This document discusses various types of bias, confounding, and causation that can occur in epidemiological studies. It defines a confounder as a variable that is associated with the exposure and affects the outcome but is not in the causal pathway. Three main types of bias are described: selection bias, information bias, and confounding. Specific biases like recall bias, observer bias, and non-respondent bias are explained. Methods for controlling confounding like matching, stratification, and multivariate analysis are also outlined. The document concludes by discussing Hill's criteria for determining a causal association and threats to the internal and external validity of experimental studies.