Classic health care research has centered on studying a group of individuals and then extrapolating the findings to the general population. In this context, evaluating the association between nonrandomized exposures and clinical outcomes can yield interesting, hypothesis-generating correlations, but assembling evidence to suggest a causal relationship has focused on testing the relationship between randomized exposures and clinical outcomes1 (Figure).
Mega JL, Sabatine MS, Antman EM. Population and Personalized Medicine in the Modern Era. JAMA. 2014;312(19):1969-1970. doi:10.1001/jama.2014.15224