|
 Researchers looked at so-called "superutilizer" patients who require
a disproportionate amount of medical resources - in part because
they're poor or don't have someone to look out for them when they
leave the hospital. Their study, published in The New England
Journal of Medicine, found these patients don't avoid
rehospitalization even if nurses, social workers and community
health workers are looking out for them after they've been
discharged.
The readmission rate was 62.3% within six months of discharge for
patients who got that extra support versus 61.7% for those who did
not.
It's the first solid test of the hotspotting concept, which has been
growing in popularity around the U.S. based on the belief that
providing extra services to the superusers will keep them out of the
hospital.
The New Jersey-based Camden Coalition of Healthcare Providers, which
created and helped spread the hotspotting concept, helped coordinate
the study.
It was led by Amy Finkelstein, a professor of economics at the
Massachusetts Institute of Technology, in Cambridge, who said it's
an example of why innovative programs that seem to save money and
benefit patients need to be tested to see if they are really
working.
The test was set up in consultation with Dr. Jeffrey Brenner when he
was director of the Coalition. He and his organization "completely
embraced this," Finkelstein told Reuters Health in a telephone
interview. "The program received a lot of accolades," she noted.
"Dr. Brenner said, 'We think what we're doing is having an impact,
but we won't know until we do a randomized evaluation. If it doesn't
work, we need to figure out why and try something else.' "
Brenner, in a phone interview with Reuters, expressed disappointment
with the findings from the trial, but said the program is evolving.
"When we started this 10 years ago, the dominant belief was this was
a coordination problem. As we got deeper into the work it became
clear that even if you did a great job of navigating and
coordinating services, there was a fundamental failure of these
services to deliver what these patients need," he said. The newest
incarnations of the program are trying to address that.
[to top of second column] |
Hotspotting was created to deal with the fact that 5% of patients -
the superutilizers - eat up 50% of the money devoted to healthcare
in the United States each year. The top 1% of these superusers
account for nearly one quarter of the spending.
All 800 patients in the study had been hospitalized twice in the
previous six months and had at least two long-term health
conditions. They also had at least two of these elements
complicating their lives: homelessness, an active drug habit, a need
to take at least five medicines, an inability to access social
services, no social support or a mental health problem.
Finkelstein said more such tests need to be done, noting that less
than 20% of the studies that examine healthcare delivery
interventions are randomized trials versus 80% of medical treatment
studies.
Doing a randomized test is "something we do as a matter of course in
medicine," she said. "We need to get to that place in healthcare
delivery."
Dr. Brenner, who is now senior vice president of integrated health
and social services at UnitedHealthcare, said he is using the
insurance company's large database to try different hotspotting
approaches in real time, setting up statistical control groups to
try to better estimate what is really working.
For example, he said, a new program to immediately place homeless
patients in a home at discharge has cut the cost of caring for these
superusers by perhaps 35% to 40% compared to the costs accrued by
patients who are placed in a homeless shelter instead.
That program is in 12 states and is going to be scaled up to 30
states, he said.
SOURCE: https://bit.ly/2QQ4Lg5 The New England Journal of Medicine,
online January 8, 2020.
[© 2020 Thomson Reuters. All rights
reserved.] Copyright 2020 Reuters. All rights reserved. This material may not be published,
broadcast, rewritten or redistributed.
Thompson Reuters is solely responsible for this content.
 |
