Reporting risk ratios instead of odds ratios is often preferable for interpretability. In a regression framework, risk ratios are most efficiently estimated through log-binomial generalized linear models; unfortunately, statistical software algorithms often fail to converge in the log-binomial setting (1). A previous publication in this journal introduced a SAS macro that can, in some settings, leverage parameter estimates from a modified Poisson regression to guide the log-binomial fitting algorithm towards convergence (2). When this approach fails to fit a log-binomial model, parameter estimates from the modified Poisson regression are reported, though such estimates are known to exhibit reduced statistical efficiency.
Here, we present …
Example data, procedure, etc.
1. Williamson T, Eliasziw M, Fick GH. Log-binomial models: Exploring failed convergence. Emerg Themes Epidemiol. 2013;10:14.
2. Spiegelman D, Hertzmark E. Easy SAS Calculations for Risk or Prevalence Ratios and Differences. Am J Epidemiol. 2005;162(3):199–200.