THE ASSOCIATION BETWEEN LACTATION AND BREAST-CANCER IN AN INTERNATIONAL CASE-CONTROL STUDY - A REANALYSIS BY MENOPAUSAL STATUS

In the large, hospital-based, international case-control study of brea st cancer conducted by MacMahon and colleagues in the 1960s, no protec tive effect of lactation was observed. However, more recent reports ha ve suggested that such an association may be limited to pre-menopausal women. Therefore, a re-analysis of the data from that original study was performed by menopausal status and with control for additional bre ast-cancer risk factors since identified. Overall, data from 4,671 par ous pre-menopausal and 7,200 parous post-menopausal women were analyze d, from 7 different sires representing areas of high risk (Glamorgan, Wales; Boston, USA), moderate risk (Athens, Greece; Slovenia, ex-Yugos lavia; Sao Paolo, Brazil), and low risk (Tokyo, Japan; Taipei, Taiwan) of breast cancer. No significant effect of lactation was found for pr e-menopausal or post-menopausal women from the high-, moderate-, or lo w-risk areas; the center-adjusted, combined odds ratio (OR) for having breast-fed was 1.05 (95% confidence interval, 0.86-1.29) among pre-me nopausal and 1.04 (0.88-1.24) among post-menopausal women. Moreover, e xamination of cumulative duration of lactation did not support an inve rse association between breast cancer and increased length of total br east-feeding, In conclusion, re-analysis of these data, by menopausal status and adjusting for age at first parity, age at menarche, age at menopause, body-mass index and years of schooling, did not reveal a pr otective effect of lactation or duration of lactation against breast-c ancer occurrence among the pre-menopausal, parous women. (C) 1997 Wile y-Liss, Inc.