Pain in children is infrequently assessed and managed by nurses. One-on-one coaching based on audit with feedback and the use of opinion leaders have been effective in changing professional health care practices. Coaching by an opinion leader for changing pediatric nurses' pain practices was tested in a clustered randomized trial in six Canadian pediatric hospitals. The rate of pain assessments, nurses' knowledge, and nonpharmacological interventions increased in the coaching group. However, there were significant site differences that could not be attributed to the coaching but to factors inherent in the sites. The context in which interventions are implemented will influence the effectiveness of individualized interventions.