From floated to conventional confidence intervals for the relative risks based on published dose-response data

Abstract

A dose-response meta-analysis of epidemiological studies can encounter different types of confidence intervals (floated vs. conventional). This paper illustrates how to back calculate conventional confidence intervals from a set of relative risks reported with floated confidence intervals or floated standard errors. Furthermore, we provide an implementation of the formulas in a user-friendly program developed for Stata software. I exemplify the point using published data about alcohol intake and endometrial cancer incidence from the Million Women Study.