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

NBME 24 Answers

nbme24/Block 4/Question#30

An investigator is studying an outbreak of ...

Only cookies are independently associated with E. coli cases

Login to comment/vote.

 +19  upvote downvote
submitted by niboonsh(173),

this question makes me want to eat an e coli cookie and hope i bleed out

Did no one notice that the Odds ratio on the top left is wrong? Am I missing something? If you calculate it, it's 6 just like the top right one....

mjmejora  thats actually really funny +  
yex  Because I said so, applies here... :-/ +  
doodimoodi  Cant believe we pay $60 for this crap +19  
aisel1787  best comment doodimoodi) +  

 +3  upvote downvote
submitted by youssefa(58),

Initially milk drinking was associated with E.coli outbreak with OR=3.9 and P<0.001 (Significant)... After stratification into ate cookies and did not eat cookies OR became 1 instead of 3.9 meaning the association disappeared. Therefore, eating cookies was a confounder and there is no real association between drinking milk and E.coli....instead, milk's (the confounder) contribution was responsible for the OR of 3.9 in the first place. This was furthered demonstrated with OR of 6 in the cookies alone group.

 +3  upvote downvote
submitted by pg32(38),

The fact that the odds ratio in the top left is incorrect makes this question very difficult. It makes it appear as if the cookies are causative but the milk had some protective factor. So obnoxious.

drpatinoire  God I thought totally the same way as you did. I stared at this question for at least 5min and asked myself what's wrong with my statistics. +1  

 +2  upvote downvote
submitted by sne(22),

OR >1 indicates increased occurrence of event. The only OR greater than 1 was in the table that indicated that the subject ate cookies but didn't drink milk. Thus, that is the only one with a significant occurrence

 +1  upvote downvote
submitted by sattanki(46),

This one there were four odds ratios, one provided under each table. The only one that had an odds ratio greater than 1.0 was the table in the top right (Odds Ratio = 6, I believe), which when you looked at the labels, led to the right answer.

 +1  upvote downvote
submitted by famylife(49),

"An odds ratio of 1 indicates that the condition or event under study is equally likely to occur in both groups. An odds ratio greater than 1 indicates that the condition or event is more likely to occur in the first group." (https://en.wikipedia.org/wiki/Odds_ratio)

 -2  upvote downvote
submitted by lilmonkey(14),

The keyword is "INDEPENDENTLY"(associated). Which in human language means "NOT ASSOCIATED".