Inclusion and exclusion criteria are a means of establishing precision in your cohort or case-control study. For example, you might suspect that a new brand of insulin tablet is more effective than an existing brand, but for some reason this seems to be true only for diabetic patients that are not overweight by more than 10%. Your professional knowledge has produced a "hunch" that has you well on your way to establishing inclusion and exclusion criteria.

Inclusion criteria are the criteria for including a patient in the study, and it is important that these criteria be clearly defined in an objective manner, so that anyone involved in the study (or anyone attempting to replicate the study) can reproduce patient inclusion decisions precisely. In other words, can you write down the inclusion criteria so that anyone else with professional knowledge similar to your own would include the same patients you included? The example above, of the new insulin tablet versus an existing one, could be translated into an inclusion criteria by specifying the two drugs (e.g. brand name, generic and manufacturer name, possibly even lot number).

Exclusion criteria are the criteria for excluding patients from the study. From the example above, the 10% overweight criterion could serve as the basis for a specific exclusion criteria statement. As in the case of inclusion criteria, is your specification precise enough that someone with knowledge similar to your own would exclude the same patients you choose to exclude?

Return to analytical designs.