Purpose: To test the hypothesis that scores of a validated measure of physician empathy are associated with clinical outcomes for patients with diabetes mellitus.
Method: This retrospective correlational study included 20,961 patients with type 1 or type 2 diabetes mellitus from a population of 284,298 adult patients in the Local Health Authority, Parma, Italy, enrolled with one of 242 primary care physicians for the entire year of 2009. Participating physicians' Jefferson Scale of Empathy scores were compared with occurrence of acute metabolic complications (hyperosmolar state, diabetic ketoacidosis, coma) in diabetes patients hospitalized in 2009.
Results: Patients of physicians with high empathy scores, compared with patients of physicians with moderate and low empathy scores, had a significantly lower rate of acute metabolic complications (4.0, 7.1, and 6.5 per 1,000 patients, respectively, P < .05). Logistic regression analysis showed physicians' empathy scores were associated with acute metabolic complications: odds ratio (OR) = 0.59 (95% confidence interval [CI], 0.37-0.95, contrasting physicians with high and low empathy scores). Patients' age (≥69 years) also contributed to the prediction of acute metabolic complications: OR = 1.7 (95% CI, 1.2-1.4). Physicians' gender and age, patients' gender, type of practice (solo, association), geographical location of practice (mountain, hills, plain), and length of time the patient had been enrolled with the physician were not associated with acute metabolic complications.
Conclusions: These results suggest that physician empathy is significantly associated with clinical outcome for patients with diabetes mellitus and should be considered an important component of clinical competence.