Andrew,
Preference seems to depend on whether your a clinician or a statistician.
Clinicians like relative risk because most folk find it easier to
understand a ratio of probabilities (relative risk) rather than a ratio of
odds (odds ratio).
Statisticians prefer odds ratios because of several better properties that
they have, e.g., it doesn't matter which way around you express the risk.
The survival odds ratio is the inverse of the mortality odds ratio; but the
survival relative risk is not the inverse of the mortality relative risk.
For example, if the treated group has a 20% mortality and the control group
a 40% mortality then:
mortality relative risk = 0.20/0.40 = 0.50
survival relative risk = 0.80/0.60 = 1.33
mortality odds ratio = (0.20/0.80)/(0.40/0.60) = 0.375
survival odds ratio = (0.80/0.20)/(0.60/0.40) = 2.67 (=1/0.375)
Of course, the most clinically relevant measure is the absolute risk
reduction or number needed to treat, which needs the patient's expected
event rate and the relative risk or odds ratio. But the transformation is a
bit easier with the relative risk.
As Eric Harvey pointed out, when the probabilities are low then the two are
very similar, but when the probabilities are high then there are not - and
you'll have to decide whether you like the easy interpretation of the
relative risk or the symmetry properties of the odds ratio. Some days I
break my eggs at the little end and somedays at the big end,
Paul Glasziou
At 08:32 AM 25/2/99 +1300, Andrew Jull wrote:
>Dear All
>
>I was recently conversing with a colleague and the question of why use ORs
>instead RRs came up. My naive response was that the choice seemed to be
>based on the individual's preference and that I had not read anything that
>suggested the use of one was more informative than another (and indeed have
>read some material that suggests ORs are misleading when the OR is high -
>but I don't want to get onto that issue).
>
>Can anyone help me with why odds ratios might be used in preference to
>relative risk or vice versa.
>
>regards
>Andrew Jull
>Clinical Nurse Consultant
>Auckland Hospital
>NEW ZEALAND
>
Paul Glasziou
Social & Preventive Medicine,
Medical School, Herston Rd,
Herston Qld 4006,
AUSTRALIA
PH 61-7-33655427 FAX: 61-7-33655442
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|