Epidemiologic studies determining causal association

The obj ective of epidemiological studies is to determine whether a causal relationship exists between a putative risk or protective factor and disease. Thus, most studies are designed to test a hypothesis Does exposure (e.g., consumption of cooked meats) cause disease (e.g., colon cancer) After the study is conducted, the data are analyzed and the magnitude of the putative association between exposure and disease is calculated. The study is then evaluated to see whether the association (if observed) could be a result of random or systematic errors or whether other exposures or factors are influencing the exposure of interest (i.e., effect modifiers). 

Not all associations represent a cause-effect relationship. Because most epidemiologic studies of drug effects do not employ random allocation, it is important to determine if a legitimate cause-effect relationship exists. A central methodologic concern in observational studies is confounding—i.e., the possibility that the apparent effect of an exposure or intervention is due wholly or partly to other factors associated with it that have their own impact on the outcome of interest. Criteria have been proposed to help determine if an association is causal. The fewer criteria that are met, the less likely it is that an association is causal. Table 9-3 is adapted from the work of Hill and Stolly. Practitioners should ask the series of questions listed in the table to interpret findings from studies to determine if an association is likely to be causal. 

---