A survey of United States rheumatologists concerning effectiveness of disease-modifying antirheumatic drugs and prednisone in the treatment of rheumatoid arthritis.

OBJECTIVE: To collect the ratings of American rheumatologists regarding relative short-term and long-term effectiveness of disease-modifying antirheumatic drug (DMARD) therapy in the treatment of rheumatoid arthritis and to compare these data with results of a meta-analysis of randomized, controlled clinical trials (RCTs) of these drugs.

METHODS: A survey was mailed to 3,200 United States rheumatologists who were members of the American College of Rheumatology during the summer of 1996. The survey was completed by 645 rheumatologists.

RESULTS: Methotrexate (MTX) was rated as substantially more effective than any other DMARD after both 1 year and 4 years of therapy. At 1 year, more than 90% of rheumatologists rated MTX as "good or excellent," compared with 35% giving a similar rating to intramuscular (IM) gold, the second-ranked DMARD. At 4 years, MTX was rated as "good or excellent" by 65%, compared with 53% for combinations, 30% for prednisone, and 11% for hydroxychloroquine, the second-ranked single DMARD. These ratings reflect results of long-term observational studies, but differ considerably from a meta-analysis of results of short-term RCTs, in which the efficacy of MTX, sulfasalazine, penicillamine, and IM gold were indistinguishable.

CONCLUSION: The ratings of US rheumatologists regarding effectiveness of DMARD therapies are in agreement with observational data, but differ substantially from results of RCTs. These findings suggest that observational studies may yield useful and important information about treatment results, which are complementary to, but not available from, the RCT model. Regulatory agencies should consider long-term observational studies as a part of the evaluation of drugs.