RESULTS OF INTERMEDIATE MEASURES FROM A POPULATION-BASED, RANDOMIZED TRIAL OF MAMMOGRAPHIC SCREENING PREVALENCE AND DETECTION OF BREAST-CARCINOMA AMONG ASIAN WOMEN - THE SINGAPORE BREAST SCREENING PROJECT

BACKGROUND, Although increasing rates of breast carcinoma incidence ha ve been observed in Asian countries, appropriate strategies for detect ing early stage breast carcinoma in such communities have been difficu lt to formulate, particularly because no large population screening tr ial specifically involving Asian women has been reported. The objectiv e of this study was to evaluate the effectiveness and quality of mammo graphy as a screening technique for Singaporean women, who are predomi nantly Chinese. METHODS. In this prospective study, 166,600 women in S ingapore ages 50-64 years were randomized to either 2-view mammography without physical examination (67,656) or observation (97,294, control s) over 2 years. RESULTS, Of these women, 28,231 (41.7%) responded and were screened; they were more likely to be married, have more formal education, be working, be Chinese, and be in a higher socioeconomic gr oup (P < 0.001 for all variables). To assess for response bias that co uld affect outcome, results were also evaluated for nonrespondents (n = 39,425). The incidence rate of cancers among nonrespondents (1 per 1 000 woman-years) was less than the 1.3 in women not invited to have sc reening (P = 0.03, relative risk [RR], 1.3; 95% confidence interval [C I], 1.0-1.7). However, cancers arising from nonrespondents did not dif fer significantly in stage distribution when compared with cancers wit hin the control group. For every 1000 women screened, 4.8 cancers were detected. The prevalence ratio (the number of cancers detected per 10 00 women at first screening divided by the corresponding incidence rat e in controls per year) was 3.6 for screened women and 2.4 for women i nvited to have screening. The majority of cancers detected through scr eening were early stage, with 64% as either ductal carcinoma in situ ( 26%) or Stage I disease (38%) and was significantly more than the corr esponding 26% in women not invited to have screening (P < 0.001). When only invasive cancers were considered, screened women still had more early cancers, with 65% having no lymph node involvement, compared wit h 47% in the group not invited to have screening (P = 0.001; RR, 1.4; 95% CI, 1.2-1.7). Women who were screened had half the risk of having Stage II or later cancers (P < 0.0001; RR, 0.5; 95% CI, 0.4-0.7) when compared with women not invited to have screening. This higher detecti on rate of early cancers through screening was accomplished with accep table recall rates of 8% for further mammographic films or physical ex amination and a biopsy rate of 1.0% (10 per 1000 women screened). The interval cancer rate was 2.1 per 10,000 women screened in the first ye ar of follow-up. CONCLUSIONS. These positive results of intermediate m easures suggest that, in Asian communities, screening mammography coul d be an important modality for detecting early stage breast carcinoma. However, the low compliance rates suggest that health education effor ts must focus on issues related to acceptability if such programs are to succeed. (C) 1998 American Cancer Society.