Associations between Caregiver Acculturation and Patient-Physician Communication and Patients’ Treatment Preferences (N = 171).

†p≤0.10. *p≤0.05. **p≤0.01. ***p≤0.001. Logistic regression models examined associations between acculturation and patient-physician relationships, treatment preferences, terminal illness acceptance, and advance care planning. Every variable that was associated (p<0.20) with both the predictor and outcome (e.g., age, ethnicity, education, health insurance, native English language, nativity, cancer type, performance status, religion, institution, survival, and caregiver relationship) was investigated as a potential confounder and retained if significant at a level of p<0.05. Models adjusted for: aestimate for measures examining respect and trust in physicians (latter not shown) could not be calculated due to near uniform response (yes), bForeign born, cperformance status, dspousal caregiver, eperformance status. fno variables met significance (p≤0.05), and g age. Missing data: treated with respect, comfort asking questions (n = 8, 4.7%); end of life discussion (n = 9, 5.3%); prognostic information (n = 12, 7.0%); life-extending care over comfort (n = 29, 17.0%); avoid death in intensive care unit (n = 14, 8.2%); and feeding tube to extend life (n = 13, 7.6%).