0
$\begingroup$

In this article, the authors used the sentence "...risk difference (i.e., treatment effect on the clinically important absolute scale)".

What did they mean by "clinically important absolute scale"?

Did they mean by "risk difference (RD) (i.e., treatment effect on the clinically important absolute scale)" = "the absolute risk reduction (ARR)"?

Or they meant: if there is a treatment effect, it should not only be statistically relevant but also clinically important?

Thank you for your help!

$\endgroup$
  • $\begingroup$ Since it's "clinically important", this should be rather a question to an epidemiologist, isn't it? $\endgroup$ – Tim Jan 7 at 13:20
  • $\begingroup$ @Tim Maybe they meant: if there is a treatment effect, it should not only be statistically relevant but also clinically important? $\endgroup$ – Ph.D.Student Jan 7 at 13:46
3
$\begingroup$

The language is a stretch.

If they cite no source, it is because they actually mean risk difference is clinically important and an absolute scale. Rather than clinically important, risk difference is conservative and less likely to be exaggerated by a clinical audience, compared with risk ratio or the nefarious odds ratio. This is especially true for rare events. If the probability of an event is 5% in group A and 3% in group B, the risk difference is 2%, not 66%.

Absolute risk reduction is a causal term, but yes measured on the same scale.

| cite | improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.