The traditional "pyramid" or sequential approach to treatment of patients with rheumatoid arthritis involved use of a nonsteroidal anti-inflammatory drug for months to years while seeking to avoid use of second-line antirheumatic drugs until evidence of joint damage was seen. This approach led to short-term reduction of inflammation and a few remissions. However, long-term remissions were rare, and most patients experienced poor long-term outcomes, including joint destruction, severe functional declines, considerable economic losses, work disability, and premature mortality. At this time, a "preventive" strategy is evolving in which early aggressive treatment with disease-modifying antirheumatic drugs is used, seeking to minimize long-term joint damage. When residual inflammation remains after maximum doses of single agents, as is usually the case, combinations of disease-modifying antirheumatic drugs appear to be a reasonable consideration for many patients. Methotrexate is the most commonly used "anchor drug" in combination therapy. Evidence from randomized, controlled clinical trials and observational studies have indicated increased efficacy and acceptable (and often lower) toxicity for combinations of methotrexate plus cyclosporine, hydroxychloroquine, sulfasalazine, leflunomide, etanercept, and infliximab. Further studies lasting 5 years or more are needed to determine the long-term effectiveness, toxicities, and optimal clinical use of disease-modifying antirheumatic drug combinations. At this time, such combinations are taken by at least some patients under care of almost all rheumatologists, and it appears likely that they will be used increasingly in the coming decades.