A meta-analysis of studies that investigated measures designed to improve health professionals' interactions with patients confirms that such efforts can produce health effects just as beneficial as taking a daily aspirin to prevent heart attack. In contrast to previous such reviews, the current report from the Empathy and Relational Science Program at Massachusetts General Hospital (MGH) only included randomized, controlled trials with more reliable results than those included in earlier studies. While it has long been believed that a good patient-clinician relationship can improve health outcomes, objective evidence to support that belief has been hard to come by.
"Although the effect we found was small, this is the first analysis of the combined results of previous studies to show that relationship factors really do make a difference in patients' health outcomes," says Helen Riess, MD, director of the Empathy and Relational Science Program in the MGH Department of Psychiatry, senior author of the report in the open-access journal PLOS ONE.
Most studies focusing on the clinician-patient relationship have been observational studies - recording aspects of clinical encounters and any potential associations with health outcomes - which cannot prove whether observed differences actually cause any outcome changes. Some studies examined how well patients understood advice they were given or how satisfied they were with their care but did not look at whether or not there were any health improvements.
To get around these limitations, the research team restricted their analysis to studies meeting a high standard of evidence. The only studies that were included were randomized, controlled trials considered the gold standard of rigorous scientific research addressing the patient-caregiver relationship in the context of treatment for specific medical conditions. The studies had to have either objective health-related outcomes measurable factors such as blood pressure reduction or weight loss or subjective outcomes that had been validated in previous studies, such as patients' reported pain scores. Only studies published in peer-reviewed journals involving interventions directed at caregivers were included.
Application of these criteria to studies in the EMBASE and MEDLINE databases led to final selection of 13 trials conducted in the U.S., Europe or Australia and published from 1997 to 2012. The studies which involved care of patients with conditions like diabetes, hypertension and osteoarthritis examined the impact of interventions training clinicians in a variety of relationship-based techniques. Some of these interventions were focused on the relationship itself, like making more eye contact with patients and paying close attention to their emotions, while others used relationship-based strategies such as motivational interviewing and goal setting to address the health issue in question. All included studies compared the outcomes in an interventional group - in which physicians, nurses or other health professionals received training - to those of a control group delivering standard care.
Application of standard methods of meta-analysis found that relationship-focused training had a small but statistically significant effect on measured health outcomes - factors such as weight loss, blood pressure, blood sugar and lipid levels, and pain - in patients with conditions such as obesity, diabetes, asthma or osteoarthritis. The authors note that the size of the effect of the interventions was greater than previously reported effects of aspirin in reducing the incidence of heart attack over five years or the influence of statins on the five-year risk of a cardiovascular event.
"Our results show that the beneficial effects of a good patient-clinician relationship on health care outcomes are of similar magnitude to many well-established medical treatments," says lead author John M. Kelley, PhD. "But many of these medical treatments, while very important, need to balance their benefits against accompanying unwanted side effects. In contrast, there are no negative side effects to a good patient-clinician relationship." Previously with the MGH Empathy and Relational Science Program, Kelley is now a psychologist in the MGH Department of Psychiatry, deputy director of the Harvard Medical School (HMS) Program in Placebo Studies at Beth Israel Deaconess Medical Center, and an associate professor of Psychology at Endicott College.
An associate professor of Psychiatry at HMS, Riess adds, "We hope our results will inspire future research to identify the specific 'active ingredients' in the relationship more clearly. The more we can understand what exactly leads to these beneficial health changes, the more targeted interventions and training we can create."