Individual physicians are widely believed to influence the kind of care their patients receive at the end of life, but to date, there is little scientific evidence to support this belief. New research from Brigham and Women’s Hospital (BWH) indicates that the individual physician a patient sees is the strongest known predictor of whether or not he or she will enroll in hospice care, outweighing other known drivers such as geographic location, patient age, race and comorbidities.
These findings are published in the June 2015 issue of Health Affairs.
“We found that the physician a patient sees is the single most important predictor we know of whether or not that patient enrolls in hospice care,” said Ziad Obermeyer, MD, a physician researcher in the Department of Emergency Medicine at BWH, Assistant Professor of Emergency Medicine at Harvard Medical School, and lead author of the paper. “This new information provides a clear policy target for improving and advancing the quality of care for patients at the end of their lives.”
Researchers used a nationally representative Medicare sample to identify people with a poor-prognosis cancer diagnosis who would have been eligible for hospice care before they died, using a palliative care screening algorithm from a large cancer center, from 2006-2011. The sample included 198,948 patients who, on average, were 78 years-old, 88 percent white and 52 percent male. The 66 percent who enrolled in hospice were more likely to be female, white and live in ZIP codes with higher median incomes, when compared to patients not enrolled in hospice.
Researchers calculated the proportion of a physician’s patients that were enrolled in hospice care, as a measure of their propensity to refer their patients to hospice. After controlling for patient, hospital, and geographic factors that predict hospice enrollment, they found that patients would be 27 percent more likely to enroll in hospice if they saw a physician in the top 10 percent of hospice use, compared to a physician in the bottom 10 percent. Additionally, researchers report that large numbers of cancer patients in this cohort were seen in a relatively small group of physicians.
“Our data show that about 10 percent of physicians cared for about half of all patients. This suggests that we can target a small group of physicians with interventions geared towards physician specialty and how often their patients enroll in hospice to improve end of life care,” Obermeyer said.
Researchers found that regional factors, greater comorbidity and physician specialty were all significantly associated with the likelihood of hospice enrollment, which generally increased over time. They also noted a new, albeit small effect on the likelihood of hospice enrollment: physicians associated with for profit hospitals were less likely to have patients enroll when compared to physicians affiliated with non-profit hospitals. Obermeyer suggests that further research exploring this link is warranted.
“As an emergency physician, I am often the first person to ask patients about what kind of care they want at the end of their life. In these situations, patients and their families often have only hours to make difficult and complex decisions,” said Obermeyer. “As physicians, we need to have these conversations earlier. We need to know what our patients really want at the end of their lives. We need to remove the barriers to having these discussions and give our patients the care they actually want.”