Vol. 3, 655- 659, December 1994 Cancer Epidemiology, Biomarkers & Prevention 655 Fatherhood and Distal Adenomas of the Large Bowel: A Study of Male Self-Defense Officials in Japan1 Judith S. Jacobson, Summon Kono, lsao Todoroki, Satoshi Honjo, Koichi Shinchi, Koji Imanishi, Hiroshi Nishikawa, Shinsaku Ogawa, Mitsuhiko Katsurada, and Alfred I. Neugut2 Division of Epidemiology, School of Public Health, College of Physicians and Surgeons, Columbia University, New York ]J. S. J., A. I. N.]; Department of Public Health, National Defense Medical College, Tokorozawa, Saitama, Japan 5. K., I. T., S. H.]; Self-Defense Forces Fukuoka Hospital, Fukuoka, Japan 1K. S., K. I.]; Self-Defense Forces Kumamoto Hospital, Kumamoto, Japan IH. N.]; and Self-Defense Forces Sapporo Hospital, Sapporo, Japan IS. 0., M. K.] Abstract Parity has been studied extensively as a risk fador for colorectal cancer but has not been definitively shown to be associated with altered risk. In a few studies, risk of coloredal cancer in childless men has been compared to risk in men with children, but results have not been consistent. We analyzed the association of fatherhood with risk of colorectal adenomas in male self-defense officials (ages 49-55) in Japan. The study participants received a preretirement health examination including flexible sigmoidoscopy at Self-Defense Forces hospitals in Japan from January 1 991 through December 1992. The examinations identified 265 cases with rectal or sigmoid adenomas and 1 480 controls with normal examinations up to 60 cm from the anus. Data on marital status, number of children, long-term work assignment away from wife and children, and other lifestyle variables were obtained by means of a self- administered questionnaire prior to physical examination. Multiple logistic regression analysis assessed the risk of adenomas in relation to number of children, marital status, long-term work assignment away from family, and military rank, with adjustment for cigarette smoking, alcohol intake, dietary variables, body mass index, and recreational physical adivity. In this relatively homogeneous group, more than 98% of both cases and controls were currently married, and more than 93% had children. The adjusted odds ratio for the association of adenomas with fatherhood was 0.4 (95% confidence interval, 0.2-0.8). Marital status and work assignment away from the family were not associated with adenoma risk. These findings suggest that coloredal adenomas and perhaps cancer risk may be associated with childlessness in men. Introdudion Reproductive and hormonal risk factors for colorectal can- cer in women have been studied extensively over the past 25 years (1 ). Some studies have found parity protective as it appears to be protective against breast cancer (2-1 1 ); others have not (1 2-21 ). Colorectal cancer has been observed to resemble breast cancer in geographical distribution and time trends (22). Women with breast cancer also have been found to be at increased risk for colorectal cancer and vice versa (23, 24), and nuns have been found to have higher risk for both cancers than women in the general population (25). A few studies of reproductive and hormonal risk factors for colorectal cancer have included male study participants (5-7, 26). Of these studies, three found parity protective in females (5-7). Of those three, two found having children protective in males as well (6, 7), one found being married protective in males (5), and one found early age at birth of the firstchild protective in males (7). A fourth, large-scale, prospective study, which found parity protective in females only at certain colonic subsites, did not find having children protective in males (26). The geographical distribution of adenomas has been found to parallel that of cancer (22, 27), and colorectal cancers appear to arise from preexisting adenomatous pol- yps (28-30). Studies of risk factors for colorectal adenomas have generally focused on risk factors associated with cob- rectal cancer, such as diet (31), family history (32-33), and lifestyle variables such as smoking and drinking (34-35). A continuing study of adenoma risk factors in male self-defense officials in Japan has reported findings on diet (36, 37), physical activity (36), obesity (38), and smoking and alcohol intake (34, 35). Study participants interviewed during 1991-1992 were asked to supply information on reproductive factors. The present study is the first to exam- me the association of coborectal adenomas with these risk factors in males. Materials and Methods Received 3/10/94; revised 7/6/94; accepted 7/6/94. 1 Supported in part by a Grant-in-Aid for Cancer Research (2-3) from the Ministry of Health and Welfare, Japan, and National Cancer Institute Train- ing Grant ST32CA09529. The 1993 Summer Institute in Japan (INT- 9305234) was supported by the United States National Science Foundation and NIH and in Japan by the Science and Technology Agency and the Center for Global Partnership. 2 To whom requests for reprints should be addressed, at Division of Epide- miology, School of Public Health, College of Physicians and Surgeons, Columbia University, 600 West 168th Street, New York, NY 10032. The study population consisted of male self-defense offi- cials admitted consecutively to Self-Defense Forces Fukuoka and Kumamoto hospitals between January 1991 and December 1 992 and to the Self-Defense Forces Sap- poro Hospital from April to December 1992 for a preretire- ment physical examination. Details of the health examina- tion have been described elsewhere (36, 37, 38). The examination included flexible sigmoidoscopic or colono- scopic examination of the rectum and sigmoid colon as a on December 3, 2021. © 1994 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from