Although a “high-quality medical decision” has no universally accepted definition (1), most would agree that, at a minimum, it reflects a choice that is medically feasible and intended to promote the values of the patient. Yet in the context of critical illness, meeting even this minimum standard is challenging because patients are often incapacitated, their prognoses are uncertain, and the decisions required are emotionally charged and highly consequential. Prevailing wisdom and professional society guidelines (2) suggest that a process-oriented benchmark for which to strive is deliberative shared decision making that promotes patient goals. However, we lack evidence on the utility or sufficiency of this process and how best to promote it. Perhaps as a result we see many undesirable outcomes of critical care: potentially inappropriate care and unwanted debility for patients, stress and depression among surrogates, and high rates of burnout among clinicians (3).