Placebo therapy in rheumatoid arthritis.

Few studies have attempted to describe the effect of placebo therapy on chronic pain. In this study pain was assessed daily, for 25 days, in 23 patients with rheumatoid arthritis. Patients received placebo during the last 9 days of the study. Pain scores during the first 16 days acted as a baseline from which each patient's response to placebo could be determined. Patients completed Form B of the Eysenck Personality Inventory on entry into the study, a pre-treatment expectation rating scale and a post-treatment assessment scale. Five patients were excluded from analysis because of trends in their baseline pain scores. Of the remaining patients, nine responded positively to placebo and five responded negatively. Four patients reported no significant change in pain and were found to be more introverted than both the negative (p less than 0.05) and positive placebo responders. Consequently, the degree of extraversion might prove a useful predictor of therapeutic response. No significant relationship was found between the patients expectation of therapy and the observed response which suggests that therapeutic outcome is not pre-determined. However, the patients assessment of therapeutic effect was closely related to the observed response (p less than 0.05). Neuroticism was greater in patients with side-effects than those without (p less than 0.05) and might be utilized as a means of detecting those patients most likely to complain of unrelated symptoms while on a new therapy. The different placebo responses observed draws attention to the difficulties in making therapeutic judgements in clinical practice and of assessing efficacy in drug trials.