysis. Mean (SD) DAS28 at visit, age and disease duration in years were: 3.11 (1.24), 59.2 (13.17), 11.2 (9.03), respectively. In total 146 (43.3%) patients had a DAS28>3.2, of whom 79 (54.1%) were satisfied with their health status if this would not change until the next visit and a majority of 96 (65.8%) did not want to change their medication. Figure 1 shows sub specifications of patients according to satisfaction an willingness to change medication. Figure 2 shows clear improvement in the predictive capacity of the models by adding additional clinical and patient related factors.

Conclusions Including a broader range of clinical and patient related parameters clearly improves the prediction of escalation of care. These results reflect that the decision process to escalate medication in daily clinical practice is more complex than is suggested by tight control guidelines. Interventions aimed at increasing tight control should address additional clinical and patient related factors rather than attained disease activity level alone.

Disclosure of Interest None Declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.