Treatment preferences of patients and physicians: influences of summary data when framing effects are controlled.

The presentation of efficacy data influences preferences for treatment options. To determine how the amount of data provided to patients influenced patient decision making after framing and labeling effects were controlled, patients and physicians were presented results of two alternative treatments for an unidentified serious medical condition, derived from summary data of lung cancer treatment after surgery (better long-term survival) or radiation therapy (better short-term survival). These data are the same as used in previous studies of framing. When summary data at one month, one year, and five years were presented in terms of both survival and mortality, patients preferred the option that would be expected if only mortality data had been presented. When more detailed data were presented (data at six discrete time points), both patients and physicians preferred the option associated with a survival frame influence in previous studies. Thus, once framing influences are controlled, preference changes can be influenced by another attribute of summary data: the amount of data presented.