Question

four risk differences and their 95% confidence intervals are shown below. which of these is most precise?

A.) -0.15 (-0.25, 0.05)

B.) -0.15 (-0.45, 0.15)

C.) -0.15 (-0.17, -0.13)

D.) -0.15 (-0.33, 0.03)

Answer 1

Confidence levels indicate the range within which the predicted means of sampled population ( here risk differences) lie.

95% Confidence level of a sample means that 95% values should lie between 1.96 standard error of the mean (SEM) above & below their sample mean (average value).

Precision of any sample increases with the increase in the sample size.

Standard error mean(SEM) = Standard deviation (SD)/ sample size (N)

So, for more precision, SD should be lower and N should be higher.

In the given case of risk differences given here, standard deviation is minimum (around 0.04, obtained by 0.17 - 0.14) for option C.

hence, C.) -0.15 (-0.17, -0.13) is the most precise

.