QUESTIONING THE NEUTRALITY OF DECISION AIDS

Purpose: To develop criteria for defining and measuring the neutrality of decision aids in light of data from behavioral economics showing the influence of framing effects on patient decision making.

Method: Systematic review of the literature. 

Result: The International Patient Decision Aids Standards Collaboration (IPDAS) has developed criteria for the development and evaluation of decision aids.  One of those criteria is that decision aids must be neutral.  What exactly neutrality means, whether neutrality is possible in light of data from behavioral economics showing the influence of framing effects on decision making, and whether neutrality is always desirable are important and unexamined questions that surround the criterion of neutrality.  Clarification on these points is especially important in light of the fact that health reform is calling for the certification of all patient decision aids.  

Conclusion: The two main dimensions of neutrality are balance and bias.  Balance is achieved when both sides are presented, and one side is not helped more than the other—the sides have equal modes and times of information presentation.  Unbiased presentation requires an absence of framing effects that induce biases such as loss aversion bias, availability bias, recency or primacy bias, anchoring bias, and default bias.  Whether a decision aid manages to achieve balance in presentation and to avoid inducing the aforementioned biases can be tested and measured.  We argue, however, that neutrality is not always desirable in decision aids, nor is it always a realistic goal. Using the example of a decision aid for medication versus surgical treatment of cardiac disease, we argue that presenting medication as the default is the ethically appropriate course of action despite the fact that it would induce the default bias, resulting in a bias towards medication, and as such not meet the IPDAS criterion of neutrality.