Dear Data Scientist, I am bringing here an epidemiology question and seeking insight on it.
I have a case cohort (<100) and try to build an age-matched control cohort to study an age-related disorder. My question is about the age matching: - How specific should it be in terms of minimum, maximum, median, mean, frequency of the age? - How could I estimate the benefit if I do the exact age matching? - We do not know if the gender and ethnicity are relevant or not to this disorder, in such a situation, would matching with gender and ethnicity beyond the age in the control cohort building still benefit? Any insights would be highly appreciated! Yixing Han
