Answers are grounded in this paper and cite their sources.
The classical correlational investigation is now commonly referred to as an ecologic study. Multiple sclerosis (MS) is particularly suited for this analytical approach. Repeated investigations have confirmed not only its marked geographic heterogeneity at the global level and predominance in Caucasians,1-4 but also remarkable geographic variations within individual countries5-12 and even within smaller sub-areas.13-15 Such findings make ecologic comparisons useful and provide excellent opportunities to a search for consistent associations as an important criterion for causality.16 The place of the ecologic study within the armamentarium of epidemiologic designs is somewhat controversial. Most modern textbooks17-20 include it with the descriptive methods. This may be correct operationally, but when the purpose of the study is considered, the ecologic study is analytical research,21 as the purpose is to venture beyond pure description by systematically considering the environment as a source of putative risk factors. The utility of the ecologic study is also a matter of debate. Most authors agree that it is a valuable tool for the generation and early support of hypotheses, but several have argued that it permits only limited conclusions because of well-recognized shortcomings (see below).17-27 Yet others28 attribute to the ecologic approach a high value in the process of causal inference in the face of the established limitations of case-control studies and the frequent impracticality of cohort studies. The low rank commonly assigned to the ecologic study in the process of causal inference refers to the single investigation. When an extended and systematic search is conducted for an association between the disease in question and a defined exposure by multiple investigations covering independent areas and populations, and when, in addition, these include smaller areas giving less rise to confounding, considerable power can be attained in the ability to test epidemiologic criteria …
No takes yet. Share an insight, caveat, or question.
Klaus Lauer (1997) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: