One of the factors that has been a major deterrent for both patients and physicians to initiation of anticoagulation in older patients has been a perception that if patients are at risk of falls, they should not be anticoagulated.  In the extreme case that is true, but what that “extreme” is has been a matter of study and debate.

An interesting analysis published several years ago showed that a patient would have to fall hundreds of times per year before optimal therapy with warfarin would represent a risk for serious bleeding (Figure 14).[16] That depends in part, of course, on the severity of the fall. If patients are falling and hitting their heads and having a subdural hemorrhage, obviously that is a different risk from someone who is merely stumbling a bit. Nonetheless physicians probably overestimate this risk because the patients may seem a bit unstable and are falling, and in doing so underestimate the risk of stroke that that older, likely frail patients have from their AFib.