P atients increasingly face treatment choices with comparable outcomes, and clinicians are confronting the challenge of how best to discuss alternative treatments while incorporating patients' values and preferences. One method involves shared decision making (SDM), "the process of interacting with patients who wish to be involved in arriving at an informed, values-based choice among two or more medically reasonable alternatives."1 Professional societies, including the most recent American Heart Association/American College of Cardiology guidelines, advocate the use of SDM within the clinical encounter.
2-4
Editorial see p 347Decision aids (DAs) are tools used during the clinical encounter to assist clinicians in engaging their patients in SDM. 5 The use of DAs to promote SDM addresses key quality goals for improving health care articulated by the Institute of Medicine (care that is evidence based and personalized).6 SDM with DAs is also prioritized in health policy initiatives, including specific provisions in the Affordable Care Act for SDM research, accreditation, and implementation.7 Although individual trials have suggested that DA use during the clinical encounter improves patient knowledge, reduces decisional conflict, and increases patient participation in choice, 8-11 the direction and magnitude of these effects across diverse patient Background-Decision aids (DAs) increase patient knowledge, reduce decisional conflict, and promote shared decision making (SDM). The extent to which they do so across diverse sociodemographic patient groups is unknown. Methods and Results-We conducted a patient-level meta-analysis of 7 randomized trials of DA versus usual care comprising 771 encounters between patients and clinicians discussing treatment options for chest pain, myocardial infarction, diabetes mellitus, and osteoporosis. Using a random effects model, we examined the impact of sociodemographic patient characteristics (age, sex, education, income, and insurance status) on the outcomes of knowledge transfer, decisional conflict, and patient involvement in SDM. Because of small numbers of people of color in the study population, we were not powered to investigate the role of race. Most patients were aged ≥65 years (61%), white (94%), and women (59%); two thirds had greater than a high school education. Compared with usual care, DA patients gained knowledge, were more likely to know their risk, and had less decisional conflict along with greater involvement in SDM. These gains were largely consistent across sociodemographic patient groups, with DAs demonstrating similar efficacy when used with vulnerable patients such as the elderly and those with less income and less formal education. Differences in efficacy were found only in knowledge of risk in 1 subgroup, with greater efficacy among those with higher education (35% versus 18%; P=0.02). Conclusions-In this patient-level meta-analysis of 7 randomized trials, DAs were efficacious across diverse sociodemographic groups as measured by knowledge transfer, decisio...