Assuming a 50% prestamp "yes" response, a 90% poststamp "yes" response, an 80% power, and an alpha level of.
The prestamp group included 398 patients; the poststamp group, 239.
Prestamp smokers had a greater number of previous quit attempts than did the poststamp group (P<.
The only variables that contributed significantly to predicting whether the physician would talk with a smoker about smoking was placement of the patient in the poststamp period rather than the prestamp period (odds ratio [OR]=5.
During the prestamp period, 50% of smokers reported that their physician had advised them to quit smoking.