Patient decision aids are known to positively impact outcomes critical to shared decision making (SDM), such as gist knowledge and decision preparedness. However, research on the potential improvement of these and other important outcomes through cultural targeting and tailoring of decision aids is very limited. This is the case despite extensive evidence supporting use of cultural targeting and tailoring to improve the effectiveness of health communications. Building on prominent psychological theory, we propose a two-stage framework incorporating cultural concepts into the design process for screening and treatment decision aids. The first phase recommends use of cultural constructs, such as collectivism and individualism, to differentially target patients whose cultures are known to vary on these dimensions. Decision aid targeting is operationalized through use of symbols and values that appeal to members of the given culture. Content dimensions within decision aids that appear particularly appropriate for targeting include surface level visual characteristics, language, beliefs, attitudes and values. The second phase of the framework is based on evidence that individuals vary in terms of how strongly cultural norms influence their approach to problem solving and decision making. In particular, the framework hypothesizes that differences in terms of access to cultural mindsets (e.g., access to interdependent versus independent self) can be measured up front and used to tailor decision aids. Thus, the second phase in the framework emphasizes the importance of not only targeting decision aid content, but also tailoring the information to the individual based on measurement of how strongly he/she is connected to dominant cultural mindsets. Overall, the framework provides a theory-based guide for researchers and practitioners who are interested in using cultural targeting and tailoring to develop and test decision aids that move beyond a “one-size fits all” approach and thereby, improve SDM in our multicultural world.