This is an archived version. For the current version, please go to training.cochrane.org/handbook/current.

The process of obtaining standard errors for ratio measures is similar to that for absolute measures, but with an additional first step. Analyses of ratio measures are performed on the natural log scale (see Chapter 9, Section 9.2.7). For a ratio measure, such as a risk ratio, odds ratio or hazard ratio (which we will denote generically as RR here), first calculate

lower limit = ln(lower confidence limit given for RR) |

upper limit = ln(upper confidence limit given for RR) |

intervention effect estimate = lnRR |

Then the formulae in Section 7.7.7.2 can be used. Note that the standard error refers to the log of the ratio measure. When using the generic inverse variance method in RevMan, the data should be entered on the natural log scale, that is as lnRR and the standard error of lnRR, as calculated here (see Chapter 9, Section 9.4.3).