Abstract
A review of literature concerning relaxation and imagery interventions for cancer pain is necessary because major review articles have excluded nursing research or were written prior to the publication of controlled studies in cancer pain conducted by nurses. This review of published nursing/medical/psychological literature of adults with cancer pain conducted over the past 14 years(1982-95) revealed few controlled studies, weak theoretical frameworks, few complete descriptions of the nature of the pain problem, and lack of control over the interventions. Most had very small sample sizes and could not demonstrate significant effects. Additionally, the intervention methods and length of the interventions were highly variable. Despite these design shortcomings, relaxation and imagery appear to reduce the sensory experience of pain, have equivocal effects on affective measures, and appear to have no effect on functional status. Suggestions for improvement include the need for more experimental studies, more complete descriptions of pain, improved statistical reporting, controls over adequacy of and compliance to the intervention, use of single interventions, and use of more complex measures of affective outcomes. Additionally, the paper contains a discussion of the problems of measuring selected outcome variables in this type of research.