Because is hard not to think of an odds ratio as a relative risk, it is
helpful to know when and how this is misleading.
Basically,
OR is always in the same direction as RR - in other words, OR <1 if
RR <1 and OR >1 if RR >1
OR exaggerates the RR - in other words, OR < RR if RR <1 and OR >RR
if RR >1
The degree of exaggeration increases with both:
prevalence (or initial risk)
the absolute difference from RR
being zero when RR =0 and zero again when
RR = 1
being greatest when RR = 0.5 and again RR
= infinity
See the graphs in the attached PPT slide (they have disappeared from where I
originally found them in Bandolier), and the different graphics in this BMJ
paper
http://www.bmj.com/content/316/7136/989.full#ref-2
But the link to the first figure Is broken - I have drawn that to the BMJ's
attention
The figures do need some thinking to understand their message, but the
effort is well worth it
Best wishes
Michael
On 5 Feb 2012, at 10:51, Jane Hartley wrote:
Can anyone suggest an easy guide to odds ratios and other basic stats
functions?
A clinical friend has asked for some help with a dissertation and
understanding source papers, this is not her area of expertise and she has
been frightened off by the scholarly texts she has been directed too - I
suspect her supervisors are not in their area of comfort with this either.
I have moved professionally away from supporting EBH and so am not up to
date with user friendly articles. Dare I admit that I looked at the
wikipedia which seemed comprehensive but impenetrable to the novice.
suggestions very welcome
Jane Hartley
|