In a randomised, placebo controlled, double blind study inosine pranobex was assessed as a possible second line drug in rheumatoid arthritis. Twenty four patients received inosine pranobex (3 g/day) and 26 patients received placebo for up to 24 weeks. Morning stiffness, articular index, grip strength, pain score, erythrocyte sedimentation rate, C reactive protein, IgG, IgM, and serum urate were assessed at weeks 0, 12, and 24. Baseline characteristics were similar except for a significantly higher C reactive protein in the placebo group. No significant improvement occurred in any variable: (a) when comparing week 0 with week 12 or week 24 for either group, (b) comparing active drug with placebo at week 12 or 24, or (c) taking all 50 patients as one group. Withdrawal from the study for lack of response or side effects was similar in both groups. Serum urate increased transiently but significantly with inosine pranobex (a recognised side effect). It is concluded that inosine pranobex has no second line activity in rheumatoid arthritis. Further, 50 patients effectively given placebo showed no spontaneous improvement in their disease activity.