Although breast self-examination (BSE) is intuitively appealing as an adjunct in breast cancer control, observed competency among women practitioners is low (20-25%), and it is not clear that training in medical practice settings results in substantial long-term improvement in BSE performance, especially for women at risk due to age. We, therefore, proposed a year long randomized controlled trial in a practice setting to determine the long-term effects of three practice-oriented teaching interventions on the lump detection abilities of women aged 40 to 60. Women added 40 to 60, who are members of an internal medicine group practice at a university teaching hospital, will undergo a pre-test to determine competency in detecting breast lumps. Each woman's detection ability will be measured by the number of detections in a limited time trial search of six manufactured breast models containing 18 simulated lumps of three different sizes (1.0, 0.5, 0.3 cm.), three hardnesses (60, 40, and 20 durometers), and two placement depths (medium, deep). Women who find less than eight lumps (a norm established by a study of physician lump detection) will be randomly allocated to one of five treatment groups or to a control group. Three different intervention strategies will be tested: (1) RN-Model, a 45 minute nursing-teaching intervention using videotapes, breast models, and methods developed by Pennypacker and associates at the University of Florida; (2) RN-ACS, a nursing-teaching intervention based on booklets, pamphlets, and methods developed by the American Cancer Society; and (3) MD, a three to four minute physician reinforcement intervention using demonstration and encouragement. These three strategies will be used alone and in combination (RN-ACS + MD, RN-Model + MD) to form five treatment groups. One year after entry into the trial, women will be re-tested to assess changes in their lump detection ability. The mean improvement in detection of all five treatment groups will be examined to determine if BSE teaching and/or physician reinforcement results in an increase of at least 33 percentage points in a group's mean improvement when compared to the mean improvement of the control group. We will also compare improvement in each treatment groups's mean number of BSE's performed/last six months (a self-report) with that of the control group to determine if the interventions resulted in an increase of at least 2.0 BSE's performed/last six months.
