Family Understanding of Seriously-ill Patient Preferences for Family Involvement in Healthcare Decision Making

BACKGROUND: Surrogate accuracy in predicting patient treatment preferences (i.e., what patients want) has been studied extensively, but it is not known whether surrogates can predict how patients want loved ones to make end-of-life decisions on their behalf. OBJECTIVE: To evaluate the ability of family members to correctly identify the preferences of seriously-ill patients regarding family involvement in decision making. DESIGN: Cross-sectional survey. PARTICIPANTS: Twenty-five pancreatic cancer and 27 amyotrophic lateral sclerosis (ALS) patients and their family members (52 dyads total). MAIN MEASURES: Patients and family members completed the Decision Control Preferences (DCP) scale regarding patient preferences for family involvement in health care decisions using conscious and unconscious scenarios. KEY RESULTS: Patient and family member agreement was 56% (29/52 dyads) for the conscious scenario (kappa 0.29) and 46% (24/52 dyads) for the unconscious scenario (kappa 0.15). Twenty-four family members identified the patient's preference as independent in the unconscious scenario, but six of these patients actually preferred shared decision making and six preferred reliant decision making. In the conscious scenario, preference for independent decision making was associated with higher odds of patient-family agreement (AOR 5.28, 1.07-26.06). In the unconscious scenario, cancer patients had a higher odds of agreement than ALS patients (AOR 3.86; 95% CI 1.02-14.54). CONCLUSION: Family members were often unable to correctly identify patient preferences for family involvement in end-of-life decision making, especially when patients desired that decisions be made using the best-interest standard. Clinicians and family members should consider explicitly eliciting patient preferences for family involvement in decision making. Additional research is still needed to identify interventions to improve family member understanding of patient preferences regarding the decision-making process itself.