This article reviews controlled trials of hypnotic treatment for chronic pain in terms of: (1) analyses comparing the effects of hypnotic treatment to six types of control conditions; (2) component analyses; and (3) predictor analyses. The findings indicate that hypnotic analgesia produces significantly greater decreases in pain relative to no-treatment and to some non-hypnotic interventions such as medication management, physical therapy, and education/advice. However, the effects of self-hypnosis training on chronic pain tend to be similar, on average, to progressive muscle relaxation and autogenic training, both of which often include hypnotic-like suggestions. None of the published studies have compared hypnosis to an equally credible placebo or minimally effective pain treatment, therefore conclusions cannot yet be made about whether hypnotic analgesia treatment is specifically effective over and above its effects on patient expectancy. Component analyses indicate that labeling versus not labeling hypnosis treatment as hypnosis, or including versus not including hand-warming suggestions, have relatively little short-term impact on outcome, although the hypnosis label may have a long-term benefit. Predictor analyses suggest that global hypnotic responsivity and ability to experience vivid images are associated with treatment outcome in hypnosis, progressive relaxation, and autogenic training treatments. The paper concludes with a discussion of the implications of the findings for future hypnosis research and for the clinical applications of hypnotic analgesia.