What's your point? Without the n evaluating the OR is a challenge no matter where you pick the thresholds.
True that a higher threshold should selectively change the odds ratio in favor of the drug response it also lowers the number of responders in the total. That number can also be helpful in understanding the effects of a drug.
If you set the threshold high so that just one of the placebo group is a responder and 10 of the treatment group responds that is an impressive response. Maybe. If the groups sizes are 1000 each then not so impressive. If the group sizes are 30 then it is a huge response and a breakthrough treatment.
The odds ratio data has to be placed in the context of the rest of the data set to be completely understood.
News 
Market Data 
Discover 
