Which statement describes the effect of a tenfold reduction in incidence in both exposed and unexposed on RR and AR?

Study for the ACVPM Epidemiology and Biostatistics Exam. Prepare with flashcards and multiple choice questions, with hints and explanations for each. Be exam-ready!

Multiple Choice

Which statement describes the effect of a tenfold reduction in incidence in both exposed and unexposed on RR and AR?

Explanation:
When both groups experience the same proportional reduction in incidence, the effect on these two measures differs because they are defined differently. Relative risk is a ratio: RR = Ie / Io. If Ie and Io are both reduced by the same factor (tenfold in this case), the factor cancels in the ratio, so RR stays the same. Attributable risk is a difference: AR = Ie − Io. If Ie and Io are both reduced by the same factor, the difference also reduces by that factor: AR' = (Ie/10) − (Io/10) = (Ie − Io)/10 = AR/10. So AR decreases tenfold. Example: start with Ie = 20 per 1000 and Io = 5 per 1000. RR = 4; AR = 15 per 1000. After a tenfold reduction, Ie' = 2 per 1000 and Io' = 0.5 per 1000. RR' = 4 (unchanged) and AR' = 1.5 per 1000 (tenfold smaller). Therefore, the statement that a tenfold reduction in incidence in both groups would reduce AR but not RR is correct.

When both groups experience the same proportional reduction in incidence, the effect on these two measures differs because they are defined differently.

Relative risk is a ratio: RR = Ie / Io. If Ie and Io are both reduced by the same factor (tenfold in this case), the factor cancels in the ratio, so RR stays the same.

Attributable risk is a difference: AR = Ie − Io. If Ie and Io are both reduced by the same factor, the difference also reduces by that factor: AR' = (Ie/10) − (Io/10) = (Ie − Io)/10 = AR/10. So AR decreases tenfold.

Example: start with Ie = 20 per 1000 and Io = 5 per 1000. RR = 4; AR = 15 per 1000. After a tenfold reduction, Ie' = 2 per 1000 and Io' = 0.5 per 1000. RR' = 4 (unchanged) and AR' = 1.5 per 1000 (tenfold smaller).

Therefore, the statement that a tenfold reduction in incidence in both groups would reduce AR but not RR is correct.

Subscribe

Get the latest from Passetra

You can unsubscribe at any time. Read our privacy policy