According to the explanation given for this question, population attributable risk percent (PAR%) is defined as: (Incidence in exposed population - incidence in whole population)/(Incidence in exposed population). I don't follow the logic here. If PAR is attributable risk as its applied to whole population, shouldn't it also follow the same pattern as (exposed - non-exposed)? I've checked this information over several web sites and they all define PAR as: (Incidence in whole population - incidence in unexposed population) and PAR% as PAR/Incidence in whole population. Can anyone else confirm this?

The ARP is used to see what % of the disease is due to a some risk factor, and the formula to be used in this kind of questions is ARP=(RR-1)/RR. Good Luck

That's true for AR%, but what I had written was about population AR (PAR).

I believe the answer provided in Biostat Review for this question is completely wrong. Since a similar question in the Step3 Qbank addresses Population attributable Risk percent, and it uses a different formula compare to this one, and would have given a different answer.

If we're talking about QID 14 under attributable risk, I also think it's wrong. Unless I'm just being dumb here, I calculated the RR as 2. Which means 2-1/2=0.5 so 50%. I'm doing RR as (a/a+b)/(c/c+d) and I keep getting 2, so if someone could explain how I'm messing up, or confirm that the answer is wrong, so I can move on in peace.

I got hung up for a long time on this question, so I'm glad to see I'm not the only one who's confused by it. It actually made me waste a lot of time changing in my head how I thought PAR% questions were done... then changing it back once I did more research.

It seems they simply substituted Incidence(unexposed) for I(population) and treated the question like a normal attributable risk % question, giving them .5-.4/.5=.2 I'm trying to figure out if I'm missing something on when to use the PAR% equation, because as far as I can tell there's no reason not to use PAR% for this question. I emailed my biostats professor to see what he'll say about it.

I believe it's asking for PAR%, which should be It-Iu/It
That gives me .4-.25/.4 = .375
PAR = 37.5%
edited by on 8/17/2016