Therapeutic trials in rheumatoid arthritis (RA), osteoarthritis, seronegative spondyloarthopathies, back pain, systemic lupus erythematosus, and systemic sclerosis are reviewed. For RA, minocycline has been proven effective in some subsets of RA, whereas tumor necrosis factor receptor IgG fusion protein appears quite effective for treating the symptoms of RA in a more resistant group. The latter trial illustrates the importance of tumor necrosis factor in RA. Also, the triple combination of hydroxychloroquine, sulfasalazine, and methotrexate is very effective even in resistant RA. In osteoarthritis, the effects of nonsteroidal anti-inflammatory drugs, intra-articular steroids, and biologics are reviewed, with generally nondifferentiable nonsteroidal anti-inflammatory drug effects and some short-term intra-articular effects of new biologics. Sulfasalazine is moderately effective for ankylosing spondylitis and psoriatic arthritis, although the large placebo response in the latter makes it more difficult to show responses. Trials in the treatment of back pain are beginning to be published, with a large cohort study over 1 year favoring surgery for early relief of pain in both sciatica and lumbar stenosis, but not showing a clear advantage in functional outcome at 1 year. Finally, early reports show the ability of dihydroepiandrosterone to decrease steroid use in systemic lupus erythematosus, whereas Relaxin appears to be effective in decreasing skin involvement in systemic sclerosis. These trials demonstrate in numerous ways the need to consider the elements of good trial design when testing therapeutic modalities in the rheumatic diseases. These key elements include 1) careful patient definition and selection; 2) removal of bias (requiring blinding, randomization, prospective studies, and often, placebo); 3) use of well-defined outcomes; and 4) careful analytic techniques.