Decisional control preferences among patients with advanced cancer: An international multicenter cross-sectional survey.

BACKGROUND: Understanding patients' decision control preferences is important in providing quality cancer care. Patients' decisional control preference can be either active (patients prefer to make decisions themselves), shared (collaborative between patient, their physician, and/or family), or passive (patients prefer that the decisions are made by either the physician and/or their family). AIM: To determine the frequency and predictors of passive decision control preferences among advanced cancer patients. We also determined the concordance between actual decision-making and decision control preferences and its association with patient satisfaction. DESIGN: In this cross-sectional survey of advanced cancer patients referred to palliative care across 11 countries, we evaluated sociodemographic variables, Control Preference Scale, and satisfaction with the decisions and care. RESULTS: A total of 1490 participants were evaluable. Shared, active, and passive decision control preferences were 33%, 44%, and 23%, respectively. Passive decision control preferences (odds ratio, p value) was more frequent in India (4.34, <0.001), Jordan (3.41, <0.001), and France (3.27, <0.001). Concordance between the actual decision-making and decision control preferences was highest in the United States ( k = 0.74) and lowest in Brazil (0.34). Passive decision control preference was significantly associated with (odds ratio per point, p value) better performance status (0.99/point, 0.017), higher education (0.64, 0.001), and country of origin (Brazil (0.26, <0.0001), Singapore (0.25, 0.0003), South Africa (0.32, 0.0002), and Jordan (2.33, 0.0037)). CONCLUSION: Passive decision control preferences were less common (23%) than shared and active decision control preference even among developing countries. Significant predictors of passive decision control preferences were performance status, education, and country of origin.