Notes
1 So, too, should the steps between primary evidence collection, drawing of inferences about the real-world meaning of any observed effects, and the formulation and roll-out of public health policy typically be slow, with adequate attention paid to complex contextual (e.g., sociocultural) factors, robust translational research, and the use of realistic models of human behavior, i.e., outside of relatively tightly-controlled clinical or experimental contexts (Earp and Darby Citation2019; Fish et al Citation2020; Ijzerman et al. Citation2020).
2 Of course, it is important that the current advice is given with ‘confidence intervals’ around it, with appropriate qualifications about the sources and strength of evidence, its limitations and generalizability, and so on. By contrast, if the current advice is given in an overly confident, definitive way—and then it needs to be updated in light of new evidence—this may contribute to the perception that science and medicine are unreliable.