The purpose of this study was to test the effect of a theoretically based nurse-delivered intervention on BSE behavior. A 2 × 2 prospective, randomized, factorial design yielded four groups: control, belief intervention, procedural intervention, and procedural/belief intervention. A total of 301 women were randomly selected from a target population. One year following intervention, significant differences in self-reported proficiency, observer-rated proficiency, and sensitivity (lump detection) were found between the Procedural and Control Group and the Procedural/Belief and Control Group. Significant increases were found on observer-rated proficiency and sensitivity for the Procedural/Belief Group when compared to the Belief Group. In addition, a significant increase was found in the Procedural/Belief Group on nodule detection, when compared to the Procedural Group alone. © 1993 John Wiley & Sons, Inc.