Related Experiment Videos
How do women compare with internal medicine residents in breast lump detection? A study with silicone models
S W Fletcher1, M S O'Malley, C A Pilgrim
1Department of Medicine, School of Medicine, University of North Carolina, Chapel Hill 27599-7300.
Objective:
Manufactured silicone breast models were used to compare the accuracy of breast examination by 300 women and 62 internal medicine residents.
Design:
The study design was cross-sectional.
Setting:
The study took place in two teaching-hospital general medicine clinics.
Patients/Participants:
Women were continuing care patients, ages 40 to 68, with no current breast complaint; 300 of 467 (66%) randomly selected women participated. Physicians were internal medicine residents with at least one-half day per week of ambulatory care practice; 62 of 64 (97%) participated.
Measurements And Main Results:
Sensitivity equalled the percentage of 18 lumps correctly detected in examination of six silicone breast models. Specificity equalled the percentage of six models examined without a false-positive detection. Women's sensitivity was lower than physicians' (40% vs. 58%), but their specificity was higher (66% vs. 52%). For both women and physicians, sensitivity varied according to lump size, hardness, and depth, with women's sensitivity lower than physicians' for each characteristic. Examination duration was the technique most strongly and consistently related to accuracy. Physicians spent more time examining models than did women (2.5 vs. 2.1 minutes per model). For both groups duration related positively to sensitivity (r = 0.46, women; 0.55, physicians) and negatively to specificity (r = -0.35, women; -0.59, physicians). After adjusting for differences in technique, women's sensitivity remained lower than physicians', whereas specificity generally remained higher. The sensitivity of physicians with prior tactile experience with breast lumps was higher than that of physicians without such experience (60% vs. 51%, p = 0.01). Too few women (2%) had prior tactile experience to permit analysis.
Conclusions:
Women's and physicians' breast examination accuracies differ, but for the two groups accuracies vary similarly by lump characteristics and examination technique. Programs to improve breast examination should focus on specificity as well as sensitivity. Training that includes tactile experience may be important.