Despite the popularity of biomarkers, their use as a tool for measuring human health effects or human exposures is not always necessary or applicable in other branches of epidemiology. One reason for this is that there are other epidemiologic methods or study designs that can serve as more appropriate alternatives. In pharmacoepidemiology, for example, clinical trials can be used not only to test the efficacy of a therapeutic measure, but also to monitor the adverse health effects from the dosing. It was mentioned in Slide 9 that in 1747, James Lind was credited with designing one of the first clinical trials. Clinical (sometimes referred to as human intervention) trials are planned experiments designed to assess the efficacy of a treatment in man by contrasting the outcomes of a group of patients or subjects treated with the test treatment with those observed in a comparable group of patients receiving a control treatment. In the case with James Lind, he used lemon juice as the treatment for scurvy, which can be caused by vitamin C deficiency.

Lind’s clinical trial was not the very first of its kind. In 1537, French army surgeon Ambroise Pare (a barber surgeon as well) applied an experimental treatment for battlefield wounds with a “digestive” made from rose oil, turpentine, and egg yolks (see e.g., Friis and Sellers, 1999; Sawyer, 2000). In addition to treatment efficacy, which actually reflects the effects of dosing or exposure, side effects are often included in a trial as outcomes of interest. Because side effects are linked to dosing, clinical trials are in fact a simplified, special version of health risk assessments.