A survey has been conducted during the past decade (starting in 1980) of reported new cases of women with breast cancer, in order to assess the interim evidence of the natural experiment of the effect of increased use of condoms on the risk of breast cancer in the US. Age-adjusted incidence rates of breast cancer (to the world standard population) per 100,000 population, and trends of changes, in percentages, were used. Prior to the 1980s, breast cancer incidence was in decline in seven out of 13 centres in the country. Following the recommendations to use condoms in ‘safer sex’ campaigns, increases of breast cancer incidence were recorded during the 1980s, as a reflection of a global phenomenon. Between the 5 year periods of 1978–82 and 1983–87, the population of the nine Surveillance, Epidemiology and End Results programme (SEER) centres (about 10% of the American population), increased by 2.9%, the number of breast cancer cases increased by 25.0%; the average annual breast cancer crude incidence rose by 21.5% (from 100.8 to 122.5 per 100,000). The increase in the average annual number of breast cancer cases during the period 1983–87 (the period of increased condom usage) was 4.5 times greater than that during the preceding period, 1978–82. During the 1980s, breast cancer trends rose significantly in the US among White women (P< 0.00001) and other ethnic groups (P< 0.005). The difference between the expected, 21%, and observed, 79%, probability of adverse effects occurring in the natural experimental trial was statistically significant (P= 0.0001). The results of the study were consistent with the conceptual framework of the hypothesis that the increase in the prevalence of condom use in the general population was most probably the primary environmental determinant of the recent rise of incidence of breast cancer. This may lend support to the argument for an alternate action of primary prevention against AIDS in the community.