Medicaid may be publicly funded, but in much of the United States, it is no longer run entirely by the government.
Instead, states often hire private insurance companies to manage care for Medicaid enrollees. Under this model, known as managed care, the state pays a private plan a fixed amount for each person it covers. The plan then builds provider networks, pays doctors and pharmacies, and uses tools such as prior authorization or claim denials to manage spending — all using public money.
That arrangement has reshaped Medicaid, which now contracts out the care of more than three- quarters of its enrollees. But it remains uncertain whether managed care preserves health care access, improves care, or even saves money.
A new study from researchers including Jacob Wallace, associate professor of public health (health policy), Chima D. Ndumele, professor of public health (health policy), and Anthony Lollo, data scientist at the Yale School of Public Health, offers some of the clearest answers yet. Published in The American Economic Review in June, the paper draws on a natural experiment in Louisiana in which nearly 100,000 Medicaid enrollees were randomly assigned either to private managed care plans or to a publicly operated fee-for-service system, where the state paid providers directly for each service. They found managed care reduced overall spending by 5.6% — or $82 per enrollee per year.
“Managed care seems to work well for some things and poorly for others, with overall savings smaller than many have assumed,” Wallace said. “The key question is who captures those savings and what are the tradeoffs from enrollee health and well-being.”
Nearly all of those savings came from a single place: prescription drugs. After Louisiana shifted drug coverage into managed care, pharmacy spending fell by roughly 25% among enrollees in private plans. The plans leaned on pharmacy claim denials to steer patients toward cheaper alternatives like generics.
But medical care was different. There, the savings were smaller, and they came with tradeoffs like a modest decline in primary and preventive care and a 14% rise in avoidable emergency department visits. Additionally, enrollees were more than three times as likely to switch out of their assigned plan — a sign, the researchers say, of lower satisfaction.
“Whether randomly assigned enrollees remain in their plans is a novel way of measuring satisfaction. When we look at it, there is clear evidence that enrollees are voting with their feet. The savings from managed care appear to come, at least partly, at the cost of enrollee satisfaction,” said Anthony Lollo, a director of data science at the Yale School of Public Health and one of the study’s authors.
People who choose one kind of health plan over another tend to differ in ways researchers cannot fully see. Sicker patients, for instance, may gravitate toward plans with fewer restrictions. That makes it hard to know whether a difference in spending reflects the plan or the people who picked it. But Louisiana’s random assignment removed much of that guesswork. Enrollees who did not select a plan during the state’s transition to managed care were assigned to one automatically, leaving the researchers with two genuinely comparable groups.
The result was a sharper picture than earlier studies could provide. Once selection bias was stripped away from the findings, managed care still saved money, but far less than some previous studies had suggested.
“Our results suggest that private managed care plans may have sharp tools for managing pharmacy benefits — where they are able to reduce spending without harming access — but their effectiveness at managing medical benefits is less clear in our setting,” the researchers wrote in their paper. “For medical care, we observe reductions in quality and more modest cost savings.”
The study suggests that managed care can help to significantly reduce costs in one area of health care — prescription drugs — but can worsen patient experiences in others. For policymakers, that points toward a more targeted use of privatization rather than an all-or-nothing choice, the researchers explain.
There are limitations, however. The authors caution that the study only examined one state Medicaid program during a specific period, from 2012 to 2014, and included a relatively young and healthy population. And data about claims cannot capture every dimension of quality, health, or patient experience.
Still, the trade-offs they found deserve a place in any decision to move more of Medicaid into managed care, they argue.
“States are going to keep contracting with private plans, since that is the reality of Medicaid today,” Wallace said. “But the real question is how to design procurement and set up competition to benefit from savings while ensuring that managed care is improving access and quality, rather than eroding it.”
The study’s other authors include Danil Agafiev Macambira, a former pre-doctoral fellow at YSPH, and Michael Geruso, an associate professor of economics at the University of Texas at Austin.