• Alexander Chaitoff, Michael B Rothberg, Amy K Windover, Leonard Calabrese, Anita D Misra-Hebert, and Kathryn A Martinez.
• Cleveland Clinic Lerner College of Medicine of Case Western Reserve University, Cleveland, OH, USA. chaitoa@ccf.org.
• J Gen Intern Med. 2019 Jan 1; 34 (1): 758175-81.

BackgroundOne widely cited study suggested a link between physician empathy and laboratory outcomes in patients with diabetes, but its findings have not been replicated. While empathy has a positive impact on patient experience, its impact on other outcomes remains unclear.ObjectiveTo assess associations between physician empathy and glycosylated hemoglobin (HgbA1c) as well as low-density lipoprotein (LDL) levels in patients with diabetes.DesignRetrospective cross-sectional study.ParticipantsPatients with diabetes who received care at a large integrated health system in the USA between January 1, 2011, and May 31, 2014, and their primary care physicians.Main MeasuresThe main independent measure was physician empathy, as measured by the Jefferson Scale of Empathy (JSE). The JSE is scored on a scale of 20-140, with higher scores indicating greater empathy. Dependent measures included patient HgbA1c and LDL. Mixed-effects linear regression models adjusting for patient sociodemographic characteristics, comorbidity index, and physician characteristics were used to assess the association between physician JSE scores and their patients' HgbA1c and LDL.Key ResultsThe sample included 4176 primary care patients who received care with one of 51 primary care physicians. Mean physician JSE score was 118.4 (standard deviation (SD) = 12). Median patient HgbA1c was 6.7% (interquartile range (IQR) = 6.2-7.5) and median LDL concentration was 83 (IQR = 66-104). In adjusted analyses, there was no association between JSE scores and HgbA1c (β = - 0.01, 95%CI = - 0.04, 0.02, p = 0.47) or LDL (β = 0.41, 95%CI = - 0.47, 1.29, p = 0.35).ConclusionPhysician empathy was not associated with HgbA1c or LDL. While interventions to increase physician empathy may result in more patient-centered care, they may not improve clinical outcomes in patients with diabetes.

26 keywords...

hide…