First, let's really define what science and engineering is.
In both cases, the underlying goal is to infer the joint distribution of a set of variables. In science, we do this by keeping all but two variables constant, choose one variable as our independent variable that we can reliably manipulate, and measure another dependent variable in order to see if it's related to the independent variable.
There is nothing
For this discussion, we will consider clinical practice high-stakes engineering.