Re: Different combined oral contraceptives and the risk of venous thrombosis: systematic review and network meta-analysis
Authors remarked that-"Unadjusted relative risks with 95% confidence intervals were reported. The requirement for crude numbers did not allow adjustment for potential confounding variables" (BMJ 2013;347:f5298).
However, Dr Philip Sedgewick says in one of his endgames (statistical question, BMJ 2013;347:f5067) that:
"It is not possible to derive adjusted relative risks—that is, adjust for confounding and allow for the simultaneous effects of other variables studied. However, it is possible to derive an odds ratio that estimates the population relative risk. Odds ratios can be adjusted for confounding using a statistical method known as logistic regression, described in a previous question(Sedgwick P. Logistic regression. BMJ2013;347:f4488). It has been proposed that the sample odds ratio is a good estimate of the population relative risk when the disease or outcome is rare in the population, typically when the prevalence is less than 10%.
Competing interests: No competing interests