A growing body of statistical tools has evolved to evaluate regional consistency in clinical trials. Approaches can have substantially different operating characteristics. To fit specific methods to particular trials, it is important to describe in advance issues such as: the specific purpose of the consistency investigation; the rationale for the definition of region to be used; the expectation and rationale for potential region differences; and the implications of failure to conclude consistency. We address whether it is sensible to view ‘region’ as a unique subgroup deserving different handling than typically applied to other important subgroups; and the extent to which ‘region’ might be an artificial construct, confounded with factors that are true effect modifiers. Such considerations can assist in choosing an appropriate approach, and in trial design, analysis, and interpretation.