Limitations to standard randomized controlled clinical trials to evaluate combination therapies in rheumatic diseases.

Randomized controlled clinical trials provide the cornerstone of modern research concerning medical therapies. However, important limitations intrinsic to clinical trials may restrict the generalizability of results to usual clinical practice. Limitations which appear particularly prominent in studies of combination therapies in inflammatory rheumatic diseases include: a) appropriate attention to group data which does not address variability of individual patient responses to specific agents; b) exclusion criteria to enter the clinical trial; c) small numbers with insufficient statistical power to identify clinically important trends; d) a time frame too short to recognize clinically meaningful differences between treatment regimens which may emerge over long periods. These limitations present problems in interpretation of studies in polymyositis, systemic lupus erythematosus, and rheumatoid arthritis. Strategies to improve accurate assessment of therapies include meta-analyses, clinical trials over at least five years, and long-term databases concerning consecutive patients seen in usual care.