Indomethacin or amylobarbitone sodium for sleep in rheumatoid arthritis, with some observations on the use of sequential analysis.

1 A double blind crossover trial in patients with rheumatoid arthritis showed that indomethacin given at night was superior to a hypnotic in providing sleep and relieving morning stiffness. 2 In a second trial, the combination of a hypnotic with indomethacin was not superior to indomethacin alone as judged by patients' preference, but further analysis suggested that the combination improved sleep without affecting morning stiffness. 3 It is argued that the use of a small closed sequential design produced too small a sample in the first trial. In the second trial, the sample was relatively large but the result was inconclusive; measurement of the response in terms of preference alone obscured the effect of treatment on one aspect of the symptomatology studied.