Researchers Zero in on Who Chooses MA vs. Traditional Medicare

Researchers from healthcare analytics company Inovalon and Harvard Medical School say they have uncovered some key differences in the characteristics of individuals who choose Medicare Advantage (MA) over traditional Medicare at age 65.

Their study looks at issues such as economic status, race, income, and health. The researchers say their report provides policymakers and health plan administrators with insights to help design value-based care programs, improve health outcomes, and address health inequities. Their research found that MA enrollees are twice as  likely to be non-white, and much more likely to be Black, Hispanic, or Asian.

Also, MA enrollees are more than 50 percent more likely than those in traditional Medicare to have been enrolled in an HMO plan immediately before turning 65. The research also found substantial differences in socioeconomic characteristics of MA and traditional Medicare enrollees. 

Healthcare Innovation recently sat down with Christie Teigland, Ph.D., vice president of research science and advanced analytics at Inovalon, and Boris Vabson, Ph.D., visiting fellow in the Department of Health Care Policy at Harvard Medical School, to discuss their findings.

HCI: Your research analyzed underlying factors influencing enrollment in Medicare Advantage vs. traditional Medicare fee for service. How much research has been done on that previously, and why is this important to explore?

Teigland: I don't think there's been any research that has looked at it the way we have, which is using the data from when these beneficiaries were age 64 and covered by a commercial health plan before they join Medicare. The work that's been done has been after people have made that choice. We're using the baseline data from before they made that choice.

Vabson: I will just add that past research has looked at folks switching from fee for service to Medicare Advantage after 65, but that's really more of an outlier case. Most people choose either Medicare Advantage or fee for service at 65, and then they stay with their choice. That is when the decision gets made. People haven't really been able to study that decision well at all because studying a decision at 65 requires having pre-65 data leading up to the decision. And that data set hasn't been assembled up to now. And I will add one additional key contribution of the study has been looking at social determinants of health characteristics in granular detail. Those are also characteristics that are just not tracked as far as standard CMS data or standard data sets. Inovalon was able to bring in really fine-grained data at the Zip-nine level, so it has really specific data on income, net worth other SDOH characteristics to be able to study this in much more detail.

HCI: This next question is perhaps a bit of a tangent: Is there good research on the broader picture of whether Medicare Advantage plans are actually saving CMS money or providing better outcomes than traditional Medicare? Is that also difficult to study — or is there a solid research base there?

Vabson: I would say that's also really hard to study. And part of the reason why it's hard to study is because the kinds of people who enroll in Medicare Advantage vs. fee for service tend to be pretty different, as we found, so if there are differences in quality or cost outcomes, it's really hard to parse out how much of it is just the program doing it vs. just the people in the program being really different to begin with. Researchers have tried doing different kinds of adjustments, but with post-65 Medicare data alone, it's hard to do. What we've been able to do is bring in this pre-65 data when folks were all under the same type of coverage, treated the same way, coded the same way. And that just allows us to, again, compare fee for service to MA much more rigorously, much more apples to apples.

HCI: Is part of what you are trying to understand is whether the motivation is largely that the MA plans can offer things like lower premiums, dental coverage, glasses or gym memberships that traditional Medicare cannot? Or because these people have had a positive experience with managed care?

Teigland: We can look at the characteristics of the plan they were enrolled in previously, but also just looking at the characteristics of the people who are choosing Medicare Advantage vs. fee for service is very telling because the people going to Medicare Advantage are much more socially disadvantaged. And that's why Medicare Advantage is probably more attractive to them. With fee for service, you are kind of on your own. It's not a one-stop shop. They don't do the care coordination that disadvantaged people really need. Some of these people have low English proficiency, they have low incomes, they have low education, so they need that care coordination, that extra benefit of really helping them through the process.

Vabson: It also has lower premiums and lower cost sharing. Past studies put the difference at $1,000 to $2,000 per year, and out-of-pocket costs in some cases even more, a couple thousand per year in cost savings for a lot of retirees. So it makes sense that folks who are more disadvantaged are going to be attracted by lower premiums and lower cost sharing.

HCI: What are some of the implications of better-off enrollees staying with traditional Medicare and lower-income people choosing Medicare Advantage? Is there a risk of creating a two-tiered system, one for wealthier people and one for everyone else?

Teigland: There are some differences in healthcare utilization and outcomes of care. I mean, no one wants to spend time in the hospital. No one wants to spend time in a nursing home, right? And we see more of that happening in fee for service. And it is probably because of the lack of coordination. You go to see a specialist and they don't have the data that your primary care doctor has. They don't know all the medications you're taking. They don't know all your other chronic conditions, so it's very disjointed. That results in people going into the hospital, going to the emergency room.

HCI: OK, say that I'm a rich guy. Why would I want fee-for-service Medicare and not Medicare Advantage?

Babson: I would say the one advantage for you would be that you wouldn't face any restrictions in terms of provider networks. You could go see pretty much any doctor and especially if you're spending time in multiple locations throughout the year, it makes it easier because some of these Medicare Advantage plans, they tend to be specific to one state, one location, although some of them do have Snowbird options.

But I would say that the way that we've looked at it has been more from the vantage point of the MA plans, because the MA plans face special challenges from the fact that they attract these more socially and economically vulnerable folks — challenges that they probably don't get full credit for from CMS and policymakers and that's where a lot of these CMS health equity efforts can really come in.

HCI: As I understand it, the Medicare Advantage payment rates are set by CMS using traditional Medicare fee for service as a benchmark. Is that going to become more challenging as the number of people on Medicare Advantage eclipses the number of people in traditional Medicare — by a lot?

Babson: Yes, a lot of things get more challenging, and I think that makes the data and analytics like Inovalon’s all the more valuable because if most of the system is going to be MA, then you really need to have access to MA data to have good visibility on this, and be able to make informed decisions.

HCI: I did an interview with Dr. Sachin Jain, the CEO of SCAN Health Plan a few weeks ago. His idea is that we should get rid of annual enrollment in Medicare Advantage, and that people should sign up for more like three years. He said the for one thing, the amount of federal taxpayer dollars going to Medicare Advantage marketing is staggering. He added that a lot of the interventions aren't going to show up in one year, they're going to show up over three years or five years, and so the motivation to invest in interventions to help people would be a lot higher if you knew that that person was going to be with you three years from now. What do you think of that idea?

Babson: I can't really comment on the value of getting rid of annual open enrollment specifically, but more generally, I would say there is less churn in the Medicare Advantage market than in other kinds of markets. People tend to stay in longer. It does make sense that some of the interventions these plans deploy have a delayed effect, and that's why in some of these studies that we're doing, we're looking at things not just over one year, but over two or three years, to really be able to pick up some of these delayed benefits.  

Teigland:  We don't see a ton of that type of churn in MA. That issue is huge for Medicaid plans, because beneficiaries do come in and out, the plans make investments, and then they fall off the rolls and they don't get paid for them for a while until they come back sick again. We have done some work with health plans to try to understand that churn and how we can reduce that.