Objective To assess the relationship between empathic communication, shared decision-making, and patient sociodemographic factors of income, education, and ethnicity in patients with diabetes. Research Design and Methods This was a cross-sectional study from five primary care practices in the Greater Toronto Area, Ontario, Canada, participating in a randomized controlled trial of a diabetes goal setting and shared decision-making plan. Participants included 30 patients with diabetes and 23 clinicians (physicians, nurses, dietitians, and pharmacists), with a sample size of 48 clinical encounters. Clinical encounter audiotapes were coded using the Empathic Communication Coding System (ECCS) and Decision Support Analysis Tool (DSAT-10). Results The most frequent empathic responses among encounters were “acknowledgement with pursuit” (28.9%) and “confirmation” (30.0%). The most frequently assessed DSAT components were “stage” (86%) and knowledge of options (82.0%). ECCS varied by education (p=0.030) and ethnicity (p=0.03), but not income. Patients with only a college degree received more empathic communication than patients with bachelor’s degrees or more, and South Asian patients received less empathic communication than Asian patients. DSAT varied with ethnicity (p=0.07) but not education or income. White patients experienced more shared decision-making than those in the “other” category. Conclusions We identified a new relationship between ECCS, education and ethnicity, as well as DSAT and ethnicity. Limitations include sample size, heterogeneity of encounters, and predominant white ethnicity. These associations may be evidence of systemic biases in healthcare, with hidden roots in medical education.