Epidemiology as a Toolkit for Clinical Scientists

improving sanitation? Under the warning that public policy is always "invasive," large forces shaping society are excluded from scientific study or intervention. In fact, private policies are also "invasive" and create health hazards for large sections ofthe world's popula? tion. To cite just one example, although tobacco smok? ing is diminishing in the United States, it is growing rapidly in many developing countries as a result of aggressive marketing by tobacco companies unfettered by "invasive" regulations. It is true that epidemiologists may have wrongly suggested policy interventions in the past (just as they have incorrectly inferred biological causes or recom? mended incorrect medical interventions), but the fact that policy recommendations have sometimes been inappropriate does not imply that they are always inappropriate. The idea that epidemiologists should never advocate public policy or health policy based on scientific data is in itself a political position. It is not a requirement of good science. Epidemiology benefits from the contributions of a wide range of disciplines, from molecular biology to economics, and not all epidemiologists need span the whole range of factors affecting people's health. Clin? ical scientists have much to contribute to epidemiol? ogy, but so do other sciences, including the social sciences, which Charlton so fears. Scientific "truth" about the causes of disease does not lie only in biology. In addition, the argument that epidemiologists should never advocate health or social policies based on their findings implies an a priori position that health should be left to individuals and that society as a whole has no role to play in protecting people's health. Indeed, when scrutinized closely, this "new" model for epide? miology is neither as scientifically objective nor as apolitical as it claims to be.