
Lauren G. answered 07/11/18
Tutor
4.8
(23)
Graduate MPH student with 2 years experience using SAS
Based on my epidemiology education, odds ratios can only be valid estimates of the relative risk when the prevalence of the disease is low. This is because the odds ratio has a "Built-in" bias factor that, when the prevalence is low, is virtually zero; as a result, it doesn't change the original relative risk formula.