|
Description
Many epidemiological studies report the effect size of an intervention as an odds ratio. While there are several good
reasons for doing so, there are also several good reasons why this is a nuisance.
One is that the effect size in a modeled health economic evaluation needs to be a relative risk instead of an
odds ratio. The usual ‘solution’ to this is to interpret the odds ratio as a relative risk, but this introduces bias: it systematically overestimates the effect size.
The OR2RR utility
implements a method to convert odds ratios into relative risks and the other way around, given disease risk and risk factor prevalence. For the conversion of relative risks into odds ratios an analytical
expression is used (see equation above), for the opposite conversion a high precision numerical method has been implemented.
Please note that OR2RR method is not valid when the odds ratios
are adjusted for confounding.
|
