A simple quality improvement approach to improving hospice length of service by oncologists yielded a doubling, to reach the national median in 1 year.— Charles F. von Gunten, MD, PhD
A quality improvement project conducted within the OhioHealth system showed that oncologists can change their behavior and refer patients earlier to hospice care. After a relatively minor intervention, 18 medical oncologists in private practice doubled the mean length of stay in hospice care for the patients they referred, according to Charles F. von Gunten, MD, PhD, Vice President of Medical Affairs, Hospice and Palliative Medicine for OhioHealth, who described the project at the 2016 Palliative Care in Oncology Symposium.1
“We needed to move an entire health system toward more routine, more systematic, and less variable care. It starts with the fact that hospice has been proven to be the best care at the end of life. We also know that enrollment in hospice lowers cost,” Dr. von Gunten said. “Therefore, for a health system, the key issue is how to reach all the eligible patients…. Our novel thought was whether we could treat referral for hospice care as a quality measure, and we concluded that we can.”
Dr. von Gunten, who is a medical oncologist himself, gave his physicians the benefit of the doubt as to their reasons for delayed referral to hospice care, rejecting the notion that their actions are driven by personal economics. “I think they are genuinely uncertain about when to refer…. They want to do the right thing. That was the basis for this project,” he revealed.
“As a group, physicians are competitive. They like data and like knowing the numbers,” he added. “So seeing their own performance and those of their peers, and being compared to benchmarks and standards, this was the core of our intervention.”
Details of the Intervention
Within the OhioHealth system, the Oncology Clinical Guidance Council, a group of medical, surgical, gynecologic, and radiation oncologists, sets standards of care for providers in central Ohio. Dr. von Gunten polled the Council members as to their opinion of the optimal length of stay in hospice care. He also obtained such information from the National Hospice and Palliative Care Organization.
Among the Council members, 67% believed the ideal duration of hospice care for cancer patients was 90 days, and 27% said it should be 45 days. The National Hospice and Palliative Care Organization national median is 44 days.
He then analyzed the median length of stay of patients referred during 2014 by the 18 medical oncologists in private practice in the OhioHealth system and found it to be considerably longer than the opinions of the Council members or the information from the National Hospice and Palliative Care Organization. At baseline, the median length of stay for 176 cancer patients referred in 2014 was only 21 days, he reported. “As expected we saw variability, but overall a big gap was seen, which was fodder for a quality improvement project,” he commented.
The next step was for the Council Chair to write these 18 oncologists, expressing the Council’s opinion about optimal length of stay and including data on length of stay from the National Hospice and Palliative Care Organization. Most important, the oncologists received a graph showing the median length of stay of their own patients and those of their peers. One year later, for calendar year 2015, the researchers again measured the stays to assess whether this feedback resulted in more appropriate hospice care.
The intervention resulted in much timelier referrals: Median duration of hospice care increased to 39.6 days for the 133 patients referred in the first 10 months of 2015. “A simple quality improvement approach to improving hospice length of service by oncologists yielded a doubling, to reach the national median in 1 year,” noted Dr. von Gunten.
Some improvements were far greater than others. One physician’s median length of stay increased from 10 days to 45 days. A few did not improve, and one actually decreased.
The major concern among these oncologists was they were referring patients too early, revealed Dr. von Gunten; therefore, they often put off these conversations. “By showing them the data, I was confirming that they were not referring patients too soon. This project helped them overcome this concern,” he continued.
Dr. von Gunten did not sense that the physicians viewed this project as punitive. “Our doctors are accustomed to getting quality data about a number of things, and this project fit the format, style, and tone of those other communications. I was surprised, but there was none of the feeling of ‘you’re attacking me; I’m doing the best I can.’”
Dr. von Gunten and his Council did not stop there. They have sent another letter with these results and plan to repeat the measurements again next year. In addition, he may extend this quality improvement measure to other specialties. “I can hardly wait to check the data later this year,” he concluded. ■
Disclosure: Dr. von Gunten reported honoraria from Otsuka and Salix and has served as a consultant or advisor for AstraZeneca and a speaker for Salix.