Physician Empathy Linked to Better Patient Outcomes

Laurie Bouck

September 13, 2012

September 13, 2012 — High levels of empathy in primary care physicians correlate with better clinical outcomes for their patients with diabetes, according to a new study of more than 20,000 patients in Italy.

The retrospective correlational study of 20,961 patients with diabetes and 242 generalist physicians who treated them in Parma, Italy, found that patients whose physicians scored highest on a validated empathy test had the lowest rates of acute metabolic complications requiring hospitalization. The research was published in the September issue of Academic Medicine.

The current study builds on a smaller study from 2011, conducted by several of the same researchers, that found that US physicians' level of empathy correlated with the ability of their patients with diabetes to manage their disease, as measured by the patients' hemoglobin A1c and low-density lipoprotein cholesterol test results. The 2011 study included 29 family physicians and 891 diabetic patients.

These studies are "the first 2 [trials] that we're aware of that have actually related empathy to tangible patient outcomes," explained Daniel Z. Louis, MS, managing director of the Center for Research in Medical Education at Jefferson Medical College in Philadelphia, Pennsylvania. Louis worked on both of the studies.

"There's real evidence that these patients with diabetes mellitus had fewer complications" when their physicians were more empathic, Louis told Medscape Medical News. "That's incredible."

In the current study, the researchers used health data from the Parma Local Health Authority of Italy's Emilia-Romagna region. They studied the effect of physician empathy on adult patients with type 1 or type 2 diabetes mellitus who were continuously enrolled for 5 years between 2002 and 2009 with a physician who participated in the National Health Service.

To measure physician empathy, the researchers mailed the physicians a validated instrument called the Jefferson Scale of Empathy, or JSE. Developed in 2001 at Jefferson Medical College's Center for Research in Medical Education and Health Care, the JSE includes 20 items that measure how well a healthcare professional understands the patient's perspective. Available in 42 languages, the JSE has been used in the United States and internationally.

The researchers measured patient outcomes by identifying ICD-9-CM codes for patients with diabetes who were hospitalized in 2009 for acute metabolic complications such as hyperosmolar state and coma. They focused on metabolic complications because "[they] can develop rather quickly," the researchers write, "and their prevention is more likely to be influenced by the primary care physician, whereas the other complications are often managed by specialists whose specific contributions are difficult to trace."

Eighty-one of the 242 physicians who completed the JSE were high scorers (with scores ranging from 112 to 137), indicating high levels of empathy. Eighty-four physicians were moderate scorers (with scores ranging from 97 to 111), and 77 physicians were low scorers (with scores ranging from 49 to 96).

The high scorers had 7224 patients with diabetes, the moderate scorers had 7303 patients with diabetes, and the low scorers had 6434 patients with diabetes. Of the 123 patients in the study hospitalized with acute metabolic complications in 2009, the complication rate per 1000 patients was significantly lower for the patients of high scorers (4.0) than for the patients of moderate and low scorers (7.1 and 6.5, respectively).

Results Move the Field Forward

"This study was well done and moves the field forward," said Kathryn I. Pollak, PhD, associate professor for the Cancer Prevention, Detection and Control Program at the Duke Cancer Institute in Durham, North Caroline. She cited the large sample size and the high 80% response rate among the physicians asked to take the JSE.

Dr. Pollak told Medscape Medical News that the authors did not control for some clinical factors that can affect disease complications. "It might be that the authors were unable to include these other measures, as they were not collected," she wrote in an email, "but it is possible (but unlikely) that patients who were sicker were seen by physicians who were less likely to self-report as empathic."

The study is unique among other empathy studies because it measures how empathy affects patient outcomes, said Richard C. Wender, MD, chair of the Department of Family and Community Medicine at Jefferson University Hospitals. Dr. Wender worked on the 2011 study but not on the 2012 study.

Dr. Wender told Medscape Medical News that physician empathy might not be as influential as the study implies. Although he thinks it's reasonable to link physician empathy with patient outcomes, it's still unclear that "what a particular patient's physician does or doesn't do is a strong determinant of whether or not an acute metabolic complication occurs," he said.

Dr. Wender recommended studying empathy in other patient populations that have a long-term relationship with a clinician, such as patients receiving dialysis or patients with cancer.

In addition, Dr. Pollak suggested studying the effect of physician empathy on patient survival among cancer patients.

Dr. Wender noted that he would also like to see studies on how physician empathy affects patients' willingness to pursue preventive care, such as cancer screenings.

Information about the Jefferson Scale of Empathy is available on their Web site.

Louis and Dr. Pollak have disclosed no relevant financial relationships. Dr. Wender worked on the 2011 study with several of the authors, but was not involved in the 2012 study.

Acad Med. 2012;87:1243-1249. Full text