Question 2

A published trial shows an Odds Ratio for effect of 0.88 [95% CI, 0.62 – 1.26].

Discuss your interpretation of this statistical statement.

This question asked for an interpretation of a statistical statement. There was good

understanding by candidates of the terms odds ratio and confidence intervals. The main

points expected for a pass included a discussion of how the Odds Ratio is more appropriate

than Risk Ratio in a retrospective case-control study where the outcome is a rare event

(<10%) and the denominator uncertain; if the outcome occurs commonly, Odds Ratio tends to

overestimate risk; an Odds ratio of 1 implies no difference.

Odds Ratio » Risk Ratio if the outcome (cells a,c) is sufficiently uncommon.

The use of the formula to illustrate how to calculate the confidence interval and the Standard

Error assisted with explaining these terms

A weakness amongst the candidates was a failure to provide a sufficient information on the

interpretation of the data. The good answers mentioned the point estimate of 0.88 implied a

reduction in the occurrence of the measured end point (eg mortality, a commonly measured

end point). However, the trial failed to reach statistical significance because the 95% CI

includes the value of 1.0 for the Odds Ratio. Candidates did not mention that the power of the

study was not given and therefore the data could not be properly interpreted and there was no

discussion on sampling errors or systemic error or bias possibly affecting the result.

Syllabus: EBM 1, 2b, c, d, g, h.

Reference Text: Statistical Methods for Anaesthesia and Intensive Care / P S Myles and T

Gin – 2001. Pages 8-11, 24-28, 74-75.