welcome redditors!to snoo-finity ... and beyond!

nbme23/Block 2/Question#28

A cohort study is conducted to examine the ...

200

+14  upvote downvote
submitted by colonelred_(46),

Attributable risk = incidence in exposed – incidence in unexposed

`= 30/1,000` (smokers) `- 30/3,000` (nonsmokers)
`= 0.03 - 0.01`
`= 0.02` (so the attributable risk is about 2%)

Applying it to a population of 10,000:

`= 0.02 * 10,000`
`= 200`

charcot_bouchard  What if i tell you this is a ques of Attributable risk % in exposed? AR= 0.02 / IR in exposed (30/1000) = 0.6667 30 case in 1000. So 300 case in 10,000 0.6667 x 300 = 200 or in another word 66% cases of 100 lung cancer cases in smokers is actually due to smoking. so in 300 cases of smokers 200 is actually due to smoking +1
charcot_bouchard  This is a mind fuck. Lemme tell u guys if any consolation while doing the ques during test i did it with AR = 0.02; NNH = 1/0.02 = 50. 50 persons smoke to cause 1 cancer. 10K smoke to cause 200 cancer. +
ls3076  Sorry if this is a stupid question. Why is it incorrect to simply apply the same proportion (30 cancer per 1000 smokers) to 10,000 smokers? +1
krewfoo99  @is3076 Thats exactly what is did. I still dont understand how that is wrong. But i guess they want us to think about it in terms of AR +

+0  upvote downvote
submitted by emh(2),

Some of the cases of cancer in the smoking population will be due to a factor other than smoking. The question is asking about the cases that are due to smoking specifically.