Thinking Through the Problem Posted on February 29, 2008 Bruce G. Link (bio) cites a lack of forethought as the biggest misstep in epidemiological studies.

I think that the main pitfalls are not really thinking through your problem to begin with, to know exactly what it is you want to investigate and to get very clear in the logic of what your idea is and how it relates to the data. That's one of the hardest things to do, and I think that's where the biggest problem is and where you need to do the most to do the most work.

Once you've got that it's not like life gets easy because then there are all the issues of how to implement it and how to make a rigorous, strong design, but for me that's the most important thing. There are some other standard things that you might say you need to worry about selection into the exposure.

You want to know what the effect of the exposure is, and then if there's any self-selection, any selection of any kind into getting the exposure, that can cloud your ability to know whether the exposure has a causal influence on the outcome. I think that for epidemiologists that might be the most difficult problem because we usually do non-experimental research, observational research, and so we're never quite certain that we've ruled out selection into the exposure.

But I think most young people get that pretty early in epidemiology classes. I think the main problem for people doing epidemiological research is really thinking through the problem to begin with.

« Back to Article