Of 12,337 participants of mass screening for breast cancer (screenees) in Gunma Prefecture between 1980 to 1985, 1,000 participants' records in Gunma Prefecture were investigated. The records of screenees who had already developed breast cancer were excluded. The rate of screenees who had a family history (FH) of cancer in first- and second-degree relatives was compared with that of 1,248 breast cancer patients (controls) using the conditional logistic regression model. The mean age of screenees (49 y.o.) was similar to that of controls (51 y.o.). The number of screenees with positive FH of cancer including other malignancies was 530 with the odds ratio of 2.68. This was statistically significant with chi-square test. The rates of screenees with positive FH of cancer were quite similar in the seven districts. Screenees had a significantly larger number of relatives with a positive FH of breast cancer compared with the controls. The rate of positive breast cancer history in parents of screenees was significantly higher than in the parents of controls. Rates of positive breast cancer history in grandmothers and aunts of screenees were also significantly higher than that in the controls. The rate of breast cancer history in sisters was not different between the two groups. Screenees had a higher rate of positive FH of cancer in parents, especially in mothers. Family histories of stomach, liver, lung, and uterus cancer in parents were more frequently observed in screenees compared with the controls and those cancer histories were also frequently observed in other family members of screenees. These data showed that a FH of cancer is one of the primary motivations of participation in the breast cancer screening program. Participants seemed to be intrinsically a high risk group of breast cancer.