Therapeutic strategies in rheumatoid arthritis over a 40-year period

To examine trends in therapeutic strategies and to identify the determinants of starting disease modifying antirheumatic drug (DMARD) therapy over a 40-year period in a population based inception cohort of patients with rheumatoid arthritis (RA).

A population based inception cohort was assembled from among all Rochester, Minnesota, residents aged > or = 18 years who were first diagnosed with RA (1987 American College of Rheumatology criteria) between January 1, 1955, and January 1, 1995. All subjects were followed longitudinally through their complete medical records until death, migration from Olmsted County, or date of abstraction (January 1, 2001, to January 1, 2003). Drug exposure data were collected on all DMARD and corticosteroid regimens. Time to DMARD initiation was examined using the Kaplan-Meier method. The influence of calendar time and disease characteristics on time from incidence to first DMARD therapy and the number of DMARD regimens were analyzed using Cox regression and proportional odds models, respectively.

The study population comprised 603 patients (73% female) with a mean age of 58 years and a mean followup of 15 years. At 2 years after RA onset, 26% of patients in the 1955-74 cohort, 40% in the 1975-84 cohort, and 70% in the 1985-94 cohort had received a DMARD (log-rank p < 0.001). Age, rheumatoid factor (RF) positivity, erythrocyte sedimentation rate, large joint swelling, rheumatoid nodules, and destructive changes on radiographs were significantly associated with time to first DMARD regimen after adjustment for calendar time and sex. Patients who were older and RF positive and who did not receive CS were more likely to have received more DMARD regimens.

Time to initiation of DMARD therapy has shortened markedly over the past 3-4 decades. These changes in management of early RA provide evidence for the translation of scientific evidence into clinical practice in rheumatology. Age and various disease characteristics are significantly associated with initiation and the number of DMARD regimens used. These should be considered as confounders when examining the effect of early DMARD treatment on disease progression and mortality.