Creating a Conversation: Insights from the Development of a Decision Aid

E vidence-based medicine requires that clinical decisions be consistent not only with the best available research evidence, but with the values and preferences of the informed patient [1]. To achieve this goal, clinicians and patients can use tools, known as decision aids, that prepare patients for decision making [2] or help clinicians assist patients in participating in making decisions [3]. Our research group has recently completed the development and testing of such a decision aid, dubbed Statin Choice, for patients with diabetes who were considering using statins (medications that lower cholesterol) to reduce their cardiovascular risk. Statin Choice sought both to help clinicians share the evidence about potential benefi ts and downsides of statins and to create a two-way conversation that would enable patients to participate in making decisions to the extent they preferred. Patients' participation could make the resulting decisions more likely to be consistent with their values and preferences. In addition, patient participation in decision making (i.e., cognitive investment in the decision and thus " ownership " of that decision) could enhance adherence to therapeutic interventions, among other potential benefi ts. Here, we present the insights that resulted from the process we followed to develop this decision aid. In sharing these insights, our intent is not prescriptive (i.e., " this is how you should develop decision aids "), but rather to inspire others to seek innovative, yet goal-directed approaches to the development of decision aids that are not only evidence-based in content but also user-centered in their design and use. Type 2 diabetes mellitus is a common chronic condition associated with very high costs, both in resources and in human suffering and lives [4]. Cardiovascular events cause most of the diabetes-related deaths and disability, making the reduction of cardiovascular risk a focus of diabetes care [5]. Clinical trials have established that lowering cholesterol with statins can reduce the risk of cardiovascular events in patients with diabetes [6,7]. Current guidelines recommend the use of statins for most patients with type 2 diabetes [8]. Despite the otherwise solid evidence base and the guideline recommendations, few patients take up statins and adhere to these medications over time [9,10]. A model for nonadherence to medications offers some explanations for this phenomenon [11,12], including use of multiple medications, discontinuity of care, and costs of medications. Limited patient knowledge [13] and participation in treatment decision making may also contribute to nonadherence. In particular, clinicians …