ObjectiveTo determine if vital statistics support a temporal association between the introduction of oral contraceptives (OC) and the incidence of, and mortality from, primary liver cancer in three countries from different regions of the world.MethodsWe usedCancer Incidence in Five Continents, volumes I-VI, for incidence data on primary liver cancer in the United States and Japan. The Centre for Epidemiology in Stockholm provided the incidence data for Sweden. We obtained mortality data for the U.S. fromVital Statistics of the United States, for Japan fromThe World Health Statistics Annual, and for Sweden from Swedish government sources. We compiled data on the prevalence of OC use in all three countries from multiple sources, including the National Survey of Family Growth, The Population Council, and original articles.ResultsDespite several hundred million woman-years of exposure to OCs, primary liver cancer incidence and mortality rates among women have not changed substantially in the United States. In Sweden, another country with extensive OC use, the primary liver cancer incidence trends in women paralleled those of men. The incidence of, and mortality rate from, primary liver cancer is gradually rising in Japan, where OC use has been negligible.ConclusionPopulation-based data from three industrialized countries with very different patterns of oral contraceptive use provide no support for a measurable effect of OCs on primary liver cancer. Although case-control studies from developed countries have suggested an increase in risk, if such an effect does exist, the public health impact appears to be negligible.