A highly publicized approach to lowering health costs failed to pass rigorous study this month, but hospitals, insurers and government health programs don’t intend to give up on the idea.
The “hotspotting” model was pioneered in Camden, New Jersey, in 2002 and inspired dozens of similar projects around the country, many financed by millions of dollars from the government and private foundations. The model focuses on people who face social barriers such as homelessness or drug addiction and use the hospital multiple times a year, typically for avoidable complications from chronic diseases. The participants work with doctors, social workers and nurses for individual help, seeking to prevent future hospital admission and extra expenses down the road.
A study in the New England Journal of Medicine published this month confirmed that repeat hospitalizations dropped for participants in the Camden program but the result was no better than the results from a group of patients who did not get the intensive care coordination.
A different study, released Wednesday, showed that a similar, multibillion-dollar experiment in California yielded even more discouraging preliminary results: Participants were hospitalized at more than double the rate of patients who were not enrolled in the program.
State and federal policymakers were attracted to the hotspotting model because they saw it as a possible solution for this intractable reality: Just 5% of patients account for half of all health spending.
The disappointing results from Camden caused some angst among health officials working on the model from Oregon to North Carolina — even before the California study was released.
“The study should give everyone who is involved in hotspotting a chance to pause and really reflect on the programs that they are doing,” said Dr. Amit Shah, chief medical officer at CareOregon, a large Medicaid managed-care insurer that for a decade has targeted some of these high-cost patients.
But “to stop now would be foolish,” said Susan Cooper, chief integration officer for Regional One Health, a Memphis, Tennessee, health system that launched a program in 2018 based on the Camden model. It is geared to patients who have the most emergency room and hospital visits because of unmet social needs such as hunger and housing.
Using nurses and social workers, the hospital pairs up these “super-utilizer” patients with community resources such as housing, transportation and meals as well as connecting them to primary care. From April 2018 through last month nearly 350 people have enrolled in the program and the hospital said it has saved more than $9 million than would have been expected to be spent on them by reducing ER visits and hospital admissions.
A Medicaid experiment similar to the Camden approach may also expand in the most populous state in the nation — at great cost and despite disappointing initial results.
As part of his 2020-21 state budget proposal, California Gov. Gavin Newsom wants to add $582.5 million to Medi-Cal, the state’s Medicaid program, to increase intensive, comprehensive care management to high-needs, high-risk patients, including homeless people and those getting out of jail or prison. If approved by the state legislature, the infusion would also fund services designed to decrease reliance on expensive hospital visits and emergency service transportation.
The move would expand the state’s $3 billion, five-year Medicaid experiment called “Whole Person Care,” which began in 2016 and provides participants with social and medical services, such as substance abuse treatment and recuperative care after hospital stays. In most cases, California is creating services for participants in addition to connecting people to existing programs.
Twenty-five counties, cities, health plans and hospitals currently participate, and nearly 150,000 patients have enrolled.
But a new report by researchers at the UCLA Center for Health Policy Research suggests that California’s experiment has resulted in more participant hospitalizations, despite offering comprehensive services.
The study found that, two years into the experiment, there was no significant change in emergency department visits for participants compared with a similar group of patients who were not part of the program. But hospitalizations went up for both groups, with participants getting admitted at more than twice the rate of the control group: 17.47 inpatient admissions per 1,000 enrollees, compared with 7.41 admissions per 1,000 for the control group.
Dana Durham, chief of the policy and medical monitoring branch of the state Department of Health Care Services, said that the data is preliminary and that she expects hospitalizations to drop the longer patients are in the program.
“There hasn’t been enough enrollment in the program to make this comparison one that we have confidence in yet,” she said.
Clemens Hong, director of the Whole Person Care pilot in Los Angeles, which has served more than 46,000 people and has a five-year budget of more than $1 billion, could not be reached for comment after the results were released Wednesday.
Prior to the release, Hong said it would take time to see positive results because the participant population is “much sicker than most people understand or realize.” He also said evaluations of these programs shouldn’t focus exclusively on hospitalizations, but should also consider outcomes such as how well they increase employment or reduce time spent living on the street.
“One of the key lessons is that it can’t just be medically focused,” Hong said about the results of the Camden study.
Camden Coalition officials agree. They say that, although their work was originally designed to help people find safety-net services, they learned that sometimes those resources were not available. The coalition now has expanded its outreach and set up more partnerships, including efforts to start a housing program.
Looking For ‘Hot Spots’
The Camden Coalition, which provided services for 194 people last year, was founded by Dr. Jeffrey Brenner, a family doctor, who was looking to improve health care in one of the poorest cities in the country.
Using medical billing data from hospitals in Camden, Brenner located “hot spots” that had larger numbers of high-cost patients — even down to the city block. These patients, he found, repeatedly showed up in emergency rooms and doctors’ offices. They often had physical health issues such as diabetes, asthma and other chronic illnesses that were compounded by psychological and social issues such as anxiety and homelessness.
Earlier studies of the initiative primarily measured how patients in the program used services before and then after being enrolled. By coordinating the medical care and social services these patients needed, Brenner and his group were able to cut down the number of rehospitalizations.
The New England Journal of Medicine study, conducted by researchers at the Massachusetts Institute of Technology and published Jan. 9, confirmed that the program over six months reduced readmissions by nearly 40%. But a group of similar patients who did not get the interventions had a similar drop.
This type of randomized-control trial, the gold standard in clinical medicine, is rarely used in health policy studies, partly due to the cost.
Shah of CareOregon said the study results prove there is no single solution that will work in every community for every type of patient.
“There is no one magic bullet,” Shah said. “The study confirmed there is a population of super utilizers that will not go away” just through these community efforts alone. He stressed local, state and federal policy changes are needed to fix social and legal challenges that influence patients’ health.
Camden Coalition officials said the results were disappointing but noted they have made major changes to the hotspotting efforts since the study trial ended in 2017. This includes efforts to help participants find stable housing and legal services to break down barriers to Medicaid and other benefits.
“We have already made some pivots,” said the coalition’s CEO, Kathleen Noonan.
She said people should not take away from the study “that focusing on social determinants of health and all those efforts are moot.” The study showed some positive benefits, including how participants were more likely to obtain food stamp benefits compared with those in the control group.
“To paraphrase Mark Twain, reports of the death of the hotspotting model are greatly exaggerated,” said Rutgers University professor Joel Cantor, director of the Center for State Health Policy, before the California results dropped. Cantor has worked closely with the coalition. “This is a single study, focusing mainly on one of many possible outcomes, in a single location.”
Brenner, who won a MacArthur Foundation fellowship — known as a “genius” grant — in 2013, said hotspotting efforts need to keep evolving.
“There are glimmers of things that work, and we need to keep at this. It is a solvable problem,” said Brenner.
When he first started the coalition, Brenner said, he hoped just helping people navigate the health system would improve their health and lower costs. But that approach faced shortcomings when services such as addiction treatment and housing assistance were lacking.
“It’s not as simple as just navigating and coordinating,” he said.