Only a few studies have examined repeat annual mammography rates, and most studies find that such regularity is low, ranging from < 5% in the general population to between 14 and 20% of first-degree relatives. The present study tested the effectiveness of a mailed intervention designed to improve compliance with breast cancer screening guidelines among women at elevated familial risk. The study used a pretest-posttest control group design; 369 twin sisters of breast cancer cases were assigned alternately to an intervention or a control group on the basis of sequential registration numbers. The intervention consisted of written materials, an audiotape, and mailed reminders. The posttest was mailed 2.5 years after the intervention in order to provide adequate time to assess the regularity of screening. The intervention and control groups were virtually identical with respect to demographic and baseline screening characteristics. Of those who returned the follow-up questionnaire, annual physician breast examinations were 12.8% higher and annual mammograms were 10.3% higher in the intervention group than in the control group. The probability of annual screening with physician breast examination and mammography was higher in the intervention group, and the probability of annual mammography continued to be higher for women over age 52 years after controlling for baseline screening, year of diagnosis, education, and status of the twin. This result is consistent with improvements found in other studies. Women who did not return follow-up questionnaires were more likely to have had fewer physician breast examinations and mammograms and more likely to be in the intervention group than those who stayed in the study. Additionally, those who dropped out of the intervention arm perceived themselves to be less susceptible and perceived screening to be less effective than did those who dropped out of the control arm. Although the intervention caused many women to be screened more regularly, those who had not been screened in the past and those who held opinions that were not conducive to screening were more likely to drop out. This mailed intervention for high-risk women increased their rate of screening. Characteristics of women resistant to such programs have been identified; alternative strategies need to be developed to reach this small percentage of high-risk women who ignore their elevated susceptibility.